iosignalfd

Today kvm_io_bus_regsiter_dev() returns void and will internally BUG_ON if it fails. We want to create dynamic MMIO/PIO entries driven from userspace later in the series, so we need to enhance the code to be more robust with the following changes:

1) Add a return value to the registration function
2) Fix up all the callsites to check the return code, handle any failures, and percolate the error up to the caller.
3) Add an unregister function that collapses holes in the array

Signed-off-by Gregory Haskins <[email protected]>

---

 arch/x86/kvm/i8254.c      |   22 ++++++++++++++++++++--
 arch/x86/kvm/i8259.c      |    9 ++++++++-
 include/linux/kvm_host.h  |    6 ++++--
 virt/kvm/coalesced_mmio.c |    8 ++++++--
 virt/kvm/ioapic.c         |    9 +++++++--
 virt/kvm/kvm_main.c       |   23 +++++++++++++++++++++--
 6 files changed, 66 insertions(+), 11 deletions(-)

arch/x86/kvm/i8254.c +20 -2

index 331705f..1c41715 100644
@@ -586,6 +586,7 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm, u32 flags)
 {
 	struct kvm_pit *pit;
 	struct kvm_kpit_state *pit_state;
+	int ret;
 
 	pit = kzalloc(sizeof(struct kvm_pit), GFP_KERNEL);
 	if (!pit)

@@ -620,14 +621,31 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm, u32 flags)
 	kvm_register_irq_mask_notifier(kvm, 0, &pit->mask_notifier);
 
 	kvm_iodevice_init(&pit->dev, &pit_dev_ops);
-	kvm_io_bus_register_dev(&kvm->pio_bus, &pit->dev);
+	ret = kvm_io_bus_register_dev(&kvm->pio_bus, &pit->dev);
+	if (ret < 0)
+		goto fail;
 
 	if (flags & KVM_PIT_SPEAKER_DUMMY) {
 		kvm_iodevice_init(&pit->speaker_dev, &speaker_dev_ops);
-		kvm_io_bus_register_dev(&kvm->pio_bus, &pit->speaker_dev);
+		ret = kvm_io_bus_register_dev(&kvm->pio_bus,
+					      &pit->speaker_dev);
+		if (ret < 0)
+			goto fail;
 	}
 
 	return pit;
+
+fail:
+	if (flags & KVM_PIT_SPEAKER_DUMMY)
+		kvm_io_bus_unregister_dev(&kvm->pio_bus, &pit->speaker_dev);
+
+	kvm_io_bus_unregister_dev(&kvm->pio_bus, &pit->dev);
+
+	if (pit->irq_source_id >= 0)
+		kvm_free_irq_source_id(kvm, pit->irq_source_id);
+
+	kfree(pit);
+	return NULL;
 }
 
 void kvm_free_pit(struct kvm *kvm)

arch/x86/kvm/i8259.c +8 -1

index 148c52a..66e37c2 100644
@@ -532,6 +532,8 @@ static const struct kvm_io_device_ops picdev_ops = {
 struct kvm_pic *kvm_create_pic(struct kvm *kvm)
 {
 	struct kvm_pic *s;
+	int ret;
+
 	s = kzalloc(sizeof(struct kvm_pic), GFP_KERNEL);
 	if (!s)
 		return NULL;

@@ -548,6 +550,11 @@ struct kvm_pic *kvm_create_pic(struct kvm *kvm)
 	 * Initialize PIO device
 	 */
 	kvm_iodevice_init(&s->dev, &picdev_ops);
-	kvm_io_bus_register_dev(&kvm->pio_bus, &s->dev);
+	ret = kvm_io_bus_register_dev(&kvm->pio_bus, &s->dev);
+	if (ret < 0) {
+		kfree(s);
+		return NULL;
+	}
+
 	return s;
 }

include/linux/kvm_host.h +4 -2

index e4e78db..eafa2b3 100644
@@ -61,8 +61,10 @@ void kvm_io_bus_init(struct kvm_io_bus *bus);
 void kvm_io_bus_destroy(struct kvm_io_bus *bus);
 struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus,
 					  gpa_t addr, int len, int is_write);
-void kvm_io_bus_register_dev(struct kvm_io_bus *bus,
-			     struct kvm_io_device *dev);
+int kvm_io_bus_register_dev(struct kvm_io_bus *bus,
+			    struct kvm_io_device *dev);
+void kvm_io_bus_unregister_dev(struct kvm_io_bus *bus,
+			    struct kvm_io_device *dev);
 
 struct kvm_vcpu {
 	struct kvm *kvm;

virt/kvm/coalesced_mmio.c +6 -2

index 397f419..3e89db8 100644
@@ -94,6 +94,7 @@ static const struct kvm_io_device_ops coalesced_mmio_ops = {
 int kvm_coalesced_mmio_init(struct kvm *kvm)
 {
 	struct kvm_coalesced_mmio_dev *dev;
+	int ret;
 
 	dev = kzalloc(sizeof(struct kvm_coalesced_mmio_dev), GFP_KERNEL);
 	if (!dev)

@@ -102,9 +103,12 @@ int kvm_coalesced_mmio_init(struct kvm *kvm)
 	kvm_iodevice_init(&dev->dev, &coalesced_mmio_ops);
 	dev->kvm = kvm;
 	kvm->coalesced_mmio_dev = dev;
-	kvm_io_bus_register_dev(&kvm->mmio_bus, &dev->dev);
 
-	return 0;
+	ret = kvm_io_bus_register_dev(&kvm->mmio_bus, &dev->dev);
+	if (ret < 0)
+		kfree(dev);
+
+	return ret;
 }
 
 int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm,

virt/kvm/ioapic.c +7 -2

index d8b2eca..28adb05 100644
@@ -335,6 +335,7 @@ static const struct kvm_io_device_ops ioapic_mmio_ops = {
 int kvm_ioapic_init(struct kvm *kvm)
 {
 	struct kvm_ioapic *ioapic;
+	int ret;
 
 	ioapic = kzalloc(sizeof(struct kvm_ioapic), GFP_KERNEL);
 	if (!ioapic)

@@ -343,7 +344,11 @@ int kvm_ioapic_init(struct kvm *kvm)
 	kvm_ioapic_reset(ioapic);
 	kvm_iodevice_init(&ioapic->dev, &ioapic_mmio_ops);
 	ioapic->kvm = kvm;
-	kvm_io_bus_register_dev(&kvm->mmio_bus, &ioapic->dev);
-	return 0;
+
+	ret = kvm_io_bus_register_dev(&kvm->mmio_bus, &ioapic->dev);
+	if (ret < 0)
+		kfree(ioapic);
+
+	return ret;
 }
 

virt/kvm/kvm_main.c +21 -3

index 9fab08e..6bd71f7 100644
@@ -2481,11 +2481,30 @@ struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus,
 	return NULL;
 }
 
-void kvm_io_bus_register_dev(struct kvm_io_bus *bus, struct kvm_io_device *dev)
+/* assumes kvm->lock held */
+int kvm_io_bus_register_dev(struct kvm_io_bus *bus, struct kvm_io_device *dev)
 {
-	BUG_ON(bus->dev_count > (NR_IOBUS_DEVS-1));
+	if (bus->dev_count > (NR_IOBUS_DEVS-1))
+		return -ENOSPC;
 
 	bus->devs[bus->dev_count++] = dev;
+
+	return 0;
+}
+
+/* assumes kvm->lock held */
+void kvm_io_bus_unregister_dev(struct kvm_io_bus *bus,
+			       struct kvm_io_device *dev)
+{
+	int i;
+
+	for (i = 0; i < bus->dev_count; i++) {
+
+		if (bus->devs[i] == dev) {
+			bus->devs[i] = bus->devs[--bus->dev_count];
+			return;
+		}
+	}
 }
 
 static struct notifier_block kvm_cpu_notifier = {
 
--

To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Quoted 9 lines · Gregory Haskins

Today kvm_io_bus_regsiter_dev() returns void and will internally BUG_ON if it fails. We want to create dynamic MMIO/PIO entries driven from userspace later in the series, so we need to enhance the code to be more robust with the following changes:

1) Add a return value to the registration function
2) Fix up all the callsites to check the return code, handle any failures, and percolate the error up to the caller.
3) Add an unregister function that collapses holes in the array

Quoted 7 lines · Gregory Haskins

Quoted 5 lines · Gregory Haskins

Today kvm_io_bus_regsiter_dev() returns void and will internally BUG_ON if it fails. We want to create dynamic MMIO/PIO entries driven from userspace later in the series, so we need to enhance the code to be more robust with the following changes:

Quoted 5 lines

Quoted 5 lines

1) Add a return value to the registration function
2) Fix up all the callsites to check the return code, handle any failures, and percolate the error up to the caller.
3) Add an unregister function that collapses holes in the array

Does not unregister need rcu bus fixes to work?

iosignalfd is a mechanism to register PIO/MMIO regions to trigger an eventfd signal when written to by a guest. Host userspace can register any arbitrary IO address with a corresponding eventfd and then pass the eventfd to a specific end-point of interest for handling.

Normal IO requires a blocking round-trip since the operation may cause side-effects in the emulated model or may return data to the caller. Therefore, an IO in KVM traps from the guest to the host, causes a VMX/SVM "heavy-weight" exit back to userspace, and is ultimately serviced by qemu's device model synchronously before returning control back to the vcpu.

However, there is a subclass of IO which acts purely as a trigger for other IO (such as to kick off an out-of-band DMA request, etc). For these patterns, the synchronous call is particularly expensive since we really only want to simply get our notification transmitted asychronously and return as quickly as possible. All the sychronous infrastructure to ensure proper data-dependencies are met in the normal IO case are just unecessary overhead for signalling. This adds additional computational load on the system, as well as latency to the signalling path.

Therefore, we provide a mechanism for registration of an in-kernel trigger point that allows the VCPU to only require a very brief, lightweight exit just long enough to signal an eventfd. This also means that any clients compatible with the eventfd interface (which includes userspace and kernelspace equally well) can now register to be notified. The end result should be a more flexible and higher performance notification API for the backend KVM hypervisor and perhipheral components.

To test this theory, we built a test-harness called "doorbell". This module has a function called "doorbell_ring()" which simply increments a counter for each time the doorbell is signaled. It supports signalling from either an eventfd, or an ioctl().

We then wired up two paths to the doorbell: One via QEMU via a registered io region and through the doorbell ioctl(). The other is direct via iosignalfd.

