lkml.org 
[lkml]   [2009]   [Jul]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: [KVM PATCH v10 2/2] KVM: add ioeventfd support
    Gregory Haskins wrote:
    > ioeventfd 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
    > ioeventfd.
    >
    > 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
    > ioeventfd-mmio: 200100 iops, 5.00us rtt
    > ioeventfd-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
    > ioeventfd-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 <ghaskins@novell.com>
    > ---
    >
    > arch/x86/kvm/x86.c | 1
    > include/linux/kvm.h | 24 ++++
    > include/linux/kvm_host.h | 10 +-
    > virt/kvm/eventfd.c | 252 ++++++++++++++++++++++++++++++++++++++++++++++
    > virt/kvm/kvm_main.c | 11 ++
    > 5 files changed, 294 insertions(+), 4 deletions(-)
    >
    > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
    > index 95fa45c..59c2d93 100644
    > --- a/arch/x86/kvm/x86.c
    > +++ b/arch/x86/kvm/x86.c
    > @@ -1212,6 +1212,7 @@ 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_IOEVENTFD:
    > case KVM_CAP_PIT2:
    > r = 1;
    > break;
    > diff --git a/include/linux/kvm.h b/include/linux/kvm.h
    > index 76c6408..22d0eb7 100644
    > --- a/include/linux/kvm.h
    > +++ b/include/linux/kvm.h
    > @@ -307,6 +307,28 @@ struct kvm_guest_debug {
    > struct kvm_guest_debug_arch arch;
    > };
    >
    > +enum {
    > + kvm_ioeventfd_flag_nr_datamatch,
    > + kvm_ioeventfd_flag_nr_pio,
    > + kvm_ioeventfd_flag_nr_deassign,
    > + kvm_ioeventfd_flag_nr_max,
    > +};
    > +
    > +#define KVM_IOEVENTFD_FLAG_DATAMATCH (1 << kvm_ioeventfd_flag_nr_datamatch)
    > +#define KVM_IOEVENTFD_FLAG_PIO (1 << kvm_ioeventfd_flag_nr_pio)
    > +#define KVM_IOEVENTFD_FLAG_DEASSIGN (1 << kvm_ioeventfd_flag_nr_deassign)
    > +
    > +#define KVM_IOEVENTFD_VALID_FLAG_MASK ((1 << kvm_ioeventfd_flag_nr_max) - 1)
    > +
    > +struct kvm_ioeventfd {
    > + __u64 datamatch;
    > + __u64 addr; /* legal pio/mmio address */
    > + __u32 len; /* 1, 2, 4, or 8 bytes */
    > + __s32 fd;
    > + __u32 flags;
    > + __u8 pad[36];
    > +};
    > +
    > #define KVM_TRC_SHIFT 16
    > /*
    > * kvm trace categories
    > @@ -409,6 +431,7 @@ struct kvm_guest_debug {
    > #define KVM_CAP_PIT2 33
    > #endif
    > #define KVM_CAP_SET_BOOT_CPU_ID 34
    > +#define KVM_CAP_IOEVENTFD 35
    >
    > #ifdef KVM_CAP_IRQ_ROUTING
    >
    > @@ -517,6 +540,7 @@ 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_IOEVENTFD _IOW(KVMIO, 0x79, struct kvm_ioeventfd)
    >
    > /*
    > * ioctls for vcpu fds
    > diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
    > index 306bc67..0347d59 100644
    > --- a/include/linux/kvm_host.h
    > +++ b/include/linux/kvm_host.h
    > @@ -154,6 +154,7 @@ struct kvm {
    > spinlock_t lock;
    > struct list_head items;
    > } irqfds;
    > + struct list_head ioeventfds;
    > #endif
    > struct kvm_vm_stat stat;
    > struct kvm_arch arch;
    > @@ -532,19 +533,24 @@ 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_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *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;
    > }
    >
    > static inline void kvm_irqfd_release(struct kvm *kvm) {}
    > +static inline int kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
    > +{
    > + return -ENOSYS;
    > +}
    >
    > #endif /* CONFIG_HAVE_KVM_EVENTFD */
    >
    > diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c
    > index 4092b8d..eee8edb 100644
    > --- a/virt/kvm/eventfd.c
    > +++ b/virt/kvm/eventfd.c
    > @@ -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>
    > @@ -28,6 +29,9 @@
    > #include <linux/file.h>
    > #include <linux/list.h>
    > #include <linux/eventfd.h>
    > +#include <linux/kernel.h>
    > +
    > +#include "iodev.h"
    >
    > /*
    > * --------------------------------------------------------------------
    > @@ -234,10 +238,11 @@ fail:
    > }
    >
    > void
    > -kvm_irqfd_init(struct kvm *kvm)
    > +kvm_eventfd_init(struct kvm *kvm)
    > {
    > spin_lock_init(&kvm->irqfds.lock);
    > INIT_LIST_HEAD(&kvm->irqfds.items);
    > + INIT_LIST_HEAD(&kvm->ioeventfds);
    > }
    >
    > /*
    > @@ -327,3 +332,248 @@ static void __exit irqfd_module_exit(void)
    >
    > module_init(irqfd_module_init);
    > module_exit(irqfd_module_exit);
    > +
    > +/*
    > + * --------------------------------------------------------------------
    > + * ioeventfd: translate a PIO/MMIO memory write to an eventfd signal.
