Merge tag 'to-linus' of git://github.com/rustyrussell/linux

* tag 'to-linus' of git://github.com/rustyrussell/linux: (24 commits)
  lguest: Make sure interrupt is allocated ok by lguest_setup_irq
  lguest: move the lguest tool to the tools directory
  lguest: switch segment-voodoo-numbers to readable symbols
  virtio: balloon: Add freeze, restore handlers to support S4
  virtio: balloon: Move vq initialization into separate function
  virtio: net: Add freeze, restore handlers to support S4
  virtio: net: Move vq and vq buf removal into separate function
  virtio: net: Move vq initialization into separate function
  virtio: blk: Add freeze, restore handlers to support S4
  virtio: blk: Move vq initialization to separate function
  virtio: console: Disable callbacks for virtqueues at start of S4 freeze
  virtio: console: Add freeze and restore handlers to support S4
  virtio: console: Move vq and vq buf removal into separate functions
  virtio: pci: add PM notification handlers for restore, freeze, thaw, poweroff
  virtio: pci: switch to new PM API
  virtio_blk: fix config handler race
  virtio: add debugging if driver doesn't kick.
  virtio: expose added descriptors immediately.
  virtio: avoid modulus operation.
  virtio: support unlocked queue kick
  ...

arch/x86/lguest/boot.c

@@ -856,18 +856,23 @@ static void __init lguest_init_IRQ(void)
 }
 
 /*
- * With CONFIG_SPARSE_IRQ, interrupt descriptors are allocated as-needed, so
- * rather than set them in lguest_init_IRQ we are called here every time an
- * lguest device needs an interrupt.
- *
- * FIXME: irq_alloc_desc_at() can fail due to lack of memory, we should
- * pass that up!
+ * Interrupt descriptors are allocated as-needed, but low-numbered ones are
+ * reserved by the generic x86 code.  So we ignore irq_alloc_desc_at if it
+ * tells us the irq is already used: other errors (ie. ENOMEM) we take
+ * seriously.
  */
-void lguest_setup_irq(unsigned int irq)
+int lguest_setup_irq(unsigned int irq)
 {
-	irq_alloc_desc_at(irq, 0);
+	int err;
+
+	/* Returns -ve error or vector number. */
+	err = irq_alloc_desc_at(irq, 0);
+	if (err < 0 && err != -EEXIST)
+		return err;
+
 	irq_set_chip_and_handler_name(irq, &lguest_irq_controller,
 				      handle_level_irq, "level");
+	return 0;
 }
 
 /*

drivers/block/virtio_blk.c

@@ -4,6 +4,7 @@
 #include <linux/blkdev.h>
 #include <linux/hdreg.h>
 #include <linux/module.h>
+#include <linux/mutex.h>
 #include <linux/virtio.h>
 #include <linux/virtio_blk.h>
 #include <linux/scatterlist.h>
@@ -36,6 +37,12 @@ struct virtio_blk
 	/* Process context for config space updates */
 	struct work_struct config_work;
 
+	/* Lock for config space updates */
+	struct mutex config_lock;
+
+	/* enable config space updates */
+	bool config_enable;
+
 	/* What host tells us, plus 2 for header & tailer. */
 	unsigned int sg_elems;
 
@@ -172,7 +179,7 @@ static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
 		}
 	}
 
-	if (virtqueue_add_buf(vblk->vq, vblk->sg, out, in, vbr) < 0) {
+	if (virtqueue_add_buf(vblk->vq, vblk->sg, out, in, vbr, GFP_ATOMIC)<0) {
 		mempool_free(vbr, vblk->pool);
 		return false;
 	}
@@ -318,6 +325,10 @@ static void virtblk_config_changed_work(struct work_struct *work)
 	char cap_str_2[10], cap_str_10[10];
 	u64 capacity, size;
 
+	mutex_lock(&vblk->config_lock);
+	if (!vblk->config_enable)
+		goto done;
+
 	/* Host must always specify the capacity. */
 	vdev->config->get(vdev, offsetof(struct virtio_blk_config, capacity),
 			  &capacity, sizeof(capacity));
@@ -340,6 +351,8 @@ static void virtblk_config_changed_work(struct work_struct *work)
 		  cap_str_10, cap_str_2);
 
 	set_capacity(vblk->disk, capacity);
+done:
+	mutex_unlock(&vblk->config_lock);
 }
 
 static void virtblk_config_changed(struct virtio_device *vdev)
@@ -349,6 +362,18 @@ static void virtblk_config_changed(struct virtio_device *vdev)
 	queue_work(virtblk_wq, &vblk->config_work);
 }
 
+static int init_vq(struct virtio_blk *vblk)
+{
+	int err = 0;
+
+	/* We expect one virtqueue, for output. */
+	vblk->vq = virtio_find_single_vq(vblk->vdev, blk_done, "requests");
+	if (IS_ERR(vblk->vq))
+		err = PTR_ERR(vblk->vq);
+
+	return err;
+}
+
 static int __devinit virtblk_probe(struct virtio_device *vdev)
 {
 	struct virtio_blk *vblk;
@@ -388,14 +413,13 @@ static int __devinit virtblk_probe(struct virtio_device *vdev)
 	vblk->vdev = vdev;
 	vblk->sg_elems = sg_elems;
 	sg_init_table(vblk->sg, vblk->sg_elems);
+	mutex_init(&vblk->config_lock);
 	INIT_WORK(&vblk->config_work, virtblk_config_changed_work);
+	vblk->config_enable = true;
 
-	/* We expect one virtqueue, for output. */
-	vblk->vq = virtio_find_single_vq(vdev, blk_done, "requests");
-	if (IS_ERR(vblk->vq)) {
-		err = PTR_ERR(vblk->vq);
+	err = init_vq(vblk);
+	if (err)
 		goto out_free_vblk;
-	}
 
 	vblk->pool = mempool_create_kmalloc_pool(1,sizeof(struct virtblk_req));
 	if (!vblk->pool) {
@@ -542,7 +566,10 @@ static void __devexit virtblk_remove(struct virtio_device *vdev)
 	struct virtio_blk *vblk = vdev->priv;
 	int index = vblk->index;
 
-	flush_work(&vblk->config_work);
+	/* Prevent config work handler from accessing the device. */
+	mutex_lock(&vblk->config_lock);
+	vblk->config_enable = false;
+	mutex_unlock(&vblk->config_lock);
 
 	/* Nothing should be pending. */
 	BUG_ON(!list_empty(&vblk->reqs));
@@ -550,6 +577,8 @@ static void __devexit virtblk_remove(struct virtio_device *vdev)
 	/* Stop all the virtqueues. */
 	vdev->config->reset(vdev);
 
+	flush_work(&vblk->config_work);
+
 	del_gendisk(vblk->disk);
 	blk_cleanup_queue(vblk->disk->queue);
 	put_disk(vblk->disk);
@@ -559,6 +588,46 @@ static void __devexit virtblk_remove(struct virtio_device *vdev)
 	ida_simple_remove(&vd_index_ida, index);
 }
 
+#ifdef CONFIG_PM
+static int virtblk_freeze(struct virtio_device *vdev)
+{
+	struct virtio_blk *vblk = vdev->priv;
+
+	/* Ensure we don't receive any more interrupts */
+	vdev->config->reset(vdev);
+
+	/* Prevent config work handler from accessing the device. */
+	mutex_lock(&vblk->config_lock);
+	vblk->config_enable = false;
+	mutex_unlock(&vblk->config_lock);
+
+	flush_work(&vblk->config_work);
+
+	spin_lock_irq(vblk->disk->queue->queue_lock);
+	blk_stop_queue(vblk->disk->queue);
+	spin_unlock_irq(vblk->disk->queue->queue_lock);
+	blk_sync_queue(vblk->disk->queue);
+
+	vdev->config->del_vqs(vdev);
+	return 0;
+}
+
+static int virtblk_restore(struct virtio_device *vdev)
+{
+	struct virtio_blk *vblk = vdev->priv;
+	int ret;
+
+	vblk->config_enable = true;
+	ret = init_vq(vdev->priv);
+	if (!ret) {
+		spin_lock_irq(vblk->disk->queue->queue_lock);
+		blk_start_queue(vblk->disk->queue);
+		spin_unlock_irq(vblk->disk->queue->queue_lock);
+	}
+	return ret;
+}
+#endif
+
 static const struct virtio_device_id id_table[] = {
 	{ VIRTIO_ID_BLOCK, VIRTIO_DEV_ANY_ID },
 	{ 0 },
@@ -584,6 +653,10 @@ static struct virtio_driver __refdata virtio_blk = {
 	.probe			= virtblk_probe,
 	.remove			= __devexit_p(virtblk_remove),
 	.config_changed		= virtblk_config_changed,
+#ifdef CONFIG_PM
+	.freeze			= virtblk_freeze,
+	.restore		= virtblk_restore,
+#endif
 };
 
 static int __init init(void)

drivers/char/hw_random/virtio-rng.c

@@ -47,7 +47,7 @@ static void register_buffer(u8 *buf, size_t size)
 	sg_init_one(&sg, buf, size);
 
