Merge branch 'stable/for-linus-3.3' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen

* 'stable/for-linus-3.3' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen: (37 commits)
  xen/pciback: Expand the warning message to include domain id.
  xen/pciback: Fix "device has been assigned to X domain!" warning
  xen/pciback: Move the PCI_DEV_FLAGS_ASSIGNED ops to the "[un|]bind"
  xen/xenbus: don't reimplement kvasprintf via a fixed size buffer
  xenbus: maximum buffer size is XENSTORE_PAYLOAD_MAX
  xen/xenbus: Reject replies with payload > XENSTORE_PAYLOAD_MAX.
  Xen: consolidate and simplify struct xenbus_driver instantiation
  xen-gntalloc: introduce missing kfree
  xen/xenbus: Fix compile error - missing header for xen_initial_domain()
  xen/netback: Enable netback on HVM guests
  xen/grant-table: Support mappings required by blkback
  xenbus: Use grant-table wrapper functions
  xenbus: Support HVM backends
  xen/xenbus-frontend: Fix compile error with randconfig
  xen/xenbus-frontend: Make error message more clear
  xen/privcmd: Remove unused support for arch specific privcmp mmap
  xen: Add xenbus_backend device
  xen: Add xenbus device driver
  xen: Add privcmd device driver
  xen/gntalloc: fix reference counts on multi-page mappings
  ...

drivers/xen/xenbus/Makefile

@@ -1,4 +1,5 @@
 obj-y	+= xenbus.o
+obj-y	+= xenbus_dev_frontend.o
 
 xenbus-objs =
 xenbus-objs += xenbus_client.o
@@ -9,4 +10,5 @@ xenbus-objs += xenbus_probe.o
 xenbus-be-objs-$(CONFIG_XEN_BACKEND) += xenbus_probe_backend.o
 xenbus-objs += $(xenbus-be-objs-y)
 
+obj-$(CONFIG_XEN_BACKEND) += xenbus_dev_backend.o
 obj-$(CONFIG_XEN_XENBUS_FRONTEND) += xenbus_probe_frontend.o

drivers/xen/xenbus/xenbus_client.c

@@ -32,15 +32,39 @@
 
 #include <linux/slab.h>
 #include <linux/types.h>
+#include <linux/spinlock.h>
 #include <linux/vmalloc.h>
 #include <linux/export.h>
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/page.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/event_channel.h>
+#include <xen/balloon.h>
 #include <xen/events.h>
 #include <xen/grant_table.h>
 #include <xen/xenbus.h>
+#include <xen/xen.h>
+
+#include "xenbus_probe.h"
+
+struct xenbus_map_node {
+	struct list_head next;
+	union {
+		struct vm_struct *area; /* PV */
+		struct page *page;     /* HVM */
+	};
+	grant_handle_t handle;
+};
+
+static DEFINE_SPINLOCK(xenbus_valloc_lock);
+static LIST_HEAD(xenbus_valloc_pages);
+
+struct xenbus_ring_ops {
+	int (*map)(struct xenbus_device *dev, int gnt, void **vaddr);
+	int (*unmap)(struct xenbus_device *dev, void *vaddr);
+};
+
+static const struct xenbus_ring_ops *ring_ops __read_mostly;
 
 const char *xenbus_strstate(enum xenbus_state state)
 {
@@ -435,21 +459,35 @@ EXPORT_SYMBOL_GPL(xenbus_free_evtchn);
  * XenbusStateClosing and the error message will be saved in XenStore.
  */
 int xenbus_map_ring_valloc(struct xenbus_device *dev, int gnt_ref, void **vaddr)
+{
+	return ring_ops->map(dev, gnt_ref, vaddr);
+}
+EXPORT_SYMBOL_GPL(xenbus_map_ring_valloc);
+
+static int xenbus_map_ring_valloc_pv(struct xenbus_device *dev,
+				     int gnt_ref, void **vaddr)
 {
 	struct gnttab_map_grant_ref op = {
 		.flags = GNTMAP_host_map | GNTMAP_contains_pte,
 		.ref   = gnt_ref,
 		.dom   = dev->otherend_id,
 	};
+	struct xenbus_map_node *node;
 	struct vm_struct *area;
 	pte_t *pte;
 