You can download this test harness here:

ftp://ftp.novell.com/dev/ghaskins/doorbell.tar.bz2

The measured results are as follows:

qemu-mmio:       110000 iops, 9.09us rtt
iosignalfd-mmio: 200100 iops, 5.00us rtt
iosignalfd-pio:  367300 iops, 2.72us rtt

I didn't measure qemu-pio, because I have to figure out how to register a PIO region with qemu's device model, and I got lazy. However, for now we can extrapolate based on the data from the NULLIO runs of +2.56us for MMIO, and -350ns for HC, we get:

qemu-pio:      153139 iops, 6.53us rtt
iosignalfd-hc: 412585 iops, 2.37us rtt

these are just for fun, for now, until I can gather more data.

Here is a graph for your convenience:

http://developer.novell.com/wiki/images/7/76/Iofd-chart.png

The conclusion to draw is that we save about 4us by skipping the userspace hop.

--------------------

Signed-off-by Gregory Haskins <[email protected]>

---

 arch/x86/kvm/x86.c       |    1 
 include/linux/kvm.h      |   15 ++
 include/linux/kvm_host.h |   10 +
 virt/kvm/eventfd.c       |  426 ++++++++++++++++++++++++++++++++++++++++++++++
 virt/kvm/kvm_main.c      |   11 +
 5 files changed, 459 insertions(+), 4 deletions(-)

arch/x86/kvm/x86.c +1 -0

index 1b91ea7..9b119e4 100644
@@ -1097,6 +1097,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_IRQ_INJECT_STATUS:
 	case KVM_CAP_ASSIGN_DEV_IRQ:
 	case KVM_CAP_IRQFD:
+	case KVM_CAP_IOSIGNALFD:
 	case KVM_CAP_PIT2:
 		r = 1;
 		break;

include/linux/kvm.h +0 -0

index 38ff31e..9de6486 100644

include/linux/kvm.h +4 -0

+};
+
 #define KVM_TRC_SHIFT           16
 /*
  * kvm trace categories

@@ -438,6 +451,7 @@ struct kvm_trace_rec {
 #define KVM_CAP_PIT2 33
 #endif
 #define KVM_CAP_SET_BOOT_CPU_ID 34
+#define KVM_CAP_IOSIGNALFD 35
 
 #ifdef KVM_CAP_IRQ_ROUTING
 

@@ -544,6 +558,7 @@ struct kvm_irqfd {
 #define KVM_IRQFD                  _IOW(KVMIO, 0x76, struct kvm_irqfd)
 #define KVM_CREATE_PIT2		   _IOW(KVMIO, 0x77, struct kvm_pit_config)
 #define KVM_SET_BOOT_CPU_ID        _IO(KVMIO, 0x78)
+#define KVM_IOSIGNALFD             _IOW(KVMIO, 0x79, struct kvm_iosignalfd)
 
 /*
  * ioctls for vcpu fds

include/linux/kvm_host.h +0 -0

index 707c4d8..6c0569a 100644

include/linux/kvm_host.h +4 -2

 	struct list_head irqfds;
+	struct list_head iosignalfds;
 #endif
 	struct kvm_vm_stat stat;
 	struct kvm_arch arch;

@@ -554,19 +555,24 @@ static inline void kvm_free_irq_routing(struct kvm *kvm) {}
 
 #ifdef CONFIG_HAVE_KVM_EVENTFD
 
-void kvm_irqfd_init(struct kvm *kvm);
+void kvm_eventfd_init(struct kvm *kvm);
 int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags);
 void kvm_irqfd_release(struct kvm *kvm);
+int kvm_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args);
 
 #else
 
-static inline void kvm_irqfd_init(struct kvm *kvm) {}
+static inline void kvm_eventfd_init(struct kvm *kvm) {}

include/linux/kvm_host.h +0 -0

 #endif /* CONFIG_HAVE_KVM_EVENTFD */
 

virt/kvm/eventfd.c +6 -1

index 2c8028c..52ac455 100644
@@ -21,6 +21,7 @@
  */
 
 #include <linux/kvm_host.h>
+#include <linux/kvm.h>
 #include <linux/workqueue.h>
 #include <linux/syscalls.h>
 #include <linux/wait.h>

@@ -29,6 +30,9 @@
 #include <linux/list.h>
 #include <linux/eventfd.h>
 #include <linux/srcu.h>
+#include <linux/kernel.h>
+
+#include "iodev.h"
 
 /*
  * --------------------------------------------------------------------

@@ -202,9 +206,10 @@ fail:
 }
 
 void
-kvm_irqfd_init(struct kvm *kvm)
+kvm_eventfd_init(struct kvm *kvm)
 {
 	INIT_LIST_HEAD(&kvm->irqfds);
+	INIT_LIST_HEAD(&kvm->iosignalfds);
 }
 
 void

virt/kvm/eventfd.c +51 -0

+ */
+
+struct _iosignalfd_item {
+	struct list_head     list;
+	struct file         *file;
+	u64                  match;
+	struct rcu_head      rcu;
+	int                  wildcard:1;
+};
+
+struct _iosignalfd_group {
+	struct list_head     list;
+	u64                  addr;
+	size_t               length;
+	size_t               count;
+	struct list_head     items;
+	struct kvm_io_device dev;
+	struct rcu_head      rcu;
+};
+
+static inline struct _iosignalfd_group *
+to_group(struct kvm_io_device *dev)
+{
+	return container_of(dev, struct _iosignalfd_group, dev);
+}
+
+static void
+iosignalfd_item_free(struct _iosignalfd_item *item)
+{
+	fput(item->file);
+	kfree(item);
+}
+
+static void
+iosignalfd_item_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_item *item;
+
+	item = container_of(rhp, struct _iosignalfd_item, rcu);
+
+	iosignalfd_item_free(item);
+}
+
+static void
+iosignalfd_group_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_group *group;
+
+	group = container_of(rhp, struct _iosignalfd_group, rcu);
+
+	kfree(group);

virt/kvm/eventfd.c +7 -0

+
+	switch (len) {
+	case 1:
+		_val = *(u8 *)val;
+		break;
+	case 2:
+		_val = *(u16 *)val;

virt/kvm/eventfd.c +24 -0

+
+	return _val == item->match;
+}
+
+/*
+ * MMIO/PIO writes trigger an event (if the data matches).
+ *
+ * This is invoked by the io_bus subsystem in response to an address match
+ * against the group.  We must then walk the list of individual items to check
+ * for a match and, if applicable, to send the appropriate signal. If the item
+ * in question does not have a "match" pointer, it is considered a wildcard
+ * and will always generate a signal.  There can be an arbitrary number
+ * of distinct matches or wildcards per group.
+ */
+static void
+iosignalfd_group_write(struct kvm_io_device *this, gpa_t addr, int len,
+		       const void *val)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(item, &group->items, list) {

virt/kvm/eventfd.c +31 -0

+static void
+iosignalfd_group_read(struct kvm_io_device *this, gpa_t addr, int len,
+		      void *val)
+{
+	memset(val, 0, len);
+}
+
+/*
+ * This function is called as KVM is completely shutting down.  We do not
+ * need to worry about locking or careful RCU dancing...just nuke anything
+ * we have as quickly as possible
+ */
+static void
+iosignalfd_group_destructor(struct kvm_io_device *this)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item, *tmp;
+
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+		list_del(&item->list);
+		iosignalfd_item_free(item);
+	}
+
+	list_del(&group->list);
+	kfree(group);
+}
+
+static const struct kvm_io_device_ops iosignalfd_ops = {
+	.read       = iosignalfd_group_read,
+	.write      = iosignalfd_group_write,
+	.in_range   = iosignalfd_group_in_range,

virt/kvm/eventfd.c +32 -0

+		if (group->addr == addr)
+			return group;
+	}
+
+	return NULL;
+}
+
+/* assumes kvm->lock is held */
+static struct _iosignalfd_group *
+iosignalfd_group_create(struct kvm *kvm, struct kvm_io_bus *bus,
+			u64 addr, size_t len)
+{
+	struct _iosignalfd_group *group;
+	int ret;
+
+	group = kzalloc(sizeof(*group), GFP_KERNEL);
+	if (!group)
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&group->list);
+	INIT_LIST_HEAD(&group->items);
+	group->addr   = addr;
+	group->length = len;
+	kvm_iodevice_init(&group->dev, &iosignalfd_ops);
+
+	ret = kvm_io_bus_register_dev(kvm, bus, &group->dev);
+	if (ret < 0) {
+		kfree(group);
+		return ERR_PTR(ret);
+	}
+
+	list_add_tail(&group->list, &kvm->iosignalfds);

virt/kvm/eventfd.c +18 -0

+		return -EINVAL;
 
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	item = kzalloc(sizeof(*item), GFP_KERNEL);
+	if (!item) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	INIT_LIST_HEAD(&item->list);
+	item->file = file;
+
+	/*
+	 * A trigger address is optional, otherwise this is a wildcard
+	 */

virt/kvm/eventfd.c +20 -0

+	if (kvm->io_device_count >= CONFIG_KVM_MAX_IO_DEVICES) {
+		ret = -ENOSPC;
+		goto unlock_fail;
+	}
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+
+		group = iosignalfd_group_create(kvm, bus,
+						args->addr, args->len);
+		if (IS_ERR(group)) {
+			ret = PTR_ERR(group);
+			goto unlock_fail;
+		}
+
+		/*
+		 * Note: We do not increment io_device_count for the first item,
+		 * as this is represented by the group device that we just
+		 * registered.  Make sure we handle this properly when we
+		 * deassign the last item