 	/* There should always be room for one buffer. */
-	if (virtqueue_add_buf(vq, &sg, 0, 1, buf) < 0)
+	if (virtqueue_add_buf(vq, &sg, 0, 1, buf, GFP_KERNEL) < 0)
 		BUG();
 
 	virtqueue_kick(vq);

drivers/char/virtio_console.c

@@ -392,7 +392,7 @@ static int add_inbuf(struct virtqueue *vq, struct port_buffer *buf)
 
 	sg_init_one(sg, buf->buf, buf->size);
 
-	ret = virtqueue_add_buf(vq, sg, 0, 1, buf);
+	ret = virtqueue_add_buf(vq, sg, 0, 1, buf, GFP_ATOMIC);
 	virtqueue_kick(vq);
 	return ret;
 }
@@ -457,7 +457,7 @@ static ssize_t __send_control_msg(struct ports_device *portdev, u32 port_id,
 	vq = portdev->c_ovq;
 
 	sg_init_one(sg, &cpkt, sizeof(cpkt));
-	if (virtqueue_add_buf(vq, sg, 1, 0, &cpkt) >= 0) {
+	if (virtqueue_add_buf(vq, sg, 1, 0, &cpkt, GFP_ATOMIC) >= 0) {
 		virtqueue_kick(vq);
 		while (!virtqueue_get_buf(vq, &len))
 			cpu_relax();
@@ -506,7 +506,7 @@ static ssize_t send_buf(struct port *port, void *in_buf, size_t in_count,
 	reclaim_consumed_buffers(port);
 
 	sg_init_one(sg, in_buf, in_count);
-	ret = virtqueue_add_buf(out_vq, sg, 1, 0, in_buf);
+	ret = virtqueue_add_buf(out_vq, sg, 1, 0, in_buf, GFP_ATOMIC);
 
 	/* Tell Host to go! */
 	virtqueue_kick(out_vq);
@@ -1271,6 +1271,20 @@ static void remove_port(struct kref *kref)
 	kfree(port);
 }
 
+static void remove_port_data(struct port *port)
+{
+	struct port_buffer *buf;
+
+	/* Remove unused data this port might have received. */
+	discard_port_data(port);
+
+	reclaim_consumed_buffers(port);
+
+	/* Remove buffers we queued up for the Host to send us data in. */
+	while ((buf = virtqueue_detach_unused_buf(port->in_vq)))
+		free_buf(buf);
+}
+
 /*
  * Port got unplugged.  Remove port from portdev's list and drop the
  * kref reference.  If no userspace has this port opened, it will
@@ -1278,8 +1292,6 @@ static void remove_port(struct kref *kref)
  */
 static void unplug_port(struct port *port)
 {
-	struct port_buffer *buf;
-
 	spin_lock_irq(&port->portdev->ports_lock);
 	list_del(&port->list);
 	spin_unlock_irq(&port->portdev->ports_lock);
@@ -1300,14 +1312,7 @@ static void unplug_port(struct port *port)
 		hvc_remove(port->cons.hvc);
 	}
 
-	/* Remove unused data this port might have received. */
-	discard_port_data(port);
-
-	reclaim_consumed_buffers(port);
-
-	/* Remove buffers we queued up for the Host to send us data in. */
-	while ((buf = virtqueue_detach_unused_buf(port->in_vq)))
-		free_buf(buf);
+	remove_port_data(port);
 
 	/*
 	 * We should just assume the device itself has gone off --
@@ -1659,6 +1664,28 @@ static const struct file_operations portdev_fops = {
 	.owner = THIS_MODULE,
 };
 
+static void remove_vqs(struct ports_device *portdev)
+{
+	portdev->vdev->config->del_vqs(portdev->vdev);
+	kfree(portdev->in_vqs);
+	kfree(portdev->out_vqs);
+}
+
+static void remove_controlq_data(struct ports_device *portdev)
+{
+	struct port_buffer *buf;
+	unsigned int len;
+
+	if (!use_multiport(portdev))
+		return;
+
+	while ((buf = virtqueue_get_buf(portdev->c_ivq, &len)))
+		free_buf(buf);
+
+	while ((buf = virtqueue_detach_unused_buf(portdev->c_ivq)))
+		free_buf(buf);
+}
+
 /*
  * Once we're further in boot, we get probed like any other virtio
  * device.
@@ -1764,9 +1791,7 @@ free_vqs:
 	/* The host might want to notify mgmt sw about device add failure */
 	__send_control_msg(portdev, VIRTIO_CONSOLE_BAD_ID,
 			   VIRTIO_CONSOLE_DEVICE_READY, 0);
-	vdev->config->del_vqs(vdev);
-	kfree(portdev->in_vqs);
-	kfree(portdev->out_vqs);
+	remove_vqs(portdev);
 free_chrdev:
 	unregister_chrdev(portdev->chr_major, "virtio-portsdev");
 free:
@@ -1804,21 +1829,8 @@ static void virtcons_remove(struct virtio_device *vdev)
 	 * have to just stop using the port, as the vqs are going
 	 * away.
 	 */
-	if (use_multiport(portdev)) {
-		struct port_buffer *buf;
-		unsigned int len;
-
-		while ((buf = virtqueue_get_buf(portdev->c_ivq, &len)))
-			free_buf(buf);
-
-		while ((buf = virtqueue_detach_unused_buf(portdev->c_ivq)))
-			free_buf(buf);
-	}
-
-	vdev->config->del_vqs(vdev);
-	kfree(portdev->in_vqs);
-	kfree(portdev->out_vqs);
-
+	remove_controlq_data(portdev);
+	remove_vqs(portdev);
 	kfree(portdev);
 }
 
@@ -1832,6 +1844,68 @@ static unsigned int features[] = {
 	VIRTIO_CONSOLE_F_MULTIPORT,
 };
 
+#ifdef CONFIG_PM
+static int virtcons_freeze(struct virtio_device *vdev)
+{
+	struct ports_device *portdev;
+	struct port *port;
+
+	portdev = vdev->priv;
+
+	vdev->config->reset(vdev);
+
+	virtqueue_disable_cb(portdev->c_ivq);
+	cancel_work_sync(&portdev->control_work);
+	/*
+	 * Once more: if control_work_handler() was running, it would
+	 * enable the cb as the last step.
+	 */
+	virtqueue_disable_cb(portdev->c_ivq);
+	remove_controlq_data(portdev);
+
+	list_for_each_entry(port, &portdev->ports, list) {
+		virtqueue_disable_cb(port->in_vq);
+		virtqueue_disable_cb(port->out_vq);
+		/*
+		 * We'll ask the host later if the new invocation has
+		 * the port opened or closed.
+		 */
+		port->host_connected = false;
+		remove_port_data(port);
+	}
+	remove_vqs(portdev);
+
+	return 0;
+}
+
+static int virtcons_restore(struct virtio_device *vdev)
+{
+	struct ports_device *portdev;
+	struct port *port;
+	int ret;
+
+	portdev = vdev->priv;
+
+	ret = init_vqs(portdev);
+	if (ret)
+		return ret;
+
+	if (use_multiport(portdev))
+		fill_queue(portdev->c_ivq, &portdev->cvq_lock);
+
+	list_for_each_entry(port, &portdev->ports, list) {
+		port->in_vq = portdev->in_vqs[port->id];
+		port->out_vq = portdev->out_vqs[port->id];
+
+		fill_queue(port->in_vq, &port->inbuf_lock);
+
+		/* Get port open/close status on the host */
+		send_control_msg(port, VIRTIO_CONSOLE_PORT_READY, 1);
+	}
+	return 0;
+}
+#endif
+
 static struct virtio_driver virtio_console = {
 	.feature_table = features,
 	.feature_table_size = ARRAY_SIZE(features),
@@ -1841,6 +1915,10 @@ static struct virtio_driver virtio_console = {
 	.probe =	virtcons_probe,
 	.remove =	virtcons_remove,
 	.config_changed = config_intr,
+#ifdef CONFIG_PM
+	.freeze =	virtcons_freeze,
+	.restore =	virtcons_restore,
+#endif
 };
 
 static int __init init(void)

drivers/lguest/Makefile

@@ -18,7 +18,7 @@ Mastery: PREFIX=M
 Beer:
 	@for f in Preparation Guest Drivers Launcher Host Switcher Mastery; do echo "{==- $$f -==}"; make -s $$f; done; echo "{==-==}"
 Preparation Preparation! Guest Drivers Launcher Host Switcher Mastery:
-	@sh ../../Documentation/virtual/lguest/extract $(PREFIX) `find ../../* -name '*.[chS]' -wholename '*lguest*'`
+	@sh ../../tools/lguest/extract $(PREFIX) `find ../../* -name '*.[chS]' -wholename '*lguest*'`
 Puppy:
 	@clear
 	@printf "      __  \n (___()'\`;\n /,    /\`\n \\\\\\\"--\\\\\\   \n"

drivers/lguest/lguest_device.c

@@ -241,7 +241,7 @@ static void lg_notify(struct virtqueue *vq)
 }
 
 /* An extern declaration inside a C file is bad form.  Don't do it. */
-extern void lguest_setup_irq(unsigned int irq);
+extern int lguest_setup_irq(unsigned int irq);
 