 	*vaddr = NULL;
 
-	area = alloc_vm_area(PAGE_SIZE, &pte);
-	if (!area)
+	node = kzalloc(sizeof(*node), GFP_KERNEL);
+	if (!node)
 		return -ENOMEM;
 
+	area = alloc_vm_area(PAGE_SIZE, &pte);
+	if (!area) {
+		kfree(node);
+		return -ENOMEM;
+	}
+
 	op.host_addr = arbitrary_virt_to_machine(pte).maddr;
 
 	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
@@ -457,19 +495,59 @@ int xenbus_map_ring_valloc(struct xenbus_device *dev, int gnt_ref, void **vaddr)
 
 	if (op.status != GNTST_okay) {
 		free_vm_area(area);
+		kfree(node);
 		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;
+	node->handle = op.handle;
+	node->area = area;
+
+	spin_lock(&xenbus_valloc_lock);
+	list_add(&node->next, &xenbus_valloc_pages);
+	spin_unlock(&xenbus_valloc_lock);
 
 	*vaddr = area->addr;
 	return 0;
 }
-EXPORT_SYMBOL_GPL(xenbus_map_ring_valloc);
+
+static int xenbus_map_ring_valloc_hvm(struct xenbus_device *dev,
+				      int gnt_ref, void **vaddr)
+{
+	struct xenbus_map_node *node;
+	int err;
+	void *addr;
+
+	*vaddr = NULL;
+
+	node = kzalloc(sizeof(*node), GFP_KERNEL);
+	if (!node)
+		return -ENOMEM;
+
+	err = alloc_xenballooned_pages(1, &node->page, false /* lowmem */);
+	if (err)
+		goto out_err;
+
+	addr = pfn_to_kaddr(page_to_pfn(node->page));
+
+	err = xenbus_map_ring(dev, gnt_ref, &node->handle, addr);
+	if (err)
+		goto out_err;
+
+	spin_lock(&xenbus_valloc_lock);
+	list_add(&node->next, &xenbus_valloc_pages);
+	spin_unlock(&xenbus_valloc_lock);
+
+	*vaddr = addr;
+	return 0;
+
+ out_err:
+	free_xenballooned_pages(1, &node->page);
+	kfree(node);
+	return err;
+}
 
 
 /**
@@ -489,12 +567,10 @@ EXPORT_SYMBOL_GPL(xenbus_map_ring_valloc);
 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,
-	};
+	struct gnttab_map_grant_ref op;
+
+	gnttab_set_map_op(&op, (phys_addr_t)vaddr, GNTMAP_host_map, gnt_ref,
+			  dev->otherend_id);
 
 	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
 		BUG();
@@ -525,32 +601,36 @@ EXPORT_SYMBOL_GPL(xenbus_map_ring);
  */
 int xenbus_unmap_ring_vfree(struct xenbus_device *dev, void *vaddr)
 {
-	struct vm_struct *area;
+	return ring_ops->unmap(dev, vaddr);
+}
+EXPORT_SYMBOL_GPL(xenbus_unmap_ring_vfree);
+
+static int xenbus_unmap_ring_vfree_pv(struct xenbus_device *dev, void *vaddr)
+{
+	struct xenbus_map_node *node;
 	struct gnttab_unmap_grant_ref op = {
 		.host_addr = (unsigned long)vaddr,
 	};
 	unsigned int level;
 
-	/* 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;
+	spin_lock(&xenbus_valloc_lock);
+	list_for_each_entry(node, &xenbus_valloc_pages, next) {
+		if (node->area->addr == vaddr) {
+			list_del(&node->next);
+			goto found;
+		}
 	}
-	read_unlock(&vmlist_lock);
+	node = NULL;
+ found:
+	spin_unlock(&xenbus_valloc_lock);
 
-	if (!area) {
+	if (!node) {
 		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;
+	op.handle = node->handle;
 	op.host_addr = arbitrary_virt_to_machine(
 		lookup_address((unsigned long)vaddr, &level)).maddr;
 
