add idl4k kernel firmware version 1.13.0.105

kernel/drivers/xen/Kconfig

@@ -0,0 +1,63 @@
+config XEN_BALLOON
+	bool "Xen memory balloon driver"
+	depends on XEN
+	default y
+	help
+	  The balloon driver allows the Xen domain to request more memory from
+	  the system to expand the domain's memory allocation, or alternatively
+	  return unneeded memory to the system.
+
+config XEN_SCRUB_PAGES
+	bool "Scrub pages before returning them to system"
+	depends on XEN_BALLOON
+	default y
+	help
+	  Scrub pages before returning them to the system for reuse by
+	  other domains.  This makes sure that any confidential data
+	  is not accidentally visible to other domains.  Is it more
+	  secure, but slightly less efficient.
+	  If in doubt, say yes.
+
+config XEN_DEV_EVTCHN
+	tristate "Xen /dev/xen/evtchn device"
+	depends on XEN
+	default y
+	help
+	  The evtchn driver allows a userspace process to triger event
+	  channels and to receive notification of an event channel
+	  firing.
+	  If in doubt, say yes.
+
+config XENFS
+	tristate "Xen filesystem"
+	depends on XEN
+	default y
+	help
+	  The xen filesystem provides a way for domains to share
+	  information with each other and with the hypervisor.
+	  For example, by reading and writing the "xenbus" file, guests
+	  may pass arbitrary information to the initial domain.
+	  If in doubt, say yes.
+
+config XEN_COMPAT_XENFS
+       bool "Create compatibility mount point /proc/xen"
+       depends on XENFS
+       default y
+       help
+         The old xenstore userspace tools expect to find "xenbus"
+         under /proc/xen, but "xenbus" is now found at the root of the
+         xenfs filesystem.  Selecting this causes the kernel to create
+         the compatibility mount point /proc/xen if it is running on
+         a xen platform.
+         If in doubt, say yes.
+
+config XEN_SYS_HYPERVISOR
+       bool "Create xen entries under /sys/hypervisor"
+       depends on XEN && SYSFS
+       select SYS_HYPERVISOR
+       default y
+       help
+         Create entries under /sys/hypervisor describing the Xen
+	 hypervisor environment.  When running native or in another
+	 virtual environment, /sys/hypervisor will still be present,
+	 but will have no xen contents.

kernel/drivers/xen/Makefile

@@ -0,0 +1,12 @@
+obj-y	+= grant-table.o features.o events.o manage.o
+obj-y	+= xenbus/
+
+nostackp := $(call cc-option, -fno-stack-protector)
+CFLAGS_features.o			:= $(nostackp)
+
+obj-$(CONFIG_HOTPLUG_CPU)	+= cpu_hotplug.o
+obj-$(CONFIG_XEN_XENCOMM)	+= xencomm.o
+obj-$(CONFIG_XEN_BALLOON)	+= balloon.o
+obj-$(CONFIG_XEN_DEV_EVTCHN)	+= evtchn.o
+obj-$(CONFIG_XENFS)		+= xenfs/
+obj-$(CONFIG_XEN_SYS_HYPERVISOR)	+= sys-hypervisor.o

kernel/drivers/xen/balloon.c

@@ -0,0 +1,578 @@
+/******************************************************************************
+ * balloon.c
+ *
+ * Xen balloon driver - enables returning/claiming memory to/from Xen.
+ *
+ * Copyright (c) 2003, B Dragovic
+ * Copyright (c) 2003-2004, M Williamson, K Fraser
+ * Copyright (c) 2005 Dan M. Smith, IBM Corporation
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/bootmem.h>
+#include <linux/pagemap.h>
+#include <linux/highmem.h>
+#include <linux/mutex.h>
+#include <linux/list.h>
+#include <linux/sysdev.h>
+
+#include <asm/page.h>
+#include <asm/pgalloc.h>
+#include <asm/pgtable.h>
+#include <asm/uaccess.h>
+#include <asm/tlb.h>
+
+#include <asm/xen/hypervisor.h>
+#include <asm/xen/hypercall.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/memory.h>
+#include <xen/xenbus.h>
+#include <xen/features.h>
+#include <xen/page.h>
+
+#define PAGES2KB(_p) ((_p)<<(PAGE_SHIFT-10))
+
+#define BALLOON_CLASS_NAME "xen_memory"
+
+struct balloon_stats {
+	/* We aim for 'current allocation' == 'target allocation'. */
+	unsigned long current_pages;
+	unsigned long target_pages;
+	/*
+	 * Drivers may alter the memory reservation independently, but they
+	 * must inform the balloon driver so we avoid hitting the hard limit.
+	 */
+	unsigned long driver_pages;
+	/* Number of pages in high- and low-memory balloons. */
+	unsigned long balloon_low;
+	unsigned long balloon_high;
+};
+
+static DEFINE_MUTEX(balloon_mutex);
+
+static struct sys_device balloon_sysdev;
+
+static int register_balloon(struct sys_device *sysdev);
+
+/*
+ * Protects atomic reservation decrease/increase against concurrent increases.
+ * Also protects non-atomic updates of current_pages and driver_pages, and
+ * balloon lists.
+ */
+static DEFINE_SPINLOCK(balloon_lock);
+
+static struct balloon_stats balloon_stats;
+
+/* We increase/decrease in batches which fit in a page */
+static unsigned long frame_list[PAGE_SIZE / sizeof(unsigned long)];
+
+#ifdef CONFIG_HIGHMEM
+#define inc_totalhigh_pages() (totalhigh_pages++)
+#define dec_totalhigh_pages() (totalhigh_pages--)
+#else
+#define inc_totalhigh_pages() do {} while(0)
+#define dec_totalhigh_pages() do {} while(0)
+#endif
+
+/* List of ballooned pages, threaded through the mem_map array. */
+static LIST_HEAD(ballooned_pages);
+
+/* Main work function, always executed in process context. */
+static void balloon_process(struct work_struct *work);
+static DECLARE_WORK(balloon_worker, balloon_process);
+static struct timer_list balloon_timer;
+
+/* When ballooning out (allocating memory to return to Xen) we don't really
+   want the kernel to try too hard since that can trigger the oom killer. */
+#define GFP_BALLOON \
+	(GFP_HIGHUSER | __GFP_NOWARN | __GFP_NORETRY | __GFP_NOMEMALLOC)
+
+static void scrub_page(struct page *page)
+{
+#ifdef CONFIG_XEN_SCRUB_PAGES
+	clear_highpage(page);
+#endif
+}
+
+/* balloon_append: add the given page to the balloon. */
+static void balloon_append(struct page *page)
+{
+	/* Lowmem is re-populated first, so highmem pages go at list tail. */
+	if (PageHighMem(page)) {
+		list_add_tail(&page->lru, &ballooned_pages);
+		balloon_stats.balloon_high++;
+		dec_totalhigh_pages();
+	} else {
+		list_add(&page->lru, &ballooned_pages);
+		balloon_stats.balloon_low++;
+	}
+
+	totalram_pages--;
+}
+
+/* balloon_retrieve: rescue a page from the balloon, if it is not empty. */
+static struct page *balloon_retrieve(void)
+{
+	struct page *page;
+
+	if (list_empty(&ballooned_pages))
+		return NULL;
+
+	page = list_entry(ballooned_pages.next, struct page, lru);
+	list_del(&page->lru);
+
+	if (PageHighMem(page)) {
+		balloon_stats.balloon_high--;
+		inc_totalhigh_pages();
+	}
+	else
+		balloon_stats.balloon_low--;
+
+	totalram_pages++;
+
+	return page;
+}
+
+static struct page *balloon_first_page(void)
+{
+	if (list_empty(&ballooned_pages))
+		return NULL;
+	return list_entry(ballooned_pages.next, struct page, lru);
+}
+
+static struct page *balloon_next_page(struct page *page)
+{
+	struct list_head *next = page->lru.next;
+	if (next == &ballooned_pages)
+		return NULL;
+	return list_entry(next, struct page, lru);
+}
+
+static void balloon_alarm(unsigned long unused)
+{
+	schedule_work(&balloon_worker);
+}
+
+static unsigned long current_target(void)
+{
+	unsigned long target = balloon_stats.target_pages;
+
+	target = min(target,
+		     balloon_stats.current_pages +
+		     balloon_stats.balloon_low +
+		     balloon_stats.balloon_high);
+
+	return target;
+}
+
+static int increase_reservation(unsigned long nr_pages)
+{
+	unsigned long  pfn, i, flags;
+	struct page   *page;
+	long           rc;
+	struct xen_memory_reservation reservation = {
+		.address_bits = 0,
+		.extent_order = 0,
+		.domid        = DOMID_SELF
+	};
+
+	if (nr_pages > ARRAY_SIZE(frame_list))
+		nr_pages = ARRAY_SIZE(frame_list);
+
+	spin_lock_irqsave(&balloon_lock, flags);
+
+	page = balloon_first_page();
+	for (i = 0; i < nr_pages; i++) {
+		BUG_ON(page == NULL);
+		frame_list[i] = page_to_pfn(page);
+		page = balloon_next_page(page);
+	}
+
+	set_xen_guest_handle(reservation.extent_start, frame_list);
+	reservation.nr_extents = nr_pages;
+	rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
+	if (rc < 0)
+		goto out;
+
+	for (i = 0; i < rc; i++) {
+		page = balloon_retrieve();
+		BUG_ON(page == NULL);
+
+		pfn = page_to_pfn(page);
+		BUG_ON(!xen_feature(XENFEAT_auto_translated_physmap) &&
+		       phys_to_machine_mapping_valid(pfn));
+
+		set_phys_to_machine(pfn, frame_list[i]);
+
+		/* Link back into the page tables if not highmem. */
+		if (pfn < max_low_pfn) {
+			int ret;
+			ret = HYPERVISOR_update_va_mapping(
+				(unsigned long)__va(pfn << PAGE_SHIFT),
+				mfn_pte(frame_list[i], PAGE_KERNEL),
+				0);
+			BUG_ON(ret);
+		}
+
+		/* Relinquish the page back to the allocator. */
+		ClearPageReserved(page);
+		init_page_count(page);
+		__free_page(page);
+	}
+
+	balloon_stats.current_pages += rc;
+
+ out:
+	spin_unlock_irqrestore(&balloon_lock, flags);
+
+	return rc < 0 ? rc : rc != nr_pages;
+}
+
+static int decrease_reservation(unsigned long nr_pages)
+{
+	unsigned long  pfn, i, flags;
+	struct page   *page;
+	int            need_sleep = 0;
+	int ret;
+	struct xen_memory_reservation reservation = {
+		.address_bits = 0,
+		.extent_order = 0,
+		.domid        = DOMID_SELF
+	};
+
+	if (nr_pages > ARRAY_SIZE(frame_list))
+		nr_pages = ARRAY_SIZE(frame_list);
+
+	for (i = 0; i < nr_pages; i++) {
+		if ((page = alloc_page(GFP_BALLOON)) == NULL) {
+			nr_pages = i;
+			need_sleep = 1;
+			break;
+		}
+
+		pfn = page_to_pfn(page);
+		frame_list[i] = pfn_to_mfn(pfn);
+
+		scrub_page(page);
+
+		if (!PageHighMem(page)) {
+			ret = HYPERVISOR_update_va_mapping(
+				(unsigned long)__va(pfn << PAGE_SHIFT),
+				__pte_ma(0), 0);
+			BUG_ON(ret);
+                }
+
+	}
+
+	/* Ensure that ballooned highmem pages don't have kmaps. */
+	kmap_flush_unused();
+	flush_tlb_all();
+
+	spin_lock_irqsave(&balloon_lock, flags);
+
+	/* No more mappings: invalidate P2M and add to balloon. */
+	for (i = 0; i < nr_pages; i++) {
+		pfn = mfn_to_pfn(frame_list[i]);
+		set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
+		balloon_append(pfn_to_page(pfn));
+	}
+
+	set_xen_guest_handle(reservation.extent_start, frame_list);
+	reservation.nr_extents   = nr_pages;
+	ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
+	BUG_ON(ret != nr_pages);
+
+	balloon_stats.current_pages -= nr_pages;
+
+	spin_unlock_irqrestore(&balloon_lock, flags);
+
+	return need_sleep;
+}
+
+/*
+ * We avoid multiple worker processes conflicting via the balloon mutex.
+ * We may of course race updates of the target counts (which are protected
+ * by the balloon lock), or with changes to the Xen hard limit, but we will
+ * recover from these in time.
+ */
+static void balloon_process(struct work_struct *work)
+{
+	int need_sleep = 0;
+	long credit;
+
+	mutex_lock(&balloon_mutex);
+
+	do {
+		credit = current_target() - balloon_stats.current_pages;
+		if (credit > 0)
+			need_sleep = (increase_reservation(credit) != 0);
+		if (credit < 0)
+			need_sleep = (decrease_reservation(-credit) != 0);
+
+#ifndef CONFIG_PREEMPT
+		if (need_resched())
+			schedule();
+#endif
+	} while ((credit != 0) && !need_sleep);
+
+	/* Schedule more work if there is some still to be done. */
+	if (current_target() != balloon_stats.current_pages)
+		mod_timer(&balloon_timer, jiffies + HZ);
+
+	mutex_unlock(&balloon_mutex);
+}
+
+/* Resets the Xen limit, sets new target, and kicks off processing. */
+static void balloon_set_new_target(unsigned long target)
+{
+	/* No need for lock. Not read-modify-write updates. */
+	balloon_stats.target_pages = target;
+	schedule_work(&balloon_worker);
+}
+
+static struct xenbus_watch target_watch =
+{
+	.node = "memory/target"
+};
+
+/* React to a change in the target key */
+static void watch_target(struct xenbus_watch *watch,
+			 const char **vec, unsigned int len)
+{
+	unsigned long long new_target;
+	int err;
+
+	err = xenbus_scanf(XBT_NIL, "memory", "target", "%llu", &new_target);
+	if (err != 1) {
+		/* This is ok (for domain0 at least) - so just return */
+		return;
+	}
+
+	/* The given memory/target value is in KiB, so it needs converting to
+	 * pages. PAGE_SHIFT converts bytes to pages, hence PAGE_SHIFT - 10.
+	 */
+	balloon_set_new_target(new_target >> (PAGE_SHIFT - 10));
+}
+
+static int balloon_init_watcher(struct notifier_block *notifier,
+				unsigned long event,
+				void *data)
+{
+	int err;
+
+	err = register_xenbus_watch(&target_watch);
+	if (err)
+		printk(KERN_ERR "Failed to set balloon watcher\n");
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block xenstore_notifier;
+
+static int __init balloon_init(void)
+{
+	unsigned long pfn;
+	struct page *page;
+
+	if (!xen_pv_domain())
+		return -ENODEV;
+
+	pr_info("xen_balloon: Initialising balloon driver.\n");
+
+	balloon_stats.current_pages = min(xen_start_info->nr_pages, max_pfn);
+	balloon_stats.target_pages  = balloon_stats.current_pages;
+	balloon_stats.balloon_low   = 0;
+	balloon_stats.balloon_high  = 0;
+	balloon_stats.driver_pages  = 0UL;
+
+	init_timer(&balloon_timer);
+	balloon_timer.data = 0;
+	balloon_timer.function = balloon_alarm;
+
+	register_balloon(&balloon_sysdev);
+
+	/* Initialise the balloon with excess memory space. */
+	for (pfn = xen_start_info->nr_pages; pfn < max_pfn; pfn++) {
+		page = pfn_to_page(pfn);
+		if (!PageReserved(page))
+			balloon_append(page);
+	}
+
+	target_watch.callback = watch_target;
+	xenstore_notifier.notifier_call = balloon_init_watcher;
+
+	register_xenstore_notifier(&xenstore_notifier);
+
+	return 0;
+}
+
+subsys_initcall(balloon_init);
+
+static void balloon_exit(void)
+{
+    /* XXX - release balloon here */
+    return;
+}
+
+module_exit(balloon_exit);
+
+#define BALLOON_SHOW(name, format, args...)				\
+	static ssize_t show_##name(struct sys_device *dev,		\
+				   struct sysdev_attribute *attr,	\
+				   char *buf)				\
+	{								\
+		return sprintf(buf, format, ##args);			\
+	}								\
+	static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL)
+
+BALLOON_SHOW(current_kb, "%lu\n", PAGES2KB(balloon_stats.current_pages));
+BALLOON_SHOW(low_kb, "%lu\n", PAGES2KB(balloon_stats.balloon_low));
+BALLOON_SHOW(high_kb, "%lu\n", PAGES2KB(balloon_stats.balloon_high));
+BALLOON_SHOW(driver_kb, "%lu\n", PAGES2KB(balloon_stats.driver_pages));
+
+static ssize_t show_target_kb(struct sys_device *dev, struct sysdev_attribute *attr,
+			      char *buf)
+{
+	return sprintf(buf, "%lu\n", PAGES2KB(balloon_stats.target_pages));
+}
+
+static ssize_t store_target_kb(struct sys_device *dev,
+			       struct sysdev_attribute *attr,
+			       const char *buf,
+			       size_t count)
+{
+	char *endchar;
+	unsigned long long target_bytes;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	target_bytes = simple_strtoull(buf, &endchar, 0) * 1024;
+
+	balloon_set_new_target(target_bytes >> PAGE_SHIFT);
+
+	return count;
+}
+
+static SYSDEV_ATTR(target_kb, S_IRUGO | S_IWUSR,
+		   show_target_kb, store_target_kb);
+
+
+static ssize_t show_target(struct sys_device *dev, struct sysdev_attribute *attr,
+			      char *buf)
+{
+	return sprintf(buf, "%llu\n",
+		       (unsigned long long)balloon_stats.target_pages
+		       << PAGE_SHIFT);
+}
+
+static ssize_t store_target(struct sys_device *dev,
+			    struct sysdev_attribute *attr,
+			    const char *buf,
+			    size_t count)
+{
+	char *endchar;
+	unsigned long long target_bytes;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	target_bytes = memparse(buf, &endchar);
+
+	balloon_set_new_target(target_bytes >> PAGE_SHIFT);
+
+	return count;
+}
+
+static SYSDEV_ATTR(target, S_IRUGO | S_IWUSR,
+		   show_target, store_target);
+
+
+static struct sysdev_attribute *balloon_attrs[] = {
+	&attr_target_kb,
+	&attr_target,
+};
+
+static struct attribute *balloon_info_attrs[] = {
+	&attr_current_kb.attr,
+	&attr_low_kb.attr,
+	&attr_high_kb.attr,
+	&attr_driver_kb.attr,
+	NULL
+};
+
+static struct attribute_group balloon_info_group = {
+	.name = "info",
+	.attrs = balloon_info_attrs,
+};
+
+static struct sysdev_class balloon_sysdev_class = {
+	.name = BALLOON_CLASS_NAME,
+};
+
+static int register_balloon(struct sys_device *sysdev)
+{
+	int i, error;
+
+	error = sysdev_class_register(&balloon_sysdev_class);
+	if (error)
+		return error;
+
+	sysdev->id = 0;
+	sysdev->cls = &balloon_sysdev_class;
+
+	error = sysdev_register(sysdev);
+	if (error) {
+		sysdev_class_unregister(&balloon_sysdev_class);
+		return error;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(balloon_attrs); i++) {
+		error = sysdev_create_file(sysdev, balloon_attrs[i]);
+		if (error)
+			goto fail;
+	}
+
+	error = sysfs_create_group(&sysdev->kobj, &balloon_info_group);
+	if (error)
+		goto fail;
+
+	return 0;
+
+ fail:
+	while (--i >= 0)
+		sysdev_remove_file(sysdev, balloon_attrs[i]);
+	sysdev_unregister(sysdev);
+	sysdev_class_unregister(&balloon_sysdev_class);
+	return error;
+}
+
+MODULE_LICENSE("GPL");

kernel/drivers/xen/cpu_hotplug.c

@@ -0,0 +1,110 @@
+#include <linux/notifier.h>
+
+#include <xen/xenbus.h>
+
+#include <asm/xen/hypervisor.h>
+#include <asm/cpu.h>
+
+static void enable_hotplug_cpu(int cpu)
+{
+	if (!cpu_present(cpu))
+		arch_register_cpu(cpu);
+
+	set_cpu_present(cpu, true);
+}
+
+static void disable_hotplug_cpu(int cpu)
+{
+	if (cpu_present(cpu))
+		arch_unregister_cpu(cpu);
+
+	set_cpu_present(cpu, false);
+}
+
+static int vcpu_online(unsigned int cpu)
+{
+	int err;
+	char dir[32], state[32];
+
+	sprintf(dir, "cpu/%u", cpu);
+	err = xenbus_scanf(XBT_NIL, dir, "availability", "%s", state);
+	if (err != 1) {
+		printk(KERN_ERR "XENBUS: Unable to read cpu state\n");
+		return err;
+	}
+
+	if (strcmp(state, "online") == 0)
+		return 1;
+	else if (strcmp(state, "offline") == 0)
+		return 0;
+
+	printk(KERN_ERR "XENBUS: unknown state(%s) on CPU%d\n", state, cpu);
+	return -EINVAL;
+}
+static void vcpu_hotplug(unsigned int cpu)
+{
+	if (!cpu_possible(cpu))
+		return;
+
+	switch (vcpu_online(cpu)) {
+	case 1:
+		enable_hotplug_cpu(cpu);
+		break;
+	case 0:
+		(void)cpu_down(cpu);
+		disable_hotplug_cpu(cpu);
+		break;
+	default:
+		break;
+	}
+}
+
+static void handle_vcpu_hotplug_event(struct xenbus_watch *watch,
+					const char **vec, unsigned int len)
+{
+	unsigned int cpu;
+	char *cpustr;
+	const char *node = vec[XS_WATCH_PATH];
+
+	cpustr = strstr(node, "cpu/");
+	if (cpustr != NULL) {
+		sscanf(cpustr, "cpu/%u", &cpu);
+		vcpu_hotplug(cpu);
+	}
+}
+
+static int setup_cpu_watcher(struct notifier_block *notifier,
+			      unsigned long event, void *data)
+{
+	int cpu;
+	static struct xenbus_watch cpu_watch = {
+		.node = "cpu",
+		.callback = handle_vcpu_hotplug_event};
+
+	(void)register_xenbus_watch(&cpu_watch);
+
+	for_each_possible_cpu(cpu) {
+		if (vcpu_online(cpu) == 0) {
+			(void)cpu_down(cpu);
+			cpu_clear(cpu, cpu_present_map);
+		}
+	}
+
+	return NOTIFY_DONE;
+}
+
+static int __init setup_vcpu_hotplug_event(void)
+{
+	static struct notifier_block xsn_cpu = {
+		.notifier_call = setup_cpu_watcher };
+
+	if (!xen_pv_domain())
+		return -ENODEV;
+
+	register_xenstore_notifier(&xsn_cpu);
+
+	return 0;
+}
+
+arch_initcall(setup_vcpu_hotplug_event);
+