 /*
  * This routine finds the Nth virtqueue described in the configuration of
@@ -292,17 +292,21 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev,
 
 	/*
 	 * OK, tell virtio_ring.c to set up a virtqueue now we know its size
-	 * and we've got a pointer to its pages.
+	 * and we've got a pointer to its pages.  Note that we set weak_barriers
+	 * to 'true': the host just a(nother) SMP CPU, so we only need inter-cpu
+	 * barriers.
 	 */
-	vq = vring_new_virtqueue(lvq->config.num, LGUEST_VRING_ALIGN,
-				 vdev, lvq->pages, lg_notify, callback, name);
+	vq = vring_new_virtqueue(lvq->config.num, LGUEST_VRING_ALIGN, vdev,
+				 true, lvq->pages, lg_notify, callback, name);
 	if (!vq) {
 		err = -ENOMEM;
 		goto unmap;
 	}
 
 	/* Make sure the interrupt is allocated. */
-	lguest_setup_irq(lvq->config.irq);
+	err = lguest_setup_irq(lvq->config.irq);
+	if (err)
+		goto destroy_vring;
 
 	/*
 	 * Tell the interrupt for this virtqueue to go to the virtio_ring
@@ -315,7 +319,7 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev,
 	err = request_irq(lvq->config.irq, vring_interrupt, IRQF_SHARED,
 			  dev_name(&vdev->dev), vq);
 	if (err)
-		goto destroy_vring;
+		goto free_desc;
 
 	/*
 	 * Last of all we hook up our 'struct lguest_vq_info" to the
@@ -324,6 +328,8 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev,
 	vq->priv = lvq;
 	return vq;
 
+free_desc:
+	irq_free_desc(lvq->config.irq);
 destroy_vring:
 	vring_del_virtqueue(vq);
 unmap:

drivers/lguest/segments.c

@@ -81,8 +81,8 @@ static void fixup_gdt_table(struct lg_cpu *cpu, unsigned start, unsigned end)
 		 * sometimes careless and leaves this as 0, even though it's
 		 * running at privilege level 1.  If so, we fix it here.
 		 */
-		if ((cpu->arch.gdt[i].b & 0x00006000) == 0)
-			cpu->arch.gdt[i].b |= (GUEST_PL << 13);
+		if (cpu->arch.gdt[i].dpl == 0)
+			cpu->arch.gdt[i].dpl |= GUEST_PL;
 
 		/*
 		 * Each descriptor has an "accessed" bit.  If we don't set it
@@ -90,7 +90,7 @@ static void fixup_gdt_table(struct lg_cpu *cpu, unsigned start, unsigned end)
 		 * that entry into a segment register.  But the GDT isn't
 		 * writable by the Guest, so bad things can happen.
 		 */
-		cpu->arch.gdt[i].b |= 0x00000100;
+		cpu->arch.gdt[i].type |= 0x1;
 	}
 }
 
@@ -114,13 +114,19 @@ void setup_default_gdt_entries(struct lguest_ro_state *state)
 
 	/*
 	 * The TSS segment refers to the TSS entry for this particular CPU.
-	 * Forgive the magic flags: the 0x8900 means the entry is Present, it's
-	 * privilege level 0 Available 386 TSS system segment, and the 0x67
-	 * means Saturn is eclipsed by Mercury in the twelfth house.
 	 */
-	gdt[GDT_ENTRY_TSS].a = 0x00000067 | (tss << 16);
-	gdt[GDT_ENTRY_TSS].b = 0x00008900 | (tss & 0xFF000000)
-		| ((tss >> 16) & 0x000000FF);
+	gdt[GDT_ENTRY_TSS].a = 0;
+	gdt[GDT_ENTRY_TSS].b = 0;
+
+	gdt[GDT_ENTRY_TSS].limit0 = 0x67;
+	gdt[GDT_ENTRY_TSS].base0  = tss & 0xFFFF;
+	gdt[GDT_ENTRY_TSS].base1  = (tss >> 16) & 0xFF;
+	gdt[GDT_ENTRY_TSS].base2  = tss >> 24;
+	gdt[GDT_ENTRY_TSS].type   = 0x9; /* 32-bit TSS (available) */
+	gdt[GDT_ENTRY_TSS].p      = 0x1; /* Entry is present */
+	gdt[GDT_ENTRY_TSS].dpl    = 0x0; /* Privilege level 0 */
+	gdt[GDT_ENTRY_TSS].s      = 0x0; /* system segment */
+
 }
 
 /*
@@ -135,8 +141,8 @@ void setup_guest_gdt(struct lg_cpu *cpu)
 	 */
 	cpu->arch.gdt[GDT_ENTRY_KERNEL_CS] = FULL_EXEC_SEGMENT;
 	cpu->arch.gdt[GDT_ENTRY_KERNEL_DS] = FULL_SEGMENT;
-	cpu->arch.gdt[GDT_ENTRY_KERNEL_CS].b |= (GUEST_PL << 13);
-	cpu->arch.gdt[GDT_ENTRY_KERNEL_DS].b |= (GUEST_PL << 13);
+	cpu->arch.gdt[GDT_ENTRY_KERNEL_CS].dpl |= GUEST_PL;
+	cpu->arch.gdt[GDT_ENTRY_KERNEL_DS].dpl |= GUEST_PL;
 }
 
 /*H:650

drivers/net/virtio_net.c

@@ -370,7 +370,7 @@ static int add_recvbuf_small(struct virtnet_info *vi, gfp_t gfp)
 
 	skb_to_sgvec(skb, vi->rx_sg + 1, 0, skb->len);
 
-	err = virtqueue_add_buf_gfp(vi->rvq, vi->rx_sg, 0, 2, skb, gfp);
+	err = virtqueue_add_buf(vi->rvq, vi->rx_sg, 0, 2, skb, gfp);
 	if (err < 0)
 		dev_kfree_skb(skb);
 
@@ -415,8 +415,8 @@ static int add_recvbuf_big(struct virtnet_info *vi, gfp_t gfp)
 
 	/* chain first in list head */
 	first->private = (unsigned long)list;
-	err = virtqueue_add_buf_gfp(vi->rvq, vi->rx_sg, 0, MAX_SKB_FRAGS + 2,
-				    first, gfp);
+	err = virtqueue_add_buf(vi->rvq, vi->rx_sg, 0, MAX_SKB_FRAGS + 2,
+				first, gfp);
 	if (err < 0)
 		give_pages(vi, first);
 
@@ -434,7 +434,7 @@ static int add_recvbuf_mergeable(struct virtnet_info *vi, gfp_t gfp)
 
 	sg_init_one(vi->rx_sg, page_address(page), PAGE_SIZE);
 
-	err = virtqueue_add_buf_gfp(vi->rvq, vi->rx_sg, 0, 1, page, gfp);
+	err = virtqueue_add_buf(vi->rvq, vi->rx_sg, 0, 1, page, gfp);
 	if (err < 0)
 		give_pages(vi, page);
 