@@ -558,16 +638,50 @@ int xenbus_unmap_ring_vfree(struct xenbus_device *dev, void *vaddr)
 		BUG();
 
 	if (op.status == GNTST_okay)
-		free_vm_area(area);
+		free_vm_area(node->area);
 	else
 		xenbus_dev_error(dev, op.status,
 				 "unmapping page at handle %d error %d",
-				 (int16_t)area->phys_addr, op.status);
+				 node->handle, op.status);
 
+	kfree(node);
 	return op.status;
 }
-EXPORT_SYMBOL_GPL(xenbus_unmap_ring_vfree);
 
+static int xenbus_unmap_ring_vfree_hvm(struct xenbus_device *dev, void *vaddr)
+{
+	int rv;
+	struct xenbus_map_node *node;
+	void *addr;
+
+	spin_lock(&xenbus_valloc_lock);
+	list_for_each_entry(node, &xenbus_valloc_pages, next) {
+		addr = pfn_to_kaddr(page_to_pfn(node->page));
+		if (addr == vaddr) {
+			list_del(&node->next);
+			goto found;
+		}
+	}
+	node = NULL;
+ found:
+	spin_unlock(&xenbus_valloc_lock);
+
+	if (!node) {
+		xenbus_dev_error(dev, -ENOENT,
+				 "can't find mapped virtual address %p", vaddr);
+		return GNTST_bad_virt_addr;
+	}
+
+	rv = xenbus_unmap_ring(dev, node->handle, addr);
+
+	if (!rv)
+		free_xenballooned_pages(1, &node->page);
+	else
+		WARN(1, "Leaking %p\n", vaddr);
+
+	kfree(node);
+	return rv;
+}
 
 /**
  * xenbus_unmap_ring
@@ -582,10 +696,9 @@ EXPORT_SYMBOL_GPL(xenbus_unmap_ring_vfree);
 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,
-	};
+	struct gnttab_unmap_grant_ref op;
+
+	gnttab_set_unmap_op(&op, (phys_addr_t)vaddr, GNTMAP_host_map, handle);
 
 	if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
 		BUG();
@@ -617,3 +730,21 @@ enum xenbus_state xenbus_read_driver_state(const char *path)
 	return result;
 }
 EXPORT_SYMBOL_GPL(xenbus_read_driver_state);
+
+static const struct xenbus_ring_ops ring_ops_pv = {
+	.map = xenbus_map_ring_valloc_pv,
+	.unmap = xenbus_unmap_ring_vfree_pv,
+};
+
+static const struct xenbus_ring_ops ring_ops_hvm = {
+	.map = xenbus_map_ring_valloc_hvm,
+	.unmap = xenbus_unmap_ring_vfree_hvm,
+};
+
+void __init xenbus_ring_ops_init(void)
+{
+	if (xen_pv_domain())
+		ring_ops = &ring_ops_pv;
+	else
+		ring_ops = &ring_ops_hvm;
+}

drivers/xen/xenbus/xenbus_comms.h

@@ -31,6 +31,8 @@
 #ifndef _XENBUS_COMMS_H
 #define _XENBUS_COMMS_H
 
+#include <linux/fs.h>
+
 int xs_init(void);
 int xb_init_comms(void);
 
@@ -43,4 +45,6 @@ int xs_input_avail(void);
 extern struct xenstore_domain_interface *xen_store_interface;
 extern int xen_store_evtchn;
 
+extern const struct file_operations xen_xenbus_fops;
+
 #endif /* _XENBUS_COMMS_H */