kernel/drivers/xen/events.c

@@ -0,0 +1,951 @@
+/*
+ * Xen event channels
+ *
+ * Xen models interrupts with abstract event channels.  Because each
+ * domain gets 1024 event channels, but NR_IRQ is not that large, we
+ * must dynamically map irqs<->event channels.  The event channels
+ * interface with the rest of the kernel by defining a xen interrupt
+ * chip.  When an event is recieved, it is mapped to an irq and sent
+ * through the normal interrupt processing path.
+ *
+ * There are four kinds of events which can be mapped to an event
+ * channel:
+ *
+ * 1. Inter-domain notifications.  This includes all the virtual
+ *    device events, since they're driven by front-ends in another domain
+ *    (typically dom0).
+ * 2. VIRQs, typically used for timers.  These are per-cpu events.
+ * 3. IPIs.
+ * 4. Hardware interrupts. Not supported at present.
+ *
+ * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007
+ */
+
+#include <linux/linkage.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/bootmem.h>
+
+#include <asm/ptrace.h>
+#include <asm/irq.h>
+#include <asm/idle.h>
+#include <asm/sync_bitops.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/hypervisor.h>
+
+#include <xen/xen-ops.h>
+#include <xen/events.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/event_channel.h>
+
+/*
+ * This lock protects updates to the following mapping and reference-count
+ * arrays. The lock does not need to be acquired to read the mapping tables.
+ */
+static DEFINE_SPINLOCK(irq_mapping_update_lock);
+
+/* IRQ <-> VIRQ mapping. */
+static DEFINE_PER_CPU(int [NR_VIRQS], virq_to_irq) = {[0 ... NR_VIRQS-1] = -1};
+
+/* IRQ <-> IPI mapping */
+static DEFINE_PER_CPU(int [XEN_NR_IPIS], ipi_to_irq) = {[0 ... XEN_NR_IPIS-1] = -1};
+
+/* Interrupt types. */
+enum xen_irq_type {
+	IRQT_UNBOUND = 0,
+	IRQT_PIRQ,
+	IRQT_VIRQ,
+	IRQT_IPI,
+	IRQT_EVTCHN
+};
+
+/*
+ * Packed IRQ information:
+ * type - enum xen_irq_type
+ * event channel - irq->event channel mapping
+ * cpu - cpu this event channel is bound to
+ * index - type-specific information:
+ *    PIRQ - vector, with MSB being "needs EIO"
+ *    VIRQ - virq number
+ *    IPI - IPI vector
+ *    EVTCHN -
+ */
+struct irq_info
+{
+	enum xen_irq_type type;	/* type */
+	unsigned short evtchn;	/* event channel */
+	unsigned short cpu;	/* cpu bound */
+
+	union {
+		unsigned short virq;
+		enum ipi_vector ipi;
+		struct {
+			unsigned short gsi;
+			unsigned short vector;
+		} pirq;
+	} u;
+};
+
+static struct irq_info irq_info[NR_IRQS];
+
+static int evtchn_to_irq[NR_EVENT_CHANNELS] = {
+	[0 ... NR_EVENT_CHANNELS-1] = -1
+};
+struct cpu_evtchn_s {
+	unsigned long bits[NR_EVENT_CHANNELS/BITS_PER_LONG];
+};
+static struct cpu_evtchn_s *cpu_evtchn_mask_p;
+static inline unsigned long *cpu_evtchn_mask(int cpu)
+{
+	return cpu_evtchn_mask_p[cpu].bits;
+}
+
+/* Xen will never allocate port zero for any purpose. */
+#define VALID_EVTCHN(chn)	((chn) != 0)
+
+static struct irq_chip xen_dynamic_chip;
+static struct irq_chip xen_percpu_chip;
+
+/* Constructor for packed IRQ information. */
+static struct irq_info mk_unbound_info(void)
+{
+	return (struct irq_info) { .type = IRQT_UNBOUND };
+}
+
+static struct irq_info mk_evtchn_info(unsigned short evtchn)
+{
+	return (struct irq_info) { .type = IRQT_EVTCHN, .evtchn = evtchn,
+			.cpu = 0 };
+}
+
+static struct irq_info mk_ipi_info(unsigned short evtchn, enum ipi_vector ipi)
+{
+	return (struct irq_info) { .type = IRQT_IPI, .evtchn = evtchn,
+			.cpu = 0, .u.ipi = ipi };
+}
+
+static struct irq_info mk_virq_info(unsigned short evtchn, unsigned short virq)
+{
+	return (struct irq_info) { .type = IRQT_VIRQ, .evtchn = evtchn,
+			.cpu = 0, .u.virq = virq };
+}
+
+static struct irq_info mk_pirq_info(unsigned short evtchn,
+				    unsigned short gsi, unsigned short vector)
+{
+	return (struct irq_info) { .type = IRQT_PIRQ, .evtchn = evtchn,
+			.cpu = 0, .u.pirq = { .gsi = gsi, .vector = vector } };
+}
+
+/*
+ * Accessors for packed IRQ information.
+ */
+static struct irq_info *info_for_irq(unsigned irq)
+{
+	return &irq_info[irq];
+}
+
+static unsigned int evtchn_from_irq(unsigned irq)
+{
+	return info_for_irq(irq)->evtchn;
+}
+
+unsigned irq_from_evtchn(unsigned int evtchn)
+{
+	return evtchn_to_irq[evtchn];
+}
+EXPORT_SYMBOL_GPL(irq_from_evtchn);
+
+static enum ipi_vector ipi_from_irq(unsigned irq)
+{
+	struct irq_info *info = info_for_irq(irq);
+
+	BUG_ON(info == NULL);
+	BUG_ON(info->type != IRQT_IPI);
+
+	return info->u.ipi;
+}
+
+static unsigned virq_from_irq(unsigned irq)
+{
+	struct irq_info *info = info_for_irq(irq);
+
+	BUG_ON(info == NULL);
+	BUG_ON(info->type != IRQT_VIRQ);
+
+	return info->u.virq;
+}
+
+static unsigned gsi_from_irq(unsigned irq)
+{
+	struct irq_info *info = info_for_irq(irq);
+
+	BUG_ON(info == NULL);
+	BUG_ON(info->type != IRQT_PIRQ);
+
+	return info->u.pirq.gsi;
+}
+
+static unsigned vector_from_irq(unsigned irq)
+{
+	struct irq_info *info = info_for_irq(irq);
+
+	BUG_ON(info == NULL);
+	BUG_ON(info->type != IRQT_PIRQ);
+
+	return info->u.pirq.vector;
+}
+
+static enum xen_irq_type type_from_irq(unsigned irq)
+{
+	return info_for_irq(irq)->type;
+}
+
+static unsigned cpu_from_irq(unsigned irq)
+{
+	return info_for_irq(irq)->cpu;
+}
+
+static unsigned int cpu_from_evtchn(unsigned int evtchn)
+{
+	int irq = evtchn_to_irq[evtchn];
+	unsigned ret = 0;
+
+	if (irq != -1)
+		ret = cpu_from_irq(irq);
+
+	return ret;
+}
+
+static inline unsigned long active_evtchns(unsigned int cpu,
+					   struct shared_info *sh,
+					   unsigned int idx)
+{
+	return (sh->evtchn_pending[idx] &
+		cpu_evtchn_mask(cpu)[idx] &
+		~sh->evtchn_mask[idx]);
+}
+
+static void bind_evtchn_to_cpu(unsigned int chn, unsigned int cpu)
+{
+	int irq = evtchn_to_irq[chn];
+
+	BUG_ON(irq == -1);
+#ifdef CONFIG_SMP
+	cpumask_copy(irq_to_desc(irq)->affinity, cpumask_of(cpu));
+#endif
+
+	__clear_bit(chn, cpu_evtchn_mask(cpu_from_irq(irq)));
+	__set_bit(chn, cpu_evtchn_mask(cpu));
+
+	irq_info[irq].cpu = cpu;
+}
+
+static void init_evtchn_cpu_bindings(void)
+{
+#ifdef CONFIG_SMP
+	struct irq_desc *desc;
+	int i;
+
+	/* By default all event channels notify CPU#0. */
+	for_each_irq_desc(i, desc) {
+		cpumask_copy(desc->affinity, cpumask_of(0));
+	}
+#endif
+
+	memset(cpu_evtchn_mask(0), ~0, sizeof(struct cpu_evtchn_s));
+}
+
+static inline void clear_evtchn(int port)
+{
+	struct shared_info *s = HYPERVISOR_shared_info;
+	sync_clear_bit(port, &s->evtchn_pending[0]);
+}
+
+static inline void set_evtchn(int port)
+{
+	struct shared_info *s = HYPERVISOR_shared_info;
+	sync_set_bit(port, &s->evtchn_pending[0]);
+}
+
+static inline int test_evtchn(int port)
+{
+	struct shared_info *s = HYPERVISOR_shared_info;
+	return sync_test_bit(port, &s->evtchn_pending[0]);
+}
+
+
+/**
+ * notify_remote_via_irq - send event to remote end of event channel via irq
+ * @irq: irq of event channel to send event to
+ *
+ * Unlike notify_remote_via_evtchn(), this is safe to use across
+ * save/restore. Notifications on a broken connection are silently
+ * dropped.
+ */
+void notify_remote_via_irq(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		notify_remote_via_evtchn(evtchn);
+}
+EXPORT_SYMBOL_GPL(notify_remote_via_irq);
+
+static void mask_evtchn(int port)
+{
+	struct shared_info *s = HYPERVISOR_shared_info;
+	sync_set_bit(port, &s->evtchn_mask[0]);
+}
+
+static void unmask_evtchn(int port)
+{
+	struct shared_info *s = HYPERVISOR_shared_info;
+	unsigned int cpu = get_cpu();
+
+	BUG_ON(!irqs_disabled());
+
+	/* Slow path (hypercall) if this is a non-local port. */
+	if (unlikely(cpu != cpu_from_evtchn(port))) {
+		struct evtchn_unmask unmask = { .port = port };
+		(void)HYPERVISOR_event_channel_op(EVTCHNOP_unmask, &unmask);
+	} else {
+		struct vcpu_info *vcpu_info = __get_cpu_var(xen_vcpu);
+
+		sync_clear_bit(port, &s->evtchn_mask[0]);
+
+		/*
+		 * The following is basically the equivalent of
+		 * 'hw_resend_irq'. Just like a real IO-APIC we 'lose
+		 * the interrupt edge' if the channel is masked.
+		 */
+		if (sync_test_bit(port, &s->evtchn_pending[0]) &&
+		    !sync_test_and_set_bit(port / BITS_PER_LONG,
+					   &vcpu_info->evtchn_pending_sel))
+			vcpu_info->evtchn_upcall_pending = 1;
+	}
+
+	put_cpu();
+}
+
+static int find_unbound_irq(void)
+{
+	int irq;
+	struct irq_desc *desc;
+
+	for (irq = 0; irq < nr_irqs; irq++)
+		if (irq_info[irq].type == IRQT_UNBOUND)
+			break;
+
+	if (irq == nr_irqs)
+		panic("No available IRQ to bind to: increase nr_irqs!\n");
+
+	desc = irq_to_desc_alloc_node(irq, 0);
+	if (WARN_ON(desc == NULL))
+		return -1;
+
+	dynamic_irq_init(irq);
+
+	return irq;
+}
+
+int bind_evtchn_to_irq(unsigned int evtchn)
+{
+	int irq;
+
+	spin_lock(&irq_mapping_update_lock);
+
+	irq = evtchn_to_irq[evtchn];
+
+	if (irq == -1) {
+		irq = find_unbound_irq();
+
+		set_irq_chip_and_handler_name(irq, &xen_dynamic_chip,
+					      handle_edge_irq, "event");
+
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_evtchn_info(evtchn);
+	}
+
+	spin_unlock(&irq_mapping_update_lock);
+
+	return irq;
+}
+EXPORT_SYMBOL_GPL(bind_evtchn_to_irq);
+
+static int bind_ipi_to_irq(unsigned int ipi, unsigned int cpu)
+{
+	struct evtchn_bind_ipi bind_ipi;
+	int evtchn, irq;
+
+	spin_lock(&irq_mapping_update_lock);
+
+	irq = per_cpu(ipi_to_irq, cpu)[ipi];
+
+	if (irq == -1) {
+		irq = find_unbound_irq();
+		if (irq < 0)
+			goto out;
+
+		set_irq_chip_and_handler_name(irq, &xen_percpu_chip,
+					      handle_percpu_irq, "ipi");
+
+		bind_ipi.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_ipi,
+						&bind_ipi) != 0)
+			BUG();
+		evtchn = bind_ipi.port;
+
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_ipi_info(evtchn, ipi);
+		per_cpu(ipi_to_irq, cpu)[ipi] = irq;
+
+		bind_evtchn_to_cpu(evtchn, cpu);
+	}
+
+ out:
+	spin_unlock(&irq_mapping_update_lock);
+	return irq;
+}
+
+
+static int bind_virq_to_irq(unsigned int virq, unsigned int cpu)
+{
+	struct evtchn_bind_virq bind_virq;
+	int evtchn, irq;
+
+	spin_lock(&irq_mapping_update_lock);
+
+	irq = per_cpu(virq_to_irq, cpu)[virq];
+
+	if (irq == -1) {
+		bind_virq.virq = virq;
+		bind_virq.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
+						&bind_virq) != 0)
+			BUG();
+		evtchn = bind_virq.port;
+
+		irq = find_unbound_irq();
+
+		set_irq_chip_and_handler_name(irq, &xen_percpu_chip,
+					      handle_percpu_irq, "virq");
+
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_virq_info(evtchn, virq);
+
+		per_cpu(virq_to_irq, cpu)[virq] = irq;
+
+		bind_evtchn_to_cpu(evtchn, cpu);
+	}
+
+	spin_unlock(&irq_mapping_update_lock);
+
+	return irq;
+}
+
+static void unbind_from_irq(unsigned int irq)
+{
+	struct evtchn_close close;
+	int evtchn = evtchn_from_irq(irq);
+
+	spin_lock(&irq_mapping_update_lock);
+
+	if (VALID_EVTCHN(evtchn)) {
+		close.port = evtchn;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_close, &close) != 0)
+			BUG();
+
+		switch (type_from_irq(irq)) {
+		case IRQT_VIRQ:
+			per_cpu(virq_to_irq, cpu_from_evtchn(evtchn))
+				[virq_from_irq(irq)] = -1;
+			break;
+		case IRQT_IPI:
+			per_cpu(ipi_to_irq, cpu_from_evtchn(evtchn))
+				[ipi_from_irq(irq)] = -1;
+			break;
+		default:
+			break;
+		}
+
+		/* Closed ports are implicitly re-bound to VCPU0. */
+		bind_evtchn_to_cpu(evtchn, 0);
+
+		evtchn_to_irq[evtchn] = -1;
+	}
+
+	if (irq_info[irq].type != IRQT_UNBOUND) {
+		irq_info[irq] = mk_unbound_info();
+
+		dynamic_irq_cleanup(irq);
+	}
+
+	spin_unlock(&irq_mapping_update_lock);
+}
+
+int bind_evtchn_to_irqhandler(unsigned int evtchn,
+			      irq_handler_t handler,
+			      unsigned long irqflags,
+			      const char *devname, void *dev_id)
+{
+	unsigned int irq;
+	int retval;
+
+	irq = bind_evtchn_to_irq(evtchn);
+	retval = request_irq(irq, handler, irqflags, devname, dev_id);
+	if (retval != 0) {
+		unbind_from_irq(irq);
+		return retval;
+	}
+
+	return irq;
+}
+EXPORT_SYMBOL_GPL(bind_evtchn_to_irqhandler);
+
+int bind_virq_to_irqhandler(unsigned int virq, unsigned int cpu,
+			    irq_handler_t handler,
+			    unsigned long irqflags, const char *devname, void *dev_id)
+{
+	unsigned int irq;
+	int retval;
+
+	irq = bind_virq_to_irq(virq, cpu);
+	retval = request_irq(irq, handler, irqflags, devname, dev_id);
+	if (retval != 0) {
+		unbind_from_irq(irq);
+		return retval;
+	}
+
+	return irq;
+}
+EXPORT_SYMBOL_GPL(bind_virq_to_irqhandler);
+
+int bind_ipi_to_irqhandler(enum ipi_vector ipi,
+			   unsigned int cpu,
+			   irq_handler_t handler,
+			   unsigned long irqflags,
+			   const char *devname,
+			   void *dev_id)
+{
+	int irq, retval;
+
+	irq = bind_ipi_to_irq(ipi, cpu);
+	if (irq < 0)
+		return irq;
+
+	irqflags |= IRQF_NO_SUSPEND | IRQF_FORCE_RESUME;
+	retval = request_irq(irq, handler, irqflags, devname, dev_id);
+	if (retval != 0) {
+		unbind_from_irq(irq);
+		return retval;
+	}
+
+	return irq;
+}
+
+void unbind_from_irqhandler(unsigned int irq, void *dev_id)
+{
+	free_irq(irq, dev_id);
+	unbind_from_irq(irq);
+}
+EXPORT_SYMBOL_GPL(unbind_from_irqhandler);
+
+void xen_send_IPI_one(unsigned int cpu, enum ipi_vector vector)
+{
+	int irq = per_cpu(ipi_to_irq, cpu)[vector];
+	BUG_ON(irq < 0);
+	notify_remote_via_irq(irq);
+}
+
+irqreturn_t xen_debug_interrupt(int irq, void *dev_id)
+{
+	struct shared_info *sh = HYPERVISOR_shared_info;
+	int cpu = smp_processor_id();
+	int i;
+	unsigned long flags;
+	static DEFINE_SPINLOCK(debug_lock);
+
+	spin_lock_irqsave(&debug_lock, flags);
+
+	printk("vcpu %d\n  ", cpu);
+
+	for_each_online_cpu(i) {
+		struct vcpu_info *v = per_cpu(xen_vcpu, i);
+		printk("%d: masked=%d pending=%d event_sel %08lx\n  ", i,
+			(get_irq_regs() && i == cpu) ? xen_irqs_disabled(get_irq_regs()) : v->evtchn_upcall_mask,
+			v->evtchn_upcall_pending,
+			v->evtchn_pending_sel);
+	}
+	printk("pending:\n   ");
+	for(i = ARRAY_SIZE(sh->evtchn_pending)-1; i >= 0; i--)
+		printk("%08lx%s", sh->evtchn_pending[i],
+			i % 8 == 0 ? "\n   " : " ");
+	printk("\nmasks:\n   ");
+	for(i = ARRAY_SIZE(sh->evtchn_mask)-1; i >= 0; i--)
+		printk("%08lx%s", sh->evtchn_mask[i],
+			i % 8 == 0 ? "\n   " : " ");
+
+	printk("\nunmasked:\n   ");
+	for(i = ARRAY_SIZE(sh->evtchn_mask)-1; i >= 0; i--)
+		printk("%08lx%s", sh->evtchn_pending[i] & ~sh->evtchn_mask[i],
+			i % 8 == 0 ? "\n   " : " ");
+
+	printk("\npending list:\n");
+	for(i = 0; i < NR_EVENT_CHANNELS; i++) {
+		if (sync_test_bit(i, sh->evtchn_pending)) {
+			printk("  %d: event %d -> irq %d\n",
+			       cpu_from_evtchn(i), i,
+			       evtchn_to_irq[i]);
+		}
+	}
+
+	spin_unlock_irqrestore(&debug_lock, flags);
+
+	return IRQ_HANDLED;
+}
+
+static DEFINE_PER_CPU(unsigned, xed_nesting_count);
+
+/*
+ * Search the CPUs pending events bitmasks.  For each one found, map
+ * the event number to an irq, and feed it into do_IRQ() for
+ * handling.
+ *
+ * Xen uses a two-level bitmap to speed searching.  The first level is
+ * a bitset of words which contain pending event bits.  The second
+ * level is a bitset of pending events themselves.
+ */
+void xen_evtchn_do_upcall(struct pt_regs *regs)
+{
+	int cpu = get_cpu();
+	struct pt_regs *old_regs = set_irq_regs(regs);
+	struct shared_info *s = HYPERVISOR_shared_info;
+	struct vcpu_info *vcpu_info = __get_cpu_var(xen_vcpu);
+ 	unsigned count;
+
+	exit_idle();
+	irq_enter();
+
+	do {
+		unsigned long pending_words;
+
+		vcpu_info->evtchn_upcall_pending = 0;
+
+		if (__get_cpu_var(xed_nesting_count)++)
+			goto out;
+
+#ifndef CONFIG_X86 /* No need for a barrier -- XCHG is a barrier on x86. */
+		/* Clear master flag /before/ clearing selector flag. */
+		wmb();
+#endif
+		pending_words = xchg(&vcpu_info->evtchn_pending_sel, 0);
+		while (pending_words != 0) {
+			unsigned long pending_bits;
+			int word_idx = __ffs(pending_words);
+			pending_words &= ~(1UL << word_idx);
+
+			while ((pending_bits = active_evtchns(cpu, s, word_idx)) != 0) {
+				int bit_idx = __ffs(pending_bits);
+				int port = (word_idx * BITS_PER_LONG) + bit_idx;
+				int irq = evtchn_to_irq[port];
+
+				if (irq != -1)
+					handle_irq(irq, regs);
+			}
+		}
+
+		BUG_ON(!irqs_disabled());
+
+		count = __get_cpu_var(xed_nesting_count);
+		__get_cpu_var(xed_nesting_count) = 0;
+	} while(count != 1);
+
+out:
+	irq_exit();
+	set_irq_regs(old_regs);
+
+	put_cpu();
+}
+
+/* Rebind a new event channel to an existing irq. */
+void rebind_evtchn_irq(int evtchn, int irq)
+{
+	struct irq_info *info = info_for_irq(irq);
+
+	/* Make sure the irq is masked, since the new event channel
+	   will also be masked. */
+	disable_irq(irq);
+
+	spin_lock(&irq_mapping_update_lock);
+
+	/* After resume the irq<->evtchn mappings are all cleared out */
+	BUG_ON(evtchn_to_irq[evtchn] != -1);
+	/* Expect irq to have been bound before,
+	   so there should be a proper type */
+	BUG_ON(info->type == IRQT_UNBOUND);
+
+	evtchn_to_irq[evtchn] = irq;
+	irq_info[irq] = mk_evtchn_info(evtchn);
+
+	spin_unlock(&irq_mapping_update_lock);
+
+	/* new event channels are always bound to cpu 0 */
+	irq_set_affinity(irq, cpumask_of(0));
+
+	/* Unmask the event channel. */
+	enable_irq(irq);
+}
+
+/* Rebind an evtchn so that it gets delivered to a specific cpu */
+static int rebind_irq_to_cpu(unsigned irq, unsigned tcpu)
+{
+	struct evtchn_bind_vcpu bind_vcpu;
+	int evtchn = evtchn_from_irq(irq);
+
+	if (!VALID_EVTCHN(evtchn))
+		return -1;
+
+	/* Send future instances of this interrupt to other vcpu. */
+	bind_vcpu.port = evtchn;
+	bind_vcpu.vcpu = tcpu;
+
+	/*
+	 * If this fails, it usually just indicates that we're dealing with a
+	 * virq or IPI channel, which don't actually need to be rebound. Ignore
+	 * it, but don't do the xenlinux-level rebind in that case.
+	 */
+	if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_vcpu, &bind_vcpu) >= 0)
+		bind_evtchn_to_cpu(evtchn, tcpu);
+
+	return 0;
+}
+
+static int set_affinity_irq(unsigned irq, const struct cpumask *dest)
+{
+	unsigned tcpu = cpumask_first(dest);
+
+	return rebind_irq_to_cpu(irq, tcpu);
+}
+
+int resend_irq_on_evtchn(unsigned int irq)
+{
+	int masked, evtchn = evtchn_from_irq(irq);
+	struct shared_info *s = HYPERVISOR_shared_info;
+
+	if (!VALID_EVTCHN(evtchn))
+		return 1;
+
+	masked = sync_test_and_set_bit(evtchn, s->evtchn_mask);
+	sync_set_bit(evtchn, s->evtchn_pending);
+	if (!masked)
+		unmask_evtchn(evtchn);
+
+	return 1;
+}
+
+static void enable_dynirq(unsigned int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		unmask_evtchn(evtchn);
+}
+
+static void disable_dynirq(unsigned int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		mask_evtchn(evtchn);
+}
+
+static void ack_dynirq(unsigned int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	move_native_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		clear_evtchn(evtchn);
+}
+
+static int retrigger_dynirq(unsigned int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+	struct shared_info *sh = HYPERVISOR_shared_info;
+	int ret = 0;
+
+	if (VALID_EVTCHN(evtchn)) {
+		int masked;
+
+		masked = sync_test_and_set_bit(evtchn, sh->evtchn_mask);
+		sync_set_bit(evtchn, sh->evtchn_pending);
+		if (!masked)
+			unmask_evtchn(evtchn);
+		ret = 1;
+	}
+
+	return ret;
+}
+
+static void restore_cpu_virqs(unsigned int cpu)
+{
+	struct evtchn_bind_virq bind_virq;
+	int virq, irq, evtchn;
+
+	for (virq = 0; virq < NR_VIRQS; virq++) {
+		if ((irq = per_cpu(virq_to_irq, cpu)[virq]) == -1)
+			continue;
+
+		BUG_ON(virq_from_irq(irq) != virq);
+
+		/* Get a new binding from Xen. */
+		bind_virq.virq = virq;
+		bind_virq.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
+						&bind_virq) != 0)
+			BUG();
+		evtchn = bind_virq.port;
+
+		/* Record the new mapping. */
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_virq_info(evtchn, virq);
+		bind_evtchn_to_cpu(evtchn, cpu);
+	}
+}
+
+static void restore_cpu_ipis(unsigned int cpu)
+{
+	struct evtchn_bind_ipi bind_ipi;
+	int ipi, irq, evtchn;
+
+	for (ipi = 0; ipi < XEN_NR_IPIS; ipi++) {
+		if ((irq = per_cpu(ipi_to_irq, cpu)[ipi]) == -1)
+			continue;
+
+		BUG_ON(ipi_from_irq(irq) != ipi);
+
+		/* Get a new binding from Xen. */
+		bind_ipi.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_ipi,
+						&bind_ipi) != 0)
+			BUG();
+		evtchn = bind_ipi.port;
+
+		/* Record the new mapping. */
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_ipi_info(evtchn, ipi);
+		bind_evtchn_to_cpu(evtchn, cpu);
+	}
+}
+
+/* Clear an irq's pending state, in preparation for polling on it */
+void xen_clear_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		clear_evtchn(evtchn);
+}
+
+void xen_set_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		set_evtchn(evtchn);
+}
+
+bool xen_test_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+	bool ret = false;
+
+	if (VALID_EVTCHN(evtchn))
+		ret = test_evtchn(evtchn);
+
+	return ret;
+}
+
+/* Poll waiting for an irq to become pending.  In the usual case, the
+   irq will be disabled so it won't deliver an interrupt. */
+void xen_poll_irq(int irq)
+{
+	evtchn_port_t evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn)) {
+		struct sched_poll poll;
+
+		poll.nr_ports = 1;
+		poll.timeout = 0;
+		set_xen_guest_handle(poll.ports, &evtchn);
+
+		if (HYPERVISOR_sched_op(SCHEDOP_poll, &poll) != 0)
+			BUG();
+	}
+}
+
+void xen_irq_resume(void)
+{
+	unsigned int cpu, irq, evtchn;
+
+	init_evtchn_cpu_bindings();
+
+	/* New event-channel space is not 'live' yet. */
+	for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
+		mask_evtchn(evtchn);
+
+	/* No IRQ <-> event-channel mappings. */
+	for (irq = 0; irq < nr_irqs; irq++)
+		irq_info[irq].evtchn = 0; /* zap event-channel binding */
+
+	for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
+		evtchn_to_irq[evtchn] = -1;
+
+	for_each_possible_cpu(cpu) {
+		restore_cpu_virqs(cpu);
+		restore_cpu_ipis(cpu);
+	}
+}
+
+static struct irq_chip xen_dynamic_chip __read_mostly = {
+	.name		= "xen-dyn",
+
+	.disable	= disable_dynirq,
+	.mask		= disable_dynirq,
+	.unmask		= enable_dynirq,
+
+	.ack		= ack_dynirq,
+	.set_affinity	= set_affinity_irq,
+	.retrigger	= retrigger_dynirq,
+};
+
+static struct irq_chip xen_percpu_chip __read_mostly = {
+	.name		= "xen-percpu",
+
+	.disable	= disable_dynirq,
+	.mask		= disable_dynirq,
+	.unmask		= enable_dynirq,
+
+	.ack		= ack_dynirq,
+};
+
+void __init xen_init_IRQ(void)
+{
+	int i;
+
+	cpu_evtchn_mask_p = kcalloc(nr_cpu_ids, sizeof(struct cpu_evtchn_s),
+				    GFP_KERNEL);
+	BUG_ON(cpu_evtchn_mask_p == NULL);
+
+	init_evtchn_cpu_bindings();
+
+	/* No event channels are 'live' right now. */
+	for (i = 0; i < NR_EVENT_CHANNELS; i++)
+		mask_evtchn(i);
+
+	irq_ctx_init(smp_processor_id());
+}