@@ -609,7 +609,7 @@ static int xmit_skb(struct virtnet_info *vi, struct sk_buff *skb)
 
 	hdr->num_sg = skb_to_sgvec(skb, vi->tx_sg + 1, 0, skb->len) + 1;
 	return virtqueue_add_buf(vi->svq, vi->tx_sg, hdr->num_sg,
-					0, skb);
+				 0, skb, GFP_ATOMIC);
 }
 
 static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
@@ -767,7 +767,7 @@ static bool virtnet_send_command(struct virtnet_info *vi, u8 class, u8 cmd,
 		sg_set_buf(&sg[i + 1], sg_virt(s), s->length);
 	sg_set_buf(&sg[out + in - 1], &status, sizeof(status));
 
-	BUG_ON(virtqueue_add_buf(vi->cvq, sg, out, in, vi) < 0);
+	BUG_ON(virtqueue_add_buf(vi->cvq, sg, out, in, vi, GFP_ATOMIC) < 0);
 
 	virtqueue_kick(vi->cvq);
 
@@ -985,15 +985,38 @@ static void virtnet_config_changed(struct virtio_device *vdev)
 	virtnet_update_status(vi);
 }
 
+static int init_vqs(struct virtnet_info *vi)
+{
+	struct virtqueue *vqs[3];
+	vq_callback_t *callbacks[] = { skb_recv_done, skb_xmit_done, NULL};
+	const char *names[] = { "input", "output", "control" };
+	int nvqs, err;
+
+	/* We expect two virtqueues, receive then send,
+	 * and optionally control. */
+	nvqs = virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ? 3 : 2;
+
+	err = vi->vdev->config->find_vqs(vi->vdev, nvqs, vqs, callbacks, names);
+	if (err)
+		return err;
+
+	vi->rvq = vqs[0];
+	vi->svq = vqs[1];
+
+	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ)) {
+		vi->cvq = vqs[2];
+
+		if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VLAN))
+			vi->dev->features |= NETIF_F_HW_VLAN_FILTER;
+	}
+	return 0;
+}
+
 static int virtnet_probe(struct virtio_device *vdev)
 {
 	int err;
 	struct net_device *dev;
 	struct virtnet_info *vi;
-	struct virtqueue *vqs[3];
-	vq_callback_t *callbacks[] = { skb_recv_done, skb_xmit_done, NULL};
-	const char *names[] = { "input", "output", "control" };
-	int nvqs;
 
 	/* Allocate ourselves a network device with room for our info */
 	dev = alloc_etherdev(sizeof(struct virtnet_info));
@@ -1065,24 +1088,10 @@ static int virtnet_probe(struct virtio_device *vdev)
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF))
 		vi->mergeable_rx_bufs = true;
 
-	/* We expect two virtqueues, receive then send,
-	 * and optionally control. */
-	nvqs = virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ? 3 : 2;
-
-	err = vdev->config->find_vqs(vdev, nvqs, vqs, callbacks, names);
+	err = init_vqs(vi);
 	if (err)
 		goto free_stats;
 
-	vi->rvq = vqs[0];
-	vi->svq = vqs[1];
-
-	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ)) {
-		vi->cvq = vqs[2];
-
-		if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VLAN))
-			dev->features |= NETIF_F_HW_VLAN_FILTER;
-	}
-
 	err = register_netdev(dev);
 	if (err) {
 		pr_debug("virtio_net: registering device failed\n");
@@ -1144,27 +1153,73 @@ static void free_unused_bufs(struct virtnet_info *vi)
 	BUG_ON(vi->num != 0);
 }
 
-static void __devexit virtnet_remove(struct virtio_device *vdev)
+static void remove_vq_common(struct virtnet_info *vi)
 {
-	struct virtnet_info *vi = vdev->priv;
-
-	/* Stop all the virtqueues. */
-	vdev->config->reset(vdev);
-
-	unregister_netdev(vi->dev);
+	vi->vdev->config->reset(vi->vdev);
 
 	/* Free unused buffers in both send and recv, if any. */
 	free_unused_bufs(vi);
 
-	vdev->config->del_vqs(vi->vdev);
+	vi->vdev->config->del_vqs(vi->vdev);
 
 	while (vi->pages)
 		__free_pages(get_a_page(vi, GFP_KERNEL), 0);
+}
+
+static void __devexit virtnet_remove(struct virtio_device *vdev)
+{
+	struct virtnet_info *vi = vdev->priv;
+
+	unregister_netdev(vi->dev);
+
+	remove_vq_common(vi);
 
 	free_percpu(vi->stats);
 	free_netdev(vi->dev);
 }
 
+#ifdef CONFIG_PM
+static int virtnet_freeze(struct virtio_device *vdev)
+{
+	struct virtnet_info *vi = vdev->priv;
+
+	virtqueue_disable_cb(vi->rvq);
+	virtqueue_disable_cb(vi->svq);
+	if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ))
+		virtqueue_disable_cb(vi->cvq);
+
+	netif_device_detach(vi->dev);
+	cancel_delayed_work_sync(&vi->refill);
+
+	if (netif_running(vi->dev))
+		napi_disable(&vi->napi);
+
+	remove_vq_common(vi);
+
+	return 0;
+}
+
+static int virtnet_restore(struct virtio_device *vdev)
+{
+	struct virtnet_info *vi = vdev->priv;
+	int err;
+
+	err = init_vqs(vi);
+	if (err)
+		return err;
+
+	if (netif_running(vi->dev))
+		virtnet_napi_enable(vi);
+
+	netif_device_attach(vi->dev);
+
+	if (!try_fill_recv(vi, GFP_KERNEL))
+		queue_delayed_work(system_nrt_wq, &vi->refill, 0);
+
+	return 0;
+}
+#endif
+
 static struct virtio_device_id id_table[] = {
 	{ VIRTIO_ID_NET, VIRTIO_DEV_ANY_ID },
 	{ 0 },
@@ -1189,6 +1244,10 @@ static struct virtio_driver virtio_net_driver = {
 	.probe =	virtnet_probe,
 	.remove =	__devexit_p(virtnet_remove),
 	.config_changed = virtnet_config_changed,
+#ifdef CONFIG_PM
+	.freeze =	virtnet_freeze,
+	.restore =	virtnet_restore,
+#endif
 };
 
 static int __init init(void)

drivers/s390/kvm/kvm_virtio.c

@@ -198,7 +198,7 @@ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev,
 		goto out;
 
 	vq = vring_new_virtqueue(config->num, KVM_S390_VIRTIO_RING_ALIGN,
-				 vdev, (void *) config->address,
+				 vdev, true, (void *) config->address,
 				 kvm_notify, callback, name);
 	if (!vq) {
 		err = -ENOMEM;

drivers/virtio/virtio_balloon.c

@@ -1,4 +1,5 @@
-/* Virtio balloon implementation, inspired by Dor Loar and Marcelo
+/*
+ * Virtio balloon implementation, inspired by Dor Laor and Marcelo
  * Tosatti's implementations.
  *
  *  Copyright 2008 Rusty Russell IBM Corporation
@@ -17,7 +18,7 @@
  *  along with this program; if not, write to the Free Software
  *  Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
-//#define DEBUG
+
 #include <linux/virtio.h>
 #include <linux/virtio_balloon.h>
 #include <linux/swap.h>
@@ -87,7 +88,7 @@ static void tell_host(struct virtio_balloon *vb, struct virtqueue *vq)
 	init_completion(&vb->acked);
 
 	/* We should always be able to add one buffer to an empty queue. */
-	if (virtqueue_add_buf(vq, &sg, 1, 0, vb) < 0)
+	if (virtqueue_add_buf(vq, &sg, 1, 0, vb, GFP_KERNEL) < 0)
 		BUG();
 	virtqueue_kick(vq);
 
@@ -149,7 +150,6 @@ static void leak_balloon(struct virtio_balloon *vb, size_t num)
 		vb->num_pages--;
 	}
 
-
 	/*
 	 * Note that if
 	 * virtio_has_feature(vdev, VIRTIO_BALLOON_F_MUST_TELL_HOST);
@@ -220,7 +220,7 @@ static void stats_handle_request(struct virtio_balloon *vb)
 
 	vq = vb->stats_vq;
 	sg_init_one(&sg, vb->stats, sizeof(vb->stats));
-	if (virtqueue_add_buf(vq, &sg, 1, 0, vb) < 0)
+	if (virtqueue_add_buf(vq, &sg, 1, 0, vb, GFP_KERNEL) < 0)
 		BUG();
 	virtqueue_kick(vq);
 }
@@ -275,32 +275,21 @@ static int balloon(void *_vballoon)
 	return 0;
 }
 
-static int virtballoon_probe(struct virtio_device *vdev)
+static int init_vqs(struct virtio_balloon *vb)
 {
-	struct virtio_balloon *vb;
 	struct virtqueue *vqs[3];
 	vq_callback_t *callbacks[] = { balloon_ack, balloon_ack, stats_request };
 	const char *names[] = { "inflate", "deflate", "stats" };
 	int err, nvqs;
 