    > + *
    > + * userspace can register a PIO/MMIO address with an eventfd for receiving
    > + * notification when the memory has been touched.
    > + * --------------------------------------------------------------------
    > + */
    > +
    > +struct _ioeventfd {
    > + struct list_head list;
    > + u64 addr;
    > + int length;
    > + struct eventfd_ctx *eventfd;
    > + u64 datamatch;
    > + struct kvm_io_device dev;
    > + bool wildcard;
    > +};
    > +
    > +static inline struct _ioeventfd *
    > +to_ioeventfd(struct kvm_io_device *dev)
    > +{
    > + return container_of(dev, struct _ioeventfd, dev);
    > +}
    > +
    > +static void
    > +ioeventfd_release(struct _ioeventfd *p)
    > +{
    > + eventfd_ctx_put(p->eventfd);
    > + list_del(&p->list);
    > + kfree(p);
    > +}
    > +
    > +static bool
    > +ioeventfd_in_range(struct _ioeventfd *p, gpa_t addr, int len, const void *val)
    > +{
    > + u64 _val;
    > +
    > + if (!(addr == p->addr && len == p->length))
    > + /* address-range must be precise for a hit */
    > + return false;
    > +
    > + if (p->wildcard)
    > + /* all else equal, wildcard is always a hit */
    > + return true;
    > +
    > + /* otherwise, we have to actually compare the data */
    > +
    > + BUG_ON(!IS_ALIGNED((unsigned long)val, len));
    > +
    > + 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;
    > + break;
    > + default:
    > + return false;
    > + }
    > +
    > + return _val == p->datamatch ? true : false;
    > +}
    > +
    > +/* MMIO/PIO writes trigger an event if the addr/val match */
    > +static int
    > +ioeventfd_write(struct kvm_io_device *this, gpa_t addr, int len,
    > + const void *val)
    > +{
    > + struct _ioeventfd *p = to_ioeventfd(this);
    > +
    > + if (!ioeventfd_in_range(p, addr, len, val))
    > + return -EOPNOTSUPP;
    > +
    > + eventfd_signal(p->eventfd, 1);
    > + return 0;
    > +}
    > +
    > +/*
    > + * This function is called as KVM is completely shutting down. We do not
    > + * need to worry about locking just nuke anything we have as quickly as possible
    > + */
    > +static void
    > +ioeventfd_destructor(struct kvm_io_device *this)
    > +{
    > + struct _ioeventfd *p = to_ioeventfd(this);
    > +
    > + ioeventfd_release(p);
    > +}
    > +
    > +static const struct kvm_io_device_ops ioeventfd_ops = {
    > + .write = ioeventfd_write,
    > + .destructor = ioeventfd_destructor,
    > +};
    > +
    > +/* assumes kvm->slots_lock held */
    > +static bool
    > +ioeventfd_check_collision(struct kvm *kvm, struct _ioeventfd *p)
    > +{
    > + struct _ioeventfd *_p;
    > +
    > + list_for_each_entry(_p, &kvm->ioeventfds, list)
    > + if (_p->addr == p->addr && _p->length == p->length &&
    > + (_p->wildcard || p->wildcard ||
    > + _p->datamatch == p->datamatch))
    > + return true;
    > +
    > + return false;
    > +}
    > +
    > +static int
    > +kvm_assign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
    > +{
    > + int pio = args->flags & KVM_IOEVENTFD_FLAG_PIO;