kernel/drivers/xen/evtchn.c

@@ -0,0 +1,506 @@
+/******************************************************************************
+ * evtchn.c
+ *
+ * Driver for receiving and demuxing event-channel signals.
+ *
+ * Copyright (c) 2004-2005, K A Fraser
+ * Multi-process extensions Copyright (c) 2004, Steven Smith
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/major.h>
+#include <linux/proc_fs.h>
+#include <linux/stat.h>
+#include <linux/poll.h>
+#include <linux/irq.h>
+#include <linux/init.h>
+#include <linux/gfp.h>
+#include <linux/mutex.h>
+#include <linux/cpu.h>
+#include <xen/events.h>
+#include <xen/evtchn.h>
+#include <asm/xen/hypervisor.h>
+
+struct per_user_data {
+	struct mutex bind_mutex; /* serialize bind/unbind operations */
+
+	/* Notification ring, accessed via /dev/xen/evtchn. */
+#define EVTCHN_RING_SIZE     (PAGE_SIZE / sizeof(evtchn_port_t))
+#define EVTCHN_RING_MASK(_i) ((_i)&(EVTCHN_RING_SIZE-1))
+	evtchn_port_t *ring;
+	unsigned int ring_cons, ring_prod, ring_overflow;
+	struct mutex ring_cons_mutex; /* protect against concurrent readers */
+
+	/* Processes wait on this queue when ring is empty. */
+	wait_queue_head_t evtchn_wait;
+	struct fasync_struct *evtchn_async_queue;
+	const char *name;
+};
+
+/* Who's bound to each port? */
+static struct per_user_data *port_user[NR_EVENT_CHANNELS];
+static DEFINE_SPINLOCK(port_user_lock); /* protects port_user[] and ring_prod */
+
+irqreturn_t evtchn_interrupt(int irq, void *data)
+{
+	unsigned int port = (unsigned long)data;
+	struct per_user_data *u;
+
+	spin_lock(&port_user_lock);
+
+	u = port_user[port];
+
+	disable_irq_nosync(irq);
+
+	if ((u->ring_prod - u->ring_cons) < EVTCHN_RING_SIZE) {
+		u->ring[EVTCHN_RING_MASK(u->ring_prod)] = port;
+		wmb(); /* Ensure ring contents visible */
+		if (u->ring_cons == u->ring_prod++) {
+			wake_up_interruptible(&u->evtchn_wait);
+			kill_fasync(&u->evtchn_async_queue,
+				    SIGIO, POLL_IN);
+		}
+	} else {
+		u->ring_overflow = 1;
+	}
+
+	spin_unlock(&port_user_lock);
+
+	return IRQ_HANDLED;
+}
+
+static ssize_t evtchn_read(struct file *file, char __user *buf,
+			   size_t count, loff_t *ppos)
+{
+	int rc;
+	unsigned int c, p, bytes1 = 0, bytes2 = 0;
+	struct per_user_data *u = file->private_data;
+
+	/* Whole number of ports. */
+	count &= ~(sizeof(evtchn_port_t)-1);
+
+	if (count == 0)
+		return 0;
+
+	if (count > PAGE_SIZE)
+		count = PAGE_SIZE;
+
+	for (;;) {
+		mutex_lock(&u->ring_cons_mutex);
+
+		rc = -EFBIG;
+		if (u->ring_overflow)
+			goto unlock_out;
+
+		c = u->ring_cons;
+		p = u->ring_prod;
+		if (c != p)
+			break;
+
+		mutex_unlock(&u->ring_cons_mutex);
+
+		if (file->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+
+		rc = wait_event_interruptible(u->evtchn_wait,
+					      u->ring_cons != u->ring_prod);
+		if (rc)
+			return rc;
+	}
+
+	/* Byte lengths of two chunks. Chunk split (if any) is at ring wrap. */
+	if (((c ^ p) & EVTCHN_RING_SIZE) != 0) {
+		bytes1 = (EVTCHN_RING_SIZE - EVTCHN_RING_MASK(c)) *
+			sizeof(evtchn_port_t);
+		bytes2 = EVTCHN_RING_MASK(p) * sizeof(evtchn_port_t);
+	} else {
+		bytes1 = (p - c) * sizeof(evtchn_port_t);
+		bytes2 = 0;
+	}
+
+	/* Truncate chunks according to caller's maximum byte count. */
+	if (bytes1 > count) {
+		bytes1 = count;
+		bytes2 = 0;
+	} else if ((bytes1 + bytes2) > count) {
+		bytes2 = count - bytes1;
+	}
+
+	rc = -EFAULT;
+	rmb(); /* Ensure that we see the port before we copy it. */
+	if (copy_to_user(buf, &u->ring[EVTCHN_RING_MASK(c)], bytes1) ||
+	    ((bytes2 != 0) &&
+	     copy_to_user(&buf[bytes1], &u->ring[0], bytes2)))
+		goto unlock_out;
+
+	u->ring_cons += (bytes1 + bytes2) / sizeof(evtchn_port_t);
+	rc = bytes1 + bytes2;
+
+ unlock_out:
+	mutex_unlock(&u->ring_cons_mutex);
+	return rc;
+}
+
+static ssize_t evtchn_write(struct file *file, const char __user *buf,
+			    size_t count, loff_t *ppos)
+{
+	int rc, i;
+	evtchn_port_t *kbuf = (evtchn_port_t *)__get_free_page(GFP_KERNEL);
+	struct per_user_data *u = file->private_data;
+
+	if (kbuf == NULL)
+		return -ENOMEM;
+
+	/* Whole number of ports. */
+	count &= ~(sizeof(evtchn_port_t)-1);
+
+	rc = 0;
+	if (count == 0)
+		goto out;
+
+	if (count > PAGE_SIZE)
+		count = PAGE_SIZE;
+
+	rc = -EFAULT;
+	if (copy_from_user(kbuf, buf, count) != 0)
+		goto out;
+
+	spin_lock_irq(&port_user_lock);
+	for (i = 0; i < (count/sizeof(evtchn_port_t)); i++)
+		if ((kbuf[i] < NR_EVENT_CHANNELS) && (port_user[kbuf[i]] == u))
+			enable_irq(irq_from_evtchn(kbuf[i]));
+	spin_unlock_irq(&port_user_lock);
+
+	rc = count;
+
+ out:
+	free_page((unsigned long)kbuf);
+	return rc;
+}
+
+static int evtchn_bind_to_user(struct per_user_data *u, int port)
+{
+	int rc = 0;
+
+	/*
+	 * Ports are never reused, so every caller should pass in a
+	 * unique port.
+	 *
+	 * (Locking not necessary because we haven't registered the
+	 * interrupt handler yet, and our caller has already
+	 * serialized bind operations.)
+	 */
+	BUG_ON(port_user[port] != NULL);
+	port_user[port] = u;
+
+	rc = bind_evtchn_to_irqhandler(port, evtchn_interrupt, IRQF_DISABLED,
+				       u->name, (void *)(unsigned long)port);
+	if (rc >= 0)
+		rc = 0;
+
+	return rc;
+}
+
+static void evtchn_unbind_from_user(struct per_user_data *u, int port)
+{
+	int irq = irq_from_evtchn(port);
+
+	unbind_from_irqhandler(irq, (void *)(unsigned long)port);
+
+	/* make sure we unbind the irq handler before clearing the port */
+	barrier();
+
+	port_user[port] = NULL;
+}
+
+static long evtchn_ioctl(struct file *file,
+			 unsigned int cmd, unsigned long arg)
+{
+	int rc;
+	struct per_user_data *u = file->private_data;
+	void __user *uarg = (void __user *) arg;
+
+	/* Prevent bind from racing with unbind */
+	mutex_lock(&u->bind_mutex);
+
+	switch (cmd) {
+	case IOCTL_EVTCHN_BIND_VIRQ: {
+		struct ioctl_evtchn_bind_virq bind;
+		struct evtchn_bind_virq bind_virq;
+
+		rc = -EFAULT;
+		if (copy_from_user(&bind, uarg, sizeof(bind)))
+			break;
+
+		bind_virq.virq = bind.virq;
+		bind_virq.vcpu = 0;
+		rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
+						 &bind_virq);
+		if (rc != 0)
+			break;
+
+		rc = evtchn_bind_to_user(u, bind_virq.port);
+		if (rc == 0)
+			rc = bind_virq.port;
+		break;
+	}
+
+	case IOCTL_EVTCHN_BIND_INTERDOMAIN: {
+		struct ioctl_evtchn_bind_interdomain bind;
+		struct evtchn_bind_interdomain bind_interdomain;
+
+		rc = -EFAULT;
+		if (copy_from_user(&bind, uarg, sizeof(bind)))
+			break;
+
+		bind_interdomain.remote_dom  = bind.remote_domain;
+		bind_interdomain.remote_port = bind.remote_port;
+		rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
+						 &bind_interdomain);
+		if (rc != 0)
+			break;
+
+		rc = evtchn_bind_to_user(u, bind_interdomain.local_port);
+		if (rc == 0)
+			rc = bind_interdomain.local_port;
+		break;
+	}
+
+	case IOCTL_EVTCHN_BIND_UNBOUND_PORT: {
+		struct ioctl_evtchn_bind_unbound_port bind;
+		struct evtchn_alloc_unbound alloc_unbound;
+
+		rc = -EFAULT;
+		if (copy_from_user(&bind, uarg, sizeof(bind)))
+			break;
+
+		alloc_unbound.dom        = DOMID_SELF;
+		alloc_unbound.remote_dom = bind.remote_domain;
+		rc = HYPERVISOR_event_channel_op(EVTCHNOP_alloc_unbound,
+						 &alloc_unbound);
+		if (rc != 0)
+			break;
+
+		rc = evtchn_bind_to_user(u, alloc_unbound.port);
+		if (rc == 0)
+			rc = alloc_unbound.port;
+		break;
+	}
+
+	case IOCTL_EVTCHN_UNBIND: {
+		struct ioctl_evtchn_unbind unbind;
+
+		rc = -EFAULT;
+		if (copy_from_user(&unbind, uarg, sizeof(unbind)))
+			break;
+
+		rc = -EINVAL;
+		if (unbind.port >= NR_EVENT_CHANNELS)
+			break;
+
+		spin_lock_irq(&port_user_lock);
+
+		rc = -ENOTCONN;
+		if (port_user[unbind.port] != u) {
+			spin_unlock_irq(&port_user_lock);
+			break;
+		}
+
+		evtchn_unbind_from_user(u, unbind.port);
+
+		spin_unlock_irq(&port_user_lock);
+
+		rc = 0;
+		break;
+	}
+
+	case IOCTL_EVTCHN_NOTIFY: {
+		struct ioctl_evtchn_notify notify;
+
+		rc = -EFAULT;
+		if (copy_from_user(&notify, uarg, sizeof(notify)))
+			break;
+
+		if (notify.port >= NR_EVENT_CHANNELS) {
+			rc = -EINVAL;
+		} else if (port_user[notify.port] != u) {
+			rc = -ENOTCONN;
+		} else {
+			notify_remote_via_evtchn(notify.port);
+			rc = 0;
+		}
+		break;
+	}
+
+	case IOCTL_EVTCHN_RESET: {
+		/* Initialise the ring to empty. Clear errors. */
+		mutex_lock(&u->ring_cons_mutex);
+		spin_lock_irq(&port_user_lock);
+		u->ring_cons = u->ring_prod = u->ring_overflow = 0;
+		spin_unlock_irq(&port_user_lock);
+		mutex_unlock(&u->ring_cons_mutex);
+		rc = 0;
+		break;
+	}
+
+	default:
+		rc = -ENOSYS;
+		break;
+	}
+	mutex_unlock(&u->bind_mutex);
+
+	return rc;
+}
+
+static unsigned int evtchn_poll(struct file *file, poll_table *wait)
+{
+	unsigned int mask = POLLOUT | POLLWRNORM;
+	struct per_user_data *u = file->private_data;
+
+	poll_wait(file, &u->evtchn_wait, wait);
+	if (u->ring_cons != u->ring_prod)
+		mask |= POLLIN | POLLRDNORM;
+	if (u->ring_overflow)
+		mask = POLLERR;
+	return mask;
+}
+
+static int evtchn_fasync(int fd, struct file *filp, int on)
+{
+	struct per_user_data *u = filp->private_data;
+	return fasync_helper(fd, filp, on, &u->evtchn_async_queue);
+}
+
+static int evtchn_open(struct inode *inode, struct file *filp)
+{
+	struct per_user_data *u;
+
+	u = kzalloc(sizeof(*u), GFP_KERNEL);
+	if (u == NULL)
+		return -ENOMEM;
+
+	u->name = kasprintf(GFP_KERNEL, "evtchn:%s", current->comm);
+	if (u->name == NULL) {
+		kfree(u);
+		return -ENOMEM;
+	}
+
+	init_waitqueue_head(&u->evtchn_wait);
+
+	u->ring = (evtchn_port_t *)__get_free_page(GFP_KERNEL);
+	if (u->ring == NULL) {
+		kfree(u->name);
+		kfree(u);
+		return -ENOMEM;
+	}
+
+	mutex_init(&u->bind_mutex);
+	mutex_init(&u->ring_cons_mutex);
+
+	filp->private_data = u;
+
+	return 0;
+}
+
+static int evtchn_release(struct inode *inode, struct file *filp)
+{
+	int i;
+	struct per_user_data *u = filp->private_data;
+
+	spin_lock_irq(&port_user_lock);
+
+	free_page((unsigned long)u->ring);
+
+	for (i = 0; i < NR_EVENT_CHANNELS; i++) {
+		if (port_user[i] != u)
+			continue;
+
+		evtchn_unbind_from_user(port_user[i], i);
+	}
+
+	spin_unlock_irq(&port_user_lock);
+
+	kfree(u->name);
+	kfree(u);
+
+	return 0;
+}
+
+static const struct file_operations evtchn_fops = {
+	.owner   = THIS_MODULE,
+	.read    = evtchn_read,
+	.write   = evtchn_write,
+	.unlocked_ioctl = evtchn_ioctl,
+	.poll    = evtchn_poll,
+	.fasync  = evtchn_fasync,
+	.open    = evtchn_open,
+	.release = evtchn_release,
+};
+
+static struct miscdevice evtchn_miscdev = {
+	.minor        = MISC_DYNAMIC_MINOR,
+	.name         = "evtchn",
+	.fops         = &evtchn_fops,
+};
+static int __init evtchn_init(void)
+{
+	int err;
+
+	if (!xen_domain())
+		return -ENODEV;
+
+	spin_lock_init(&port_user_lock);
+	memset(port_user, 0, sizeof(port_user));
+
+	/* Create '/dev/misc/evtchn'. */
+	err = misc_register(&evtchn_miscdev);
+	if (err != 0) {
+		printk(KERN_ALERT "Could not register /dev/misc/evtchn\n");
+		return err;
+	}
+
+	printk(KERN_INFO "Event-channel device installed.\n");
+
+	return 0;
+}
+
+static void __exit evtchn_cleanup(void)
+{
+	misc_deregister(&evtchn_miscdev);
+}
+
+module_init(evtchn_init);
+module_exit(evtchn_cleanup);
+
+MODULE_LICENSE("GPL");

kernel/drivers/xen/features.c

@@ -0,0 +1,33 @@
+/******************************************************************************
+ * features.c
+ *
+ * Xen feature flags.
+ *
+ * Copyright (c) 2006, Ian Campbell, XenSource Inc.
+ */
+#include <linux/types.h>
+#include <linux/cache.h>
+#include <linux/module.h>
+
+#include <asm/xen/hypercall.h>
+
+#include <xen/interface/xen.h>
+#include <xen/interface/version.h>
+#include <xen/features.h>
+
+u8 xen_features[XENFEAT_NR_SUBMAPS * 32] __read_mostly;
+EXPORT_SYMBOL_GPL(xen_features);
+
+void xen_setup_features(void)
+{
+	struct xen_feature_info fi;
+	int i, j;
+
+	for (i = 0; i < XENFEAT_NR_SUBMAPS; i++) {
+		fi.submap_idx = i;
+		if (HYPERVISOR_xen_version(XENVER_get_features, &fi) < 0)
+			break;
+		for (j = 0; j < 32; j++)
+			xen_features[i * 32 + j] = !!(fi.submap & 1<<j);
+	}
+}

kernel/drivers/xen/grant-table.c

@@ -0,0 +1,558 @@
+/******************************************************************************
+ * grant_table.c
+ *
+ * Granting foreign access to our memory reservation.
+ *
+ * Copyright (c) 2005-2006, Christopher Clark
+ * Copyright (c) 2004-2005, K A Fraser
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
+#include <linux/uaccess.h>
+
+#include <xen/interface/xen.h>
+#include <xen/page.h>
+#include <xen/grant_table.h>
+#include <asm/xen/hypercall.h>
+
+#include <asm/pgtable.h>
+#include <asm/sync_bitops.h>
+
+
+/* External tools reserve first few grant table entries. */
+#define NR_RESERVED_ENTRIES 8
+#define GNTTAB_LIST_END 0xffffffff
+#define GREFS_PER_GRANT_FRAME (PAGE_SIZE / sizeof(struct grant_entry))
+
+static grant_ref_t **gnttab_list;
+static unsigned int nr_grant_frames;
+static unsigned int boot_max_nr_grant_frames;
+static int gnttab_free_count;
+static grant_ref_t gnttab_free_head;
+static DEFINE_SPINLOCK(gnttab_list_lock);
+
+static struct grant_entry *shared;
+
+static struct gnttab_free_callback *gnttab_free_callback_list;
+
+static int gnttab_expand(unsigned int req_entries);
+
+#define RPP (PAGE_SIZE / sizeof(grant_ref_t))
+
+static inline grant_ref_t *__gnttab_entry(grant_ref_t entry)
+{
+	return &gnttab_list[(entry) / RPP][(entry) % RPP];
+}
+/* This can be used as an l-value */
+#define gnttab_entry(entry) (*__gnttab_entry(entry))
+
+static int get_free_entries(unsigned count)
+{
+	unsigned long flags;
+	int ref, rc;
+	grant_ref_t head;
+
+	spin_lock_irqsave(&gnttab_list_lock, flags);
+
+	if ((gnttab_free_count < count) &&
+	    ((rc = gnttab_expand(count - gnttab_free_count)) < 0)) {
+		spin_unlock_irqrestore(&gnttab_list_lock, flags);
+		return rc;
+	}
+
+	ref = head = gnttab_free_head;
+	gnttab_free_count -= count;
+	while (count-- > 1)
+		head = gnttab_entry(head);
+	gnttab_free_head = gnttab_entry(head);
+	gnttab_entry(head) = GNTTAB_LIST_END;
+
+	spin_unlock_irqrestore(&gnttab_list_lock, flags);
+
+	return ref;
+}
+
+static void do_free_callbacks(void)
+{
+	struct gnttab_free_callback *callback, *next;
+
+	callback = gnttab_free_callback_list;
+	gnttab_free_callback_list = NULL;
+
+	while (callback != NULL) {
+		next = callback->next;
+		if (gnttab_free_count >= callback->count) {
+			callback->next = NULL;
+			callback->fn(callback->arg);
+		} else {
+			callback->next = gnttab_free_callback_list;
+			gnttab_free_callback_list = callback;
+		}
+		callback = next;
+	}
+}
+
+static inline void check_free_callbacks(void)
+{
+	if (unlikely(gnttab_free_callback_list))
+		do_free_callbacks();
+}
+
+static void put_free_entry(grant_ref_t ref)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&gnttab_list_lock, flags);
+	gnttab_entry(ref) = gnttab_free_head;
+	gnttab_free_head = ref;
+	gnttab_free_count++;
+	check_free_callbacks();
+	spin_unlock_irqrestore(&gnttab_list_lock, flags);
+}
+
+static void update_grant_entry(grant_ref_t ref, domid_t domid,
+			       unsigned long frame, unsigned flags)
+{
+	/*
+	 * Introducing a valid entry into the grant table:
+	 *  1. Write ent->domid.
+	 *  2. Write ent->frame:
+	 *      GTF_permit_access:   Frame to which access is permitted.
+	 *      GTF_accept_transfer: Pseudo-phys frame slot being filled by new
+	 *                           frame, or zero if none.
+	 *  3. Write memory barrier (WMB).
+	 *  4. Write ent->flags, inc. valid type.
+	 */
+	shared[ref].frame = frame;
+	shared[ref].domid = domid;
+	wmb();
+	shared[ref].flags = flags;
+}
+
+/*
+ * Public grant-issuing interface functions
+ */
+void gnttab_grant_foreign_access_ref(grant_ref_t ref, domid_t domid,
+				     unsigned long frame, int readonly)
+{
+	update_grant_entry(ref, domid, frame,
+			   GTF_permit_access | (readonly ? GTF_readonly : 0));
+}
+EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access_ref);
+
+int gnttab_grant_foreign_access(domid_t domid, unsigned long frame,
+				int readonly)
+{
+	int ref;
+
+	ref = get_free_entries(1);
+	if (unlikely(ref < 0))
+		return -ENOSPC;
+
+	gnttab_grant_foreign_access_ref(ref, domid, frame, readonly);
+
+	return ref;
+}
+EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access);
+
+int gnttab_query_foreign_access(grant_ref_t ref)
+{
+	u16 nflags;
+
+	nflags = shared[ref].flags;
+
+	return (nflags & (GTF_reading|GTF_writing));
+}
+EXPORT_SYMBOL_GPL(gnttab_query_foreign_access);
+
+int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly)
+{
+	u16 flags, nflags;
+
+	nflags = shared[ref].flags;
+	do {
+		flags = nflags;
+		if (flags & (GTF_reading|GTF_writing)) {
+			printk(KERN_ALERT "WARNING: g.e. still in use!\n");
+			return 0;
+		}
+	} while ((nflags = sync_cmpxchg(&shared[ref].flags, flags, 0)) != flags);
+
+	return 1;
+}
+EXPORT_SYMBOL_GPL(gnttab_end_foreign_access_ref);
+
+void gnttab_end_foreign_access(grant_ref_t ref, int readonly,
+			       unsigned long page)
+{
+	if (gnttab_end_foreign_access_ref(ref, readonly)) {
+		put_free_entry(ref);
+		if (page != 0)
+			free_page(page);
+	} else {
+		/* XXX This needs to be fixed so that the ref and page are
+		   placed on a list to be freed up later. */
+		printk(KERN_WARNING
+		       "WARNING: leaking g.e. and page still in use!\n");
+	}
+}
+EXPORT_SYMBOL_GPL(gnttab_end_foreign_access);
+
+int gnttab_grant_foreign_transfer(domid_t domid, unsigned long pfn)
+{
+	int ref;
+
+	ref = get_free_entries(1);
+	if (unlikely(ref < 0))
+		return -ENOSPC;
+	gnttab_grant_foreign_transfer_ref(ref, domid, pfn);
+
+	return ref;
+}
+EXPORT_SYMBOL_GPL(gnttab_grant_foreign_transfer);
+
+void gnttab_grant_foreign_transfer_ref(grant_ref_t ref, domid_t domid,
+				       unsigned long pfn)
+{
+	update_grant_entry(ref, domid, pfn, GTF_accept_transfer);
+}
+EXPORT_SYMBOL_GPL(gnttab_grant_foreign_transfer_ref);
+
+unsigned long gnttab_end_foreign_transfer_ref(grant_ref_t ref)
+{
+	unsigned long frame;
+	u16           flags;
+
+	/*
+	 * If a transfer is not even yet started, try to reclaim the grant
+	 * reference and return failure (== 0).
+	 */
+	while (!((flags = shared[ref].flags) & GTF_transfer_committed)) {
+		if (sync_cmpxchg(&shared[ref].flags, flags, 0) == flags)
+			return 0;
+		cpu_relax();
+	}
+
+	/* If a transfer is in progress then wait until it is completed. */
+	while (!(flags & GTF_transfer_completed)) {
+		flags = shared[ref].flags;
+		cpu_relax();
+	}
+
+	rmb();	/* Read the frame number /after/ reading completion status. */
+	frame = shared[ref].frame;
+	BUG_ON(frame == 0);
+
+	return frame;
+}
+EXPORT_SYMBOL_GPL(gnttab_end_foreign_transfer_ref);
+
+unsigned long gnttab_end_foreign_transfer(grant_ref_t ref)
+{
+	unsigned long frame = gnttab_end_foreign_transfer_ref(ref);
+	put_free_entry(ref);
+	return frame;
+}
+EXPORT_SYMBOL_GPL(gnttab_end_foreign_transfer);
+
+void gnttab_free_grant_reference(grant_ref_t ref)
+{
+	put_free_entry(ref);
+}
+EXPORT_SYMBOL_GPL(gnttab_free_grant_reference);
+
+void gnttab_free_grant_references(grant_ref_t head)
+{
+	grant_ref_t ref;
+	unsigned long flags;
+	int count = 1;
+	if (head == GNTTAB_LIST_END)
+		return;
+	spin_lock_irqsave(&gnttab_list_lock, flags);
+	ref = head;
+	while (gnttab_entry(ref) != GNTTAB_LIST_END) {
+		ref = gnttab_entry(ref);
+		count++;
+	}
+	gnttab_entry(ref) = gnttab_free_head;
+	gnttab_free_head = head;
+	gnttab_free_count += count;
+	check_free_callbacks();
+	spin_unlock_irqrestore(&gnttab_list_lock, flags);
+}
+EXPORT_SYMBOL_GPL(gnttab_free_grant_references);
+
+int gnttab_alloc_grant_references(u16 count, grant_ref_t *head)
+{
+	int h = get_free_entries(count);
+
+	if (h < 0)
+		return -ENOSPC;
+
+	*head = h;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gnttab_alloc_grant_references);
+
+int gnttab_empty_grant_references(const grant_ref_t *private_head)
+{
+	return (*private_head == GNTTAB_LIST_END);
+}
+EXPORT_SYMBOL_GPL(gnttab_empty_grant_references);
+
+int gnttab_claim_grant_reference(grant_ref_t *private_head)
+{
+	grant_ref_t g = *private_head;
+	if (unlikely(g == GNTTAB_LIST_END))
+		return -ENOSPC;
+	*private_head = gnttab_entry(g);
+	return g;
+}
+EXPORT_SYMBOL_GPL(gnttab_claim_grant_reference);
+
+void gnttab_release_grant_reference(grant_ref_t *private_head,
+				    grant_ref_t release)
+{
+	gnttab_entry(release) = *private_head;
+	*private_head = release;
+}
+EXPORT_SYMBOL_GPL(gnttab_release_grant_reference);
+
+void gnttab_request_free_callback(struct gnttab_free_callback *callback,
+				  void (*fn)(void *), void *arg, u16 count)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&gnttab_list_lock, flags);
+	if (callback->next)
+		goto out;
+	callback->fn = fn;
+	callback->arg = arg;
+	callback->count = count;
+	callback->next = gnttab_free_callback_list;
+	gnttab_free_callback_list = callback;
+	check_free_callbacks();
+out:
+	spin_unlock_irqrestore(&gnttab_list_lock, flags);
+}
+EXPORT_SYMBOL_GPL(gnttab_request_free_callback);
+
+void gnttab_cancel_free_callback(struct gnttab_free_callback *callback)
+{
+	struct gnttab_free_callback **pcb;
+	unsigned long flags;
+
+	spin_lock_irqsave(&gnttab_list_lock, flags);
+	for (pcb = &gnttab_free_callback_list; *pcb; pcb = &(*pcb)->next) {
+		if (*pcb == callback) {
+			*pcb = callback->next;
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&gnttab_list_lock, flags);
+}
+EXPORT_SYMBOL_GPL(gnttab_cancel_free_callback);
+
+static int grow_gnttab_list(unsigned int more_frames)
+{
+	unsigned int new_nr_grant_frames, extra_entries, i;
+	unsigned int nr_glist_frames, new_nr_glist_frames;
+
+	new_nr_grant_frames = nr_grant_frames + more_frames;
+	extra_entries       = more_frames * GREFS_PER_GRANT_FRAME;
+
+	nr_glist_frames = (nr_grant_frames * GREFS_PER_GRANT_FRAME + RPP - 1) / RPP;
+	new_nr_glist_frames =
+		(new_nr_grant_frames * GREFS_PER_GRANT_FRAME + RPP - 1) / RPP;
+	for (i = nr_glist_frames; i < new_nr_glist_frames; i++) {
+		gnttab_list[i] = (grant_ref_t *)__get_free_page(GFP_ATOMIC);
+		if (!gnttab_list[i])
+			goto grow_nomem;
+	}
+
+
+	for (i = GREFS_PER_GRANT_FRAME * nr_grant_frames;
+	     i < GREFS_PER_GRANT_FRAME * new_nr_grant_frames - 1; i++)
+		gnttab_entry(i) = i + 1;
+
+	gnttab_entry(i) = gnttab_free_head;
+	gnttab_free_head = GREFS_PER_GRANT_FRAME * nr_grant_frames;
+	gnttab_free_count += extra_entries;
+
+	nr_grant_frames = new_nr_grant_frames;
+
+	check_free_callbacks();
+
+	return 0;
+
+grow_nomem:
+	for ( ; i >= nr_glist_frames; i--)
+		free_page((unsigned long) gnttab_list[i]);
+	return -ENOMEM;
+}
+
+static unsigned int __max_nr_grant_frames(void)
+{
+	struct gnttab_query_size query;
+	int rc;
+
+	query.dom = DOMID_SELF;
+
+	rc = HYPERVISOR_grant_table_op(GNTTABOP_query_size, &query, 1);
+	if ((rc < 0) || (query.status != GNTST_okay))
+		return 4; /* Legacy max supported number of frames */
+
+	return query.max_nr_frames;
+}
+
+static inline unsigned int max_nr_grant_frames(void)
+{
+	unsigned int xen_max = __max_nr_grant_frames();
+
+	if (xen_max > boot_max_nr_grant_frames)
+		return boot_max_nr_grant_frames;
+	return xen_max;
+}
+
+static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
+{
+	struct gnttab_setup_table setup;
+	unsigned long *frames;
+	unsigned int nr_gframes = end_idx + 1;
+	int rc;
+
+	frames = kmalloc(nr_gframes * sizeof(unsigned long), GFP_ATOMIC);
+	if (!frames)
+		return -ENOMEM;
+
+	setup.dom        = DOMID_SELF;
+	setup.nr_frames  = nr_gframes;
+	set_xen_guest_handle(setup.frame_list, frames);
+
+	rc = HYPERVISOR_grant_table_op(GNTTABOP_setup_table, &setup, 1);
+	if (rc == -ENOSYS) {
+		kfree(frames);
+		return -ENOSYS;
+	}
+
+	BUG_ON(rc || setup.status);
+
+	rc = arch_gnttab_map_shared(frames, nr_gframes, max_nr_grant_frames(),
+				    &shared);
+	BUG_ON(rc);
+
+	kfree(frames);
+
+	return 0;
+}
+
+int gnttab_resume(void)
+{
+	if (max_nr_grant_frames() < nr_grant_frames)
+		return -ENOSYS;
+	return gnttab_map(0, nr_grant_frames - 1);
+}
+
+int gnttab_suspend(void)
+{
+	arch_gnttab_unmap_shared(shared, nr_grant_frames);
+	return 0;
+}
+
+static int gnttab_expand(unsigned int req_entries)
+{
+	int rc;
+	unsigned int cur, extra;
+
+	cur = nr_grant_frames;
+	extra = ((req_entries + (GREFS_PER_GRANT_FRAME-1)) /
+		 GREFS_PER_GRANT_FRAME);
+	if (cur + extra > max_nr_grant_frames())
+		return -ENOSPC;
+
+	rc = gnttab_map(cur, cur + extra - 1);
+	if (rc == 0)
+		rc = grow_gnttab_list(extra);
+
+	return rc;
+}
+
+static int __devinit gnttab_init(void)
+{
+	int i;
+	unsigned int max_nr_glist_frames, nr_glist_frames;
+	unsigned int nr_init_grefs;
+
+	if (!xen_domain())
+		return -ENODEV;
+
+	nr_grant_frames = 1;
+	boot_max_nr_grant_frames = __max_nr_grant_frames();
+
+	/* Determine the maximum number of frames required for the
+	 * grant reference free list on the current hypervisor.
+	 */
+	max_nr_glist_frames = (boot_max_nr_grant_frames *
+			       GREFS_PER_GRANT_FRAME / RPP);
+
+	gnttab_list = kmalloc(max_nr_glist_frames * sizeof(grant_ref_t *),
+			      GFP_KERNEL);
+	if (gnttab_list == NULL)
+		return -ENOMEM;
+
+	nr_glist_frames = (nr_grant_frames * GREFS_PER_GRANT_FRAME + RPP - 1) / RPP;
+	for (i = 0; i < nr_glist_frames; i++) {
+		gnttab_list[i] = (grant_ref_t *)__get_free_page(GFP_KERNEL);
+		if (gnttab_list[i] == NULL)
+			goto ini_nomem;
+	}
+
+	if (gnttab_resume() < 0)
+		return -ENODEV;
+
+	nr_init_grefs = nr_grant_frames * GREFS_PER_GRANT_FRAME;
+
+	for (i = NR_RESERVED_ENTRIES; i < nr_init_grefs - 1; i++)
+		gnttab_entry(i) = i + 1;
+
+	gnttab_entry(nr_init_grefs - 1) = GNTTAB_LIST_END;
+	gnttab_free_count = nr_init_grefs - NR_RESERVED_ENTRIES;
+	gnttab_free_head  = NR_RESERVED_ENTRIES;
+
+	printk("Grant table initialized\n");
+	return 0;
+
+ ini_nomem:
+	for (i--; i >= 0; i--)
+		free_page((unsigned long)gnttab_list[i]);
+	kfree(gnttab_list);
+	return -ENOMEM;
+}
+
+core_initcall(gnttab_init);