virt/kvm/eventfd.c +93 -0

+		}
+
+		kvm->io_device_count++;
+	}
+
+	/*
+	 * Note: We are committed to succeed at this point since we have
+	 * (potentially) published a new group-device.  Any failure handling
+	 * added in the future after this point will need to be carefully
+	 * considered.
+	 */
+
+	list_add_tail_rcu(&item->list, &group->items);
+	group->count++;
+
+	mutex_unlock(&kvm->lock);
+
+	return 0;
+
+unlock_fail:
+	mutex_unlock(&kvm->lock);
+fail:
+	if (item)
+		/*
+		 * it would have never made it to the group->items list
+		 * in the failure path, so we dont need to worry about removing
+		 * it
+		 */
+		kfree(item);
+
+	fput(file);
+
+	return ret;
+}
+
+
+static int
+kvm_deassign_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	int                       pio = args->flags & KVM_IOSIGNALFD_FLAG_PIO;
+	struct kvm_io_bus        *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
+	struct _iosignalfd_group *group;
+	struct _iosignalfd_item  *item, *tmp;
+	struct file              *file;
+	int                       ret = 0;
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	mutex_lock(&kvm->lock);
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * Exhaustively search our group->items list for any items that might
+	 * match the specified fd, and (carefully) remove each one found.
+	 */
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+
+		if (item->file != file)
+			continue;
+
+		list_del_rcu(&item->list);
+		group->count--;
+		if (group->count)
+			/*
+			 * We only decrement the global count if this is *not*
+			 * the last item.  The last item will be accounted for
+			 * by the io_bus_unregister
+			 */
+			kvm->io_device_count--;
+
+		/*
+		 * The item may be still referenced inside our group->write()
+		 * path's RCU read-side CS, so defer the actual free to the
+		 * next grace
+		 */
+		call_rcu(&item->rcu, iosignalfd_item_deferred_free);
+	}
+
+	/*
+	 * Check if the group is now completely vacated as a result of
+	 * removing the items.  If so, unregister/delete it
+	 */
+	if (!group->count) {
+
+		kvm_io_bus_unregister_dev(kvm, bus, &group->dev);
+

virt/kvm/eventfd.c +18 -0

+	}
+
+out:
+	mutex_unlock(&kvm->lock);
+
+	fput(file);
+
+	return ret;
+}
+
+int
+kvm_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	if (args->flags & KVM_IOSIGNALFD_FLAG_DEASSIGN)
+		return kvm_deassign_iosignalfd(kvm, args);
+
+	return kvm_assign_iosignalfd(kvm, args);
+}

virt/kvm/kvm_main.c +10 -2

index 42cbea7..e6495d4 100644
@@ -971,7 +971,7 @@ static struct kvm *kvm_create_vm(void)
 	atomic_inc(&kvm->mm->mm_count);
 	spin_lock_init(&kvm->mmu_lock);
 	kvm_io_bus_init(&kvm->pio_bus);
-	kvm_irqfd_init(kvm);
+	kvm_eventfd_init(kvm);
 	mutex_init(&kvm->lock);
 	mutex_init(&kvm->irq_lock);
 	kvm_io_bus_init(&kvm->mmio_bus);

@@ -2227,6 +2227,15 @@ static long kvm_vm_ioctl(struct file *filp,
 		r = kvm_irqfd(kvm, data.fd, data.gsi, data.flags);
 		break;
 	}
+	case KVM_IOSIGNALFD: {
+		struct kvm_iosignalfd data;
+
+		r = -EFAULT;
+		if (copy_from_user(&data, argp, sizeof data))
+			goto out;
+		r = kvm_iosignalfd(kvm, &data);
+		break;
+	}
 #ifdef CONFIG_KVM_APIC_ARCHITECTURE
 	case KVM_SET_BOOT_CPU_ID:
 		r = 0;
 
--

To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Some comments below:

Quoted 96 lines · Gregory Haskins

iosignalfd is a mechanism to register PIO/MMIO regions to trigger an eventfd signal when written to by a guest. Host userspace can register any arbitrary IO address with a corresponding eventfd and then pass the eventfd to a specific end-point of interest for handling.

Normal IO requires a blocking round-trip since the operation may cause side-effects in the emulated model or may return data to the caller. Therefore, an IO in KVM traps from the guest to the host, causes a VMX/SVM "heavy-weight" exit back to userspace, and is ultimately serviced by qemu's device model synchronously before returning control back to the vcpu.

However, there is a subclass of IO which acts purely as a trigger for other IO (such as to kick off an out-of-band DMA request, etc). For these patterns, the synchronous call is particularly expensive since we really only want to simply get our notification transmitted asychronously and return as quickly as possible. All the sychronous infrastructure to ensure proper data-dependencies are met in the normal IO case are just unecessary overhead for signalling. This adds additional computational load on the system, as well as latency to the signalling path.

Therefore, we provide a mechanism for registration of an in-kernel trigger point that allows the VCPU to only require a very brief, lightweight exit just long enough to signal an eventfd. This also means that any clients compatible with the eventfd interface (which includes userspace and kernelspace equally well) can now register to be notified. The end result should be a more flexible and higher performance notification API for the backend KVM hypervisor and perhipheral components.

To test this theory, we built a test-harness called "doorbell". This module has a function called "doorbell_ring()" which simply increments a counter for each time the doorbell is signaled. It supports signalling from either an eventfd, or an ioctl().

We then wired up two paths to the doorbell: One via QEMU via a registered io region and through the doorbell ioctl(). The other is direct via iosignalfd.

You can download this test harness here:

ftp://ftp.novell.com/dev/ghaskins/doorbell.tar.bz2

The measured results are as follows:

qemu-mmio:       110000 iops, 9.09us rtt
iosignalfd-mmio: 200100 iops, 5.00us rtt
iosignalfd-pio:  367300 iops, 2.72us rtt

I didn't measure qemu-pio, because I have to figure out how to register a PIO region with qemu's device model, and I got lazy. However, for now we can extrapolate based on the data from the NULLIO runs of +2.56us for MMIO, and -350ns for HC, we get:

qemu-pio:      153139 iops, 6.53us rtt
iosignalfd-hc: 412585 iops, 2.37us rtt

these are just for fun, for now, until I can gather more data.

Here is a graph for your convenience:

http://developer.novell.com/wiki/images/7/76/Iofd-chart.png

The conclusion to draw is that we save about 4us by skipping the userspace hop.

--------------------

Signed-off-by Gregory Haskins <[email protected]>

---

 arch/x86/kvm/x86.c       |    1 
 include/linux/kvm.h      |   15 ++
 include/linux/kvm_host.h |   10 +
 virt/kvm/eventfd.c       |  426 ++++++++++++++++++++++++++++++++++++++++++++++
 virt/kvm/kvm_main.c      |   11 +
 5 files changed, 459 insertions(+), 4 deletions(-)

arch/x86/kvm/x86.c +1 -0

index 1b91ea7..9b119e4 100644
@@ -1097,6 +1097,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_IRQ_INJECT_STATUS:
 	case KVM_CAP_ASSIGN_DEV_IRQ:
 	case KVM_CAP_IRQFD:
+	case KVM_CAP_IOSIGNALFD:
 	case KVM_CAP_PIT2:
 		r = 1;
 		break;

include/linux/kvm.h +0 -0

index 38ff31e..9de6486 100644
@@ -307,6 +307,19 @@ struct kvm_guest_debug {
 	struct kvm_guest_debug_arch arch;
 };
 

include/linux/kvm.h struct kvm_guest_debug {

Quoted 4 lines

include/linux/kvm.h +3 -0

+#define KVM_IOSIGNALFD_FLAG_TRIGGER   (1 << 0)
+#define KVM_IOSIGNALFD_FLAG_PIO       (1 << 1)
+#define KVM_IOSIGNALFD_FLAG_DEASSIGN  (1 << 2)
 

This is exposed to userspace authors.
So - some documentation for flags?

include/linux/kvm.h struct kvm_guest_debug {

Quoted 5 lines

include/linux/kvm.h +3 -0

+
+struct kvm_iosignalfd {
+	__u64 trigger;
 

Is value, or trigger_value, a better name?

Which bits are used if len is e.g. 1
and what happens with unused bits?

include/linux/kvm.h struct kvm_guest_debug {

Quoted 4 lines

include/linux/kvm.h +2 -0

+	__u64 addr;
+	__u32 len;
 

What are the legal values for len/addr?
Probably should document them.

include/linux/kvm.h struct kvm_guest_debug { · line 318

Quoted 2 lines

include/linux/kvm.h +1 -0

+	__u32 fd;

fd should probably be signed? You cast it to int later.

include/linux/kvm.h struct kvm_guest_debug { · line 320

Quoted 3 lines

include/linux/kvm.h +2 -0

+	__u32 flags;
+	__u8  pad[36];

4 byte padding not enough?

Quoted 19 lines

include/linux/kvm.h +2 -0

+};
+

#define KVM_TRC_SHIFT 16
/*
* kvm trace categories

include/linux/kvm.h +2 -0

@@ -438,6 +451,7 @@ struct kvm_trace_rec {
 #define KVM_CAP_PIT2 33
 #endif
 #define KVM_CAP_SET_BOOT_CPU_ID 34
+#define KVM_CAP_IOSIGNALFD 35
 
 #ifdef KVM_CAP_IRQ_ROUTING
 

@@ -544,6 +558,7 @@ struct kvm_irqfd {
 #define KVM_IRQFD                  _IOW(KVMIO, 0x76, struct kvm_irqfd)
 #define KVM_CREATE_PIT2		   _IOW(KVMIO, 0x77, struct kvm_pit_config)
 #define KVM_SET_BOOT_CPU_ID        _IO(KVMIO, 0x78)
+#define KVM_IOSIGNALFD             _IOW(KVMIO, 0x79, struct kvm_iosignalfd)
 

include/linux/kvm_host.h struct kvm { · line 148

Quoted 10 lines

include/linux/kvm.h +0 -0

 /*
  * ioctls for vcpu fds

include/linux/kvm_host.h +0 -0

index 707c4d8..6c0569a 100644
@@ -146,6 +146,7 @@ struct kvm {
 	struct kvm_io_bus pio_bus;
 #ifdef CONFIG_HAVE_KVM_EVENTFD
 

BTW, do we want this config option?

Quoted 25 lines

include/linux/kvm_host.h +1 -0

 	struct list_head irqfds;
+	struct list_head iosignalfds;
 #endif
 	struct kvm_vm_stat stat;

include/linux/kvm_host.h +3 -2

 	struct kvm_arch arch;

@@ -554,19 +555,24 @@ static inline void kvm_free_irq_routing(struct kvm *kvm) {}
 
 #ifdef CONFIG_HAVE_KVM_EVENTFD
 
-void kvm_irqfd_init(struct kvm *kvm);
+void kvm_eventfd_init(struct kvm *kvm);
 int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags);
 void kvm_irqfd_release(struct kvm *kvm);
+int kvm_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args);
 
 #else
 
-static inline void kvm_irqfd_init(struct kvm *kvm) {}
+static inline void kvm_eventfd_init(struct kvm *kvm) {}
 static inline int kvm_irqfd(struct kvm *kvm, int fd, int gsi, int flags)
 {
 	return -EINVAL;
 }
 

include/linux/kvm_host.h static inline void kvm_free_irq_routing(struct kvm *kvm) {} · line 576

Quoted 6 lines

include/linux/kvm_host.h +4 -0

 static inline void kvm_irqfd_release(struct kvm *kvm) {}
+static inline int kvm_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	return -EINVAL;
+}
 

I thought ENOTTY is the accepted error for an unsupported ioctl?