-	vdev->priv = vb = kmalloc(sizeof(*vb), GFP_KERNEL);
-	if (!vb) {
-		err = -ENOMEM;
-		goto out;
-	}
-
-	INIT_LIST_HEAD(&vb->pages);
-	vb->num_pages = 0;
-	init_waitqueue_head(&vb->config_change);
-	vb->vdev = vdev;
-	vb->need_stats_update = 0;
-
-	/* We expect two virtqueues: inflate and deflate,
-	 * and optionally stat. */
+	/*
+	 * We expect two virtqueues: inflate and deflate, and
+	 * optionally stat.
+	 */
 	nvqs = virtio_has_feature(vb->vdev, VIRTIO_BALLOON_F_STATS_VQ) ? 3 : 2;
-	err = vdev->config->find_vqs(vdev, nvqs, vqs, callbacks, names);
+	err = vb->vdev->config->find_vqs(vb->vdev, nvqs, vqs, callbacks, names);
 	if (err)
-		goto out_free_vb;
+		return err;
 
 	vb->inflate_vq = vqs[0];
 	vb->deflate_vq = vqs[1];
@@ -313,10 +302,34 @@ static int virtballoon_probe(struct virtio_device *vdev)
 		 * use it to signal us later.
 		 */
 		sg_init_one(&sg, vb->stats, sizeof vb->stats);
-		if (virtqueue_add_buf(vb->stats_vq, &sg, 1, 0, vb) < 0)
+		if (virtqueue_add_buf(vb->stats_vq, &sg, 1, 0, vb, GFP_KERNEL)
+		    < 0)
 			BUG();
 		virtqueue_kick(vb->stats_vq);
 	}
+	return 0;
+}
+
+static int virtballoon_probe(struct virtio_device *vdev)
+{
+	struct virtio_balloon *vb;
+	int err;
+
+	vdev->priv = vb = kmalloc(sizeof(*vb), GFP_KERNEL);
+	if (!vb) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	INIT_LIST_HEAD(&vb->pages);
+	vb->num_pages = 0;
+	init_waitqueue_head(&vb->config_change);
+	vb->vdev = vdev;
+	vb->need_stats_update = 0;
+
+	err = init_vqs(vb);
+	if (err)
+		goto out_free_vb;
 
 	vb->thread = kthread_run(balloon, vb, "vballoon");
 	if (IS_ERR(vb->thread)) {
@@ -351,6 +364,48 @@ static void __devexit virtballoon_remove(struct virtio_device *vdev)
 	kfree(vb);
 }
 
+#ifdef CONFIG_PM
+static int virtballoon_freeze(struct virtio_device *vdev)
+{
+	/*
+	 * The kthread is already frozen by the PM core before this
+	 * function is called.
+	 */
+
+	/* Ensure we don't get any more requests from the host */
+	vdev->config->reset(vdev);
+	vdev->config->del_vqs(vdev);
+	return 0;
+}
+
+static int virtballoon_thaw(struct virtio_device *vdev)
+{
+	return init_vqs(vdev->priv);
+}
+
+static int virtballoon_restore(struct virtio_device *vdev)
+{
+	struct virtio_balloon *vb = vdev->priv;
+	struct page *page, *page2;
+
+	/* We're starting from a clean slate */
+	vb->num_pages = 0;
+
+	/*
+	 * If a request wasn't complete at the time of freezing, this
+	 * could have been set.
+	 */
+	vb->need_stats_update = 0;
+
+	/* We don't have these pages in the balloon anymore! */
+	list_for_each_entry_safe(page, page2, &vb->pages, lru) {
+		list_del(&page->lru);
+		totalram_pages++;
+	}
+	return init_vqs(vdev->priv);
+}
+#endif
+
 static unsigned int features[] = {
 	VIRTIO_BALLOON_F_MUST_TELL_HOST,
 	VIRTIO_BALLOON_F_STATS_VQ,
@@ -365,6 +420,11 @@ static struct virtio_driver virtio_balloon_driver = {
 	.probe =	virtballoon_probe,
 	.remove =	__devexit_p(virtballoon_remove),
 	.config_changed = virtballoon_changed,
+#ifdef CONFIG_PM
+	.freeze	=	virtballoon_freeze,
+	.restore =	virtballoon_restore,
+	.thaw =		virtballoon_thaw,
+#endif
 };
 
 static int __init init(void)

drivers/virtio/virtio_mmio.c

@@ -310,8 +310,8 @@ static struct virtqueue *vm_setup_vq(struct virtio_device *vdev, unsigned index,
 			vm_dev->base + VIRTIO_MMIO_QUEUE_PFN);
 
 	/* Create the vring */
-	vq = vring_new_virtqueue(info->num, VIRTIO_MMIO_VRING_ALIGN,
-				 vdev, info->queue, vm_notify, callback, name);
+	vq = vring_new_virtqueue(info->num, VIRTIO_MMIO_VRING_ALIGN, vdev,
+				 true, info->queue, vm_notify, callback, name);
 	if (!vq) {
 		err = -ENOMEM;
 		goto error_new_virtqueue;

drivers/virtio/virtio_pci.c

@@ -55,6 +55,10 @@ struct virtio_pci_device
 	unsigned msix_vectors;
 	/* Vectors allocated, excluding per-vq vectors if any */
 	unsigned msix_used_vectors;
+
+	/* Status saved during hibernate/restore */
+	u8 saved_status;
+
 	/* Whether we have vector per vq */
 	bool per_vq_vectors;
 };
@@ -414,8 +418,8 @@ static struct virtqueue *setup_vq(struct virtio_device *vdev, unsigned index,
 		  vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN);
 
 	/* create the vring */
-	vq = vring_new_virtqueue(info->num, VIRTIO_PCI_VRING_ALIGN,
-				 vdev, info->queue, vp_notify, callback, name);
+	vq = vring_new_virtqueue(info->num, VIRTIO_PCI_VRING_ALIGN, vdev,
+				 true, info->queue, vp_notify, callback, name);
 	if (!vq) {
 		err = -ENOMEM;
 		goto out_activate_queue;
@@ -716,19 +720,114 @@ static void __devexit virtio_pci_remove(struct pci_dev *pci_dev)
 }
 
 #ifdef CONFIG_PM
-static int virtio_pci_suspend(struct pci_dev *pci_dev, pm_message_t state)
+static int virtio_pci_suspend(struct device *dev)
 {
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+
 	pci_save_state(pci_dev);
 	pci_set_power_state(pci_dev, PCI_D3hot);
 	return 0;
 }
 
-static int virtio_pci_resume(struct pci_dev *pci_dev)
+static int virtio_pci_resume(struct device *dev)
 {
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+
 	pci_restore_state(pci_dev);
 	pci_set_power_state(pci_dev, PCI_D0);
 	return 0;
 }
+
+static int virtio_pci_freeze(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct virtio_pci_device *vp_dev = pci_get_drvdata(pci_dev);
+	struct virtio_driver *drv;
+	int ret;
+
+	drv = container_of(vp_dev->vdev.dev.driver,
+			   struct virtio_driver, driver);
+
+	ret = 0;
+	vp_dev->saved_status = vp_get_status(&vp_dev->vdev);
+	if (drv && drv->freeze)
+		ret = drv->freeze(&vp_dev->vdev);
+
+	if (!ret)
+		pci_disable_device(pci_dev);
+	return ret;
+}
+
+static int restore_common(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct virtio_pci_device *vp_dev = pci_get_drvdata(pci_dev);
+	int ret;
+
+	ret = pci_enable_device(pci_dev);
+	if (ret)
+		return ret;
+	pci_set_master(pci_dev);
+	vp_finalize_features(&vp_dev->vdev);
+
+	return ret;
+}
+
+static int virtio_pci_thaw(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct virtio_pci_device *vp_dev = pci_get_drvdata(pci_dev);
+	struct virtio_driver *drv;
+	int ret;
+
+	ret = restore_common(dev);
+	if (ret)
+		return ret;
+
+	drv = container_of(vp_dev->vdev.dev.driver,
+			   struct virtio_driver, driver);
+
+	if (drv && drv->thaw)
+		ret = drv->thaw(&vp_dev->vdev);
+	else if (drv && drv->restore)
+		ret = drv->restore(&vp_dev->vdev);
+
+	/* Finally, tell the device we're all set */
+	if (!ret)
+		vp_set_status(&vp_dev->vdev, vp_dev->saved_status);
+
+	return ret;
+}
+
+static int virtio_pci_restore(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct virtio_pci_device *vp_dev = pci_get_drvdata(pci_dev);
+	struct virtio_driver *drv;
+	int ret;
+
+	drv = container_of(vp_dev->vdev.dev.driver,
+			   struct virtio_driver, driver);
+
+	ret = restore_common(dev);
+	if (!ret && drv && drv->restore)
+		ret = drv->restore(&vp_dev->vdev);
+
+	/* Finally, tell the device we're all set */
+	if (!ret)
+		vp_set_status(&vp_dev->vdev, vp_dev->saved_status);
+
+	return ret;
+}
+
+static const struct dev_pm_ops virtio_pci_pm_ops = {
+	.suspend	= virtio_pci_suspend,
+	.resume		= virtio_pci_resume,
+	.freeze		= virtio_pci_freeze,
+	.thaw		= virtio_pci_thaw,
+	.restore	= virtio_pci_restore,
+	.poweroff	= virtio_pci_suspend,
+};
 #endif
 