kernel/drivers/xen/manage.c

@@ -0,0 +1,273 @@
+/*
+ * Handle extern requests for shutdown, reboot and sysrq
+ */
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/reboot.h>
+#include <linux/sysrq.h>
+#include <linux/stop_machine.h>
+#include <linux/freezer.h>
+
+#include <xen/xenbus.h>
+#include <xen/grant_table.h>
+#include <xen/events.h>
+#include <xen/hvc-console.h>
+#include <xen/xen-ops.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+enum shutdown_state {
+	SHUTDOWN_INVALID = -1,
+	SHUTDOWN_POWEROFF = 0,
+	SHUTDOWN_SUSPEND = 2,
+	/* Code 3 is SHUTDOWN_CRASH, which we don't use because the domain can only
+	   report a crash, not be instructed to crash!
+	   HALT is the same as POWEROFF, as far as we're concerned.  The tools use
+	   the distinction when we return the reason code to them.  */
+	 SHUTDOWN_HALT = 4,
+};
+
+/* Ignore multiple shutdown requests. */
+static enum shutdown_state shutting_down = SHUTDOWN_INVALID;
+
+#ifdef CONFIG_PM_SLEEP
+static int xen_suspend(void *data)
+{
+	int *cancelled = data;
+	int err;
+
+	BUG_ON(!irqs_disabled());
+
+	err = sysdev_suspend(PMSG_SUSPEND);
+	if (err) {
+		printk(KERN_ERR "xen_suspend: sysdev_suspend failed: %d\n",
+			err);
+		return err;
+	}
+
+	xen_mm_pin_all();
+	gnttab_suspend();
+	xen_pre_suspend();
+
+	/*
+	 * This hypercall returns 1 if suspend was cancelled
+	 * or the domain was merely checkpointed, and 0 if it
+	 * is resuming in a new domain.
+	 */
+	*cancelled = HYPERVISOR_suspend(virt_to_mfn(xen_start_info));
+
+	xen_post_suspend(*cancelled);
+	gnttab_resume();
+	xen_mm_unpin_all();
+
+	if (!*cancelled) {
+		xen_irq_resume();
+		xen_console_resume();
+		xen_timer_resume();
+	}
+
+	sysdev_resume();
+
+	return 0;
+}
+
+static void do_suspend(void)
+{
+	int err;
+	int cancelled = 1;
+
+	shutting_down = SHUTDOWN_SUSPEND;
+
+	err = stop_machine_create();
+	if (err) {
+		printk(KERN_ERR "xen suspend: failed to setup stop_machine %d\n", err);
+		goto out;
+	}
+
+#ifdef CONFIG_PREEMPT
+	/* If the kernel is preemptible, we need to freeze all the processes
+	   to prevent them from being in the middle of a pagetable update
+	   during suspend. */
+	err = freeze_processes();
+	if (err) {
+		printk(KERN_ERR "xen suspend: freeze failed %d\n", err);
+		goto out_destroy_sm;
+	}
+#endif
+
+	err = dpm_suspend_start(PMSG_SUSPEND);
+	if (err) {
+		printk(KERN_ERR "xen suspend: dpm_suspend_start %d\n", err);
+		goto out_thaw;
+	}
+
+	printk(KERN_DEBUG "suspending xenstore...\n");
+	xs_suspend();
+
+	err = dpm_suspend_noirq(PMSG_SUSPEND);
+	if (err) {
+		printk(KERN_ERR "dpm_suspend_noirq failed: %d\n", err);
+		goto out_resume;
+	}
+
+	err = stop_machine(xen_suspend, &cancelled, cpumask_of(0));
+
+	dpm_resume_noirq(PMSG_RESUME);
+
+	if (err) {
+		printk(KERN_ERR "failed to start xen_suspend: %d\n", err);
+		cancelled = 1;
+	}
+
+out_resume:
+	if (!cancelled) {
+		xen_arch_resume();
+		xs_resume();
+	} else
+		xs_suspend_cancel();
+
+	dpm_resume_end(PMSG_RESUME);
+
+	/* Make sure timer events get retriggered on all CPUs */
+	clock_was_set();
+
+out_thaw:
+#ifdef CONFIG_PREEMPT
+	thaw_processes();
+
+out_destroy_sm:
+#endif
+	stop_machine_destroy();
+
+out:
+	shutting_down = SHUTDOWN_INVALID;
+}
+#endif	/* CONFIG_PM_SLEEP */
+
+static void shutdown_handler(struct xenbus_watch *watch,
+			     const char **vec, unsigned int len)
+{
+	char *str;
+	struct xenbus_transaction xbt;
+	int err;
+
+	if (shutting_down != SHUTDOWN_INVALID)
+		return;
+
+ again:
+	err = xenbus_transaction_start(&xbt);
+	if (err)
+		return;
+
+	str = (char *)xenbus_read(xbt, "control", "shutdown", NULL);
+	/* Ignore read errors and empty reads. */
+	if (XENBUS_IS_ERR_READ(str)) {
+		xenbus_transaction_end(xbt, 1);
+		return;
+	}
+
+	xenbus_write(xbt, "control", "shutdown", "");
+
+	err = xenbus_transaction_end(xbt, 0);
+	if (err == -EAGAIN) {
+		kfree(str);
+		goto again;
+	}
+
+	if (strcmp(str, "poweroff") == 0 ||
+	    strcmp(str, "halt") == 0) {
+		shutting_down = SHUTDOWN_POWEROFF;
+		orderly_poweroff(false);
+	} else if (strcmp(str, "reboot") == 0) {
+		shutting_down = SHUTDOWN_POWEROFF; /* ? */
+		ctrl_alt_del();
+#ifdef CONFIG_PM_SLEEP
+	} else if (strcmp(str, "suspend") == 0) {
+		do_suspend();
+#endif
+	} else {
+		printk(KERN_INFO "Ignoring shutdown request: %s\n", str);
+		shutting_down = SHUTDOWN_INVALID;
+	}
+
+	kfree(str);
+}
+
+static void sysrq_handler(struct xenbus_watch *watch, const char **vec,
+			  unsigned int len)
+{
+	char sysrq_key = '\0';
+	struct xenbus_transaction xbt;
+	int err;
+
+ again:
+	err = xenbus_transaction_start(&xbt);
+	if (err)
+		return;
+	if (!xenbus_scanf(xbt, "control", "sysrq", "%c", &sysrq_key)) {
+		printk(KERN_ERR "Unable to read sysrq code in "
+		       "control/sysrq\n");
+		xenbus_transaction_end(xbt, 1);
+		return;
+	}
+
+	if (sysrq_key != '\0')
+		xenbus_printf(xbt, "control", "sysrq", "%c", '\0');
+
+	err = xenbus_transaction_end(xbt, 0);
+	if (err == -EAGAIN)
+		goto again;
+
+	if (sysrq_key != '\0')
+		handle_sysrq(sysrq_key, NULL);
+}
+
+static struct xenbus_watch shutdown_watch = {
+	.node = "control/shutdown",
+	.callback = shutdown_handler
+};
+
+static struct xenbus_watch sysrq_watch = {
+	.node = "control/sysrq",
+	.callback = sysrq_handler
+};
+
+static int setup_shutdown_watcher(void)
+{
+	int err;
+
+	err = register_xenbus_watch(&shutdown_watch);
+	if (err) {
+		printk(KERN_ERR "Failed to set shutdown watcher\n");
+		return err;
+	}
+
+	err = register_xenbus_watch(&sysrq_watch);
+	if (err) {
+		printk(KERN_ERR "Failed to set sysrq watcher\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int shutdown_event(struct notifier_block *notifier,
+			  unsigned long event,
+			  void *data)
+{
+	setup_shutdown_watcher();
+	return NOTIFY_DONE;
+}
+
+static int __init setup_shutdown_event(void)
+{
+	static struct notifier_block xenstore_notifier = {
+		.notifier_call = shutdown_event
+	};
+	register_xenstore_notifier(&xenstore_notifier);
+
+	return 0;
+}
+
+subsys_initcall(setup_shutdown_event);

kernel/drivers/xen/sys-hypervisor.c

@@ -0,0 +1,445 @@
+/*
+ *  copyright (c) 2006 IBM Corporation
+ *  Authored by: Mike D. Day <ncmike@us.ibm.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.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/kobject.h>
+
+#include <asm/xen/hypervisor.h>
+#include <asm/xen/hypercall.h>
+
+#include <xen/xenbus.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/version.h>
+
+#define HYPERVISOR_ATTR_RO(_name) \
+static struct hyp_sysfs_attr  _name##_attr = __ATTR_RO(_name)
+
+#define HYPERVISOR_ATTR_RW(_name) \
+static struct hyp_sysfs_attr _name##_attr = \
+	__ATTR(_name, 0644, _name##_show, _name##_store)
+
+struct hyp_sysfs_attr {
+	struct attribute attr;
+	ssize_t (*show)(struct hyp_sysfs_attr *, char *);
+	ssize_t (*store)(struct hyp_sysfs_attr *, const char *, size_t);
+	void *hyp_attr_data;
+};
+
+static ssize_t type_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	return sprintf(buffer, "xen\n");
+}
+
+HYPERVISOR_ATTR_RO(type);
+
+static int __init xen_sysfs_type_init(void)
+{
+	return sysfs_create_file(hypervisor_kobj, &type_attr.attr);
+}
+
+static void xen_sysfs_type_destroy(void)
+{
+	sysfs_remove_file(hypervisor_kobj, &type_attr.attr);
+}
+
+/* xen version attributes */
+static ssize_t major_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int version = HYPERVISOR_xen_version(XENVER_version, NULL);
+	if (version)
+		return sprintf(buffer, "%d\n", version >> 16);
+	return -ENODEV;
+}
+
+HYPERVISOR_ATTR_RO(major);
+
+static ssize_t minor_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int version = HYPERVISOR_xen_version(XENVER_version, NULL);
+	if (version)
+		return sprintf(buffer, "%d\n", version & 0xff);
+	return -ENODEV;
+}
+
+HYPERVISOR_ATTR_RO(minor);
+
+static ssize_t extra_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret = -ENOMEM;
+	char *extra;
+
+	extra = kmalloc(XEN_EXTRAVERSION_LEN, GFP_KERNEL);
+	if (extra) {
+		ret = HYPERVISOR_xen_version(XENVER_extraversion, extra);
+		if (!ret)
+			ret = sprintf(buffer, "%s\n", extra);
+		kfree(extra);
+	}
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(extra);
+
+static struct attribute *version_attrs[] = {
+	&major_attr.attr,
+	&minor_attr.attr,
+	&extra_attr.attr,
+	NULL
+};
+
+static struct attribute_group version_group = {
+	.name = "version",
+	.attrs = version_attrs,
+};
+
+static int __init xen_sysfs_version_init(void)
+{
+	return sysfs_create_group(hypervisor_kobj, &version_group);
+}
+
+static void xen_sysfs_version_destroy(void)
+{
+	sysfs_remove_group(hypervisor_kobj, &version_group);
+}
+
+/* UUID */
+
+static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	char *vm, *val;
+	int ret;
+	extern int xenstored_ready;
+
+	if (!xenstored_ready)
+		return -EBUSY;
+
+	vm = xenbus_read(XBT_NIL, "vm", "", NULL);
+	if (IS_ERR(vm))
+		return PTR_ERR(vm);
+	val = xenbus_read(XBT_NIL, vm, "uuid", NULL);
+	kfree(vm);
+	if (IS_ERR(val))
+		return PTR_ERR(val);
+	ret = sprintf(buffer, "%s\n", val);
+	kfree(val);
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(uuid);
+
+static int __init xen_sysfs_uuid_init(void)
+{
+	return sysfs_create_file(hypervisor_kobj, &uuid_attr.attr);
+}
+
+static void xen_sysfs_uuid_destroy(void)
+{
+	sysfs_remove_file(hypervisor_kobj, &uuid_attr.attr);
+}
+
+/* xen compilation attributes */
+
+static ssize_t compiler_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret = -ENOMEM;
+	struct xen_compile_info *info;
+
+	info = kmalloc(sizeof(struct xen_compile_info), GFP_KERNEL);
+	if (info) {
+		ret = HYPERVISOR_xen_version(XENVER_compile_info, info);
+		if (!ret)
+			ret = sprintf(buffer, "%s\n", info->compiler);
+		kfree(info);
+	}
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(compiler);
+
+static ssize_t compiled_by_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret = -ENOMEM;
+	struct xen_compile_info *info;
+
+	info = kmalloc(sizeof(struct xen_compile_info), GFP_KERNEL);
+	if (info) {
+		ret = HYPERVISOR_xen_version(XENVER_compile_info, info);
+		if (!ret)
+			ret = sprintf(buffer, "%s\n", info->compile_by);
+		kfree(info);
+	}
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(compiled_by);
+
+static ssize_t compile_date_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret = -ENOMEM;
+	struct xen_compile_info *info;
+
+	info = kmalloc(sizeof(struct xen_compile_info), GFP_KERNEL);
+	if (info) {
+		ret = HYPERVISOR_xen_version(XENVER_compile_info, info);
+		if (!ret)
+			ret = sprintf(buffer, "%s\n", info->compile_date);
+		kfree(info);
+	}
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(compile_date);
+
+static struct attribute *xen_compile_attrs[] = {
+	&compiler_attr.attr,
+	&compiled_by_attr.attr,
+	&compile_date_attr.attr,
+	NULL
+};
+
+static struct attribute_group xen_compilation_group = {
+	.name = "compilation",
+	.attrs = xen_compile_attrs,
+};
+
+int __init static xen_compilation_init(void)
+{
+	return sysfs_create_group(hypervisor_kobj, &xen_compilation_group);
+}
+
+static void xen_compilation_destroy(void)
+{
+	sysfs_remove_group(hypervisor_kobj, &xen_compilation_group);
+}
+
+/* xen properties info */
+
+static ssize_t capabilities_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret = -ENOMEM;
+	char *caps;
+
+	caps = kmalloc(XEN_CAPABILITIES_INFO_LEN, GFP_KERNEL);
+	if (caps) {
+		ret = HYPERVISOR_xen_version(XENVER_capabilities, caps);
+		if (!ret)
+			ret = sprintf(buffer, "%s\n", caps);
+		kfree(caps);
+	}
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(capabilities);
+
+static ssize_t changeset_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret = -ENOMEM;
+	char *cset;
+
+	cset = kmalloc(XEN_CHANGESET_INFO_LEN, GFP_KERNEL);
+	if (cset) {
+		ret = HYPERVISOR_xen_version(XENVER_changeset, cset);
+		if (!ret)
+			ret = sprintf(buffer, "%s\n", cset);
+		kfree(cset);
+	}
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(changeset);
+
+static ssize_t virtual_start_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret = -ENOMEM;
+	struct xen_platform_parameters *parms;
+
+	parms = kmalloc(sizeof(struct xen_platform_parameters), GFP_KERNEL);
+	if (parms) {
+		ret = HYPERVISOR_xen_version(XENVER_platform_parameters,
+					     parms);
+		if (!ret)
+			ret = sprintf(buffer, "%lx\n", parms->virt_start);
+		kfree(parms);
+	}
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(virtual_start);
+
+static ssize_t pagesize_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	int ret;
+
+	ret = HYPERVISOR_xen_version(XENVER_pagesize, NULL);
+	if (ret > 0)
+		ret = sprintf(buffer, "%x\n", ret);
+
+	return ret;
+}
+
+HYPERVISOR_ATTR_RO(pagesize);
+
+static ssize_t xen_feature_show(int index, char *buffer)
+{
+	ssize_t ret;
+	struct xen_feature_info info;
+
+	info.submap_idx = index;
+	ret = HYPERVISOR_xen_version(XENVER_get_features, &info);
+	if (!ret)
+		ret = sprintf(buffer, "%08x", info.submap);
+
+	return ret;
+}
+
+static ssize_t features_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	ssize_t len;
+	int i;
+
+	len = 0;
+	for (i = XENFEAT_NR_SUBMAPS-1; i >= 0; i--) {
+		int ret = xen_feature_show(i, buffer + len);
+		if (ret < 0) {
+			if (len == 0)
+				len = ret;
+			break;
+		}
+		len += ret;
+	}
+	if (len > 0)
+		buffer[len++] = '\n';
+
+	return len;
+}
+
+HYPERVISOR_ATTR_RO(features);
+
+static struct attribute *xen_properties_attrs[] = {
+	&capabilities_attr.attr,
+	&changeset_attr.attr,
+	&virtual_start_attr.attr,
+	&pagesize_attr.attr,
+	&features_attr.attr,
+	NULL
+};
+
+static struct attribute_group xen_properties_group = {
+	.name = "properties",
+	.attrs = xen_properties_attrs,
+};
+
+static int __init xen_properties_init(void)
+{
+	return sysfs_create_group(hypervisor_kobj, &xen_properties_group);
+}
+
+static void xen_properties_destroy(void)
+{
+	sysfs_remove_group(hypervisor_kobj, &xen_properties_group);
+}
+
+static int __init hyper_sysfs_init(void)
+{
+	int ret;
+
+	if (!xen_domain())
+		return -ENODEV;
+
+	ret = xen_sysfs_type_init();
+	if (ret)
+		goto out;
+	ret = xen_sysfs_version_init();
+	if (ret)
+		goto version_out;
+	ret = xen_compilation_init();
+	if (ret)
+		goto comp_out;
+	ret = xen_sysfs_uuid_init();
+	if (ret)
+		goto uuid_out;
+	ret = xen_properties_init();
+	if (ret)
+		goto prop_out;
+
+	goto out;
+
+prop_out:
+	xen_sysfs_uuid_destroy();
+uuid_out:
+	xen_compilation_destroy();
+comp_out:
+	xen_sysfs_version_destroy();
+version_out:
+	xen_sysfs_type_destroy();
+out:
+	return ret;
+}
+
+static void __exit hyper_sysfs_exit(void)
+{
+	xen_properties_destroy();
+	xen_compilation_destroy();
+	xen_sysfs_uuid_destroy();
+	xen_sysfs_version_destroy();
+	xen_sysfs_type_destroy();
+
+}
+module_init(hyper_sysfs_init);
+module_exit(hyper_sysfs_exit);
+
+static ssize_t hyp_sysfs_show(struct kobject *kobj,
+			      struct attribute *attr,
+			      char *buffer)
+{
+	struct hyp_sysfs_attr *hyp_attr;
+	hyp_attr = container_of(attr, struct hyp_sysfs_attr, attr);
+	if (hyp_attr->show)
+		return hyp_attr->show(hyp_attr, buffer);
+	return 0;
+}
+
+static ssize_t hyp_sysfs_store(struct kobject *kobj,
+			       struct attribute *attr,
+			       const char *buffer,
+			       size_t len)
+{
+	struct hyp_sysfs_attr *hyp_attr;
+	hyp_attr = container_of(attr, struct hyp_sysfs_attr, attr);
+	if (hyp_attr->store)
+		return hyp_attr->store(hyp_attr, buffer, len);
+	return 0;
+}
+
+static struct sysfs_ops hyp_sysfs_ops = {
+	.show = hyp_sysfs_show,
+	.store = hyp_sysfs_store,
+};
+
+static struct kobj_type hyp_sysfs_kobj_type = {
+	.sysfs_ops = &hyp_sysfs_ops,
+};
+
+static int __init hypervisor_subsys_init(void)
+{
+	if (!xen_domain())
+		return -ENODEV;
+
+	hypervisor_kobj->ktype = &hyp_sysfs_kobj_type;
+	return 0;
+}
+device_initcall(hypervisor_subsys_init);

kernel/drivers/xen/xenbus/Makefile

@@ -0,0 +1,7 @@
+obj-y	+= xenbus.o
+
+xenbus-objs =
+xenbus-objs += xenbus_client.o
+xenbus-objs += xenbus_comms.o
+xenbus-objs += xenbus_xs.o
+xenbus-objs += xenbus_probe.o