drivers/xen/xenbus/xenbus_dev_backend.c

@@ -0,0 +1,90 @@
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/capability.h>
+
+#include <xen/xen.h>
+#include <xen/page.h>
+#include <xen/xenbus_dev.h>
+
+#include "xenbus_comms.h"
+
+MODULE_LICENSE("GPL");
+
+static int xenbus_backend_open(struct inode *inode, struct file *filp)
+{
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	return nonseekable_open(inode, filp);
+}
+
+static long xenbus_backend_ioctl(struct file *file, unsigned int cmd, unsigned long data)
+{
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	switch (cmd) {
+		case IOCTL_XENBUS_BACKEND_EVTCHN:
+			if (xen_store_evtchn > 0)
+				return xen_store_evtchn;
+			return -ENODEV;
+
+		default:
+			return -ENOTTY;
+	}
+}
+
+static int xenbus_backend_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	size_t size = vma->vm_end - vma->vm_start;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if ((size > PAGE_SIZE) || (vma->vm_pgoff != 0))
+		return -EINVAL;
+
+	if (remap_pfn_range(vma, vma->vm_start,
+			    virt_to_pfn(xen_store_interface),
+			    size, vma->vm_page_prot))
+		return -EAGAIN;
+
+	return 0;
+}
+
+const struct file_operations xenbus_backend_fops = {
+	.open = xenbus_backend_open,
+	.mmap = xenbus_backend_mmap,
+	.unlocked_ioctl = xenbus_backend_ioctl,
+};
+
+static struct miscdevice xenbus_backend_dev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "xen/xenbus_backend",
+	.fops = &xenbus_backend_fops,
+};
+
+static int __init xenbus_backend_init(void)
+{
+	int err;
+
+	if (!xen_initial_domain())
+		return -ENODEV;
+
+	err = misc_register(&xenbus_backend_dev);
+	if (err)
+		printk(KERN_ERR "Could not register xenbus backend device\n");
+	return err;
+}
+
+static void __exit xenbus_backend_exit(void)
+{
+	misc_deregister(&xenbus_backend_dev);
+}
+
+module_init(xenbus_backend_init);
+module_exit(xenbus_backend_exit);

drivers/xen/xenbus/xenbus_dev_frontend.c

@@ -0,0 +1,625 @@
+/*
+ * 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 <linux/slab.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+
+#include "xenbus_comms.h"
+
+#include <xen/xenbus.h>
+#include <xen/xen.h>
+#include <asm/xen/hypervisor.h>
+
+MODULE_LICENSE("GPL");
+
+/*
+ * 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[XENSTORE_PAYLOAD_MAX];
+	} 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);
+again:
+	while (list_empty(&u->read_buffers)) {
+		mutex_unlock(&u->reply_mutex);
+		if (filp->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+
+		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 != 0) {
+			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);
+		}
+	}
+	if (i == 0)
+		goto again;
+
+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));
+		wake_up(&u->read_waitq);
+		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 != 0) {
+		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_WATCH:
+	case XS_UNWATCH:
+		/* (Un)Ask for some path to be watched for changes */
+		ret = xenbus_write_watch(msg_type, u);
+		break;
+
+	default:
+		/* Send out a transaction */
+		ret = xenbus_write_transaction(msg_type, u);
+		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;
+	struct read_buffer *rb, *tmp_rb;
+
+	/*
+	 * 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);
+	}
+
+	list_for_each_entry_safe(rb, tmp_rb, &u->read_buffers, list) {
+		list_del(&rb->list);
+		kfree(rb);
+	}
+	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 xen_xenbus_fops = {
+	.read = xenbus_file_read,
+	.write = xenbus_file_write,
+	.open = xenbus_file_open,
+	.release = xenbus_file_release,
+	.poll = xenbus_file_poll,
+	.llseek = no_llseek,
+};
+EXPORT_SYMBOL_GPL(xen_xenbus_fops);
+
+static struct miscdevice xenbus_dev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "xen/xenbus",
+	.fops = &xen_xenbus_fops,
+};
+
+static int __init xenbus_init(void)
+{
+	int err;
+
+	if (!xen_domain())
+		return -ENODEV;
+
+	err = misc_register(&xenbus_dev);
+	if (err)
+		printk(KERN_ERR "Could not register xenbus frontend device\n");
+	return err;
+}
+
+static void __exit xenbus_exit(void)
+{
+	misc_deregister(&xenbus_dev);
+}
+
+module_init(xenbus_init);
+module_exit(xenbus_exit);

drivers/xen/xenbus/xenbus_probe.c

@@ -291,14 +291,9 @@ void xenbus_dev_shutdown(struct device *_dev)
 EXPORT_SYMBOL_GPL(xenbus_dev_shutdown);
 
 int xenbus_register_driver_common(struct xenbus_driver *drv,
-				  struct xen_bus_type *bus,
-				  struct module *owner,
-				  const char *mod_name)
+				  struct xen_bus_type *bus)
 {
-	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);
 }
@@ -730,6 +725,8 @@ static int __init xenbus_init(void)
 	if (!xen_domain())
 		return -ENODEV;
 