 static struct pci_driver virtio_pci_driver = {
@@ -737,8 +836,7 @@ static struct pci_driver virtio_pci_driver = {
 	.probe		= virtio_pci_probe,
 	.remove		= __devexit_p(virtio_pci_remove),
 #ifdef CONFIG_PM
-	.suspend	= virtio_pci_suspend,
-	.resume		= virtio_pci_resume,
+	.driver.pm	= &virtio_pci_pm_ops,
 #endif
 };
 

drivers/virtio/virtio_ring.c

@@ -22,23 +22,27 @@
 #include <linux/device.h>
 #include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/hrtimer.h>
 
 /* virtio guest is communicating with a virtual "device" that actually runs on
  * a host processor.  Memory barriers are used to control SMP effects. */
 #ifdef CONFIG_SMP
 /* Where possible, use SMP barriers which are more lightweight than mandatory
  * barriers, because mandatory barriers control MMIO effects on accesses
- * through relaxed memory I/O windows (which virtio does not use). */
-#define virtio_mb() smp_mb()
-#define virtio_rmb() smp_rmb()
-#define virtio_wmb() smp_wmb()
+ * through relaxed memory I/O windows (which virtio-pci does not use). */
+#define virtio_mb(vq) \
+	do { if ((vq)->weak_barriers) smp_mb(); else mb(); } while(0)
+#define virtio_rmb(vq) \
+	do { if ((vq)->weak_barriers) smp_rmb(); else rmb(); } while(0)
+#define virtio_wmb(vq) \
+	do { if ((vq)->weak_barriers) smp_rmb(); else rmb(); } while(0)
 #else
 /* We must force memory ordering even if guest is UP since host could be
  * running on another CPU, but SMP barriers are defined to barrier() in that
  * configuration. So fall back to mandatory barriers instead. */
-#define virtio_mb() mb()
-#define virtio_rmb() rmb()
-#define virtio_wmb() wmb()
+#define virtio_mb(vq) mb()
+#define virtio_rmb(vq) rmb()
+#define virtio_wmb(vq) wmb()
 #endif
 
 #ifdef DEBUG
@@ -77,6 +81,9 @@ struct vring_virtqueue
 	/* Actual memory layout for this queue */
 	struct vring vring;
 
+	/* Can we use weak barriers? */
+	bool weak_barriers;
+
 	/* Other side has made a mess, don't try any more. */
 	bool broken;
 
@@ -102,6 +109,10 @@ struct vring_virtqueue
 #ifdef DEBUG
 	/* They're supposed to lock for us. */
 	unsigned int in_use;
+
+	/* Figure out if their kicks are too delayed. */
+	bool last_add_time_valid;
+	ktime_t last_add_time;
 #endif
 
 	/* Tokens for callbacks. */
@@ -160,12 +171,29 @@ static int vring_add_indirect(struct vring_virtqueue *vq,
 	return head;
 }
 
-int virtqueue_add_buf_gfp(struct virtqueue *_vq,
-			  struct scatterlist sg[],
-			  unsigned int out,
-			  unsigned int in,
-			  void *data,
-			  gfp_t gfp)
+/**
+ * virtqueue_add_buf - expose buffer to other end
+ * @vq: the struct virtqueue we're talking about.
+ * @sg: the description of the buffer(s).
+ * @out_num: the number of sg readable by other side
+ * @in_num: the number of sg which are writable (after readable ones)
+ * @data: the token identifying the buffer.
+ * @gfp: how to do memory allocations (if necessary).
+ *
+ * Caller must ensure we don't call this with other virtqueue operations
+ * at the same time (except where noted).
+ *
+ * Returns remaining capacity of queue or a negative error
+ * (ie. ENOSPC).  Note that it only really makes sense to treat all
+ * positive return values as "available": indirect buffers mean that
+ * we can put an entire sg[] array inside a single queue entry.
+ */
+int virtqueue_add_buf(struct virtqueue *_vq,
+		      struct scatterlist sg[],
+		      unsigned int out,
+		      unsigned int in,
+		      void *data,
+		      gfp_t gfp)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
 	unsigned int i, avail, uninitialized_var(prev);
@@ -175,6 +203,19 @@ int virtqueue_add_buf_gfp(struct virtqueue *_vq,
 
 	BUG_ON(data == NULL);
 
+#ifdef DEBUG
+	{
+		ktime_t now = ktime_get();
+
+		/* No kick or get, with .1 second between?  Warn. */
+		if (vq->last_add_time_valid)
+			WARN_ON(ktime_to_ms(ktime_sub(now, vq->last_add_time))
+					    > 100);
+		vq->last_add_time = now;
+		vq->last_add_time_valid = true;
+	}
+#endif
+
 	/* If the host supports indirect descriptor tables, and we have multiple
 	 * buffers, then go indirect. FIXME: tune this threshold */
 	if (vq->indirect && (out + in) > 1 && vq->num_free) {
@@ -227,40 +268,102 @@ add_head:
 	vq->data[head] = data;
 
 	/* Put entry in available array (but don't update avail->idx until they
-	 * do sync).  FIXME: avoid modulus here? */
-	avail = (vq->vring.avail->idx + vq->num_added++) % vq->vring.num;
+	 * do sync). */
+	avail = (vq->vring.avail->idx & (vq->vring.num-1));
 	vq->vring.avail->ring[avail] = head;
 
+	/* Descriptors and available array need to be set before we expose the
+	 * new available array entries. */
+	virtio_wmb(vq);
+	vq->vring.avail->idx++;
+	vq->num_added++;
+
+	/* This is very unlikely, but theoretically possible.  Kick
+	 * just in case. */
+	if (unlikely(vq->num_added == (1 << 16) - 1))
+		virtqueue_kick(_vq);
+
 	pr_debug("Added buffer head %i to %p\n", head, vq);
 	END_USE(vq);
 
 	return vq->num_free;
 }
-EXPORT_SYMBOL_GPL(virtqueue_add_buf_gfp);
+EXPORT_SYMBOL_GPL(virtqueue_add_buf);
 
-void virtqueue_kick(struct virtqueue *_vq)
+/**
+ * virtqueue_kick_prepare - first half of split virtqueue_kick call.
+ * @vq: the struct virtqueue
+ *
+ * Instead of virtqueue_kick(), you can do:
+ *	if (virtqueue_kick_prepare(vq))
+ *		virtqueue_notify(vq);
+ *
+ * This is sometimes useful because the virtqueue_kick_prepare() needs
+ * to be serialized, but the actual virtqueue_notify() call does not.
+ */
+bool virtqueue_kick_prepare(struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
 	u16 new, old;
+	bool needs_kick;
+
 	START_USE(vq);
 	/* Descriptors and available array need to be set before we expose the
 	 * new available array entries. */
-	virtio_wmb();
+	virtio_wmb(vq);
 
-	old = vq->vring.avail->idx;
-	new = vq->vring.avail->idx = old + vq->num_added;
+	old = vq->vring.avail->idx - vq->num_added;
+	new = vq->vring.avail->idx;
 	vq->num_added = 0;
 
-	/* Need to update avail index before checking if we should notify */
-	virtio_mb();
-
-	if (vq->event ?
-	    vring_need_event(vring_avail_event(&vq->vring), new, old) :
-	    !(vq->vring.used->flags & VRING_USED_F_NO_NOTIFY))
-		/* Prod other side to tell it about changes. */
-		vq->notify(&vq->vq);
+#ifdef DEBUG
+	if (vq->last_add_time_valid) {
+		WARN_ON(ktime_to_ms(ktime_sub(ktime_get(),
+					      vq->last_add_time)) > 100);
+	}
+	vq->last_add_time_valid = false;
+#endif
 