kernel/drivers/xen/xenbus/xenbus_client.c

@@ -0,0 +1,568 @@
+/******************************************************************************
+ * Client-facing interface for the Xenbus driver.  In other words, the
+ * interface between the Xenbus and the device-specific code, be it the
+ * frontend or the backend of that driver.
+ *
+ * Copyright (C) 2005 XenSource Ltd
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/types.h>
+#include <linux/vmalloc.h>
+#include <asm/xen/hypervisor.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/event_channel.h>
+#include <xen/events.h>
+#include <xen/grant_table.h>
+#include <xen/xenbus.h>
+
+const char *xenbus_strstate(enum xenbus_state state)
+{
+	static const char *const name[] = {
+		[ XenbusStateUnknown      ] = "Unknown",
+		[ XenbusStateInitialising ] = "Initialising",
+		[ XenbusStateInitWait     ] = "InitWait",
+		[ XenbusStateInitialised  ] = "Initialised",
+		[ XenbusStateConnected    ] = "Connected",
+		[ XenbusStateClosing      ] = "Closing",
+		[ XenbusStateClosed	  ] = "Closed",
+	};
+	return (state < ARRAY_SIZE(name)) ? name[state] : "INVALID";
+}
+EXPORT_SYMBOL_GPL(xenbus_strstate);
+
+/**
+ * xenbus_watch_path - register a watch
+ * @dev: xenbus device
+ * @path: path to watch
+ * @watch: watch to register
+ * @callback: callback to register
+ *
+ * Register a @watch on the given path, using the given xenbus_watch structure
+ * for storage, and the given @callback function as the callback.  Return 0 on
+ * success, or -errno on error.  On success, the given @path will be saved as
+ * @watch->node, and remains the caller's to free.  On error, @watch->node will
+ * be NULL, the device will switch to %XenbusStateClosing, and the error will
+ * be saved in the store.
+ */
+int xenbus_watch_path(struct xenbus_device *dev, const char *path,
+		      struct xenbus_watch *watch,
+		      void (*callback)(struct xenbus_watch *,
+				       const char **, unsigned int))
+{
+	int err;
+
+	watch->node = path;
+	watch->callback = callback;
+
+	err = register_xenbus_watch(watch);
+
+	if (err) {
+		watch->node = NULL;
+		watch->callback = NULL;
+		xenbus_dev_fatal(dev, err, "adding watch on %s", path);
+	}
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_watch_path);
+
+
+/**
+ * xenbus_watch_pathfmt - register a watch on a sprintf-formatted path
+ * @dev: xenbus device
+ * @watch: watch to register
+ * @callback: callback to register
+ * @pathfmt: format of path to watch
+ *
+ * Register a watch on the given @path, using the given xenbus_watch
+ * structure for storage, and the given @callback function as the callback.
+ * Return 0 on success, or -errno on error.  On success, the watched path
+ * (@path/@path2) will be saved as @watch->node, and becomes the caller's to
+ * kfree().  On error, watch->node will be NULL, so the caller has nothing to
+ * free, the device will switch to %XenbusStateClosing, and the error will be
+ * saved in the store.
+ */
+int xenbus_watch_pathfmt(struct xenbus_device *dev,
+			 struct xenbus_watch *watch,
+			 void (*callback)(struct xenbus_watch *,
+					const char **, unsigned int),
+			 const char *pathfmt, ...)
+{
+	int err;
+	va_list ap;
+	char *path;
+
+	va_start(ap, pathfmt);
+	path = kvasprintf(GFP_NOIO | __GFP_HIGH, pathfmt, ap);
+	va_end(ap);
+
+	if (!path) {
+		xenbus_dev_fatal(dev, -ENOMEM, "allocating path for watch");
+		return -ENOMEM;
+	}
+	err = xenbus_watch_path(dev, path, watch, callback);
+
+	if (err)
+		kfree(path);
+	return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_watch_pathfmt);
+
+
+/**
+ * xenbus_switch_state
+ * @dev: xenbus device
+ * @state: new state
+ *
+ * Advertise in the store a change of the given driver to the given new_state.
+ * Return 0 on success, or -errno on error.  On error, the device will switch
+ * to XenbusStateClosing, and the error will be saved in the store.
+ */
+int xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state state)
+{
+	/* We check whether the state is currently set to the given value, and
+	   if not, then the state is set.  We don't want to unconditionally
+	   write the given state, because we don't want to fire watches
+	   unnecessarily.  Furthermore, if the node has gone, we don't write
+	   to it, as the device will be tearing down, and we don't want to
+	   resurrect that directory.
+
+	   Note that, because of this cached value of our state, this function
+	   will not work inside a Xenstore transaction (something it was
+	   trying to in the past) because dev->state would not get reset if
+	   the transaction was aborted.
+
+	 */
+
+	int current_state;
+	int err;
+
+	if (state == dev->state)
+		return 0;
+
+	err = xenbus_scanf(XBT_NIL, dev->nodename, "state", "%d",
+			   &current_state);
+	if (err != 1)
+		return 0;
+
+	err = xenbus_printf(XBT_NIL, dev->nodename, "state", "%d", state);
+	if (err) {
+		if (state != XenbusStateClosing) /* Avoid looping */
+			xenbus_dev_fatal(dev, err, "writing new state");
+		return err;
+	}
+
+	dev->state = state;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xenbus_switch_state);
+
+int xenbus_frontend_closed(struct xenbus_device *dev)
+{
+	xenbus_switch_state(dev, XenbusStateClosed);
+	complete(&dev->down);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xenbus_frontend_closed);
+
+/**
+ * Return the path to the error node for the given device, or NULL on failure.
+ * If the value returned is non-NULL, then it is the caller's to kfree.
+ */
+static char *error_path(struct xenbus_device *dev)
+{
+	return kasprintf(GFP_KERNEL, "error/%s", dev->nodename);
+}
+
+
+static void xenbus_va_dev_error(struct xenbus_device *dev, int err,
+				const char *fmt, va_list ap)
+{
+	int ret;
+	unsigned int len;
+	char *printf_buffer = NULL;
+	char *path_buffer = NULL;
+
+#define PRINTF_BUFFER_SIZE 4096
+	printf_buffer = kmalloc(PRINTF_BUFFER_SIZE, GFP_KERNEL);
+	if (printf_buffer == NULL)
+		goto fail;
+
+	len = sprintf(printf_buffer, "%i ", -err);
+	ret = vsnprintf(printf_buffer+len, PRINTF_BUFFER_SIZE-len, fmt, ap);
+
+	BUG_ON(len + ret > PRINTF_BUFFER_SIZE-1);
+
+	dev_err(&dev->dev, "%s\n", printf_buffer);
+
+	path_buffer = error_path(dev);
+
+	if (path_buffer == NULL) {
+		dev_err(&dev->dev, "failed to write error node for %s (%s)\n",
+		       dev->nodename, printf_buffer);
+		goto fail;
+	}
+
+	if (xenbus_write(XBT_NIL, path_buffer, "error", printf_buffer) != 0) {
+		dev_err(&dev->dev, "failed to write error node for %s (%s)\n",
+		       dev->nodename, printf_buffer);
+		goto fail;
+	}
+
+fail:
+	kfree(printf_buffer);
+	kfree(path_buffer);
+}
+
+
+/**
+ * xenbus_dev_error
+ * @dev: xenbus device
+ * @err: error to report
+ * @fmt: error message format
+ *
+ * Report the given negative errno into the store, along with the given
+ * formatted message.
+ */
+void xenbus_dev_error(struct xenbus_device *dev, int err, const char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	xenbus_va_dev_error(dev, err, fmt, ap);
+	va_end(ap);
+}
+EXPORT_SYMBOL_GPL(xenbus_dev_error);
+
+/**
+ * xenbus_dev_fatal
+ * @dev: xenbus device
+ * @err: error to report
+ * @fmt: error message format
+ *
+ * Equivalent to xenbus_dev_error(dev, err, fmt, args), followed by
+ * xenbus_switch_state(dev, XenbusStateClosing) to schedule an orderly
+ * closedown of this driver and its peer.
+ */
+
+void xenbus_dev_fatal(struct xenbus_device *dev, int err, const char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	xenbus_va_dev_error(dev, err, fmt, ap);
+	va_end(ap);
+
+	xenbus_switch_state(dev, XenbusStateClosing);
+}
+EXPORT_SYMBOL_GPL(xenbus_dev_fatal);
+
+/**
+ * xenbus_grant_ring
+ * @dev: xenbus device
+ * @ring_mfn: mfn of ring to grant
+
+ * Grant access to the given @ring_mfn to the peer of the given device.  Return
+ * 0 on success, or -errno on error.  On error, the device will switch to
+ * XenbusStateClosing, and the error will be saved in the store.
+ */
+int xenbus_grant_ring(struct xenbus_device *dev, unsigned long ring_mfn)
+{
+	int err = gnttab_grant_foreign_access(dev->otherend_id, ring_mfn, 0);
+	if (err < 0)
+		xenbus_dev_fatal(dev, err, "granting access to ring page");
+	return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_grant_ring);
+
+
+/**
+ * Allocate an event channel for the given xenbus_device, assigning the newly
+ * created local port to *port.  Return 0 on success, or -errno on error.  On
+ * error, the device will switch to XenbusStateClosing, and the error will be
+ * saved in the store.
+ */
+int xenbus_alloc_evtchn(struct xenbus_device *dev, int *port)
+{
+	struct evtchn_alloc_unbound alloc_unbound;
+	int err;
+
+	alloc_unbound.dom = DOMID_SELF;
+	alloc_unbound.remote_dom = dev->otherend_id;
+
+	err = HYPERVISOR_event_channel_op(EVTCHNOP_alloc_unbound,
+					  &alloc_unbound);
+	if (err)
+		xenbus_dev_fatal(dev, err, "allocating event channel");
+	else
+		*port = alloc_unbound.port;
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_alloc_evtchn);
+
+
+/**
+ * Bind to an existing interdomain event channel in another domain. Returns 0
+ * on success and stores the local port in *port. On error, returns -errno,
+ * switches the device to XenbusStateClosing, and saves the error in XenStore.
+ */
+int xenbus_bind_evtchn(struct xenbus_device *dev, int remote_port, int *port)
+{
+	struct evtchn_bind_interdomain bind_interdomain;
+	int err;
+
+	bind_interdomain.remote_dom = dev->otherend_id;
+	bind_interdomain.remote_port = remote_port;
+
+	err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
+					  &bind_interdomain);
+	if (err)
+		xenbus_dev_fatal(dev, err,
+				 "binding to event channel %d from domain %d",
+				 remote_port, dev->otherend_id);
+	else
+		*port = bind_interdomain.local_port;
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_bind_evtchn);
+
+
+/**
+ * Free an existing event channel. Returns 0 on success or -errno on error.
+ */
+int xenbus_free_evtchn(struct xenbus_device *dev, int port)
+{
+	struct evtchn_close close;
+	int err;
+
+	close.port = port;
+
+	err = HYPERVISOR_event_channel_op(EVTCHNOP_close, &close);
+	if (err)
+		xenbus_dev_error(dev, err, "freeing event channel %d", port);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_free_evtchn);
+
+
+/**
+ * xenbus_map_ring_valloc
+ * @dev: xenbus device
+ * @gnt_ref: grant reference
+ * @vaddr: pointer to address to be filled out by mapping
+ *
+ * Based on Rusty Russell's skeleton driver's map_page.
+ * Map a page of memory into this domain from another domain's grant table.
+ * xenbus_map_ring_valloc allocates a page of virtual address space, maps the
+ * page to that address, and sets *vaddr to that address.
+ * Returns 0 on success, and GNTST_* (see xen/include/interface/grant_table.h)
+ * or -ENOMEM on error. If an error is returned, device will switch to
+ * XenbusStateClosing and the error message will be saved in XenStore.
+ */
+int xenbus_map_ring_valloc(struct xenbus_device *dev, int gnt_ref, void **vaddr)
+{
+	struct gnttab_map_grant_ref op = {
+		.flags = GNTMAP_host_map,
+		.ref   = gnt_ref,
+		.dom   = dev->otherend_id,
+	};
+	struct vm_struct *area;
+
+	*vaddr = NULL;
+
+	area = xen_alloc_vm_area(PAGE_SIZE);
+	if (!area)
+		return -ENOMEM;
+
+	op.host_addr = (unsigned long)area->addr;
+
+	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
+		BUG();
+
+	if (op.status != GNTST_okay) {
+		xen_free_vm_area(area);
+		xenbus_dev_fatal(dev, op.status,
+				 "mapping in shared page %d from domain %d",
+				 gnt_ref, dev->otherend_id);
+		return op.status;
+	}
+
+	/* Stuff the handle in an unused field */
+	area->phys_addr = (unsigned long)op.handle;
+
+	*vaddr = area->addr;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xenbus_map_ring_valloc);
+
+
+/**
+ * xenbus_map_ring
+ * @dev: xenbus device
+ * @gnt_ref: grant reference
+ * @handle: pointer to grant handle to be filled
+ * @vaddr: address to be mapped to
+ *
+ * Map a page of memory into this domain from another domain's grant table.
+ * xenbus_map_ring does not allocate the virtual address space (you must do
+ * this yourself!). It only maps in the page to the specified address.
+ * Returns 0 on success, and GNTST_* (see xen/include/interface/grant_table.h)
+ * or -ENOMEM on error. If an error is returned, device will switch to
+ * XenbusStateClosing and the error message will be saved in XenStore.
+ */
+int xenbus_map_ring(struct xenbus_device *dev, int gnt_ref,
+		    grant_handle_t *handle, void *vaddr)
+{
+	struct gnttab_map_grant_ref op = {
+		.host_addr = (unsigned long)vaddr,
+		.flags     = GNTMAP_host_map,
+		.ref       = gnt_ref,
+		.dom       = dev->otherend_id,
+	};
+
+	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
+		BUG();
+
+	if (op.status != GNTST_okay) {
+		xenbus_dev_fatal(dev, op.status,
+				 "mapping in shared page %d from domain %d",
+				 gnt_ref, dev->otherend_id);
+	} else
+		*handle = op.handle;
+
+	return op.status;
+}
+EXPORT_SYMBOL_GPL(xenbus_map_ring);
+
+
+/**
+ * xenbus_unmap_ring_vfree
+ * @dev: xenbus device
+ * @vaddr: addr to unmap
+ *
+ * Based on Rusty Russell's skeleton driver's unmap_page.
+ * Unmap a page of memory in this domain that was imported from another domain.
+ * Use xenbus_unmap_ring_vfree if you mapped in your memory with
+ * xenbus_map_ring_valloc (it will free the virtual address space).
+ * Returns 0 on success and returns GNTST_* on error
+ * (see xen/include/interface/grant_table.h).
+ */
+int xenbus_unmap_ring_vfree(struct xenbus_device *dev, void *vaddr)
+{
+	struct vm_struct *area;
+	struct gnttab_unmap_grant_ref op = {
+		.host_addr = (unsigned long)vaddr,
+	};
+
+	/* It'd be nice if linux/vmalloc.h provided a find_vm_area(void *addr)
+	 * method so that we don't have to muck with vmalloc internals here.
+	 * We could force the user to hang on to their struct vm_struct from
+	 * xenbus_map_ring_valloc, but these 6 lines considerably simplify
+	 * this API.
+	 */
+	read_lock(&vmlist_lock);
+	for (area = vmlist; area != NULL; area = area->next) {
+		if (area->addr == vaddr)
+			break;
+	}
+	read_unlock(&vmlist_lock);
+
+	if (!area) {
+		xenbus_dev_error(dev, -ENOENT,
+				 "can't find mapped virtual address %p", vaddr);
+		return GNTST_bad_virt_addr;
+	}
+
+	op.handle = (grant_handle_t)area->phys_addr;
+
+	if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
+		BUG();
+
+	if (op.status == GNTST_okay)
+		xen_free_vm_area(area);
+	else
+		xenbus_dev_error(dev, op.status,
+				 "unmapping page at handle %d error %d",
+				 (int16_t)area->phys_addr, op.status);
+
+	return op.status;
+}
+EXPORT_SYMBOL_GPL(xenbus_unmap_ring_vfree);
+
+
+/**
+ * xenbus_unmap_ring
+ * @dev: xenbus device
+ * @handle: grant handle
+ * @vaddr: addr to unmap
+ *
+ * Unmap a page of memory in this domain that was imported from another domain.
+ * Returns 0 on success and returns GNTST_* on error
+ * (see xen/include/interface/grant_table.h).
+ */
+int xenbus_unmap_ring(struct xenbus_device *dev,
+		      grant_handle_t handle, void *vaddr)
+{
+	struct gnttab_unmap_grant_ref op = {
+		.host_addr = (unsigned long)vaddr,
+		.handle    = handle,
+	};
+
+	if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
+		BUG();
+
+	if (op.status != GNTST_okay)
+		xenbus_dev_error(dev, op.status,
+				 "unmapping page at handle %d error %d",
+				 handle, op.status);
+
+	return op.status;
+}
+EXPORT_SYMBOL_GPL(xenbus_unmap_ring);
+
+
+/**
+ * xenbus_read_driver_state
+ * @path: path for driver
+ *
+ * Return the state of the driver rooted at the given store path, or
+ * XenbusStateUnknown if no state can be read.
+ */
+enum xenbus_state xenbus_read_driver_state(const char *path)
+{
+	enum xenbus_state result;
+	int err = xenbus_gather(XBT_NIL, path, "state", "%d", &result, NULL);
+	if (err)
+		result = XenbusStateUnknown;
+
+	return result;
+}
+EXPORT_SYMBOL_GPL(xenbus_read_driver_state);

kernel/drivers/xen/xenbus/xenbus_comms.c

@@ -0,0 +1,234 @@
+/******************************************************************************
+ * xenbus_comms.c
+ *
+ * Low level code to talks to Xen Store: ringbuffer and event channel.
+ *
+ * Copyright (C) 2005 Rusty Russell, IBM Corporation
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/wait.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <linux/err.h>
+#include <xen/xenbus.h>
+#include <asm/xen/hypervisor.h>
+#include <xen/events.h>
+#include <xen/page.h>
+#include "xenbus_comms.h"
+
+static int xenbus_irq;
+
+static DECLARE_WORK(probe_work, xenbus_probe);
+
+static DECLARE_WAIT_QUEUE_HEAD(xb_waitq);
+
+static irqreturn_t wake_waiting(int irq, void *unused)
+{
+	if (unlikely(xenstored_ready == 0)) {
+		xenstored_ready = 1;
+		schedule_work(&probe_work);
+	}
+
+	wake_up(&xb_waitq);
+	return IRQ_HANDLED;
+}
+
+static int check_indexes(XENSTORE_RING_IDX cons, XENSTORE_RING_IDX prod)
+{
+	return ((prod - cons) <= XENSTORE_RING_SIZE);
+}
+
+static void *get_output_chunk(XENSTORE_RING_IDX cons,
+			      XENSTORE_RING_IDX prod,
+			      char *buf, uint32_t *len)
+{
+	*len = XENSTORE_RING_SIZE - MASK_XENSTORE_IDX(prod);
+	if ((XENSTORE_RING_SIZE - (prod - cons)) < *len)
+		*len = XENSTORE_RING_SIZE - (prod - cons);
+	return buf + MASK_XENSTORE_IDX(prod);
+}
+
+static const void *get_input_chunk(XENSTORE_RING_IDX cons,
+				   XENSTORE_RING_IDX prod,
+				   const char *buf, uint32_t *len)
+{
+	*len = XENSTORE_RING_SIZE - MASK_XENSTORE_IDX(cons);
+	if ((prod - cons) < *len)
+		*len = prod - cons;
+	return buf + MASK_XENSTORE_IDX(cons);
+}
+
+/**
+ * xb_write - low level write
+ * @data: buffer to send
+ * @len: length of buffer
+ *
+ * Returns 0 on success, error otherwise.
+ */
+int xb_write(const void *data, unsigned len)
+{
+	struct xenstore_domain_interface *intf = xen_store_interface;
+	XENSTORE_RING_IDX cons, prod;
+	int rc;
+
+	while (len != 0) {
+		void *dst;
+		unsigned int avail;
+
+		rc = wait_event_interruptible(
+			xb_waitq,
+			(intf->req_prod - intf->req_cons) !=
+			XENSTORE_RING_SIZE);
+		if (rc < 0)
+			return rc;
+
+		/* Read indexes, then verify. */
+		cons = intf->req_cons;
+		prod = intf->req_prod;
+		if (!check_indexes(cons, prod)) {
+			intf->req_cons = intf->req_prod = 0;
+			return -EIO;
+		}
+
+		dst = get_output_chunk(cons, prod, intf->req, &avail);
+		if (avail == 0)
+			continue;
+		if (avail > len)
+			avail = len;
+
+		/* Must write data /after/ reading the consumer index. */
+		mb();
+
+		memcpy(dst, data, avail);
+		data += avail;
+		len -= avail;
+
+		/* Other side must not see new producer until data is there. */
+		wmb();
+		intf->req_prod += avail;
+
+		/* Implies mb(): other side will see the updated producer. */
+		notify_remote_via_evtchn(xen_store_evtchn);
+	}
+
+	return 0;
+}
+
+int xb_data_to_read(void)
+{
+	struct xenstore_domain_interface *intf = xen_store_interface;
+	return (intf->rsp_cons != intf->rsp_prod);
+}
+
+int xb_wait_for_data_to_read(void)
+{
+	return wait_event_interruptible(xb_waitq, xb_data_to_read());
+}
+
+int xb_read(void *data, unsigned len)
+{
+	struct xenstore_domain_interface *intf = xen_store_interface;
+	XENSTORE_RING_IDX cons, prod;
+	int rc;
+
+	while (len != 0) {
+		unsigned int avail;
+		const char *src;
+
+		rc = xb_wait_for_data_to_read();
+		if (rc < 0)
+			return rc;
+
+		/* Read indexes, then verify. */
+		cons = intf->rsp_cons;
+		prod = intf->rsp_prod;
+		if (!check_indexes(cons, prod)) {
+			intf->rsp_cons = intf->rsp_prod = 0;
+			return -EIO;
+		}
+
+		src = get_input_chunk(cons, prod, intf->rsp, &avail);
+		if (avail == 0)
+			continue;
+		if (avail > len)
+			avail = len;
+
+		/* Must read data /after/ reading the producer index. */
+		rmb();
+
+		memcpy(data, src, avail);
+		data += avail;
+		len -= avail;
+
+		/* Other side must not see free space until we've copied out */
+		mb();
+		intf->rsp_cons += avail;
+
+		pr_debug("Finished read of %i bytes (%i to go)\n", avail, len);
+
+		/* Implies mb(): other side will see the updated consumer. */
+		notify_remote_via_evtchn(xen_store_evtchn);
+	}
+
+	return 0;
+}
+
+/**
+ * xb_init_comms - Set up interrupt handler off store event channel.
+ */
+int xb_init_comms(void)
+{
+	struct xenstore_domain_interface *intf = xen_store_interface;
+
+	if (intf->req_prod != intf->req_cons)
+		printk(KERN_ERR "XENBUS request ring is not quiescent "
+		       "(%08x:%08x)!\n", intf->req_cons, intf->req_prod);
+
+	if (intf->rsp_prod != intf->rsp_cons) {
+		printk(KERN_WARNING "XENBUS response ring is not quiescent "
+		       "(%08x:%08x): fixing up\n",
+		       intf->rsp_cons, intf->rsp_prod);
+		intf->rsp_cons = intf->rsp_prod;
+	}
+
+	if (xenbus_irq) {
+		/* Already have an irq; assume we're resuming */
+		rebind_evtchn_irq(xen_store_evtchn, xenbus_irq);
+	} else {
+		int err;
+		err = bind_evtchn_to_irqhandler(xen_store_evtchn, wake_waiting,
+						0, "xenbus", &xb_waitq);
+		if (err <= 0) {
+			printk(KERN_ERR "XENBUS request irq failed %i\n", err);
+			return err;
+		}
+
+		xenbus_irq = err;
+	}
+
+	return 0;
+}

kernel/drivers/xen/xenbus/xenbus_comms.h

@@ -0,0 +1,46 @@
+/*
+ * Private include for xenbus communications.
+ *
+ * Copyright (C) 2005 Rusty Russell, IBM Corporation
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef _XENBUS_COMMS_H
+#define _XENBUS_COMMS_H
+
+int xs_init(void);
+int xb_init_comms(void);
+
+/* Low level routines. */
+int xb_write(const void *data, unsigned len);
+int xb_read(void *data, unsigned len);
+int xb_data_to_read(void);
+int xb_wait_for_data_to_read(void);
+int xs_input_avail(void);
+extern struct xenstore_domain_interface *xen_store_interface;
+extern int xen_store_evtchn;
+
+#endif /* _XENBUS_COMMS_H */