Quoted 36 lines

include/linux/kvm_host.h +0 -0

 #endif /* CONFIG_HAVE_KVM_EVENTFD */
 

virt/kvm/eventfd.c +6 -1

index 2c8028c..52ac455 100644
@@ -21,6 +21,7 @@
  */
 
 #include <linux/kvm_host.h>
+#include <linux/kvm.h>
 #include <linux/workqueue.h>
 #include <linux/syscalls.h>
 #include <linux/wait.h>

@@ -29,6 +30,9 @@
 #include <linux/list.h>
 #include <linux/eventfd.h>
 #include <linux/srcu.h>
+#include <linux/kernel.h>
+
+#include "iodev.h"
 
 /*
  * --------------------------------------------------------------------

@@ -202,9 +206,10 @@ fail:
 }
 
 void
-kvm_irqfd_init(struct kvm *kvm)
+kvm_eventfd_init(struct kvm *kvm)
 {
 	INIT_LIST_HEAD(&kvm->irqfds);
+	INIT_LIST_HEAD(&kvm->iosignalfds);
 }
 

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 228

Quoted 12 lines

virt/kvm/eventfd.c +6 -0

 void

@@ -215,3 +220,422 @@ kvm_irqfd_release(struct kvm *kvm)
 	list_for_each_entry_safe(irqfd, tmp, &kvm->irqfds, list)
 		irqfd_disconnect(irqfd);
 }
+
+/*
+ * --------------------------------------------------------------------
+ * iosignalfd: translate a PIO/MMIO memory write to an eventfd signal.
+ *
+ * userspace can register a PIO/MMIO address with an eventfd for recieving

typo

Quoted 4 lines

virt/kvm/eventfd.c +2 -0

+ * notification when the memory has been touched.
+ * --------------------------------------------------------------------

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 240

Quoted 10 lines

virt/kvm/eventfd.c +10 -0

+ */
+
+/*
+ * Design note: We create one PIO/MMIO device (iosignalfd_group) which
+ * aggregates  one or more iosignalfd_items.  Each item points to exactly one
+ * eventfd, and can be registered to trigger on any write to the group
+ * (wildcard), or to a write of a specific value.  If more than one item is to
+ * be supported, the addr/len ranges must all be identical in the group.  If a
+ * trigger value is to be supported on a particular item, the group range must
+ * be exactly the width of the trigger.

Some duplicate spaces in the text above, apparently at random places.

Quoted 56 lines

virt/kvm/eventfd.c +56 -0

+ */
+
+struct _iosignalfd_item {
+	struct list_head     list;
+	struct file         *file;
+	u64                  match;
+	struct rcu_head      rcu;
+	int                  wildcard:1;
+};
+
+struct _iosignalfd_group {
+	struct list_head     list;
+	u64                  addr;
+	size_t               length;
+	size_t               count;
+	struct list_head     items;
+	struct kvm_io_device dev;
+	struct rcu_head      rcu;
+};
+
+static inline struct _iosignalfd_group *
+to_group(struct kvm_io_device *dev)
+{
+	return container_of(dev, struct _iosignalfd_group, dev);
+}
+
+static void
+iosignalfd_item_free(struct _iosignalfd_item *item)
+{
+	fput(item->file);
+	kfree(item);
+}
+
+static void
+iosignalfd_item_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_item *item;
+
+	item = container_of(rhp, struct _iosignalfd_item, rcu);
+
+	iosignalfd_item_free(item);
+}
+
+static void
+iosignalfd_group_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_group *group;
+
+	group = container_of(rhp, struct _iosignalfd_group, rcu);
+
+	kfree(group);
+}
+
+static int
+iosignalfd_group_in_range(struct kvm_io_device *this, gpa_t addr, int len,
+			  int is_write)

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 301

Quoted 6 lines

virt/kvm/eventfd.c +5 -0

+{
+	struct _iosignalfd_group *p = to_group(this);
+
+	return ((addr >= p->addr && (addr < p->addr + p->length)));
+}

What does this test? len is ignored ...

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 303

Quoted 3 lines

virt/kvm/eventfd.c +2 -0

+
+static int

This seems to be returning bool ...

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 313

Quoted 12 lines

virt/kvm/eventfd.c +10 -0

+iosignalfd_is_match(struct _iosignalfd_group *group,
+		    struct _iosignalfd_item *item,
+		    const void *val,
+		    int len)
+{
+	u64 _val;
+
+	if (len != group->length)
+		/* mis-matched length is always a miss */
+		return false;

Why is that? what if there's 8 byte write which covers a 4 byte group?

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 323

Quoted 12 lines

virt/kvm/eventfd.c +10 -0

+
+	if (item->wildcard)
+		/* wildcard is always a hit */
+		return true;
+
+	/* otherwise, we have to actually compare the data */
+
+	if (!IS_ALIGNED((unsigned long)val, len))
+		/* protect against this request causing a SIGBUS */
+		return false;

Could you explain what this does please?

I thought misaligned accesses are allowed.

Quoted 14 lines

virt/kvm/eventfd.c +13 -0

+
+	switch (len) {
+	case 1:
+		_val = *(u8 *)val;
+		break;
+	case 2:
+		_val = *(u16 *)val;
+		break;
+	case 4:
+		_val = *(u32 *)val;
+		break;
+	case 8:
+		_val = *(u64 *)val;

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 340

Quoted 5 lines

virt/kvm/eventfd.c +4 -0

+		break;
+	default:
+		return false;
+	}

So legal values for len are 1,2,4 and 8?
Might be a good idea to document this.

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 342

Quoted 3 lines

virt/kvm/eventfd.c +2 -0

+
+	return _val == item->match;

Quoted 28 lines

virt/kvm/eventfd.c +28 -0

+}
+
+/*
+ * MMIO/PIO writes trigger an event (if the data matches).
+ *
+ * This is invoked by the io_bus subsystem in response to an address match
+ * against the group.  We must then walk the list of individual items to check
+ * for a match and, if applicable, to send the appropriate signal. If the item
+ * in question does not have a "match" pointer, it is considered a wildcard
+ * and will always generate a signal.  There can be an arbitrary number
+ * of distinct matches or wildcards per group.
+ */
+static void
+iosignalfd_group_write(struct kvm_io_device *this, gpa_t addr, int len,
+		       const void *val)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(item, &group->items, list) {
+		if (iosignalfd_is_match(group, item, val, len))
+			eventfd_signal(item->file, 1);
+	}
+
+	rcu_read_unlock();
+}

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 374

Quoted 5 lines

virt/kvm/eventfd.c +4 -0

+
+/*
+ * MMIO/PIO reads against the group indiscriminately return all zeros
+ */

Does it have to be so? It would be better to bounce reads to userspace...

Quoted 36 lines

virt/kvm/eventfd.c +36 -0

+static void
+iosignalfd_group_read(struct kvm_io_device *this, gpa_t addr, int len,
+		      void *val)
+{
+	memset(val, 0, len);
+}
+
+/*
+ * This function is called as KVM is completely shutting down.  We do not
+ * need to worry about locking or careful RCU dancing...just nuke anything
+ * we have as quickly as possible
+ */
+static void
+iosignalfd_group_destructor(struct kvm_io_device *this)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item, *tmp;
+
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+		list_del(&item->list);
+		iosignalfd_item_free(item);
+	}
+
+	list_del(&group->list);
+	kfree(group);
+}
+
+static const struct kvm_io_device_ops iosignalfd_ops = {
+	.read       = iosignalfd_group_read,
+	.write      = iosignalfd_group_write,
+	.in_range   = iosignalfd_group_in_range,
+	.destructor = iosignalfd_group_destructor,
+};
+
+/* assumes kvm->lock held */
+static struct _iosignalfd_group *

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 415

Quoted 6 lines

virt/kvm/eventfd.c +5 -0

+iosignalfd_group_find(struct kvm *kvm, u64 addr)
+{
+	struct _iosignalfd_group *group;
+
+	list_for_each_entry(group, &kvm->iosignalfds, list) {

{} not needed here

Quoted 34 lines

virt/kvm/eventfd.c +32 -0

+		if (group->addr == addr)
+			return group;
+	}
+
+	return NULL;
+}
+
+/* assumes kvm->lock is held */
+static struct _iosignalfd_group *
+iosignalfd_group_create(struct kvm *kvm, struct kvm_io_bus *bus,
+			u64 addr, size_t len)
+{
+	struct _iosignalfd_group *group;
+	int ret;
+
+	group = kzalloc(sizeof(*group), GFP_KERNEL);
+	if (!group)
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&group->list);
+	INIT_LIST_HEAD(&group->items);
+	group->addr   = addr;
+	group->length = len;
+	kvm_iodevice_init(&group->dev, &iosignalfd_ops);
+
+	ret = kvm_io_bus_register_dev(kvm, bus, &group->dev);
+	if (ret < 0) {
+		kfree(group);
+		return ERR_PTR(ret);
+	}
+
+	list_add_tail(&group->list, &kvm->iosignalfds);

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 457

Quoted 10 lines

virt/kvm/eventfd.c +10 -0

+
+	return group;
+}
+
+static int
+kvm_assign_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	int                       pio = args->flags & KVM_IOSIGNALFD_FLAG_PIO;
+	struct kvm_io_bus        *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
+	struct _iosignalfd_group *group = NULL;

why does group need to be initialized?

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 458

Quoted 2 lines

virt/kvm/eventfd.c +1 -0

+	struct _iosignalfd_item  *item = NULL;

Why does item need to be initialized?

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 462

Quoted 5 lines

virt/kvm/eventfd.c +4 -0

+	struct file              *file;
+	int                       ret;
+
+	if (args->len > sizeof(u64))

Is e.g. value 3 legal?