+	if (vq->event) {
+		needs_kick = vring_need_event(vring_avail_event(&vq->vring),
+					      new, old);
+	} else {
+		needs_kick = !(vq->vring.used->flags & VRING_USED_F_NO_NOTIFY);
+	}
 	END_USE(vq);
+	return needs_kick;
+}
+EXPORT_SYMBOL_GPL(virtqueue_kick_prepare);
+
+/**
+ * virtqueue_notify - second half of split virtqueue_kick call.
+ * @vq: the struct virtqueue
+ *
+ * This does not need to be serialized.
+ */
+void virtqueue_notify(struct virtqueue *_vq)
+{
+	struct vring_virtqueue *vq = to_vvq(_vq);
+
+	/* Prod other side to tell it about changes. */
+	vq->notify(_vq);
+}
+EXPORT_SYMBOL_GPL(virtqueue_notify);
+
+/**
+ * virtqueue_kick - update after add_buf
+ * @vq: the struct virtqueue
+ *
+ * After one or more virtqueue_add_buf calls, invoke this to kick
+ * the other side.
+ *
+ * Caller must ensure we don't call this with other virtqueue
+ * operations at the same time (except where noted).
+ */
+void virtqueue_kick(struct virtqueue *vq)
+{
+	if (virtqueue_kick_prepare(vq))
+		virtqueue_notify(vq);
 }
 EXPORT_SYMBOL_GPL(virtqueue_kick);
 
@@ -294,11 +397,28 @@ static inline bool more_used(const struct vring_virtqueue *vq)
 	return vq->last_used_idx != vq->vring.used->idx;
 }
 
+/**
+ * virtqueue_get_buf - get the next used buffer
+ * @vq: the struct virtqueue we're talking about.
+ * @len: the length written into the buffer
+ *
+ * If the driver wrote data into the buffer, @len will be set to the
+ * amount written.  This means you don't need to clear the buffer
+ * beforehand to ensure there's no data leakage in the case of short
+ * writes.
+ *
+ * Caller must ensure we don't call this with other virtqueue
+ * operations at the same time (except where noted).
+ *
+ * Returns NULL if there are no used buffers, or the "data" token
+ * handed to virtqueue_add_buf().
+ */
 void *virtqueue_get_buf(struct virtqueue *_vq, unsigned int *len)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
 	void *ret;
 	unsigned int i;
+	u16 last_used;
 
 	START_USE(vq);
 
@@ -314,10 +434,11 @@ void *virtqueue_get_buf(struct virtqueue *_vq, unsigned int *len)
 	}
 
 	/* Only get used array entries after they have been exposed by host. */
-	virtio_rmb();
+	virtio_rmb(vq);
 
-	i = vq->vring.used->ring[vq->last_used_idx%vq->vring.num].id;
-	*len = vq->vring.used->ring[vq->last_used_idx%vq->vring.num].len;
+	last_used = (vq->last_used_idx & (vq->vring.num - 1));
+	i = vq->vring.used->ring[last_used].id;
+	*len = vq->vring.used->ring[last_used].len;
 
 	if (unlikely(i >= vq->vring.num)) {
 		BAD_RING(vq, "id %u out of range\n", i);
@@ -337,14 +458,27 @@ void *virtqueue_get_buf(struct virtqueue *_vq, unsigned int *len)
 	 * the read in the next get_buf call. */
 	if (!(vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)) {
 		vring_used_event(&vq->vring) = vq->last_used_idx;
-		virtio_mb();
+		virtio_mb(vq);
 	}
 
+#ifdef DEBUG
+	vq->last_add_time_valid = false;
+#endif
+
 	END_USE(vq);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(virtqueue_get_buf);
 
+/**
+ * virtqueue_disable_cb - disable callbacks
+ * @vq: the struct virtqueue we're talking about.
+ *
+ * Note that this is not necessarily synchronous, hence unreliable and only
+ * useful as an optimization.
+ *
+ * Unlike other operations, this need not be serialized.
+ */
 void virtqueue_disable_cb(struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
@@ -353,6 +487,17 @@ void virtqueue_disable_cb(struct virtqueue *_vq)
 }
 EXPORT_SYMBOL_GPL(virtqueue_disable_cb);
 
+/**
+ * virtqueue_enable_cb - restart callbacks after disable_cb.
+ * @vq: the struct virtqueue we're talking about.
+ *
+ * This re-enables callbacks; it returns "false" if there are pending
+ * buffers in the queue, to detect a possible race between the driver
+ * checking for more work, and enabling callbacks.
+ *
+ * Caller must ensure we don't call this with other virtqueue
+ * operations at the same time (except where noted).
+ */
 bool virtqueue_enable_cb(struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
@@ -366,7 +511,7 @@ bool virtqueue_enable_cb(struct virtqueue *_vq)
 	 * entry. Always do both to keep code simple. */
 	vq->vring.avail->flags &= ~VRING_AVAIL_F_NO_INTERRUPT;
 	vring_used_event(&vq->vring) = vq->last_used_idx;
-	virtio_mb();
+	virtio_mb(vq);
 	if (unlikely(more_used(vq))) {
 		END_USE(vq);
 		return false;
@@ -377,6 +522,19 @@ bool virtqueue_enable_cb(struct virtqueue *_vq)
 }
 EXPORT_SYMBOL_GPL(virtqueue_enable_cb);
 
+/**
+ * virtqueue_enable_cb_delayed - restart callbacks after disable_cb.
+ * @vq: the struct virtqueue we're talking about.
+ *
+ * This re-enables callbacks but hints to the other side to delay
+ * interrupts until most of the available buffers have been processed;
+ * it returns "false" if there are many pending buffers in the queue,
+ * to detect a possible race between the driver checking for more work,
+ * and enabling callbacks.
+ *
+ * Caller must ensure we don't call this with other virtqueue
+ * operations at the same time (except where noted).
+ */
 bool virtqueue_enable_cb_delayed(struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
@@ -393,7 +551,7 @@ bool virtqueue_enable_cb_delayed(struct virtqueue *_vq)
 	/* TODO: tune this threshold */
 	bufs = (u16)(vq->vring.avail->idx - vq->last_used_idx) * 3 / 4;
 	vring_used_event(&vq->vring) = vq->last_used_idx + bufs;
-	virtio_mb();
+	virtio_mb(vq);
 	if (unlikely((u16)(vq->vring.used->idx - vq->last_used_idx) > bufs)) {
 		END_USE(vq);
 		return false;
@@ -404,6 +562,14 @@ bool virtqueue_enable_cb_delayed(struct virtqueue *_vq)
 }
 EXPORT_SYMBOL_GPL(virtqueue_enable_cb_delayed);
 
+/**
+ * virtqueue_detach_unused_buf - detach first unused buffer
+ * @vq: the struct virtqueue we're talking about.
+ *
+ * Returns NULL or the "data" token handed to virtqueue_add_buf().
+ * This is not valid on an active queue; it is useful only for device
+ * shutdown.
+ */
 void *virtqueue_detach_unused_buf(struct virtqueue *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
@@ -453,6 +619,7 @@ EXPORT_SYMBOL_GPL(vring_interrupt);
 struct virtqueue *vring_new_virtqueue(unsigned int num,
 				      unsigned int vring_align,
 				      struct virtio_device *vdev,
+				      bool weak_barriers,
 				      void *pages,
 				      void (*notify)(struct virtqueue *),
 				      void (*callback)(struct virtqueue *),
@@ -476,12 +643,14 @@ struct virtqueue *vring_new_virtqueue(unsigned int num,
 	vq->vq.vdev = vdev;
 	vq->vq.name = name;
 	vq->notify = notify;
+	vq->weak_barriers = weak_barriers;
 	vq->broken = false;
 	vq->last_used_idx = 0;
 	vq->num_added = 0;
 	list_add_tail(&vq->vq.list, &vdev->vqs);
 #ifdef DEBUG
 	vq->in_use = false;
+	vq->last_add_time_valid = false;
 #endif
 
 	vq->indirect = virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC);
@@ -530,7 +699,13 @@ void vring_transport_features(struct virtio_device *vdev)
 }
 EXPORT_SYMBOL_GPL(vring_transport_features);
 