kernel/drivers/xen/xenbus/xenbus_probe.c

@@ -0,0 +1,956 @@
+/******************************************************************************
+ * Talks to Xen Store to figure out what devices we have.
+ *
+ * Copyright (C) 2005 Rusty Russell, IBM Corporation
+ * Copyright (C) 2005 Mike Wray, Hewlett-Packard
+ * Copyright (C) 2005, 2006 XenSource Ltd
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#define DPRINTK(fmt, args...)				\
+	pr_debug("xenbus_probe (%s:%d) " fmt ".\n",	\
+		 __func__, __LINE__, ##args)
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/fcntl.h>
+#include <linux/mm.h>
+#include <linux/proc_fs.h>
+#include <linux/notifier.h>
+#include <linux/kthread.h>
+#include <linux/mutex.h>
+#include <linux/io.h>
+
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/xen/hypervisor.h>
+#include <xen/xenbus.h>
+#include <xen/events.h>
+#include <xen/page.h>
+
+#include "xenbus_comms.h"
+#include "xenbus_probe.h"
+
+
+int xen_store_evtchn;
+EXPORT_SYMBOL(xen_store_evtchn);
+
+struct xenstore_domain_interface *xen_store_interface;
+static unsigned long xen_store_mfn;
+
+static BLOCKING_NOTIFIER_HEAD(xenstore_chain);
+
+static void wait_for_devices(struct xenbus_driver *xendrv);
+
+static int xenbus_probe_frontend(const char *type, const char *name);
+
+static void xenbus_dev_shutdown(struct device *_dev);
+
+static int xenbus_dev_suspend(struct device *dev, pm_message_t state);
+static int xenbus_dev_resume(struct device *dev);
+
+/* If something in array of ids matches this device, return it. */
+static const struct xenbus_device_id *
+match_device(const struct xenbus_device_id *arr, struct xenbus_device *dev)
+{
+	for (; *arr->devicetype != '\0'; arr++) {
+		if (!strcmp(arr->devicetype, dev->devicetype))
+			return arr;
+	}
+	return NULL;
+}
+
+int xenbus_match(struct device *_dev, struct device_driver *_drv)
+{
+	struct xenbus_driver *drv = to_xenbus_driver(_drv);
+
+	if (!drv->ids)
+		return 0;
+
+	return match_device(drv->ids, to_xenbus_device(_dev)) != NULL;
+}
+
+static int xenbus_uevent(struct device *_dev, struct kobj_uevent_env *env)
+{
+	struct xenbus_device *dev = to_xenbus_device(_dev);
+
+	if (add_uevent_var(env, "MODALIAS=xen:%s", dev->devicetype))
+		return -ENOMEM;
+
+	return 0;
+}
+
+/* device/<type>/<id> => <type>-<id> */
+static int frontend_bus_id(char bus_id[XEN_BUS_ID_SIZE], const char *nodename)
+{
+	nodename = strchr(nodename, '/');
+	if (!nodename || strlen(nodename + 1) >= XEN_BUS_ID_SIZE) {
+		printk(KERN_WARNING "XENBUS: bad frontend %s\n", nodename);
+		return -EINVAL;
+	}
+
+	strlcpy(bus_id, nodename + 1, XEN_BUS_ID_SIZE);
+	if (!strchr(bus_id, '/')) {
+		printk(KERN_WARNING "XENBUS: bus_id %s no slash\n", bus_id);
+		return -EINVAL;
+	}
+	*strchr(bus_id, '/') = '-';
+	return 0;
+}
+
+
+static void free_otherend_details(struct xenbus_device *dev)
+{
+	kfree(dev->otherend);
+	dev->otherend = NULL;
+}
+
+
+static void free_otherend_watch(struct xenbus_device *dev)
+{
+	if (dev->otherend_watch.node) {
+		unregister_xenbus_watch(&dev->otherend_watch);
+		kfree(dev->otherend_watch.node);
+		dev->otherend_watch.node = NULL;
+	}
+}
+
+
+int read_otherend_details(struct xenbus_device *xendev,
+				 char *id_node, char *path_node)
+{
+	int err = xenbus_gather(XBT_NIL, xendev->nodename,
+				id_node, "%i", &xendev->otherend_id,
+				path_node, NULL, &xendev->otherend,
+				NULL);
+	if (err) {
+		xenbus_dev_fatal(xendev, err,
+				 "reading other end details from %s",
+				 xendev->nodename);
+		return err;
+	}
+	if (strlen(xendev->otherend) == 0 ||
+	    !xenbus_exists(XBT_NIL, xendev->otherend, "")) {
+		xenbus_dev_fatal(xendev, -ENOENT,
+				 "unable to read other end from %s.  "
+				 "missing or inaccessible.",
+				 xendev->nodename);
+		free_otherend_details(xendev);
+		return -ENOENT;
+	}
+
+	return 0;
+}
+
+
+static int read_backend_details(struct xenbus_device *xendev)
+{
+	return read_otherend_details(xendev, "backend-id", "backend");
+}
+
+static struct device_attribute xenbus_dev_attrs[] = {
+	__ATTR_NULL
+};
+
+/* Bus type for frontend drivers. */
+static struct xen_bus_type xenbus_frontend = {
+	.root = "device",
+	.levels = 2, 		/* device/type/<id> */
+	.get_bus_id = frontend_bus_id,
+	.probe = xenbus_probe_frontend,
+	.bus = {
+		.name      = "xen",
+		.match     = xenbus_match,
+		.uevent    = xenbus_uevent,
+		.probe     = xenbus_dev_probe,
+		.remove    = xenbus_dev_remove,
+		.shutdown  = xenbus_dev_shutdown,
+		.dev_attrs = xenbus_dev_attrs,
+
+		.suspend   = xenbus_dev_suspend,
+		.resume    = xenbus_dev_resume,
+	},
+};
+
+static void otherend_changed(struct xenbus_watch *watch,
+			     const char **vec, unsigned int len)
+{
+	struct xenbus_device *dev =
+		container_of(watch, struct xenbus_device, otherend_watch);
+	struct xenbus_driver *drv = to_xenbus_driver(dev->dev.driver);
+	enum xenbus_state state;
+
+	/* Protect us against watches firing on old details when the otherend
+	   details change, say immediately after a resume. */
+	if (!dev->otherend ||
+	    strncmp(dev->otherend, vec[XS_WATCH_PATH],
+		    strlen(dev->otherend))) {
+		dev_dbg(&dev->dev, "Ignoring watch at %s\n",
+			vec[XS_WATCH_PATH]);
+		return;
+	}
+
+	state = xenbus_read_driver_state(dev->otherend);
+
+	dev_dbg(&dev->dev, "state is %d, (%s), %s, %s\n",
+		state, xenbus_strstate(state), dev->otherend_watch.node,
+		vec[XS_WATCH_PATH]);
+
+	/*
+	 * Ignore xenbus transitions during shutdown. This prevents us doing
+	 * work that can fail e.g., when the rootfs is gone.
+	 */
+	if (system_state > SYSTEM_RUNNING) {
+		struct xen_bus_type *bus = bus;
+		bus = container_of(dev->dev.bus, struct xen_bus_type, bus);
+		/* If we're frontend, drive the state machine to Closed. */
+		/* This should cause the backend to release our resources. */
+		if ((bus == &xenbus_frontend) && (state == XenbusStateClosing))
+			xenbus_frontend_closed(dev);
+		return;
+	}
+
+	if (drv->otherend_changed)
+		drv->otherend_changed(dev, state);
+}
+
+
+static int talk_to_otherend(struct xenbus_device *dev)
+{
+	struct xenbus_driver *drv = to_xenbus_driver(dev->dev.driver);
+
+	free_otherend_watch(dev);
+	free_otherend_details(dev);
+
+	return drv->read_otherend_details(dev);
+}
+
+
+static int watch_otherend(struct xenbus_device *dev)
+{
+	return xenbus_watch_pathfmt(dev, &dev->otherend_watch, otherend_changed,
+				    "%s/%s", dev->otherend, "state");
+}
+
+
+int xenbus_dev_probe(struct device *_dev)
+{
+	struct xenbus_device *dev = to_xenbus_device(_dev);
+	struct xenbus_driver *drv = to_xenbus_driver(_dev->driver);
+	const struct xenbus_device_id *id;
+	int err;
+
+	DPRINTK("%s", dev->nodename);
+
+	if (!drv->probe) {
+		err = -ENODEV;
+		goto fail;
+	}
+
+	id = match_device(drv->ids, dev);
+	if (!id) {
+		err = -ENODEV;
+		goto fail;
+	}
+
+	err = talk_to_otherend(dev);
+	if (err) {
+		dev_warn(&dev->dev, "talk_to_otherend on %s failed.\n",
+			 dev->nodename);
+		return err;
+	}
+
+	err = drv->probe(dev, id);
+	if (err)
+		goto fail;
+
+	err = watch_otherend(dev);
+	if (err) {
+		dev_warn(&dev->dev, "watch_otherend on %s failed.\n",
+		       dev->nodename);
+		return err;
+	}
+
+	return 0;
+fail:
+	xenbus_dev_error(dev, err, "xenbus_dev_probe on %s", dev->nodename);
+	xenbus_switch_state(dev, XenbusStateClosed);
+	return -ENODEV;
+}
+
+int xenbus_dev_remove(struct device *_dev)
+{
+	struct xenbus_device *dev = to_xenbus_device(_dev);
+	struct xenbus_driver *drv = to_xenbus_driver(_dev->driver);
+
+	DPRINTK("%s", dev->nodename);
+
+	free_otherend_watch(dev);
+	free_otherend_details(dev);
+
+	if (drv->remove)
+		drv->remove(dev);
+
+	xenbus_switch_state(dev, XenbusStateClosed);
+	return 0;
+}
+
+static void xenbus_dev_shutdown(struct device *_dev)
+{
+	struct xenbus_device *dev = to_xenbus_device(_dev);
+	unsigned long timeout = 5*HZ;
+
+	DPRINTK("%s", dev->nodename);
+
+	get_device(&dev->dev);
+	if (dev->state != XenbusStateConnected) {
+		printk(KERN_INFO "%s: %s: %s != Connected, skipping\n", __func__,
+		       dev->nodename, xenbus_strstate(dev->state));
+		goto out;
+	}
+	xenbus_switch_state(dev, XenbusStateClosing);
+	timeout = wait_for_completion_timeout(&dev->down, timeout);
+	if (!timeout)
+		printk(KERN_INFO "%s: %s timeout closing device\n",
+		       __func__, dev->nodename);
+ out:
+	put_device(&dev->dev);
+}
+
+int xenbus_register_driver_common(struct xenbus_driver *drv,
+				  struct xen_bus_type *bus,
+				  struct module *owner,
+				  const char *mod_name)
+{
+	drv->driver.name = drv->name;
+	drv->driver.bus = &bus->bus;
+	drv->driver.owner = owner;
+	drv->driver.mod_name = mod_name;
+
+	return driver_register(&drv->driver);
+}
+
+int __xenbus_register_frontend(struct xenbus_driver *drv,
+			       struct module *owner, const char *mod_name)
+{
+	int ret;
+
+	drv->read_otherend_details = read_backend_details;
+
+	ret = xenbus_register_driver_common(drv, &xenbus_frontend,
+					    owner, mod_name);
+	if (ret)
+		return ret;
+
+	/* If this driver is loaded as a module wait for devices to attach. */
+	wait_for_devices(drv);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(__xenbus_register_frontend);
+
+void xenbus_unregister_driver(struct xenbus_driver *drv)
+{
+	driver_unregister(&drv->driver);
+}
+EXPORT_SYMBOL_GPL(xenbus_unregister_driver);
+
+struct xb_find_info
+{
+	struct xenbus_device *dev;
+	const char *nodename;
+};
+
+static int cmp_dev(struct device *dev, void *data)
+{
+	struct xenbus_device *xendev = to_xenbus_device(dev);
+	struct xb_find_info *info = data;
+
+	if (!strcmp(xendev->nodename, info->nodename)) {
+		info->dev = xendev;
+		get_device(dev);
+		return 1;
+	}
+	return 0;
+}
+
+struct xenbus_device *xenbus_device_find(const char *nodename,
+					 struct bus_type *bus)
+{
+	struct xb_find_info info = { .dev = NULL, .nodename = nodename };
+
+	bus_for_each_dev(bus, NULL, &info, cmp_dev);
+	return info.dev;
+}
+
+static int cleanup_dev(struct device *dev, void *data)
+{
+	struct xenbus_device *xendev = to_xenbus_device(dev);
+	struct xb_find_info *info = data;
+	int len = strlen(info->nodename);
+
+	DPRINTK("%s", info->nodename);
+
+	/* Match the info->nodename path, or any subdirectory of that path. */
+	if (strncmp(xendev->nodename, info->nodename, len))
+		return 0;
+
+	/* If the node name is longer, ensure it really is a subdirectory. */
+	if ((strlen(xendev->nodename) > len) && (xendev->nodename[len] != '/'))
+		return 0;
+
+	info->dev = xendev;
+	get_device(dev);
+	return 1;
+}
+
+static void xenbus_cleanup_devices(const char *path, struct bus_type *bus)
+{
+	struct xb_find_info info = { .nodename = path };
+
+	do {
+		info.dev = NULL;
+		bus_for_each_dev(bus, NULL, &info, cleanup_dev);
+		if (info.dev) {
+			device_unregister(&info.dev->dev);
+			put_device(&info.dev->dev);
+		}
+	} while (info.dev);
+}
+
+static void xenbus_dev_release(struct device *dev)
+{
+	if (dev)
+		kfree(to_xenbus_device(dev));
+}
+
+static ssize_t xendev_show_nodename(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%s\n", to_xenbus_device(dev)->nodename);
+}
+static DEVICE_ATTR(nodename, S_IRUSR | S_IRGRP | S_IROTH, xendev_show_nodename, NULL);
+
+static ssize_t xendev_show_devtype(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%s\n", to_xenbus_device(dev)->devicetype);
+}
+static DEVICE_ATTR(devtype, S_IRUSR | S_IRGRP | S_IROTH, xendev_show_devtype, NULL);
+
+static ssize_t xendev_show_modalias(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "xen:%s\n", to_xenbus_device(dev)->devicetype);
+}
+static DEVICE_ATTR(modalias, S_IRUSR | S_IRGRP | S_IROTH, xendev_show_modalias, NULL);
+
+int xenbus_probe_node(struct xen_bus_type *bus,
+		      const char *type,
+		      const char *nodename)
+{
+	char devname[XEN_BUS_ID_SIZE];
+	int err;
+	struct xenbus_device *xendev;
+	size_t stringlen;
+	char *tmpstring;
+
+	enum xenbus_state state = xenbus_read_driver_state(nodename);
+
+	if (state != XenbusStateInitialising) {
+		/* Device is not new, so ignore it.  This can happen if a
+		   device is going away after switching to Closed.  */
+		return 0;
+	}
+
+	stringlen = strlen(nodename) + 1 + strlen(type) + 1;
+	xendev = kzalloc(sizeof(*xendev) + stringlen, GFP_KERNEL);
+	if (!xendev)
+		return -ENOMEM;
+
+	xendev->state = XenbusStateInitialising;
+
+	/* Copy the strings into the extra space. */
+
+	tmpstring = (char *)(xendev + 1);
+	strcpy(tmpstring, nodename);
+	xendev->nodename = tmpstring;
+
+	tmpstring += strlen(tmpstring) + 1;
+	strcpy(tmpstring, type);
+	xendev->devicetype = tmpstring;
+	init_completion(&xendev->down);
+
+	xendev->dev.bus = &bus->bus;
+	xendev->dev.release = xenbus_dev_release;
+
+	err = bus->get_bus_id(devname, xendev->nodename);
+	if (err)
+		goto fail;
+
+	dev_set_name(&xendev->dev, devname);
+
+	/* Register with generic device framework. */
+	err = device_register(&xendev->dev);
+	if (err)
+		goto fail;
+
+	err = device_create_file(&xendev->dev, &dev_attr_nodename);
+	if (err)
+		goto fail_unregister;
+
+	err = device_create_file(&xendev->dev, &dev_attr_devtype);
+	if (err)
+		goto fail_remove_nodename;
+
+	err = device_create_file(&xendev->dev, &dev_attr_modalias);
+	if (err)
+		goto fail_remove_devtype;
+
+	return 0;
+fail_remove_devtype:
+	device_remove_file(&xendev->dev, &dev_attr_devtype);
+fail_remove_nodename:
+	device_remove_file(&xendev->dev, &dev_attr_nodename);
+fail_unregister:
+	device_unregister(&xendev->dev);
+fail:
+	kfree(xendev);
+	return err;
+}
+
+/* device/<typename>/<name> */
+static int xenbus_probe_frontend(const char *type, const char *name)
+{
+	char *nodename;
+	int err;
+
+	nodename = kasprintf(GFP_KERNEL, "%s/%s/%s",
+			     xenbus_frontend.root, type, name);
+	if (!nodename)
+		return -ENOMEM;
+
+	DPRINTK("%s", nodename);
+
+	err = xenbus_probe_node(&xenbus_frontend, type, nodename);
+	kfree(nodename);
+	return err;
+}
+
+static int xenbus_probe_device_type(struct xen_bus_type *bus, const char *type)
+{
+	int err = 0;
+	char **dir;
+	unsigned int dir_n = 0;
+	int i;
+
+	dir = xenbus_directory(XBT_NIL, bus->root, type, &dir_n);
+	if (IS_ERR(dir))
+		return PTR_ERR(dir);
+
+	for (i = 0; i < dir_n; i++) {
+		err = bus->probe(type, dir[i]);
+		if (err)
+			break;
+	}
+	kfree(dir);
+	return err;
+}
+
+int xenbus_probe_devices(struct xen_bus_type *bus)
+{
+	int err = 0;
+	char **dir;
+	unsigned int i, dir_n;
+
+	dir = xenbus_directory(XBT_NIL, bus->root, "", &dir_n);
+	if (IS_ERR(dir))
+		return PTR_ERR(dir);
+
+	for (i = 0; i < dir_n; i++) {
+		err = xenbus_probe_device_type(bus, dir[i]);
+		if (err)
+			break;
+	}
+	kfree(dir);
+	return err;
+}
+
+static unsigned int char_count(const char *str, char c)
+{
+	unsigned int i, ret = 0;
+
+	for (i = 0; str[i]; i++)
+		if (str[i] == c)
+			ret++;
+	return ret;
+}
+
+static int strsep_len(const char *str, char c, unsigned int len)
+{
+	unsigned int i;
+
+	for (i = 0; str[i]; i++)
+		if (str[i] == c) {
+			if (len == 0)
+				return i;
+			len--;
+		}
+	return (len == 0) ? i : -ERANGE;
+}
+
+void xenbus_dev_changed(const char *node, struct xen_bus_type *bus)
+{
+	int exists, rootlen;
+	struct xenbus_device *dev;
+	char type[XEN_BUS_ID_SIZE];
+	const char *p, *root;
+
+	if (char_count(node, '/') < 2)
+		return;
+
+	exists = xenbus_exists(XBT_NIL, node, "");
+	if (!exists) {
+		xenbus_cleanup_devices(node, &bus->bus);
+		return;
+	}
+
+	/* backend/<type>/... or device/<type>/... */
+	p = strchr(node, '/') + 1;
+	snprintf(type, XEN_BUS_ID_SIZE, "%.*s", (int)strcspn(p, "/"), p);
+	type[XEN_BUS_ID_SIZE-1] = '\0';
+
+	rootlen = strsep_len(node, '/', bus->levels);
+	if (rootlen < 0)
+		return;
+	root = kasprintf(GFP_KERNEL, "%.*s", rootlen, node);
+	if (!root)
+		return;
+
+	dev = xenbus_device_find(root, &bus->bus);
+	if (!dev)
+		xenbus_probe_node(bus, type, root);
+	else
+		put_device(&dev->dev);
+
+	kfree(root);
+}
+EXPORT_SYMBOL_GPL(xenbus_dev_changed);
+
+static void frontend_changed(struct xenbus_watch *watch,
+			     const char **vec, unsigned int len)
+{
+	DPRINTK("");
+
+	xenbus_dev_changed(vec[XS_WATCH_PATH], &xenbus_frontend);
+}
+
+/* We watch for devices appearing and vanishing. */
+static struct xenbus_watch fe_watch = {
+	.node = "device",
+	.callback = frontend_changed,
+};
+
+static int xenbus_dev_suspend(struct device *dev, pm_message_t state)
+{
+	int err = 0;
+	struct xenbus_driver *drv;
+	struct xenbus_device *xdev;
+
+	DPRINTK("");
+
+	if (dev->driver == NULL)
+		return 0;
+	drv = to_xenbus_driver(dev->driver);
+	xdev = container_of(dev, struct xenbus_device, dev);
+	if (drv->suspend)
+		err = drv->suspend(xdev, state);
+	if (err)
+		printk(KERN_WARNING
+		       "xenbus: suspend %s failed: %i\n", dev_name(dev), err);
+	return 0;
+}
+
+static int xenbus_dev_resume(struct device *dev)
+{
+	int err;
+	struct xenbus_driver *drv;
+	struct xenbus_device *xdev;
+
+	DPRINTK("");
+
+	if (dev->driver == NULL)
+		return 0;
+
+	drv = to_xenbus_driver(dev->driver);
+	xdev = container_of(dev, struct xenbus_device, dev);
+
+	err = talk_to_otherend(xdev);
+	if (err) {
+		printk(KERN_WARNING
+		       "xenbus: resume (talk_to_otherend) %s failed: %i\n",
+		       dev_name(dev), err);
+		return err;
+	}
+
+	xdev->state = XenbusStateInitialising;
+
+	if (drv->resume) {
+		err = drv->resume(xdev);
+		if (err) {
+			printk(KERN_WARNING
+			       "xenbus: resume %s failed: %i\n",
+			       dev_name(dev), err);
+			return err;
+		}
+	}
+
+	err = watch_otherend(xdev);
+	if (err) {
+		printk(KERN_WARNING
+		       "xenbus_probe: resume (watch_otherend) %s failed: "
+		       "%d.\n", dev_name(dev), err);
+		return err;
+	}
+
+	return 0;
+}
+
+/* A flag to determine if xenstored is 'ready' (i.e. has started) */
+int xenstored_ready = 0;
+
+
+int register_xenstore_notifier(struct notifier_block *nb)
+{
+	int ret = 0;
+
+	if (xenstored_ready > 0)
+		ret = nb->notifier_call(nb, 0, NULL);
+	else
+		blocking_notifier_chain_register(&xenstore_chain, nb);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(register_xenstore_notifier);
+
+void unregister_xenstore_notifier(struct notifier_block *nb)
+{
+	blocking_notifier_chain_unregister(&xenstore_chain, nb);
+}
+EXPORT_SYMBOL_GPL(unregister_xenstore_notifier);
+
+void xenbus_probe(struct work_struct *unused)
+{
+	BUG_ON((xenstored_ready <= 0));
+
+	/* Enumerate devices in xenstore and watch for changes. */
+	xenbus_probe_devices(&xenbus_frontend);
+	register_xenbus_watch(&fe_watch);
+	xenbus_backend_probe_and_watch();
+
+	/* Notify others that xenstore is up */
+	blocking_notifier_call_chain(&xenstore_chain, 0, NULL);
+}
+
+static int __init xenbus_probe_init(void)
+{
+	int err = 0;
+
+	DPRINTK("");
+
+	err = -ENODEV;
+	if (!xen_domain())
+		goto out_error;
+
+	/* Register ourselves with the kernel bus subsystem */
+	err = bus_register(&xenbus_frontend.bus);
+	if (err)
+		goto out_error;
+
+	err = xenbus_backend_bus_register();
+	if (err)
+		goto out_unreg_front;
+
+	/*
+	 * Domain0 doesn't have a store_evtchn or store_mfn yet.
+	 */
+	if (xen_initial_domain()) {
+		/* dom0 not yet supported */
+	} else {
+		xenstored_ready = 1;
+		xen_store_evtchn = xen_start_info->store_evtchn;
+		xen_store_mfn = xen_start_info->store_mfn;
+	}
+	xen_store_interface = mfn_to_virt(xen_store_mfn);
+
+	/* Initialize the interface to xenstore. */
+	err = xs_init();
+	if (err) {
+		printk(KERN_WARNING
+		       "XENBUS: Error initializing xenstore comms: %i\n", err);
+		goto out_unreg_back;
+	}
+
+	if (!xen_initial_domain())
+		xenbus_probe(NULL);
+
+#ifdef CONFIG_XEN_COMPAT_XENFS
+	/*
+	 * Create xenfs mountpoint in /proc for compatibility with
+	 * utilities that expect to find "xenbus" under "/proc/xen".
+	 */
+	proc_mkdir("xen", NULL);
+#endif
+
+	return 0;
+
+  out_unreg_back:
+	xenbus_backend_bus_unregister();
+
+  out_unreg_front:
+	bus_unregister(&xenbus_frontend.bus);
+
+  out_error:
+	return err;
+}
+
+postcore_initcall(xenbus_probe_init);
+
+MODULE_LICENSE("GPL");
+
+static int is_device_connecting(struct device *dev, void *data)
+{
+	struct xenbus_device *xendev = to_xenbus_device(dev);
+	struct device_driver *drv = data;
+	struct xenbus_driver *xendrv;
+
+	/*
+	 * A device with no driver will never connect. We care only about
+	 * devices which should currently be in the process of connecting.
+	 */
+	if (!dev->driver)
+		return 0;
+
+	/* Is this search limited to a particular driver? */
+	if (drv && (dev->driver != drv))
+		return 0;
+
+	xendrv = to_xenbus_driver(dev->driver);
+	return (xendev->state < XenbusStateConnected ||
+		(xendev->state == XenbusStateConnected &&
+		 xendrv->is_ready && !xendrv->is_ready(xendev)));
+}
+
+static int exists_connecting_device(struct device_driver *drv)
+{
+	return bus_for_each_dev(&xenbus_frontend.bus, NULL, drv,
+				is_device_connecting);
+}
+
+static int print_device_status(struct device *dev, void *data)
+{
+	struct xenbus_device *xendev = to_xenbus_device(dev);
+	struct device_driver *drv = data;
+
+	/* Is this operation limited to a particular driver? */
+	if (drv && (dev->driver != drv))
+		return 0;
+
+	if (!dev->driver) {
+		/* Information only: is this too noisy? */
+		printk(KERN_INFO "XENBUS: Device with no driver: %s\n",
+		       xendev->nodename);
+	} else if (xendev->state < XenbusStateConnected) {
+		enum xenbus_state rstate = XenbusStateUnknown;
+		if (xendev->otherend)
+			rstate = xenbus_read_driver_state(xendev->otherend);
+		printk(KERN_WARNING "XENBUS: Timeout connecting "
+		       "to device: %s (local state %d, remote state %d)\n",
+		       xendev->nodename, xendev->state, rstate);
+	}
+
+	return 0;
+}
+
+/* We only wait for device setup after most initcalls have run. */
+static int ready_to_wait_for_devices;
+
+/*
+ * On a 5-minute timeout, wait for all devices currently configured.  We need
+ * to do this to guarantee that the filesystems and / or network devices
+ * needed for boot are available, before we can allow the boot to proceed.
+ *
+ * This needs to be on a late_initcall, to happen after the frontend device
+ * drivers have been initialised, but before the root fs is mounted.
+ *
+ * A possible improvement here would be to have the tools add a per-device
+ * flag to the store entry, indicating whether it is needed at boot time.
+ * This would allow people who knew what they were doing to accelerate their
+ * boot slightly, but of course needs tools or manual intervention to set up
+ * those flags correctly.
+ */
+static void wait_for_devices(struct xenbus_driver *xendrv)
+{
+	unsigned long start = jiffies;
+	struct device_driver *drv = xendrv ? &xendrv->driver : NULL;
+	unsigned int seconds_waited = 0;
+
+	if (!ready_to_wait_for_devices || !xen_domain())
+		return;
+
+	while (exists_connecting_device(drv)) {
+		if (time_after(jiffies, start + (seconds_waited+5)*HZ)) {
+			if (!seconds_waited)
+				printk(KERN_WARNING "XENBUS: Waiting for "
+				       "devices to initialise: ");
+			seconds_waited += 5;
+			printk("%us...", 300 - seconds_waited);
+			if (seconds_waited == 300)
+				break;
+		}
+
+		schedule_timeout_interruptible(HZ/10);
+	}
+
+	if (seconds_waited)
+		printk("\n");
+
+	bus_for_each_dev(&xenbus_frontend.bus, NULL, drv,
+			 print_device_status);
+}
+
+#ifndef MODULE
+static int __init boot_wait_for_devices(void)
+{
+	ready_to_wait_for_devices = 1;
+	wait_for_devices(NULL);
+	return 0;
+}
+
+late_initcall(boot_wait_for_devices);
+#endif