Quoted 27 lines

virt/kvm/eventfd.c +23 -0

+		return -EINVAL;
 
 
 
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	item = kzalloc(sizeof(*item), GFP_KERNEL);
+	if (!item) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	INIT_LIST_HEAD(&item->list);
+	item->file = file;
+
+	/*
+	 * A trigger address is optional, otherwise this is a wildcard
+	 */
+	if (args->flags & KVM_IOSIGNALFD_FLAG_TRIGGER)
+		item->match = args->trigger;
+	else
+		item->wildcard = true;
+

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 490

Quoted 6 lines

virt/kvm/eventfd.c +5 -0

+	mutex_lock(&kvm->lock);
+
+	/*
+	 * Put an upper limit on the number of items we support
+	 */

Groups and items, actually, right?

Quoted 27 lines

virt/kvm/eventfd.c +27 -0

+	if (kvm->io_device_count >= CONFIG_KVM_MAX_IO_DEVICES) {
+		ret = -ENOSPC;
+		goto unlock_fail;
+	}
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+
+		group = iosignalfd_group_create(kvm, bus,
+						args->addr, args->len);
+		if (IS_ERR(group)) {
+			ret = PTR_ERR(group);
+			goto unlock_fail;
+		}
+
+		/*
+		 * Note: We do not increment io_device_count for the first item,
+		 * as this is represented by the group device that we just
+		 * registered.  Make sure we handle this properly when we
+		 * deassign the last item
+		 */
+	} else {
+
+		if (group->length != args->len) {
+			/*
+			 * Existing groups must have the same addr/len tuple
+			 * or we reject the request

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 520

Quoted 4 lines

virt/kvm/eventfd.c +3 -0

+			 */
+			ret = -EINVAL;
+			goto unlock_fail;

Most errors seem to trigger EINVAL. Applications will be easier to debug if different errors are returned on different mistakes.

E.g. here EBUSY might be good. And same
in other places.

Quoted 101 lines

virt/kvm/eventfd.c +101 -0

+		}
+
+		kvm->io_device_count++;
+	}
+
+	/*
+	 * Note: We are committed to succeed at this point since we have
+	 * (potentially) published a new group-device.  Any failure handling
+	 * added in the future after this point will need to be carefully
+	 * considered.
+	 */
+
+	list_add_tail_rcu(&item->list, &group->items);
+	group->count++;
+
+	mutex_unlock(&kvm->lock);
+
+	return 0;
+
+unlock_fail:
+	mutex_unlock(&kvm->lock);
+fail:
+	if (item)
+		/*
+		 * it would have never made it to the group->items list
+		 * in the failure path, so we dont need to worry about removing
+		 * it
+		 */
+		kfree(item);
+
+	fput(file);
+
+	return ret;
+}
+
+
+static int
+kvm_deassign_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	int                       pio = args->flags & KVM_IOSIGNALFD_FLAG_PIO;
+	struct kvm_io_bus        *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
+	struct _iosignalfd_group *group;
+	struct _iosignalfd_item  *item, *tmp;
+	struct file              *file;
+	int                       ret = 0;
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	mutex_lock(&kvm->lock);
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * Exhaustively search our group->items list for any items that might
+	 * match the specified fd, and (carefully) remove each one found.
+	 */
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+
+		if (item->file != file)
+			continue;
+
+		list_del_rcu(&item->list);
+		group->count--;
+		if (group->count)
+			/*
+			 * We only decrement the global count if this is *not*
+			 * the last item.  The last item will be accounted for
+			 * by the io_bus_unregister
+			 */
+			kvm->io_device_count--;
+
+		/*
+		 * The item may be still referenced inside our group->write()
+		 * path's RCU read-side CS, so defer the actual free to the
+		 * next grace
+		 */
+		call_rcu(&item->rcu, iosignalfd_item_deferred_free);
+	}
+
+	/*
+	 * Check if the group is now completely vacated as a result of
+	 * removing the items.  If so, unregister/delete it
+	 */
+	if (!group->count) {
+
+		kvm_io_bus_unregister_dev(kvm, bus, &group->dev);
+
+		/*
+		 * Like the item, the group may also still be referenced as
+		 * per above.  However, the kvm->iosignalfds list is not
+		 * RCU protected (its protected by kvm->lock instead) so
+		 * we can just plain-vanilla remove it.  What needs to be
+		 * done carefully is the actual freeing of the group pointer
+		 * since we walk the group->items list within the RCU CS.
+		 */

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 623

Quoted 3 lines

virt/kvm/eventfd.c +2 -0

+		list_del(&group->list);
+		call_rcu(&group->rcu, iosignalfd_group_deferred_free);

This is a deferred call, is it not, with no guarantee on when it will run? If correct I think synchronize_rcu might be better here:
- can the module go away while iosignalfd_group_deferred_free is running?

- can eventfd be signalled *after* ioctl exits? If yes this might confuse applications if they use the eventfd for something else.

Quoted 49 lines

virt/kvm/eventfd.c +18 -0

+	}
+
+out:
+	mutex_unlock(&kvm->lock);
+
+	fput(file);
+
+	return ret;
+}
+
+int
+kvm_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	if (args->flags & KVM_IOSIGNALFD_FLAG_DEASSIGN)
+		return kvm_deassign_iosignalfd(kvm, args);
+
+	return kvm_assign_iosignalfd(kvm, args);
+}

virt/kvm/kvm_main.c +10 -1

index 42cbea7..e6495d4 100644
@@ -971,7 +971,7 @@ static struct kvm *kvm_create_vm(void)
 	atomic_inc(&kvm->mm->mm_count);
 	spin_lock_init(&kvm->mmu_lock);
 	kvm_io_bus_init(&kvm->pio_bus);
-	kvm_irqfd_init(kvm);
+	kvm_eventfd_init(kvm);
 	mutex_init(&kvm->lock);
 	mutex_init(&kvm->irq_lock);
 	kvm_io_bus_init(&kvm->mmio_bus);

@@ -2227,6 +2227,15 @@ static long kvm_vm_ioctl(struct file *filp,
 		r = kvm_irqfd(kvm, data.fd, data.gsi, data.flags);
 		break;
 	}
+	case KVM_IOSIGNALFD: {
+		struct kvm_iosignalfd data;
+
+		r = -EFAULT;
+		if (copy_from_user(&data, argp, sizeof data))
+			goto out;
+		r = kvm_iosignalfd(kvm, &data);
+		break;
+	}
 #ifdef CONFIG_KVM_APIC_ARCHITECTURE
 	case KVM_SET_BOOT_CPU_ID:
 		r = 0;

Quoted 5 lines

--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Quoted 9 lines

Quoted 9 lines

+ * Design note: We create one PIO/MMIO device (iosignalfd_group) which
+ * aggregates one or more iosignalfd_items. Each item points to exactly one
+ * eventfd, and can be registered to trigger on any write to the group
+ * (wildcard), or to a write of a specific value. If more than one item is to
+ * be supported, the addr/len ranges must all be identical in the group. If a
+ * trigger value is to be supported on a particular item, the group range must
+ * be exactly the width of the trigger.

Quoted 1 line

Some duplicate spaces in the text above, apparently at random places.

-ENOPARSE ;)

Can you elaborate?

Quoted 7 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 235

Quoted 7 lines

+ * notification when the memory has been touched.
+ * --------------------------------------------------------------------
+ */

virt/kvm/eventfd.c +4 -0

+
+/*
+ * Design note: We create one PIO/MMIO device (iosignalfd_group) which
+ * aggregates  one or more iosignalfd_items.  Each item points to exactly one

Quoted 2 lines

Quoted 2 lines

+ * eventfd, and can be registered to trigger on any write to the group
+ * (wildcard), or to a write of a specific value. If more than one item is to

Quoted 1 line

Quoted 1 line

+ * be supported, the addr/len ranges must all be identical in the group. If a

Quoted 7 lines

Quoted 3 lines

+ * trigger value is to be supported on a particular item, the group range must
+ * be exactly the width of the trigger.

Some duplicate spaces in the text above, apparently at random places.

-ENOPARSE ;)

Can you elaborate?

Quoted 57 lines

Quoted 57 lines

+ */

virt/kvm/eventfd.c +55 -0

+
+struct _iosignalfd_item {
+	struct list_head     list;
+	struct file         *file;
+	u64                  match;
+	struct rcu_head      rcu;
+	int                  wildcard:1;
+};
+
+struct _iosignalfd_group {
+	struct list_head     list;
+	u64                  addr;
+	size_t               length;
+	size_t               count;
+	struct list_head     items;
+	struct kvm_io_device dev;
+	struct rcu_head      rcu;
+};
+
+static inline struct _iosignalfd_group *
+to_group(struct kvm_io_device *dev)
+{
+	return container_of(dev, struct _iosignalfd_group, dev);
+}
+
+static void
+iosignalfd_item_free(struct _iosignalfd_item *item)
+{
+	fput(item->file);
+	kfree(item);
+}
+
+static void
+iosignalfd_item_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_item *item;
+
+	item = container_of(rhp, struct _iosignalfd_item, rcu);
+
+	iosignalfd_item_free(item);
+}
+
+static void
+iosignalfd_group_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_group *group;
+
+	group = container_of(rhp, struct _iosignalfd_group, rcu);
+
+	kfree(group);
+}
+
+static int
+iosignalfd_group_in_range(struct kvm_io_device *this, gpa_t addr, int len,
+			  int is_write)

Quoted 10 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 301

Quoted 6 lines

virt/kvm/eventfd.c +5 -0

+{
+	struct _iosignalfd_group *p = to_group(this);
+
+	return ((addr >= p->addr && (addr < p->addr + p->length)));
+}

What does this test? len is ignored ...

Yeah, I was following precedent with other IO devices. However, this *is* sloppy, I agree. Will fix.

Quoted 6 lines

Review

Quoted 3 lines

(quoted diff) +2 -0

+
+static int

This seems to be returning bool ...

Ack

Quoted 16 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 313

Quoted 12 lines

virt/kvm/eventfd.c +10 -0

+iosignalfd_is_match(struct _iosignalfd_group *group,
+		    struct _iosignalfd_item *item,
+		    const void *val,
+		    int len)
+{
+	u64 _val;
+
+	if (len != group->length)
+		/* mis-matched length is always a miss */
+		return false;

Why is that? what if there's 8 byte write which covers a 4 byte group?

v7 and earlier used to allow that for wildcards, actually. It of
course would never make sense to allow mis-matched writes for non-wildcards, since the idea is to match the value exactly. However, the feedback I got from Avi was that we should make the wildcard vs non-wildcard access symmetrical and ensure they both conform to the size.