-/* return the size of the vring within the virtqueue */
+/**
+ * virtqueue_get_vring_size - return the size of the virtqueue's vring
+ * @vq: the struct virtqueue containing the vring of interest.
+ *
+ * Returns the size of the vring.  This is mainly used for boasting to
+ * userspace.  Unlike other operations, this need not be serialized.
+ */
 unsigned int virtqueue_get_vring_size(struct virtqueue *_vq)
 {
 

include/linux/virtio.h

@@ -25,70 +25,18 @@ struct virtqueue {
 	void *priv;
 };
 
-/**
- * operations for virtqueue
- * virtqueue_add_buf: expose buffer to other end
- *	vq: the struct virtqueue we're talking about.
- *	sg: the description of the buffer(s).
- *	out_num: the number of sg readable by other side
- *	in_num: the number of sg which are writable (after readable ones)
- *	data: the token identifying the buffer.
- *	gfp: how to do memory allocations (if necessary).
- *      Returns remaining capacity of queue (sg segments) or a negative error.
- * virtqueue_kick: update after add_buf
- *	vq: the struct virtqueue
- *	After one or more add_buf calls, invoke this to kick the other side.
- * virtqueue_get_buf: get the next used buffer
- *	vq: the struct virtqueue we're talking about.
- *	len: the length written into the buffer
- *	Returns NULL or the "data" token handed to add_buf.
- * virtqueue_disable_cb: disable callbacks
- *	vq: the struct virtqueue we're talking about.
- *	Note that this is not necessarily synchronous, hence unreliable and only
- *	useful as an optimization.
- * virtqueue_enable_cb: restart callbacks after disable_cb.
- *	vq: the struct virtqueue we're talking about.
- *	This re-enables callbacks; it returns "false" if there are pending
- *	buffers in the queue, to detect a possible race between the driver
- *	checking for more work, and enabling callbacks.
- * virtqueue_enable_cb_delayed: restart callbacks after disable_cb.
- *	vq: the struct virtqueue we're talking about.
- *	This re-enables callbacks but hints to the other side to delay
- *	interrupts until most of the available buffers have been processed;
- *	it returns "false" if there are many pending buffers in the queue,
- *	to detect a possible race between the driver checking for more work,
- *	and enabling callbacks.
- * virtqueue_detach_unused_buf: detach first unused buffer
- * 	vq: the struct virtqueue we're talking about.
- * 	Returns NULL or the "data" token handed to add_buf
- * virtqueue_get_vring_size: return the size of the virtqueue's vring
- *	vq: the struct virtqueue containing the vring of interest.
- *	Returns the size of the vring.
- *
- * Locking rules are straightforward: the driver is responsible for
- * locking.  No two operations may be invoked simultaneously, with the exception
- * of virtqueue_disable_cb.
- *
- * All operations can be called in any context.
- */
+int virtqueue_add_buf(struct virtqueue *vq,
+		      struct scatterlist sg[],
+		      unsigned int out_num,
+		      unsigned int in_num,
+		      void *data,
+		      gfp_t gfp);
 
-int virtqueue_add_buf_gfp(struct virtqueue *vq,
-			  struct scatterlist sg[],
-			  unsigned int out_num,
-			  unsigned int in_num,
-			  void *data,
-			  gfp_t gfp);
+void virtqueue_kick(struct virtqueue *vq);
 
-static inline int virtqueue_add_buf(struct virtqueue *vq,
-				    struct scatterlist sg[],
-				    unsigned int out_num,
-				    unsigned int in_num,
-				    void *data)
-{
-	return virtqueue_add_buf_gfp(vq, sg, out_num, in_num, data, GFP_ATOMIC);
-}
+bool virtqueue_kick_prepare(struct virtqueue *vq);
 
-void virtqueue_kick(struct virtqueue *vq);
+void virtqueue_notify(struct virtqueue *vq);
 
 void *virtqueue_get_buf(struct virtqueue *vq, unsigned int *len);
 
@@ -146,6 +94,11 @@ struct virtio_driver {
 	int (*probe)(struct virtio_device *dev);
 	void (*remove)(struct virtio_device *dev);
 	void (*config_changed)(struct virtio_device *dev);
+#ifdef CONFIG_PM
+	int (*freeze)(struct virtio_device *dev);
+	int (*thaw)(struct virtio_device *dev);
+	int (*restore)(struct virtio_device *dev);
+#endif
 };
 
 int register_virtio_driver(struct virtio_driver *drv);

include/linux/virtio_ring.h

@@ -168,6 +168,7 @@ struct virtqueue;
 struct virtqueue *vring_new_virtqueue(unsigned int num,
 				      unsigned int vring_align,
 				      struct virtio_device *vdev,
+				      bool weak_barriers,
 				      void *pages,
 				      void (*notify)(struct virtqueue *vq),
 				      void (*callback)(struct virtqueue *vq),

net/9p/trans_virtio.c

@@ -272,7 +272,8 @@ req_retry:
 	in = pack_sg_list(chan->sg, out,
 			  VIRTQUEUE_NUM, req->rc->sdata, req->rc->capacity);
 
-	err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc);
+	err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc,
+				GFP_ATOMIC);
 	if (err < 0) {
 		if (err == -ENOSPC) {
 			chan->ring_bufs_avail = 0;
@@ -414,7 +415,8 @@ req_retry_pinned:
 		in += pack_sg_list_p(chan->sg, out + in, VIRTQUEUE_NUM,
 				     in_pages, in_nr_pages, uidata, inlen);
 
-	err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc);
+	err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc,
+				GFP_ATOMIC);
 	if (err < 0) {
 		if (err == -ENOSPC) {
 			chan->ring_bufs_avail = 0;

Documentation/virtual/lguest/lguest.c → tools/lguest/lguest.c

@@ -49,7 +49,7 @@
 #include <linux/virtio_rng.h>
 #include <linux/virtio_ring.h>
 #include <asm/bootparam.h>
-#include "../../../include/linux/lguest_launcher.h"
+#include "../../include/linux/lguest_launcher.h"
 /*L:110
  * We can ignore the 43 include files we need for this program, but I do want
  * to draw attention to the use of kernel-style types.

tools/virtio/linux/virtio.h

@@ -186,21 +186,12 @@ struct virtqueue {
 #endif
 
 /* Interfaces exported by virtio_ring. */
-int virtqueue_add_buf_gfp(struct virtqueue *vq,
-			  struct scatterlist sg[],
-			  unsigned int out_num,
-			  unsigned int in_num,
-			  void *data,
-			  gfp_t gfp);
-
-static inline int virtqueue_add_buf(struct virtqueue *vq,
-				    struct scatterlist sg[],
-				    unsigned int out_num,
-				    unsigned int in_num,
-				    void *data)
-{
-	return virtqueue_add_buf_gfp(vq, sg, out_num, in_num, data, GFP_ATOMIC);
-}
+int virtqueue_add_buf(struct virtqueue *vq,
+		      struct scatterlist sg[],
+		      unsigned int out_num,
+		      unsigned int in_num,
+		      void *data,
+		      gfp_t gfp);
 
 void virtqueue_kick(struct virtqueue *vq);
 
@@ -214,6 +205,7 @@ void *virtqueue_detach_unused_buf(struct virtqueue *vq);
 struct virtqueue *vring_new_virtqueue(unsigned int num,
 				      unsigned int vring_align,
 				      struct virtio_device *vdev,
+				      bool weak_barriers,
 				      void *pages,
 				      void (*notify)(struct virtqueue *vq),
 				      void (*callback)(struct virtqueue *vq),

tools/virtio/virtio_test.c

@@ -92,7 +92,8 @@ static void vq_info_add(struct vdev_info *dev, int num)
 	assert(r >= 0);
 	memset(info->ring, 0, vring_size(num, 4096));
 	vring_init(&info->vring, num, info->ring, 4096);
-	info->vq = vring_new_virtqueue(info->vring.num, 4096, &dev->vdev, info->ring,
+	info->vq = vring_new_virtqueue(info->vring.num, 4096, &dev->vdev,
+				       true, info->ring,
 				       vq_notify, vq_callback, "test");
 	assert(info->vq);
 	info->vq->priv = info;
@@ -160,7 +161,8 @@ static void run_test(struct vdev_info *dev, struct vq_info *vq, int bufs)
 			if (started < bufs) {
 				sg_init_one(&sl, dev->buf, dev->buf_size);
 				r = virtqueue_add_buf(vq->vq, &sl, 1, 0,
-						      dev->buf + started);
+						      dev->buf + started,
+						      GFP_ATOMIC);
 				if (likely(r >= 0)) {
 					++started;
 					virtqueue_kick(vq->vq);