kernel/drivers/xen/xenbus/xenbus_probe.h

@@ -0,0 +1,76 @@
+/******************************************************************************
+ * xenbus_probe.h
+ *
+ * Talks to Xen Store to figure out what devices we have.
+ *
+ * Copyright (C) 2005 Rusty Russell, IBM Corporation
+ * Copyright (C) 2005 XenSource Ltd.
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef _XENBUS_PROBE_H
+#define _XENBUS_PROBE_H
+
+#define XEN_BUS_ID_SIZE			20
+
+#ifdef CONFIG_XEN_BACKEND
+extern void xenbus_backend_suspend(int (*fn)(struct device *, void *));
+extern void xenbus_backend_resume(int (*fn)(struct device *, void *));
+extern void xenbus_backend_probe_and_watch(void);
+extern int xenbus_backend_bus_register(void);
+extern void xenbus_backend_bus_unregister(void);
+#else
+static inline void xenbus_backend_suspend(int (*fn)(struct device *, void *)) {}
+static inline void xenbus_backend_resume(int (*fn)(struct device *, void *)) {}
+static inline void xenbus_backend_probe_and_watch(void) {}
+static inline int xenbus_backend_bus_register(void) { return 0; }
+static inline void xenbus_backend_bus_unregister(void) {}
+#endif
+
+struct xen_bus_type
+{
+	char *root;
+	unsigned int levels;
+	int (*get_bus_id)(char bus_id[XEN_BUS_ID_SIZE], const char *nodename);
+	int (*probe)(const char *type, const char *dir);
+	struct bus_type bus;
+};
+
+extern int xenbus_match(struct device *_dev, struct device_driver *_drv);
+extern int xenbus_dev_probe(struct device *_dev);
+extern int xenbus_dev_remove(struct device *_dev);
+extern int xenbus_register_driver_common(struct xenbus_driver *drv,
+					 struct xen_bus_type *bus,
+					 struct module *owner,
+					 const char *mod_name);
+extern int xenbus_probe_node(struct xen_bus_type *bus,
+			     const char *type,
+			     const char *nodename);
+extern int xenbus_probe_devices(struct xen_bus_type *bus);
+
+extern void xenbus_dev_changed(const char *node, struct xen_bus_type *bus);
+
+#endif

kernel/drivers/xen/xenbus/xenbus_xs.c

@@ -0,0 +1,864 @@
+/******************************************************************************
+ * xenbus_xs.c
+ *
+ * This is the kernel equivalent of the "xs" library.  We don't need everything
+ * and we use xenbus_comms for communication.
+ *
+ * Copyright (C) 2005 Rusty Russell, IBM Corporation
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/unistd.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/uio.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/err.h>
+#include <linux/slab.h>
+#include <linux/fcntl.h>
+#include <linux/kthread.h>
+#include <linux/rwsem.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <xen/xenbus.h>
+#include "xenbus_comms.h"
+
+struct xs_stored_msg {
+	struct list_head list;
+
+	struct xsd_sockmsg hdr;
+
+	union {
+		/* Queued replies. */
+		struct {
+			char *body;
+		} reply;
+
+		/* Queued watch events. */
+		struct {
+			struct xenbus_watch *handle;
+			char **vec;
+			unsigned int vec_size;
+		} watch;
+	} u;
+};
+
+struct xs_handle {
+	/* A list of replies. Currently only one will ever be outstanding. */
+	struct list_head reply_list;
+	spinlock_t reply_lock;
+	wait_queue_head_t reply_waitq;
+
+	/*
+	 * Mutex ordering: transaction_mutex -> watch_mutex -> request_mutex.
+	 * response_mutex is never taken simultaneously with the other three.
+	 */
+
+	/* One request at a time. */
+	struct mutex request_mutex;
+
+	/* Protect xenbus reader thread against save/restore. */
+	struct mutex response_mutex;
+
+	/* Protect transactions against save/restore. */
+	struct rw_semaphore transaction_mutex;
+
+	/* Protect watch (de)register against save/restore. */
+	struct rw_semaphore watch_mutex;
+};
+
+static struct xs_handle xs_state;
+
+/* List of registered watches, and a lock to protect it. */
+static LIST_HEAD(watches);
+static DEFINE_SPINLOCK(watches_lock);
+
+/* List of pending watch callback events, and a lock to protect it. */
+static LIST_HEAD(watch_events);
+static DEFINE_SPINLOCK(watch_events_lock);
+
+/*
+ * Details of the xenwatch callback kernel thread. The thread waits on the
+ * watch_events_waitq for work to do (queued on watch_events list). When it
+ * wakes up it acquires the xenwatch_mutex before reading the list and
+ * carrying out work.
+ */
+static pid_t xenwatch_pid;
+static DEFINE_MUTEX(xenwatch_mutex);
+static DECLARE_WAIT_QUEUE_HEAD(watch_events_waitq);
+
+static int get_error(const char *errorstring)
+{
+	unsigned int i;
+
+	for (i = 0; strcmp(errorstring, xsd_errors[i].errstring) != 0; i++) {
+		if (i == ARRAY_SIZE(xsd_errors) - 1) {
+			printk(KERN_WARNING
+			       "XENBUS xen store gave: unknown error %s",
+			       errorstring);
+			return EINVAL;
+		}
+	}
+	return xsd_errors[i].errnum;
+}
+
+static void *read_reply(enum xsd_sockmsg_type *type, unsigned int *len)
+{
+	struct xs_stored_msg *msg;
+	char *body;
+
+	spin_lock(&xs_state.reply_lock);
+
+	while (list_empty(&xs_state.reply_list)) {
+		spin_unlock(&xs_state.reply_lock);
+		/* XXX FIXME: Avoid synchronous wait for response here. */
+		wait_event(xs_state.reply_waitq,
+			   !list_empty(&xs_state.reply_list));
+		spin_lock(&xs_state.reply_lock);
+	}
+
+	msg = list_entry(xs_state.reply_list.next,
+			 struct xs_stored_msg, list);
+	list_del(&msg->list);
+
+	spin_unlock(&xs_state.reply_lock);
+
+	*type = msg->hdr.type;
+	if (len)
+		*len = msg->hdr.len;
+	body = msg->u.reply.body;
+
+	kfree(msg);
+
+	return body;
+}
+
+void *xenbus_dev_request_and_reply(struct xsd_sockmsg *msg)
+{
+	void *ret;
+	struct xsd_sockmsg req_msg = *msg;
+	int err;
+
+	if (req_msg.type == XS_TRANSACTION_START)
+		down_read(&xs_state.transaction_mutex);
+
+	mutex_lock(&xs_state.request_mutex);
+
+	err = xb_write(msg, sizeof(*msg) + msg->len);
+	if (err) {
+		msg->type = XS_ERROR;
+		ret = ERR_PTR(err);
+	} else
+		ret = read_reply(&msg->type, &msg->len);
+
+	mutex_unlock(&xs_state.request_mutex);
+
+	if ((msg->type == XS_TRANSACTION_END) ||
+	    ((req_msg.type == XS_TRANSACTION_START) &&
+	     (msg->type == XS_ERROR)))
+		up_read(&xs_state.transaction_mutex);
+
+	return ret;
+}
+EXPORT_SYMBOL(xenbus_dev_request_and_reply);
+
+/* Send message to xs, get kmalloc'ed reply.  ERR_PTR() on error. */
+static void *xs_talkv(struct xenbus_transaction t,
+		      enum xsd_sockmsg_type type,
+		      const struct kvec *iovec,
+		      unsigned int num_vecs,
+		      unsigned int *len)
+{
+	struct xsd_sockmsg msg;
+	void *ret = NULL;
+	unsigned int i;
+	int err;
+
+	msg.tx_id = t.id;
+	msg.req_id = 0;
+	msg.type = type;
+	msg.len = 0;
+	for (i = 0; i < num_vecs; i++)
+		msg.len += iovec[i].iov_len;
+
+	mutex_lock(&xs_state.request_mutex);
+
+	err = xb_write(&msg, sizeof(msg));
+	if (err) {
+		mutex_unlock(&xs_state.request_mutex);
+		return ERR_PTR(err);
+	}
+
+	for (i = 0; i < num_vecs; i++) {
+		err = xb_write(iovec[i].iov_base, iovec[i].iov_len);
+		if (err) {
+			mutex_unlock(&xs_state.request_mutex);
+			return ERR_PTR(err);
+		}
+	}
+
+	ret = read_reply(&msg.type, len);
+
+	mutex_unlock(&xs_state.request_mutex);
+
+	if (IS_ERR(ret))
+		return ret;
+
+	if (msg.type == XS_ERROR) {
+		err = get_error(ret);
+		kfree(ret);
+		return ERR_PTR(-err);
+	}
+
+	if (msg.type != type) {
+		if (printk_ratelimit())
+			printk(KERN_WARNING
+			       "XENBUS unexpected type [%d], expected [%d]\n",
+			       msg.type, type);
+		kfree(ret);
+		return ERR_PTR(-EINVAL);
+	}
+	return ret;
+}
+
+/* Simplified version of xs_talkv: single message. */
+static void *xs_single(struct xenbus_transaction t,
+		       enum xsd_sockmsg_type type,
+		       const char *string,
+		       unsigned int *len)
+{
+	struct kvec iovec;
+
+	iovec.iov_base = (void *)string;
+	iovec.iov_len = strlen(string) + 1;
+	return xs_talkv(t, type, &iovec, 1, len);
+}
+
+/* Many commands only need an ack, don't care what it says. */
+static int xs_error(char *reply)
+{
+	if (IS_ERR(reply))
+		return PTR_ERR(reply);
+	kfree(reply);
+	return 0;
+}
+
+static unsigned int count_strings(const char *strings, unsigned int len)
+{
+	unsigned int num;
+	const char *p;
+
+	for (p = strings, num = 0; p < strings + len; p += strlen(p) + 1)
+		num++;
+
+	return num;
+}
+
+/* Return the path to dir with /name appended. Buffer must be kfree()'ed. */
+static char *join(const char *dir, const char *name)
+{
+	char *buffer;
+
+	if (strlen(name) == 0)
+		buffer = kasprintf(GFP_NOIO | __GFP_HIGH, "%s", dir);
+	else
+		buffer = kasprintf(GFP_NOIO | __GFP_HIGH, "%s/%s", dir, name);
+	return (!buffer) ? ERR_PTR(-ENOMEM) : buffer;
+}
+
+static char **split(char *strings, unsigned int len, unsigned int *num)
+{
+	char *p, **ret;
+
+	/* Count the strings. */
+	*num = count_strings(strings, len);
+
+	/* Transfer to one big alloc for easy freeing. */
+	ret = kmalloc(*num * sizeof(char *) + len, GFP_NOIO | __GFP_HIGH);
+	if (!ret) {
+		kfree(strings);
+		return ERR_PTR(-ENOMEM);
+	}
+	memcpy(&ret[*num], strings, len);
+	kfree(strings);
+
+	strings = (char *)&ret[*num];
+	for (p = strings, *num = 0; p < strings + len; p += strlen(p) + 1)
+		ret[(*num)++] = p;
+
+	return ret;
+}
+
+char **xenbus_directory(struct xenbus_transaction t,
+			const char *dir, const char *node, unsigned int *num)
+{
+	char *strings, *path;
+	unsigned int len;
+
+	path = join(dir, node);
+	if (IS_ERR(path))
+		return (char **)path;
+
+	strings = xs_single(t, XS_DIRECTORY, path, &len);
+	kfree(path);
+	if (IS_ERR(strings))
+		return (char **)strings;
+
+	return split(strings, len, num);
+}
+EXPORT_SYMBOL_GPL(xenbus_directory);
+
+/* Check if a path exists. Return 1 if it does. */
+int xenbus_exists(struct xenbus_transaction t,
+		  const char *dir, const char *node)
+{
+	char **d;
+	int dir_n;
+
+	d = xenbus_directory(t, dir, node, &dir_n);
+	if (IS_ERR(d))
+		return 0;
+	kfree(d);
+	return 1;
+}
+EXPORT_SYMBOL_GPL(xenbus_exists);
+
+/* Get the value of a single file.
+ * Returns a kmalloced value: call free() on it after use.
+ * len indicates length in bytes.
+ */
+void *xenbus_read(struct xenbus_transaction t,
+		  const char *dir, const char *node, unsigned int *len)
+{
+	char *path;
+	void *ret;
+
+	path = join(dir, node);
+	if (IS_ERR(path))
+		return (void *)path;
+
+	ret = xs_single(t, XS_READ, path, len);
+	kfree(path);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xenbus_read);
+
+/* Write the value of a single file.
+ * Returns -err on failure.
+ */
+int xenbus_write(struct xenbus_transaction t,
+		 const char *dir, const char *node, const char *string)
+{
+	const char *path;
+	struct kvec iovec[2];
+	int ret;
+
+	path = join(dir, node);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	iovec[0].iov_base = (void *)path;
+	iovec[0].iov_len = strlen(path) + 1;
+	iovec[1].iov_base = (void *)string;
+	iovec[1].iov_len = strlen(string);
+
+	ret = xs_error(xs_talkv(t, XS_WRITE, iovec, ARRAY_SIZE(iovec), NULL));
+	kfree(path);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xenbus_write);
+
+/* Create a new directory. */
+int xenbus_mkdir(struct xenbus_transaction t,
+		 const char *dir, const char *node)
+{
+	char *path;
+	int ret;
+
+	path = join(dir, node);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	ret = xs_error(xs_single(t, XS_MKDIR, path, NULL));
+	kfree(path);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xenbus_mkdir);
+
+/* Destroy a file or directory (directories must be empty). */
+int xenbus_rm(struct xenbus_transaction t, const char *dir, const char *node)
+{
+	char *path;
+	int ret;
+
+	path = join(dir, node);
+	if (IS_ERR(path))
+		return PTR_ERR(path);
+
+	ret = xs_error(xs_single(t, XS_RM, path, NULL));
+	kfree(path);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xenbus_rm);
+
+/* Start a transaction: changes by others will not be seen during this
+ * transaction, and changes will not be visible to others until end.
+ */
+int xenbus_transaction_start(struct xenbus_transaction *t)
+{
+	char *id_str;
+
+	down_read(&xs_state.transaction_mutex);
+
+	id_str = xs_single(XBT_NIL, XS_TRANSACTION_START, "", NULL);
+	if (IS_ERR(id_str)) {
+		up_read(&xs_state.transaction_mutex);
+		return PTR_ERR(id_str);
+	}
+
+	t->id = simple_strtoul(id_str, NULL, 0);
+	kfree(id_str);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xenbus_transaction_start);
+
+/* End a transaction.
+ * If abandon is true, transaction is discarded instead of committed.
+ */
+int xenbus_transaction_end(struct xenbus_transaction t, int abort)
+{
+	char abortstr[2];
+	int err;
+
+	if (abort)
+		strcpy(abortstr, "F");
+	else
+		strcpy(abortstr, "T");
+
+	err = xs_error(xs_single(t, XS_TRANSACTION_END, abortstr, NULL));
+
+	up_read(&xs_state.transaction_mutex);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(xenbus_transaction_end);
+
+/* Single read and scanf: returns -errno or num scanned. */
+int xenbus_scanf(struct xenbus_transaction t,
+		 const char *dir, const char *node, const char *fmt, ...)
+{
+	va_list ap;
+	int ret;
+	char *val;
+
+	val = xenbus_read(t, dir, node, NULL);
+	if (IS_ERR(val))
+		return PTR_ERR(val);
+
+	va_start(ap, fmt);
+	ret = vsscanf(val, fmt, ap);
+	va_end(ap);
+	kfree(val);
+	/* Distinctive errno. */
+	if (ret == 0)
+		return -ERANGE;
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xenbus_scanf);
+
+/* Single printf and write: returns -errno or 0. */
+int xenbus_printf(struct xenbus_transaction t,
+		  const char *dir, const char *node, const char *fmt, ...)
+{
+	va_list ap;
+	int ret;
+#define PRINTF_BUFFER_SIZE 4096
+	char *printf_buffer;
+
+	printf_buffer = kmalloc(PRINTF_BUFFER_SIZE, GFP_NOIO | __GFP_HIGH);
+	if (printf_buffer == NULL)
+		return -ENOMEM;
+
+	va_start(ap, fmt);
+	ret = vsnprintf(printf_buffer, PRINTF_BUFFER_SIZE, fmt, ap);
+	va_end(ap);
+
+	BUG_ON(ret > PRINTF_BUFFER_SIZE-1);
+	ret = xenbus_write(t, dir, node, printf_buffer);
+
+	kfree(printf_buffer);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xenbus_printf);
+
+/* Takes tuples of names, scanf-style args, and void **, NULL terminated. */
+int xenbus_gather(struct xenbus_transaction t, const char *dir, ...)
+{
+	va_list ap;
+	const char *name;
+	int ret = 0;
+
+	va_start(ap, dir);
+	while (ret == 0 && (name = va_arg(ap, char *)) != NULL) {
+		const char *fmt = va_arg(ap, char *);
+		void *result = va_arg(ap, void *);
+		char *p;
+
+		p = xenbus_read(t, dir, name, NULL);
+		if (IS_ERR(p)) {
+			ret = PTR_ERR(p);
+			break;
+		}
+		if (fmt) {
+			if (sscanf(p, fmt, result) == 0)
+				ret = -EINVAL;
+			kfree(p);
+		} else
+			*(char **)result = p;
+	}
+	va_end(ap);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xenbus_gather);
+
+static int xs_watch(const char *path, const char *token)
+{
+	struct kvec iov[2];
+
+	iov[0].iov_base = (void *)path;
+	iov[0].iov_len = strlen(path) + 1;
+	iov[1].iov_base = (void *)token;
+	iov[1].iov_len = strlen(token) + 1;
+
+	return xs_error(xs_talkv(XBT_NIL, XS_WATCH, iov,
+				 ARRAY_SIZE(iov), NULL));
+}
+
+static int xs_unwatch(const char *path, const char *token)
+{
+	struct kvec iov[2];
+
+	iov[0].iov_base = (char *)path;
+	iov[0].iov_len = strlen(path) + 1;
+	iov[1].iov_base = (char *)token;
+	iov[1].iov_len = strlen(token) + 1;
+
+	return xs_error(xs_talkv(XBT_NIL, XS_UNWATCH, iov,
+				 ARRAY_SIZE(iov), NULL));
+}
+
+static struct xenbus_watch *find_watch(const char *token)
+{
+	struct xenbus_watch *i, *cmp;
+
+	cmp = (void *)simple_strtoul(token, NULL, 16);
+
+	list_for_each_entry(i, &watches, list)
+		if (i == cmp)
+			return i;
+
+	return NULL;
+}
+
+/* Register callback to watch this node. */
+int register_xenbus_watch(struct xenbus_watch *watch)
+{
+	/* Pointer in ascii is the token. */
+	char token[sizeof(watch) * 2 + 1];
+	int err;
+
+	sprintf(token, "%lX", (long)watch);
+
+	down_read(&xs_state.watch_mutex);
+
+	spin_lock(&watches_lock);
+	BUG_ON(find_watch(token));
+	list_add(&watch->list, &watches);
+	spin_unlock(&watches_lock);
+
+	err = xs_watch(watch->node, token);
+
+	/* Ignore errors due to multiple registration. */
+	if ((err != 0) && (err != -EEXIST)) {
+		spin_lock(&watches_lock);
+		list_del(&watch->list);
+		spin_unlock(&watches_lock);
+	}
+
+	up_read(&xs_state.watch_mutex);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(register_xenbus_watch);
+
+void unregister_xenbus_watch(struct xenbus_watch *watch)
+{
+	struct xs_stored_msg *msg, *tmp;
+	char token[sizeof(watch) * 2 + 1];
+	int err;
+
+	sprintf(token, "%lX", (long)watch);
+
+	down_read(&xs_state.watch_mutex);
+
+	spin_lock(&watches_lock);
+	BUG_ON(!find_watch(token));
+	list_del(&watch->list);
+	spin_unlock(&watches_lock);
+
+	err = xs_unwatch(watch->node, token);
+	if (err)
+		printk(KERN_WARNING
+		       "XENBUS Failed to release watch %s: %i\n",
+		       watch->node, err);
+
+	up_read(&xs_state.watch_mutex);
+
+	/* Make sure there are no callbacks running currently (unless
+	   its us) */
+	if (current->pid != xenwatch_pid)
+		mutex_lock(&xenwatch_mutex);
+
+	/* Cancel pending watch events. */
+	spin_lock(&watch_events_lock);
+	list_for_each_entry_safe(msg, tmp, &watch_events, list) {
+		if (msg->u.watch.handle != watch)
+			continue;
+		list_del(&msg->list);
+		kfree(msg->u.watch.vec);
+		kfree(msg);
+	}
+	spin_unlock(&watch_events_lock);
+
+	if (current->pid != xenwatch_pid)
+		mutex_unlock(&xenwatch_mutex);
+}
+EXPORT_SYMBOL_GPL(unregister_xenbus_watch);
+
+void xs_suspend(void)
+{
+	down_write(&xs_state.transaction_mutex);
+	down_write(&xs_state.watch_mutex);
+	mutex_lock(&xs_state.request_mutex);
+	mutex_lock(&xs_state.response_mutex);
+}
+
+void xs_resume(void)
+{
+	struct xenbus_watch *watch;
+	char token[sizeof(watch) * 2 + 1];
+
+	xb_init_comms();
+
+	mutex_unlock(&xs_state.response_mutex);
+	mutex_unlock(&xs_state.request_mutex);
+	up_write(&xs_state.transaction_mutex);
+
+	/* No need for watches_lock: the watch_mutex is sufficient. */
+	list_for_each_entry(watch, &watches, list) {
+		sprintf(token, "%lX", (long)watch);
+		xs_watch(watch->node, token);
+	}
+
+	up_write(&xs_state.watch_mutex);
+}
+
+void xs_suspend_cancel(void)
+{
+	mutex_unlock(&xs_state.response_mutex);
+	mutex_unlock(&xs_state.request_mutex);
+	up_write(&xs_state.watch_mutex);
+	up_write(&xs_state.transaction_mutex);
+}
+
+static int xenwatch_thread(void *unused)
+{
+	struct list_head *ent;
+	struct xs_stored_msg *msg;
+
+	for (;;) {
+		wait_event_interruptible(watch_events_waitq,
+					 !list_empty(&watch_events));
+
+		if (kthread_should_stop())
+			break;
+
+		mutex_lock(&xenwatch_mutex);
+
+		spin_lock(&watch_events_lock);
+		ent = watch_events.next;
+		if (ent != &watch_events)
+			list_del(ent);
+		spin_unlock(&watch_events_lock);
+
+		if (ent != &watch_events) {
+			msg = list_entry(ent, struct xs_stored_msg, list);
+			msg->u.watch.handle->callback(
+				msg->u.watch.handle,
+				(const char **)msg->u.watch.vec,
+				msg->u.watch.vec_size);
+			kfree(msg->u.watch.vec);
+			kfree(msg);
+		}
+
+		mutex_unlock(&xenwatch_mutex);
+	}
+
+	return 0;
+}
+
+static int process_msg(void)
+{
+	struct xs_stored_msg *msg;
+	char *body;
+	int err;
+
+	/*
+	 * We must disallow save/restore while reading a xenstore message.
+	 * A partial read across s/r leaves us out of sync with xenstored.
+	 */
+	for (;;) {
+		err = xb_wait_for_data_to_read();
+		if (err)
+			return err;
+		mutex_lock(&xs_state.response_mutex);
+		if (xb_data_to_read())
+			break;
+		/* We raced with save/restore: pending data 'disappeared'. */
+		mutex_unlock(&xs_state.response_mutex);
+	}
+
+
+	msg = kmalloc(sizeof(*msg), GFP_NOIO | __GFP_HIGH);
+	if (msg == NULL) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	err = xb_read(&msg->hdr, sizeof(msg->hdr));
+	if (err) {
+		kfree(msg);
+		goto out;
+	}
+
+	body = kmalloc(msg->hdr.len + 1, GFP_NOIO | __GFP_HIGH);
+	if (body == NULL) {
+		kfree(msg);
+		err = -ENOMEM;
+		goto out;
+	}
+
+	err = xb_read(body, msg->hdr.len);
+	if (err) {
+		kfree(body);
+		kfree(msg);
+		goto out;
+	}
+	body[msg->hdr.len] = '\0';
+
+	if (msg->hdr.type == XS_WATCH_EVENT) {
+		msg->u.watch.vec = split(body, msg->hdr.len,
+					 &msg->u.watch.vec_size);
+		if (IS_ERR(msg->u.watch.vec)) {
+			err = PTR_ERR(msg->u.watch.vec);
+			kfree(msg);
+			goto out;
+		}
+
+		spin_lock(&watches_lock);
+		msg->u.watch.handle = find_watch(
+			msg->u.watch.vec[XS_WATCH_TOKEN]);
+		if (msg->u.watch.handle != NULL) {
+			spin_lock(&watch_events_lock);
+			list_add_tail(&msg->list, &watch_events);
+			wake_up(&watch_events_waitq);
+			spin_unlock(&watch_events_lock);
+		} else {
+			kfree(msg->u.watch.vec);
+			kfree(msg);
+		}
+		spin_unlock(&watches_lock);
+	} else {
+		msg->u.reply.body = body;
+		spin_lock(&xs_state.reply_lock);
+		list_add_tail(&msg->list, &xs_state.reply_list);
+		spin_unlock(&xs_state.reply_lock);
+		wake_up(&xs_state.reply_waitq);
+	}
+
+ out:
+	mutex_unlock(&xs_state.response_mutex);
+	return err;
+}
+
+static int xenbus_thread(void *unused)
+{
+	int err;
+
+	for (;;) {
+		err = process_msg();
+		if (err)
+			printk(KERN_WARNING "XENBUS error %d while reading "
+			       "message\n", err);
+		if (kthread_should_stop())
+			break;
+	}
+
+	return 0;
+}
+
+int xs_init(void)
+{
+	int err;
+	struct task_struct *task;
+
+	INIT_LIST_HEAD(&xs_state.reply_list);
+	spin_lock_init(&xs_state.reply_lock);
+	init_waitqueue_head(&xs_state.reply_waitq);
+
+	mutex_init(&xs_state.request_mutex);
+	mutex_init(&xs_state.response_mutex);
+	init_rwsem(&xs_state.transaction_mutex);
+	init_rwsem(&xs_state.watch_mutex);
+
+	/* Initialize the shared memory rings to talk to xenstored */
+	err = xb_init_comms();
+	if (err)
+		return err;
+
+	task = kthread_run(xenwatch_thread, NULL, "xenwatch");
+	if (IS_ERR(task))
+		return PTR_ERR(task);
+	xenwatch_pid = task->pid;
+
+	task = kthread_run(xenbus_thread, NULL, "xenbus");
+	if (IS_ERR(task))
+		return PTR_ERR(task);
+
+	return 0;
+}