Quoted 15 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 323

Quoted 12 lines

virt/kvm/eventfd.c +10 -0

+
+	if (item->wildcard)
+		/* wildcard is always a hit */
+		return true;
+
+	/* otherwise, we have to actually compare the data */
+
+	if (!IS_ALIGNED((unsigned long)val, len))
+		/* protect against this request causing a SIGBUS */
+		return false;

Could you explain what this does please?

Sure: item->match is a fixed u64 to represent all group->length values. So it might have a 1, 2, 4, or 8 byte value in it. When I write arrives, we need to cast the data-register (in this case represented by (void*)val) into a u64 so the equality check (see [A], below) can be done. However, you can't cast an unaligned pointer, or it will SIGBUS on many (most?) architectures.

Quoted 2 lines

I thought misaligned accesses are allowed.

If thats true, we are in trouble ;)

Quoted 14 lines

Quoted 14 lines

virt/kvm/eventfd.c +13 -0

+
+	switch (len) {
+	case 1:
+		_val = *(u8 *)val;
+		break;
+	case 2:
+		_val = *(u16 *)val;
+		break;
+	case 4:
+		_val = *(u32 *)val;
+		break;
+	case 8:
+		_val = *(u64 *)val;

Quoted 10 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 340

Quoted 5 lines

virt/kvm/eventfd.c +4 -0

+		break;
+	default:
+		return false;
+	}

So legal values for len are 1,2,4 and 8?
Might be a good idea to document this.

Ack

Quoted 3 lines

Review

Quoted 3 lines

(quoted diff) +2 -0

+
+	return _val == item->match;

[A]

Quoted 28 lines

Quoted 28 lines

virt/kvm/eventfd.c +28 -0

+}
+
+/*
+ * MMIO/PIO writes trigger an event (if the data matches).
+ *
+ * This is invoked by the io_bus subsystem in response to an address match
+ * against the group.  We must then walk the list of individual items to check
+ * for a match and, if applicable, to send the appropriate signal. If the item
+ * in question does not have a "match" pointer, it is considered a wildcard
+ * and will always generate a signal.  There can be an arbitrary number
+ * of distinct matches or wildcards per group.
+ */
+static void
+iosignalfd_group_write(struct kvm_io_device *this, gpa_t addr, int len,
+		       const void *val)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(item, &group->items, list) {
+		if (iosignalfd_is_match(group, item, val, len))
+			eventfd_signal(item->file, 1);
+	}
+
+	rcu_read_unlock();
+}

Quoted 10 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 374

Quoted 5 lines

virt/kvm/eventfd.c +4 -0

+
+/*
+ * MMIO/PIO reads against the group indiscriminately return all zeros
+ */

Does it have to be so? It would be better to bounce reads to userspace...

Good idea. I can set is_write = false and I should never get this function called.

Quoted 36 lines

Quoted 36 lines

virt/kvm/eventfd.c +36 -0

+static void
+iosignalfd_group_read(struct kvm_io_device *this, gpa_t addr, int len,
+		      void *val)
+{
+	memset(val, 0, len);
+}
+
+/*
+ * This function is called as KVM is completely shutting down.  We do not
+ * need to worry about locking or careful RCU dancing...just nuke anything
+ * we have as quickly as possible
+ */
+static void
+iosignalfd_group_destructor(struct kvm_io_device *this)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item, *tmp;
+
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+		list_del(&item->list);
+		iosignalfd_item_free(item);
+	}
+
+	list_del(&group->list);
+	kfree(group);
+}
+
+static const struct kvm_io_device_ops iosignalfd_ops = {
+	.read       = iosignalfd_group_read,
+	.write      = iosignalfd_group_write,
+	.in_range   = iosignalfd_group_in_range,
+	.destructor = iosignalfd_group_destructor,
+};
+
+/* assumes kvm->lock held */
+static struct _iosignalfd_group *

Quoted 9 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 415

Quoted 6 lines

virt/kvm/eventfd.c +5 -0

+iosignalfd_group_find(struct kvm *kvm, u64 addr)
+{
+	struct _iosignalfd_group *group;
+
+	list_for_each_entry(group, &kvm->iosignalfds, list) {

{} not needed here

Ack

Quoted 48 lines

Quoted 34 lines

virt/kvm/eventfd.c +32 -0

+		if (group->addr == addr)
+			return group;
+	}
+
+	return NULL;
+}
+
+/* assumes kvm->lock is held */
+static struct _iosignalfd_group *
+iosignalfd_group_create(struct kvm *kvm, struct kvm_io_bus *bus,
+			u64 addr, size_t len)
+{
+	struct _iosignalfd_group *group;
+	int ret;
+
+	group = kzalloc(sizeof(*group), GFP_KERNEL);
+	if (!group)
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&group->list);
+	INIT_LIST_HEAD(&group->items);
+	group->addr   = addr;
+	group->length = len;
+	kvm_iodevice_init(&group->dev, &iosignalfd_ops);
+
+	ret = kvm_io_bus_register_dev(kvm, bus, &group->dev);
+	if (ret < 0) {
+		kfree(group);
+		return ERR_PTR(ret);
+	}
+
+	list_add_tail(&group->list, &kvm->iosignalfds);

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 457

Quoted 10 lines

virt/kvm/eventfd.c +10 -0

+
+	return group;
+}
+
+static int
+kvm_assign_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	int                       pio = args->flags & KVM_IOSIGNALFD_FLAG_PIO;
+	struct kvm_io_bus        *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
+	struct _iosignalfd_group *group = NULL;

why does group need to be initialized?

Quoted 6 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 458

Quoted 2 lines

virt/kvm/eventfd.c +1 -0

+	struct _iosignalfd_item  *item = NULL;

Why does item need to be initialized?

Probably leftover from versions prior to v8. Will fix.

Quoted 8 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 462

Quoted 5 lines

virt/kvm/eventfd.c +4 -0

+	struct file              *file;
+	int                       ret;
+
+	if (args->len > sizeof(u64))

Is e.g. value 3 legal?

Ack. Will check against legal values.

Quoted 27 lines

Quoted 27 lines

virt/kvm/eventfd.c +23 -0

+		return -EINVAL;
 
 
 
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	item = kzalloc(sizeof(*item), GFP_KERNEL);
+	if (!item) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	INIT_LIST_HEAD(&item->list);
+	item->file = file;
+
+	/*
+	 * A trigger address is optional, otherwise this is a wildcard
+	 */
+	if (args->flags & KVM_IOSIGNALFD_FLAG_TRIGGER)
+		item->match = args->trigger;
+	else
+		item->wildcard = true;
+

Quoted 10 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 490

Quoted 6 lines

virt/kvm/eventfd.c +5 -0

+	mutex_lock(&kvm->lock);
+
+	/*
+	 * Put an upper limit on the number of items we support
+	 */

Groups and items, actually, right?

Yeah, though technically that is implicit when you say "items", since each group always has at least one item. I will try to make this clearer, though.

Quoted 27 lines

Quoted 27 lines

virt/kvm/eventfd.c +27 -0

+	if (kvm->io_device_count >= CONFIG_KVM_MAX_IO_DEVICES) {
+		ret = -ENOSPC;
+		goto unlock_fail;
+	}
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+
+		group = iosignalfd_group_create(kvm, bus,
+						args->addr, args->len);
+		if (IS_ERR(group)) {
+			ret = PTR_ERR(group);
+			goto unlock_fail;
+		}
+
+		/*
+		 * Note: We do not increment io_device_count for the first item,
+		 * as this is represented by the group device that we just
+		 * registered.  Make sure we handle this properly when we
+		 * deassign the last item
+		 */
+	} else {
+
+		if (group->length != args->len) {
+			/*
+			 * Existing groups must have the same addr/len tuple
+			 * or we reject the request

Quoted 8 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 520

Quoted 4 lines

virt/kvm/eventfd.c +3 -0

+			 */
+			ret = -EINVAL;
+			goto unlock_fail;

Most errors seem to trigger EINVAL. Applications will be easier to debug if different errors are returned on different mistakes.

Yeah, agreed. Will try to differentiate some errors here.

Quoted 4 lines

E.g. here EBUSY might be good. And same
in other places.

Actually, I think EBUSY is supposed to be a transitory error, and would not be appropriate to use here. That said, your point is taken: Find more appropriate and descriptive errors.

Quoted 101 lines

Quoted 101 lines

virt/kvm/eventfd.c +101 -0

+		}
+
+		kvm->io_device_count++;
+	}
+
+	/*
+	 * Note: We are committed to succeed at this point since we have
+	 * (potentially) published a new group-device.  Any failure handling
+	 * added in the future after this point will need to be carefully
+	 * considered.
+	 */
+
+	list_add_tail_rcu(&item->list, &group->items);
+	group->count++;
+
+	mutex_unlock(&kvm->lock);
+
+	return 0;
+
+unlock_fail:
+	mutex_unlock(&kvm->lock);
+fail:
+	if (item)
+		/*
+		 * it would have never made it to the group->items list
+		 * in the failure path, so we dont need to worry about removing
+		 * it
+		 */
+		kfree(item);
+
+	fput(file);
+
+	return ret;
+}
+
+
+static int
+kvm_deassign_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	int                       pio = args->flags & KVM_IOSIGNALFD_FLAG_PIO;
+	struct kvm_io_bus        *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
+	struct _iosignalfd_group *group;
+	struct _iosignalfd_item  *item, *tmp;
+	struct file              *file;
+	int                       ret = 0;
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	mutex_lock(&kvm->lock);
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * Exhaustively search our group->items list for any items that might
+	 * match the specified fd, and (carefully) remove each one found.
+	 */
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+
+		if (item->file != file)
+			continue;
+
+		list_del_rcu(&item->list);
+		group->count--;
+		if (group->count)
+			/*
+			 * We only decrement the global count if this is *not*
+			 * the last item.  The last item will be accounted for
+			 * by the io_bus_unregister
+			 */
+			kvm->io_device_count--;
+
+		/*
+		 * The item may be still referenced inside our group->write()
+		 * path's RCU read-side CS, so defer the actual free to the
+		 * next grace
+		 */
+		call_rcu(&item->rcu, iosignalfd_item_deferred_free);
+	}
+
+	/*
+	 * Check if the group is now completely vacated as a result of
+	 * removing the items.  If so, unregister/delete it
+	 */
+	if (!group->count) {
+
+		kvm_io_bus_unregister_dev(kvm, bus, &group->dev);
+
+		/*
+		 * Like the item, the group may also still be referenced as
+		 * per above.  However, the kvm->iosignalfds list is not
+		 * RCU protected (its protected by kvm->lock instead) so
+		 * we can just plain-vanilla remove it.  What needs to be
+		 * done carefully is the actual freeing of the group pointer
+		 * since we walk the group->items list within the RCU CS.
+		 */

Quoted 9 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 623

Quoted 3 lines

virt/kvm/eventfd.c +2 -0

+		list_del(&group->list);
+		call_rcu(&group->rcu, iosignalfd_group_deferred_free);

This is a deferred call, is it not, with no guarantee on when it will run? If correct I think synchronize_rcu might be better here:
- can the module go away while iosignalfd_group_deferred_free is running?