+	xenbus_ring_ops_init();
+
 	if (xen_hvm_domain()) {
 		uint64_t v = 0;
 		err = hvm_get_parameter(HVM_PARAM_STORE_EVTCHN, &v);

drivers/xen/xenbus/xenbus_probe.h

@@ -53,9 +53,7 @@ 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);
+					 struct xen_bus_type *bus);
 extern int xenbus_probe_node(struct xen_bus_type *bus,
 			     const char *type,
 			     const char *nodename);
@@ -76,4 +74,6 @@ extern void xenbus_otherend_changed(struct xenbus_watch *watch,
 extern int xenbus_read_otherend_details(struct xenbus_device *xendev,
 					char *id_node, char *path_node);
 
+void xenbus_ring_ops_init(void);
+
 #endif

drivers/xen/xenbus/xenbus_probe_backend.c

@@ -232,15 +232,13 @@ int xenbus_dev_is_online(struct xenbus_device *dev)
 }
 EXPORT_SYMBOL_GPL(xenbus_dev_is_online);
 
-int __xenbus_register_backend(struct xenbus_driver *drv,
-			      struct module *owner, const char *mod_name)
+int xenbus_register_backend(struct xenbus_driver *drv)
 {
 	drv->read_otherend_details = read_frontend_details;
 
-	return xenbus_register_driver_common(drv, &xenbus_backend,
-					     owner, mod_name);
+	return xenbus_register_driver_common(drv, &xenbus_backend);
 }
-EXPORT_SYMBOL_GPL(__xenbus_register_backend);
+EXPORT_SYMBOL_GPL(xenbus_register_backend);
 
 static int backend_probe_and_watch(struct notifier_block *notifier,
 				   unsigned long event,

drivers/xen/xenbus/xenbus_probe_frontend.c

@@ -230,15 +230,13 @@ static void wait_for_devices(struct xenbus_driver *xendrv)
 			 print_device_status);
 }
 
-int __xenbus_register_frontend(struct xenbus_driver *drv,
-			       struct module *owner, const char *mod_name)
+int xenbus_register_frontend(struct xenbus_driver *drv)
 {
 	int ret;
 
 	drv->read_otherend_details = read_backend_details;
 
-	ret = xenbus_register_driver_common(drv, &xenbus_frontend,
-					    owner, mod_name);
+	ret = xenbus_register_driver_common(drv, &xenbus_frontend);
 	if (ret)
 		return ret;
 
@@ -247,7 +245,7 @@ int __xenbus_register_frontend(struct xenbus_driver *drv,
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(__xenbus_register_frontend);
+EXPORT_SYMBOL_GPL(xenbus_register_frontend);
 
 static DECLARE_WAIT_QUEUE_HEAD(backend_state_wq);
 static int backend_state;

drivers/xen/xenbus/xenbus_xs.c

@@ -532,21 +532,18 @@ int xenbus_printf(struct xenbus_transaction t,
 {
 	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;
+	char *buf;
 
 	va_start(ap, fmt);
-	ret = vsnprintf(printf_buffer, PRINTF_BUFFER_SIZE, fmt, ap);
+	buf = kvasprintf(GFP_NOIO | __GFP_HIGH, fmt, ap);
 	va_end(ap);
 
-	BUG_ON(ret > PRINTF_BUFFER_SIZE-1);
-	ret = xenbus_write(t, dir, node, printf_buffer);
+	if (!buf)
+		return -ENOMEM;
 
-	kfree(printf_buffer);
+	ret = xenbus_write(t, dir, node, buf);
+
+	kfree(buf);
 
 	return ret;
 }
@@ -801,6 +798,12 @@ static int process_msg(void)
 		goto out;
 	}
 
+	if (msg->hdr.len > XENSTORE_PAYLOAD_MAX) {
+		kfree(msg);
+		err = -EINVAL;
+		goto out;
+	}
+
 	body = kmalloc(msg->hdr.len + 1, GFP_NOIO | __GFP_HIGH);
 	if (body == NULL) {
 		kfree(msg);