kernel/drivers/xen/xencomm.c

@@ -0,0 +1,217 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * 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.
+ *
+ * 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
+ *
+ * Copyright (C) IBM Corp. 2006
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#include <linux/gfp.h>
+#include <linux/mm.h>
+#include <asm/page.h>
+#include <xen/xencomm.h>
+#include <xen/interface/xen.h>
+#include <asm/xen/xencomm.h>	/* for xencomm_is_phys_contiguous() */
+
+static int xencomm_init(struct xencomm_desc *desc,
+			void *buffer, unsigned long bytes)
+{
+	unsigned long recorded = 0;
+	int i = 0;
+
+	while ((recorded < bytes) && (i < desc->nr_addrs)) {
+		unsigned long vaddr = (unsigned long)buffer + recorded;
+		unsigned long paddr;
+		int offset;
+		int chunksz;
+
+		offset = vaddr % PAGE_SIZE; /* handle partial pages */
+		chunksz = min(PAGE_SIZE - offset, bytes - recorded);
+
+		paddr = xencomm_vtop(vaddr);
+		if (paddr == ~0UL) {
+			printk(KERN_DEBUG "%s: couldn't translate vaddr %lx\n",
+			       __func__, vaddr);
+			return -EINVAL;
+		}
+
+		desc->address[i++] = paddr;
+		recorded += chunksz;
+	}
+
+	if (recorded < bytes) {
+		printk(KERN_DEBUG
+		       "%s: could only translate %ld of %ld bytes\n",
+		       __func__, recorded, bytes);
+		return -ENOSPC;
+	}
+
+	/* mark remaining addresses invalid (just for safety) */
+	while (i < desc->nr_addrs)
+		desc->address[i++] = XENCOMM_INVALID;
+
+	desc->magic = XENCOMM_MAGIC;
+
+	return 0;
+}
+
+static struct xencomm_desc *xencomm_alloc(gfp_t gfp_mask,
+					  void *buffer, unsigned long bytes)
+{
+	struct xencomm_desc *desc;
+	unsigned long buffer_ulong = (unsigned long)buffer;
+	unsigned long start = buffer_ulong & PAGE_MASK;
+	unsigned long end = (buffer_ulong + bytes) | ~PAGE_MASK;
+	unsigned long nr_addrs = (end - start + 1) >> PAGE_SHIFT;
+	unsigned long size = sizeof(*desc) +
+		sizeof(desc->address[0]) * nr_addrs;
+
+	/*
+	 * slab allocator returns at least sizeof(void*) aligned pointer.
+	 * When sizeof(*desc) > sizeof(void*), struct xencomm_desc might
+	 * cross page boundary.
+	 */
+	if (sizeof(*desc) > sizeof(void *)) {
+		unsigned long order = get_order(size);
+		desc = (struct xencomm_desc *)__get_free_pages(gfp_mask,
+							       order);
+		if (desc == NULL)
+			return NULL;
+
+		desc->nr_addrs =
+			((PAGE_SIZE << order) - sizeof(struct xencomm_desc)) /
+			sizeof(*desc->address);
+	} else {
+		desc = kmalloc(size, gfp_mask);
+		if (desc == NULL)
+			return NULL;
+
+		desc->nr_addrs = nr_addrs;
+	}
+	return desc;
+}
+
+void xencomm_free(struct xencomm_handle *desc)
+{
+	if (desc && !((ulong)desc & XENCOMM_INLINE_FLAG)) {
+		struct xencomm_desc *desc__ = (struct xencomm_desc *)desc;
+		if (sizeof(*desc__) > sizeof(void *)) {
+			unsigned long size = sizeof(*desc__) +
+				sizeof(desc__->address[0]) * desc__->nr_addrs;
+			unsigned long order = get_order(size);
+			free_pages((unsigned long)__va(desc), order);
+		} else
+			kfree(__va(desc));
+	}
+}
+
+static int xencomm_create(void *buffer, unsigned long bytes,
+			  struct xencomm_desc **ret, gfp_t gfp_mask)
+{
+	struct xencomm_desc *desc;
+	int rc;
+
+	pr_debug("%s: %p[%ld]\n", __func__, buffer, bytes);
+
+	if (bytes == 0) {
+		/* don't create a descriptor; Xen recognizes NULL. */
+		BUG_ON(buffer != NULL);
+		*ret = NULL;
+		return 0;
+	}
+
+	BUG_ON(buffer == NULL); /* 'bytes' is non-zero */
+
+	desc = xencomm_alloc(gfp_mask, buffer, bytes);
+	if (!desc) {
+		printk(KERN_DEBUG "%s failure\n", "xencomm_alloc");
+		return -ENOMEM;
+	}
+
+	rc = xencomm_init(desc, buffer, bytes);
+	if (rc) {
+		printk(KERN_DEBUG "%s failure: %d\n", "xencomm_init", rc);
+		xencomm_free((struct xencomm_handle *)__pa(desc));
+		return rc;
+	}
+
+	*ret = desc;
+	return 0;
+}
+
+static struct xencomm_handle *xencomm_create_inline(void *ptr)
+{
+	unsigned long paddr;
+
+	BUG_ON(!xencomm_is_phys_contiguous((unsigned long)ptr));
+
+	paddr = (unsigned long)xencomm_pa(ptr);
+	BUG_ON(paddr & XENCOMM_INLINE_FLAG);
+	return (struct xencomm_handle *)(paddr | XENCOMM_INLINE_FLAG);
+}
+
+/* "mini" routine, for stack-based communications: */
+static int xencomm_create_mini(void *buffer,
+	unsigned long bytes, struct xencomm_mini *xc_desc,
+	struct xencomm_desc **ret)
+{
+	int rc = 0;
+	struct xencomm_desc *desc;
+	BUG_ON(((unsigned long)xc_desc) % sizeof(*xc_desc) != 0);
+
+	desc = (void *)xc_desc;
+
+	desc->nr_addrs = XENCOMM_MINI_ADDRS;
+
+	rc = xencomm_init(desc, buffer, bytes);
+	if (!rc)
+		*ret = desc;
+
+	return rc;
+}
+
+struct xencomm_handle *xencomm_map(void *ptr, unsigned long bytes)
+{
+	int rc;
+	struct xencomm_desc *desc;
+
+	if (xencomm_is_phys_contiguous((unsigned long)ptr))
+		return xencomm_create_inline(ptr);
+
+	rc = xencomm_create(ptr, bytes, &desc, GFP_KERNEL);
+
+	if (rc || desc == NULL)
+		return NULL;
+
+	return xencomm_pa(desc);
+}
+
+struct xencomm_handle *__xencomm_map_no_alloc(void *ptr, unsigned long bytes,
+			struct xencomm_mini *xc_desc)
+{
+	int rc;
+	struct xencomm_desc *desc = NULL;
+
+	if (xencomm_is_phys_contiguous((unsigned long)ptr))
+		return xencomm_create_inline(ptr);
+
+	rc = xencomm_create_mini(ptr, bytes, xc_desc,
+				&desc);
+
+	if (rc)
+		return NULL;
+
+	return xencomm_pa(desc);
+}

kernel/drivers/xen/xenfs/Makefile

@@ -0,0 +1,3 @@
+obj-$(CONFIG_XENFS) += xenfs.o
+
+xenfs-objs = super.o xenbus.o

kernel/drivers/xen/xenfs/super.c

@@ -0,0 +1,81 @@
+/*
+ *  xenfs.c - a filesystem for passing info between the a domain and
+ *  the hypervisor.
+ *
+ * 2008-10-07  Alex Zeffertt    Replaced /proc/xen/xenbus with xenfs filesystem
+ *                              and /proc/xen compatibility mount point.
+ *                              Turned xenfs into a loadable module.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/magic.h>
+
+#include "xenfs.h"
+
+#include <asm/xen/hypervisor.h>
+
+MODULE_DESCRIPTION("Xen filesystem");
+MODULE_LICENSE("GPL");
+
+static ssize_t capabilities_read(struct file *file, char __user *buf,
+				 size_t size, loff_t *off)
+{
+	char *tmp = "";
+
+	if (xen_initial_domain())
+		tmp = "control_d\n";
+
+	return simple_read_from_buffer(buf, size, off, tmp, strlen(tmp));
+}
+
+static const struct file_operations capabilities_file_ops = {
+	.read = capabilities_read,
+};
+
+static int xenfs_fill_super(struct super_block *sb, void *data, int silent)
+{
+	static struct tree_descr xenfs_files[] = {
+		[1] = {},
+		{ "xenbus", &xenbus_file_ops, S_IRUSR|S_IWUSR },
+		{ "capabilities", &capabilities_file_ops, S_IRUGO },
+		{""},
+	};
+
+	return simple_fill_super(sb, XENFS_SUPER_MAGIC, xenfs_files);
+}
+
+static int xenfs_get_sb(struct file_system_type *fs_type,
+			int flags, const char *dev_name,
+			void *data, struct vfsmount *mnt)
+{
+	return get_sb_single(fs_type, flags, data, xenfs_fill_super, mnt);
+}
+
+static struct file_system_type xenfs_type = {
+	.owner =	THIS_MODULE,
+	.name =		"xenfs",
+	.get_sb =	xenfs_get_sb,
+	.kill_sb =	kill_litter_super,
+};
+
+static int __init xenfs_init(void)
+{
+	if (xen_pv_domain())
+		return register_filesystem(&xenfs_type);
+
+	printk(KERN_INFO "XENFS: not registering filesystem on non-xen platform\n");
+	return 0;
+}
+
+static void __exit xenfs_exit(void)
+{
+	if (xen_pv_domain())
+		unregister_filesystem(&xenfs_type);
+}
+
+module_init(xenfs_init);
+module_exit(xenfs_exit);
+

kernel/drivers/xen/xenfs/xenbus.c

@@ -0,0 +1,593 @@
+/*
+ * Driver giving user-space access to the kernel's xenbus connection
+ * to xenstore.
+ *
+ * Copyright (c) 2005, Christian Limpach
+ * Copyright (c) 2005, Rusty Russell, IBM Corporation
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Changes:
+ * 2008-10-07  Alex Zeffertt    Replaced /proc/xen/xenbus with xenfs filesystem
+ *                              and /proc/xen compatibility mount point.
+ *                              Turned xenfs into a loadable module.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/uio.h>
+#include <linux/notifier.h>
+#include <linux/wait.h>
+#include <linux/fs.h>
+#include <linux/poll.h>
+#include <linux/mutex.h>
+#include <linux/sched.h>
+#include <linux/spinlock.h>
+#include <linux/mount.h>
+#include <linux/pagemap.h>
+#include <linux/uaccess.h>
+#include <linux/init.h>
+#include <linux/namei.h>
+#include <linux/string.h>
+
+#include "xenfs.h"
+#include "../xenbus/xenbus_comms.h"
+
+#include <xen/xenbus.h>
+#include <asm/xen/hypervisor.h>
+
+/*
+ * An element of a list of outstanding transactions, for which we're
+ * still waiting a reply.
+ */
+struct xenbus_transaction_holder {
+	struct list_head list;
+	struct xenbus_transaction handle;
+};
+
+/*
+ * A buffer of data on the queue.
+ */
+struct read_buffer {
+	struct list_head list;
+	unsigned int cons;
+	unsigned int len;
+	char msg[];
+};
+
+struct xenbus_file_priv {
+	/*
+	 * msgbuffer_mutex is held while partial requests are built up
+	 * and complete requests are acted on.  It therefore protects
+	 * the "transactions" and "watches" lists, and the partial
+	 * request length and buffer.
+	 *
+	 * reply_mutex protects the reply being built up to return to
+	 * usermode.  It nests inside msgbuffer_mutex but may be held
+	 * alone during a watch callback.
+	 */
+	struct mutex msgbuffer_mutex;
+
+	/* In-progress transactions */
+	struct list_head transactions;
+
+	/* Active watches. */
+	struct list_head watches;
+
+	/* Partial request. */
+	unsigned int len;
+	union {
+		struct xsd_sockmsg msg;
+		char buffer[PAGE_SIZE];
+	} u;
+
+	/* Response queue. */
+	struct mutex reply_mutex;
+	struct list_head read_buffers;
+	wait_queue_head_t read_waitq;
+
+};
+
+/* Read out any raw xenbus messages queued up. */
+static ssize_t xenbus_file_read(struct file *filp,
+			       char __user *ubuf,
+			       size_t len, loff_t *ppos)
+{
+	struct xenbus_file_priv *u = filp->private_data;
+	struct read_buffer *rb;
+	unsigned i;
+	int ret;
+
+	mutex_lock(&u->reply_mutex);
+	while (list_empty(&u->read_buffers)) {
+		mutex_unlock(&u->reply_mutex);
+		ret = wait_event_interruptible(u->read_waitq,
+					       !list_empty(&u->read_buffers));
+		if (ret)
+			return ret;
+		mutex_lock(&u->reply_mutex);
+	}
+
+	rb = list_entry(u->read_buffers.next, struct read_buffer, list);
+	i = 0;
+	while (i < len) {
+		unsigned sz = min((unsigned)len - i, rb->len - rb->cons);
+
+		ret = copy_to_user(ubuf + i, &rb->msg[rb->cons], sz);
+
+		i += sz - ret;
+		rb->cons += sz - ret;
+
+		if (ret != sz) {
+			if (i == 0)
+				i = -EFAULT;
+			goto out;
+		}
+
+		/* Clear out buffer if it has been consumed */
+		if (rb->cons == rb->len) {
+			list_del(&rb->list);
+			kfree(rb);
+			if (list_empty(&u->read_buffers))
+				break;
+			rb = list_entry(u->read_buffers.next,
+					struct read_buffer, list);
+		}
+	}
+
+out:
+	mutex_unlock(&u->reply_mutex);
+	return i;
+}
+
+/*
+ * Add a buffer to the queue.  Caller must hold the appropriate lock
+ * if the queue is not local.  (Commonly the caller will build up
+ * multiple queued buffers on a temporary local list, and then add it
+ * to the appropriate list under lock once all the buffers have een
+ * successfully allocated.)
+ */
+static int queue_reply(struct list_head *queue, const void *data, size_t len)
+{
+	struct read_buffer *rb;
+
+	if (len == 0)
+		return 0;
+
+	rb = kmalloc(sizeof(*rb) + len, GFP_KERNEL);
+	if (rb == NULL)
+		return -ENOMEM;
+
+	rb->cons = 0;
+	rb->len = len;
+
+	memcpy(rb->msg, data, len);
+
+	list_add_tail(&rb->list, queue);
+	return 0;
+}
+
+/*
+ * Free all the read_buffer s on a list.
+ * Caller must have sole reference to list.
+ */
+static void queue_cleanup(struct list_head *list)
+{
+	struct read_buffer *rb;
+
+	while (!list_empty(list)) {
+		rb = list_entry(list->next, struct read_buffer, list);
+		list_del(list->next);
+		kfree(rb);
+	}
+}
+
+struct watch_adapter {
+	struct list_head list;
+	struct xenbus_watch watch;
+	struct xenbus_file_priv *dev_data;
+	char *token;
+};
+
+static void free_watch_adapter(struct watch_adapter *watch)
+{
+	kfree(watch->watch.node);
+	kfree(watch->token);
+	kfree(watch);
+}
+
+static struct watch_adapter *alloc_watch_adapter(const char *path,
+						 const char *token)
+{
+	struct watch_adapter *watch;
+
+	watch = kzalloc(sizeof(*watch), GFP_KERNEL);
+	if (watch == NULL)
+		goto out_fail;
+
+	watch->watch.node = kstrdup(path, GFP_KERNEL);
+	if (watch->watch.node == NULL)
+		goto out_free;
+
+	watch->token = kstrdup(token, GFP_KERNEL);
+	if (watch->token == NULL)
+		goto out_free;
+
+	return watch;
+
+out_free:
+	free_watch_adapter(watch);
+
+out_fail:
+	return NULL;
+}
+
+static void watch_fired(struct xenbus_watch *watch,
+			const char **vec,
+			unsigned int len)
+{
+	struct watch_adapter *adap;
+	struct xsd_sockmsg hdr;
+	const char *path, *token;
+	int path_len, tok_len, body_len, data_len = 0;
+	int ret;
+	LIST_HEAD(staging_q);
+
+	adap = container_of(watch, struct watch_adapter, watch);
+
+	path = vec[XS_WATCH_PATH];
+	token = adap->token;
+
+	path_len = strlen(path) + 1;
+	tok_len = strlen(token) + 1;
+	if (len > 2)
+		data_len = vec[len] - vec[2] + 1;
+	body_len = path_len + tok_len + data_len;
+
+	hdr.type = XS_WATCH_EVENT;
+	hdr.len = body_len;
+
+	mutex_lock(&adap->dev_data->reply_mutex);
+
+	ret = queue_reply(&staging_q, &hdr, sizeof(hdr));
+	if (!ret)
+		ret = queue_reply(&staging_q, path, path_len);
+	if (!ret)
+		ret = queue_reply(&staging_q, token, tok_len);
+	if (!ret && len > 2)
+		ret = queue_reply(&staging_q, vec[2], data_len);
+
+	if (!ret) {
+		/* success: pass reply list onto watcher */
+		list_splice_tail(&staging_q, &adap->dev_data->read_buffers);
+		wake_up(&adap->dev_data->read_waitq);
+	} else
+		queue_cleanup(&staging_q);
+
+	mutex_unlock(&adap->dev_data->reply_mutex);
+}
+
+static int xenbus_write_transaction(unsigned msg_type,
+				    struct xenbus_file_priv *u)
+{
+	int rc;
+	void *reply;
+	struct xenbus_transaction_holder *trans = NULL;
+	LIST_HEAD(staging_q);
+
+	if (msg_type == XS_TRANSACTION_START) {
+		trans = kmalloc(sizeof(*trans), GFP_KERNEL);
+		if (!trans) {
+			rc = -ENOMEM;
+			goto out;
+		}
+	}
+
+	reply = xenbus_dev_request_and_reply(&u->u.msg);
+	if (IS_ERR(reply)) {
+		kfree(trans);
+		rc = PTR_ERR(reply);
+		goto out;
+	}
+
+	if (msg_type == XS_TRANSACTION_START) {
+		trans->handle.id = simple_strtoul(reply, NULL, 0);
+
+		list_add(&trans->list, &u->transactions);
+	} else if (msg_type == XS_TRANSACTION_END) {
+		list_for_each_entry(trans, &u->transactions, list)
+			if (trans->handle.id == u->u.msg.tx_id)
+				break;
+		BUG_ON(&trans->list == &u->transactions);
+		list_del(&trans->list);
+
+		kfree(trans);
+	}
+
+	mutex_lock(&u->reply_mutex);
+	rc = queue_reply(&staging_q, &u->u.msg, sizeof(u->u.msg));
+	if (!rc)
+		rc = queue_reply(&staging_q, reply, u->u.msg.len);
+	if (!rc) {
+		list_splice_tail(&staging_q, &u->read_buffers);
+		wake_up(&u->read_waitq);
+	} else {
+		queue_cleanup(&staging_q);
+	}
+	mutex_unlock(&u->reply_mutex);
+
+	kfree(reply);
+
+out:
+	return rc;
+}
+
+static int xenbus_write_watch(unsigned msg_type, struct xenbus_file_priv *u)
+{
+	struct watch_adapter *watch, *tmp_watch;
+	char *path, *token;
+	int err, rc;
+	LIST_HEAD(staging_q);
+
+	path = u->u.buffer + sizeof(u->u.msg);
+	token = memchr(path, 0, u->u.msg.len);
+	if (token == NULL) {
+		rc = -EILSEQ;
+		goto out;
+	}
+	token++;
+
+	if (msg_type == XS_WATCH) {
+		watch = alloc_watch_adapter(path, token);
+		if (watch == NULL) {
+			rc = -ENOMEM;
+			goto out;
+		}
+
+		watch->watch.callback = watch_fired;
+		watch->dev_data = u;
+
+		err = register_xenbus_watch(&watch->watch);
+		if (err) {
+			free_watch_adapter(watch);
+			rc = err;
+			goto out;
+		}
+		list_add(&watch->list, &u->watches);
+	} else {
+		list_for_each_entry_safe(watch, tmp_watch, &u->watches, list) {
+			if (!strcmp(watch->token, token) &&
+			    !strcmp(watch->watch.node, path)) {
+				unregister_xenbus_watch(&watch->watch);
+				list_del(&watch->list);
+				free_watch_adapter(watch);
+				break;
+			}
+		}
+	}
+
+	/* Success.  Synthesize a reply to say all is OK. */
+	{
+		struct {
+			struct xsd_sockmsg hdr;
+			char body[3];
+		} __packed reply = {
+			{
+				.type = msg_type,
+				.len = sizeof(reply.body)
+			},
+			"OK"
+		};
+
+		mutex_lock(&u->reply_mutex);
+		rc = queue_reply(&u->read_buffers, &reply, sizeof(reply));
+		mutex_unlock(&u->reply_mutex);
+	}
+
+out:
+	return rc;
+}
+
+static ssize_t xenbus_file_write(struct file *filp,
+				const char __user *ubuf,
+				size_t len, loff_t *ppos)
+{
+	struct xenbus_file_priv *u = filp->private_data;
+	uint32_t msg_type;
+	int rc = len;
+	int ret;
+	LIST_HEAD(staging_q);
+
+	/*
+	 * We're expecting usermode to be writing properly formed
+	 * xenbus messages.  If they write an incomplete message we
+	 * buffer it up.  Once it is complete, we act on it.
+	 */
+
+	/*
+	 * Make sure concurrent writers can't stomp all over each
+	 * other's messages and make a mess of our partial message
+	 * buffer.  We don't make any attemppt to stop multiple
+	 * writers from making a mess of each other's incomplete
+	 * messages; we're just trying to guarantee our own internal
+	 * consistency and make sure that single writes are handled
+	 * atomically.
+	 */
+	mutex_lock(&u->msgbuffer_mutex);
+
+	/* Get this out of the way early to avoid confusion */
+	if (len == 0)
+		goto out;
+
+	/* Can't write a xenbus message larger we can buffer */
+	if ((len + u->len) > sizeof(u->u.buffer)) {
+		/* On error, dump existing buffer */
+		u->len = 0;
+		rc = -EINVAL;
+		goto out;
+	}
+
+	ret = copy_from_user(u->u.buffer + u->len, ubuf, len);
+
+	if (ret == len) {
+		rc = -EFAULT;
+		goto out;
+	}
+
+	/* Deal with a partial copy. */
+	len -= ret;
+	rc = len;
+
+	u->len += len;
+
+	/* Return if we haven't got a full message yet */
+	if (u->len < sizeof(u->u.msg))
+		goto out;	/* not even the header yet */
+
+	/* If we're expecting a message that's larger than we can
+	   possibly send, dump what we have and return an error. */
+	if ((sizeof(u->u.msg) + u->u.msg.len) > sizeof(u->u.buffer)) {
+		rc = -E2BIG;
+		u->len = 0;
+		goto out;
+	}
+
+	if (u->len < (sizeof(u->u.msg) + u->u.msg.len))
+		goto out;	/* incomplete data portion */
+
+	/*
+	 * OK, now we have a complete message.  Do something with it.
+	 */
+
+	msg_type = u->u.msg.type;
+
+	switch (msg_type) {
+	case XS_TRANSACTION_START:
+	case XS_TRANSACTION_END:
+	case XS_DIRECTORY:
+	case XS_READ:
+	case XS_GET_PERMS:
+	case XS_RELEASE:
+	case XS_GET_DOMAIN_PATH:
+	case XS_WRITE:
+	case XS_MKDIR:
+	case XS_RM:
+	case XS_SET_PERMS:
+		/* Send out a transaction */
+		ret = xenbus_write_transaction(msg_type, u);
+		break;
+
+	case XS_WATCH:
+	case XS_UNWATCH:
+		/* (Un)Ask for some path to be watched for changes */
+		ret = xenbus_write_watch(msg_type, u);
+		break;
+
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	if (ret != 0)
+		rc = ret;
+
+	/* Buffered message consumed */
+	u->len = 0;
+
+ out:
+	mutex_unlock(&u->msgbuffer_mutex);
+	return rc;
+}
+
+static int xenbus_file_open(struct inode *inode, struct file *filp)
+{
+	struct xenbus_file_priv *u;
+
+	if (xen_store_evtchn == 0)
+		return -ENOENT;
+
+	nonseekable_open(inode, filp);
+
+	u = kzalloc(sizeof(*u), GFP_KERNEL);
+	if (u == NULL)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&u->transactions);
+	INIT_LIST_HEAD(&u->watches);
+	INIT_LIST_HEAD(&u->read_buffers);
+	init_waitqueue_head(&u->read_waitq);
+
+	mutex_init(&u->reply_mutex);
+	mutex_init(&u->msgbuffer_mutex);
+
+	filp->private_data = u;
+
+	return 0;
+}
+
+static int xenbus_file_release(struct inode *inode, struct file *filp)
+{
+	struct xenbus_file_priv *u = filp->private_data;
+	struct xenbus_transaction_holder *trans, *tmp;
+	struct watch_adapter *watch, *tmp_watch;
+
+	/*
+	 * No need for locking here because there are no other users,
+	 * by definition.
+	 */
+
+	list_for_each_entry_safe(trans, tmp, &u->transactions, list) {
+		xenbus_transaction_end(trans->handle, 1);
+		list_del(&trans->list);
+		kfree(trans);
+	}
+
+	list_for_each_entry_safe(watch, tmp_watch, &u->watches, list) {
+		unregister_xenbus_watch(&watch->watch);
+		list_del(&watch->list);
+		free_watch_adapter(watch);
+	}
+
+	kfree(u);
+
+	return 0;
+}
+
+static unsigned int xenbus_file_poll(struct file *file, poll_table *wait)
+{
+	struct xenbus_file_priv *u = file->private_data;
+
+	poll_wait(file, &u->read_waitq, wait);
+	if (!list_empty(&u->read_buffers))
+		return POLLIN | POLLRDNORM;
+	return 0;
+}
+
+const struct file_operations xenbus_file_ops = {
+	.read = xenbus_file_read,
+	.write = xenbus_file_write,
+	.open = xenbus_file_open,
+	.release = xenbus_file_release,
+	.poll = xenbus_file_poll,
+};

kernel/drivers/xen/xenfs/xenfs.h

@@ -0,0 +1,6 @@
+#ifndef _XENFS_XENBUS_H
+#define _XENFS_XENBUS_H
+
+extern const struct file_operations xenbus_file_ops;
+
+#endif	/* _XENFS_XENBUS_H */