Good catch. Once I go this route it will be easy to use SRCU instead of RCU, too. So I will fix this up.

Quoted 4 lines

- can eventfd be signalled *after* ioctl exits? If yes this might confuse applications if they use the eventfd for something else.

Not by iosignalfd. Once this function completes, we synchronously guarantee that no more IO activity will generate an event on the affected eventfds. Of course, this has no bearing on whether some other producer wont signal, but that is beyond the scope of iosignalfd.

Quoted 49 lines

Quoted 49 lines

virt/kvm/eventfd.c +18 -0

+	}
+
+out:
+	mutex_unlock(&kvm->lock);
+
+	fput(file);
+
+	return ret;
+}
+
+int
+kvm_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	if (args->flags & KVM_IOSIGNALFD_FLAG_DEASSIGN)
+		return kvm_deassign_iosignalfd(kvm, args);
+
+	return kvm_assign_iosignalfd(kvm, args);
+}

virt/kvm/kvm_main.c +10 -1

index 42cbea7..e6495d4 100644
@@ -971,7 +971,7 @@ static struct kvm *kvm_create_vm(void)
 	atomic_inc(&kvm->mm->mm_count);
 	spin_lock_init(&kvm->mmu_lock);
 	kvm_io_bus_init(&kvm->pio_bus);
-	kvm_irqfd_init(kvm);
+	kvm_eventfd_init(kvm);
 	mutex_init(&kvm->lock);
 	mutex_init(&kvm->irq_lock);
 	kvm_io_bus_init(&kvm->mmio_bus);

@@ -2227,6 +2227,15 @@ static long kvm_vm_ioctl(struct file *filp,
 		r = kvm_irqfd(kvm, data.fd, data.gsi, data.flags);
 		break;
 	}
+	case KVM_IOSIGNALFD: {
+		struct kvm_iosignalfd data;
+
+		r = -EFAULT;
+		if (copy_from_user(&data, argp, sizeof data))
+			goto out;
+		r = kvm_iosignalfd(kvm, &data);
+		break;
+	}
 #ifdef CONFIG_KVM_APIC_ARCHITECTURE
 	case KVM_SET_BOOT_CPU_ID:
 		r = 0;

Quoted 5 lines

Quoted 5 lines

--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Thanks Michael,
-Greg

Quoted 8 lines · Gregory Haskins

Quoted 8 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 235

Quoted 8 lines

+ * notification when the memory has been touched.
+ * --------------------------------------------------------------------
+ */

virt/kvm/eventfd.c +4 -0

+
+/*
+ * Design note: We create one PIO/MMIO device (iosignalfd_group) which
+ * aggregates  one or more iosignalfd_items.  Each item points to exactly one

^^ ^^

Quoted 3 lines

Quoted 3 lines

Quoted 3 lines

+ * eventfd, and can be registered to trigger on any write to the group
+ * (wildcard), or to a write of a specific value. If more than one item is to

                                                       ^^

Quoted 2 lines

Quoted 2 lines

Quoted 2 lines

+ * be supported, the addr/len ranges must all be identical in the group. If a

                                                                              ^^

Quoted 8 lines

Quoted 8 lines

Quoted 4 lines

+ * trigger value is to be supported on a particular item, the group range must
+ * be exactly the width of the trigger.

Some duplicate spaces in the text above, apparently at random places.

Quoted 4 lines

-ENOPARSE ;)

Can you elaborate?

Marked with ^^

Quoted 122 lines

Quoted 57 lines

Quoted 57 lines

+ */

virt/kvm/eventfd.c +55 -0

+
+struct _iosignalfd_item {
+	struct list_head     list;
+	struct file         *file;
+	u64                  match;
+	struct rcu_head      rcu;
+	int                  wildcard:1;
+};
+
+struct _iosignalfd_group {
+	struct list_head     list;
+	u64                  addr;
+	size_t               length;
+	size_t               count;
+	struct list_head     items;
+	struct kvm_io_device dev;
+	struct rcu_head      rcu;
+};
+
+static inline struct _iosignalfd_group *
+to_group(struct kvm_io_device *dev)
+{
+	return container_of(dev, struct _iosignalfd_group, dev);
+}
+
+static void
+iosignalfd_item_free(struct _iosignalfd_item *item)
+{
+	fput(item->file);
+	kfree(item);
+}
+
+static void
+iosignalfd_item_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_item *item;
+
+	item = container_of(rhp, struct _iosignalfd_item, rcu);
+
+	iosignalfd_item_free(item);
+}
+
+static void
+iosignalfd_group_deferred_free(struct rcu_head *rhp)
+{
+	struct _iosignalfd_group *group;
+
+	group = container_of(rhp, struct _iosignalfd_group, rcu);
+
+	kfree(group);
+}
+
+static int
+iosignalfd_group_in_range(struct kvm_io_device *this, gpa_t addr, int len,
+			  int is_write)

Quoted 11 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 301

Quoted 7 lines

virt/kvm/eventfd.c +5 -0

+{
+	struct _iosignalfd_group *p = to_group(this);
+
+	return ((addr >= p->addr && (addr < p->addr + p->length)));
+}

What does this test? len is ignored ...

Yeah, I was following precedent with other IO devices. However, this *is* sloppy, I agree. Will fix.

Quoted 7 lines

Review

Quoted 4 lines

(quoted diff) +2 -0

+
+static int

This seems to be returning bool ...

Ack

Quoted 18 lines

Quoted 4 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 313

Quoted 10 lines

virt/kvm/eventfd.c +10 -0

+iosignalfd_is_match(struct _iosignalfd_group *group,
+		    struct _iosignalfd_item *item,
+		    const void *val,
+		    int len)
+{
+	u64 _val;
+
+	if (len != group->length)
+		/* mis-matched length is always a miss */
+		return false;

Why is that? what if there's 8 byte write which covers a 4 byte group?

v7 and earlier used to allow that for wildcards, actually. It of
course would never make sense to allow mis-matched writes for non-wildcards, since the idea is to match the value exactly. However, the feedback I got from Avi was that we should make the wildcard vs non-wildcard access symmetrical and ensure they both conform to the size.

Quoted 17 lines

Quoted 4 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 323

Quoted 10 lines

virt/kvm/eventfd.c +10 -0

+
+	if (item->wildcard)
+		/* wildcard is always a hit */
+		return true;
+
+	/* otherwise, we have to actually compare the data */
+
+	if (!IS_ALIGNED((unsigned long)val, len))
+		/* protect against this request causing a SIGBUS */
+		return false;

Could you explain what this does please?

Quoted 7 lines

Sure: item->match is a fixed u64 to represent all group->length values. So it might have a 1, 2, 4, or 8 byte value in it. When I write arrives, we need to cast the data-register (in this case represented by (void*)val) into a u64 so the equality check (see [A], below) can be done. However, you can't cast an unaligned pointer, or it will SIGBUS on many (most?) architectures.

I mean guest access. Does it have to be aligned?
You could memcpy the value...

Quoted 5 lines

Quoted 3 lines

I thought misaligned accesses are allowed.

If thats true, we are in trouble ;)

I think it works at least on x86:
http://en.wikipedia.org/wiki/Packed#x86_and_x86-64

Quoted 469 lines

Quoted 15 lines

Quoted 15 lines

virt/kvm/eventfd.c +13 -0

+
+	switch (len) {
+	case 1:
+		_val = *(u8 *)val;
+		break;
+	case 2:
+		_val = *(u16 *)val;
+		break;
+	case 4:
+		_val = *(u32 *)val;
+		break;
+	case 8:
+		_val = *(u64 *)val;

Quoted 11 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 340

Quoted 6 lines

virt/kvm/eventfd.c +4 -0

+		break;
+	default:
+		return false;
+	}

So legal values for len are 1,2,4 and 8?
Might be a good idea to document this.

Ack

Quoted 4 lines

Review

Quoted 4 lines

(quoted diff) +2 -0

+
+	return _val == item->match;

[A]

Quoted 28 lines

Quoted 28 lines

virt/kvm/eventfd.c +28 -0

+}
+
+/*
+ * MMIO/PIO writes trigger an event (if the data matches).
+ *
+ * This is invoked by the io_bus subsystem in response to an address match
+ * against the group.  We must then walk the list of individual items to check
+ * for a match and, if applicable, to send the appropriate signal. If the item
+ * in question does not have a "match" pointer, it is considered a wildcard
+ * and will always generate a signal.  There can be an arbitrary number
+ * of distinct matches or wildcards per group.
+ */
+static void
+iosignalfd_group_write(struct kvm_io_device *this, gpa_t addr, int len,
+		       const void *val)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(item, &group->items, list) {
+		if (iosignalfd_is_match(group, item, val, len))
+			eventfd_signal(item->file, 1);
+	}
+
+	rcu_read_unlock();
+}

Quoted 11 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 374

Quoted 6 lines

virt/kvm/eventfd.c +4 -0

+
+/*
+ * MMIO/PIO reads against the group indiscriminately return all zeros
+ */

Does it have to be so? It would be better to bounce reads to userspace...

Good idea. I can set is_write = false and I should never get this function called.

Quoted 36 lines

Quoted 36 lines

virt/kvm/eventfd.c +36 -0

+static void
+iosignalfd_group_read(struct kvm_io_device *this, gpa_t addr, int len,
+		      void *val)
+{
+	memset(val, 0, len);
+}
+
+/*
+ * This function is called as KVM is completely shutting down.  We do not
+ * need to worry about locking or careful RCU dancing...just nuke anything
+ * we have as quickly as possible
+ */
+static void
+iosignalfd_group_destructor(struct kvm_io_device *this)
+{
+	struct _iosignalfd_group *group = to_group(this);
+	struct _iosignalfd_item *item, *tmp;
+
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+		list_del(&item->list);
+		iosignalfd_item_free(item);
+	}
+
+	list_del(&group->list);
+	kfree(group);
+}
+
+static const struct kvm_io_device_ops iosignalfd_ops = {
+	.read       = iosignalfd_group_read,
+	.write      = iosignalfd_group_write,
+	.in_range   = iosignalfd_group_in_range,
+	.destructor = iosignalfd_group_destructor,
+};
+
+/* assumes kvm->lock held */
+static struct _iosignalfd_group *

Quoted 10 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 415

Quoted 7 lines

virt/kvm/eventfd.c +5 -0

+iosignalfd_group_find(struct kvm *kvm, u64 addr)
+{
+	struct _iosignalfd_group *group;
+
+	list_for_each_entry(group, &kvm->iosignalfds, list) {

{} not needed here

Ack

Quoted 50 lines

Quoted 36 lines

virt/kvm/eventfd.c +32 -0

+		if (group->addr == addr)
+			return group;
+	}
+
+	return NULL;
+}
+
+/* assumes kvm->lock is held */
+static struct _iosignalfd_group *
+iosignalfd_group_create(struct kvm *kvm, struct kvm_io_bus *bus,
+			u64 addr, size_t len)
+{
+	struct _iosignalfd_group *group;
+	int ret;
+
+	group = kzalloc(sizeof(*group), GFP_KERNEL);
+	if (!group)
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&group->list);
+	INIT_LIST_HEAD(&group->items);
+	group->addr   = addr;
+	group->length = len;
+	kvm_iodevice_init(&group->dev, &iosignalfd_ops);
+
+	ret = kvm_io_bus_register_dev(kvm, bus, &group->dev);
+	if (ret < 0) {
+		kfree(group);
+		return ERR_PTR(ret);
+	}
+
+	list_add_tail(&group->list, &kvm->iosignalfds);

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 457

Quoted 10 lines

virt/kvm/eventfd.c +10 -0

+
+	return group;
+}
+
+static int
+kvm_assign_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	int                       pio = args->flags & KVM_IOSIGNALFD_FLAG_PIO;
+	struct kvm_io_bus        *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
+	struct _iosignalfd_group *group = NULL;

why does group need to be initialized?

Quoted 7 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 458

Quoted 3 lines

virt/kvm/eventfd.c +1 -0

+	struct _iosignalfd_item  *item = NULL;

Why does item need to be initialized?

Probably leftover from versions prior to v8. Will fix.

Quoted 9 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 462

Quoted 6 lines

virt/kvm/eventfd.c +4 -0

+	struct file              *file;
+	int                       ret;
+
+	if (args->len > sizeof(u64))

Is e.g. value 3 legal?

Ack. Will check against legal values.

Quoted 29 lines

Quoted 29 lines

virt/kvm/eventfd.c +23 -0

+		return -EINVAL;
 
 
 
 
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	item = kzalloc(sizeof(*item), GFP_KERNEL);
+	if (!item) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	INIT_LIST_HEAD(&item->list);
+	item->file = file;
+
+	/*
+	 * A trigger address is optional, otherwise this is a wildcard
+	 */
+	if (args->flags & KVM_IOSIGNALFD_FLAG_TRIGGER)
+		item->match = args->trigger;
+	else
+		item->wildcard = true;
+

Quoted 11 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 490

Quoted 7 lines

virt/kvm/eventfd.c +5 -0

+	mutex_lock(&kvm->lock);
+
+	/*
+	 * Put an upper limit on the number of items we support
+	 */

Groups and items, actually, right?

Yeah, though technically that is implicit when you say "items", since each group always has at least one item. I will try to make this clearer, though.

Quoted 27 lines

Quoted 27 lines

virt/kvm/eventfd.c +27 -0

+	if (kvm->io_device_count >= CONFIG_KVM_MAX_IO_DEVICES) {
+		ret = -ENOSPC;
+		goto unlock_fail;
+	}
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+
+		group = iosignalfd_group_create(kvm, bus,
+						args->addr, args->len);
+		if (IS_ERR(group)) {
+			ret = PTR_ERR(group);
+			goto unlock_fail;
+		}
+
+		/*
+		 * Note: We do not increment io_device_count for the first item,
+		 * as this is represented by the group device that we just
+		 * registered.  Make sure we handle this properly when we
+		 * deassign the last item
+		 */
+	} else {
+
+		if (group->length != args->len) {
+			/*
+			 * Existing groups must have the same addr/len tuple
+			 * or we reject the request

Quoted 9 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 520

Quoted 5 lines

virt/kvm/eventfd.c +3 -0

+			 */
+			ret = -EINVAL;
+			goto unlock_fail;

Most errors seem to trigger EINVAL. Applications will be easier to debug if different errors are returned on different mistakes.

Yeah, agreed. Will try to differentiate some errors here.

Quoted 5 lines

E.g. here EBUSY might be good. And same
in other places.

Actually, I think EBUSY is supposed to be a transitory error, and would not be appropriate to use here. That said, your point is taken: Find more appropriate and descriptive errors.

Quoted 101 lines

Quoted 101 lines

virt/kvm/eventfd.c +101 -0

+		}
+
+		kvm->io_device_count++;
+	}
+
+	/*
+	 * Note: We are committed to succeed at this point since we have
+	 * (potentially) published a new group-device.  Any failure handling
+	 * added in the future after this point will need to be carefully
+	 * considered.
+	 */
+
+	list_add_tail_rcu(&item->list, &group->items);
+	group->count++;
+
+	mutex_unlock(&kvm->lock);
+
+	return 0;
+
+unlock_fail:
+	mutex_unlock(&kvm->lock);
+fail:
+	if (item)
+		/*
+		 * it would have never made it to the group->items list
+		 * in the failure path, so we dont need to worry about removing
+		 * it
+		 */
+		kfree(item);
+
+	fput(file);
+
+	return ret;
+}
+
+
+static int
+kvm_deassign_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	int                       pio = args->flags & KVM_IOSIGNALFD_FLAG_PIO;
+	struct kvm_io_bus        *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
+	struct _iosignalfd_group *group;
+	struct _iosignalfd_item  *item, *tmp;
+	struct file              *file;
+	int                       ret = 0;
+
+	file = eventfd_fget(args->fd);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	mutex_lock(&kvm->lock);
+
+	group = iosignalfd_group_find(kvm, args->addr);
+	if (!group) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * Exhaustively search our group->items list for any items that might
+	 * match the specified fd, and (carefully) remove each one found.
+	 */
+	list_for_each_entry_safe(item, tmp, &group->items, list) {
+
+		if (item->file != file)
+			continue;
+
+		list_del_rcu(&item->list);
+		group->count--;
+		if (group->count)
+			/*
+			 * We only decrement the global count if this is *not*
+			 * the last item.  The last item will be accounted for
+			 * by the io_bus_unregister
+			 */
+			kvm->io_device_count--;
+
+		/*
+		 * The item may be still referenced inside our group->write()
+		 * path's RCU read-side CS, so defer the actual free to the
+		 * next grace
+		 */
+		call_rcu(&item->rcu, iosignalfd_item_deferred_free);
+	}
+
+	/*
+	 * Check if the group is now completely vacated as a result of
+	 * removing the items.  If so, unregister/delete it
+	 */
+	if (!group->count) {
+
+		kvm_io_bus_unregister_dev(kvm, bus, &group->dev);
+
+		/*
+		 * Like the item, the group may also still be referenced as
+		 * per above.  However, the kvm->iosignalfds list is not
+		 * RCU protected (its protected by kvm->lock instead) so
+		 * we can just plain-vanilla remove it.  What needs to be
+		 * done carefully is the actual freeing of the group pointer
+		 * since we walk the group->items list within the RCU CS.
+		 */

Quoted 10 lines

virt/kvm/eventfd.c kvm_irqfd_release(struct kvm *kvm) · line 623

Quoted 4 lines

virt/kvm/eventfd.c +2 -0

+		list_del(&group->list);
+		call_rcu(&group->rcu, iosignalfd_group_deferred_free);

This is a deferred call, is it not, with no guarantee on when it will run? If correct I think synchronize_rcu might be better here:
- can the module go away while iosignalfd_group_deferred_free is running?

Good catch. Once I go this route it will be easy to use SRCU instead of RCU, too. So I will fix this up.

Quoted 5 lines

- can eventfd be signalled *after* ioctl exits? If yes this might confuse applications if they use the eventfd for something else.

Not by iosignalfd. Once this function completes, we synchronously guarantee that no more IO activity will generate an event on the affected eventfds. Of course, this has no bearing on whether some other producer wont signal, but that is beyond the scope of iosignalfd.

Quoted 50 lines

Quoted 50 lines

virt/kvm/eventfd.c +18 -0

+	}
+
+out:
+	mutex_unlock(&kvm->lock);
+
+	fput(file);
+
+	return ret;
+}
+
+int
+kvm_iosignalfd(struct kvm *kvm, struct kvm_iosignalfd *args)
+{
+	if (args->flags & KVM_IOSIGNALFD_FLAG_DEASSIGN)
+		return kvm_deassign_iosignalfd(kvm, args);
+
+	return kvm_assign_iosignalfd(kvm, args);
+}

virt/kvm/kvm_main.c +10 -1

index 42cbea7..e6495d4 100644
@@ -971,7 +971,7 @@ static struct kvm *kvm_create_vm(void)
 	atomic_inc(&kvm->mm->mm_count);
 	spin_lock_init(&kvm->mmu_lock);
 	kvm_io_bus_init(&kvm->pio_bus);
-	kvm_irqfd_init(kvm);
+	kvm_eventfd_init(kvm);
 	mutex_init(&kvm->lock);
 	mutex_init(&kvm->irq_lock);
 	kvm_io_bus_init(&kvm->mmio_bus);

@@ -2227,6 +2227,15 @@ static long kvm_vm_ioctl(struct file *filp,
 		r = kvm_irqfd(kvm, data.fd, data.gsi, data.flags);
 		break;
 	}
+	case KVM_IOSIGNALFD: {
+		struct kvm_iosignalfd data;
+
+		r = -EFAULT;
+		if (copy_from_user(&data, argp, sizeof data))
+			goto out;
+		r = kvm_iosignalfd(kvm, &data);
+		break;
+	}
 #ifdef CONFIG_KVM_APIC_ARCHITECTURE
 	case KVM_SET_BOOT_CPU_ID:
 		r = 0;

Quoted 11 lines

Quoted 6 lines

Quoted 6 lines

--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Thanks Michael,
-Greg

--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html