    > + struct kvm_io_bus *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
    > + struct _ioeventfd *p;
    > + struct eventfd_ctx *eventfd;
    > + int ret;
    > +
    > + /* must be natural-word sized */
    > + switch (args->len) {
    > + case 1:
    > + case 2:
    > + case 4:
    > + case 8:
    > + break;
    > + default:
    > + return -EINVAL;
    > + }
    > +
    > + /* check for range overflow */
    > + if (args->addr + args->len < args->addr)
    > + return -EINVAL;
    > +
    > + /* check for extra flags that we don't understand */
    > + if (args->flags & ~KVM_IOEVENTFD_VALID_FLAG_MASK)
    > + return -EINVAL;
    > +
    > + eventfd = eventfd_ctx_fdget(args->fd);
    > + if (IS_ERR(eventfd))
    > + return PTR_ERR(eventfd);
    > +
    > + p = kzalloc(sizeof(*p), GFP_KERNEL);
    > + if (!p) {
    > + ret = -ENOMEM;
    > + goto fail;
    > + }
    > +
    > + INIT_LIST_HEAD(&p->list);
    > + p->addr = args->addr;
    > + p->length = args->len;
    > + p->eventfd = eventfd;
    > +
    > + /* The datamatch feature is optional, otherwise this is a wildcard */
    > + if (args->flags & KVM_IOEVENTFD_FLAG_DATAMATCH)
    > + p->datamatch = args->datamatch;
    > + else
    > + p->wildcard = true;
    > +
    > + down_write(&kvm->slots_lock);
    > +
    > + /* Verify that there isnt a match already */
    > + if (ioeventfd_check_collision(kvm, p)) {
    > + ret = -EEXIST;
    > + goto unlock_fail;
    > + }
    > +
    > + kvm_iodevice_init(&p->dev, &ioeventfd_ops);
    > +
    > + ret = __kvm_io_bus_register_dev(bus, &p->dev);
    > + if (ret < 0)
    > + goto unlock_fail;
    > +
    > + list_add_tail(&p->list, &kvm->ioeventfds);
    > +
    > + up_write(&kvm->slots_lock);
    > +
    > + return 0;
    > +
    > +unlock_fail:
    > + up_write(&kvm->slots_lock);
    > +
    > +fail:
    > + kfree(p);
    > + eventfd_ctx_put(eventfd);
    > +
    > + return ret;
    > +}
    > +
    > +static int
    > +kvm_deassign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
    > +{
    > + int pio = args->flags & KVM_IOEVENTFD_FLAG_PIO;
    > + struct kvm_io_bus *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus;
    > + struct _ioeventfd *p, *tmp;
    > + struct eventfd_ctx *eventfd;
    > + int ret = -ENOENT;
    > +
    > + eventfd = eventfd_ctx_fdget(args->fd);
    > + if (IS_ERR(eventfd))
    > + return PTR_ERR(eventfd);
    > +
    > + down_write(&kvm->slots_lock);
    > +
    > + list_for_each_entry_safe(p, tmp, &kvm->ioeventfds, list) {
    > + bool wildcard = args->flags & KVM_IOEVENTFD_FLAG_DATAMATCH ?
    > + true : false;
    >
    Doh! Inverted logic. "wildcard" should be "false" if DATAMATCH is
    defined in the flags.

    Avi, please reverse true/false here before accepting (assuming a v11
    respin isn't needed). Of course, if you would prefer me to just submit
    a new patch, that is fine too.

    -Greg

    > +
    > + if (p->eventfd != eventfd ||
    > + p->addr != args->addr ||
    > + p->length != args->len ||
    > + p->wildcard != wildcard)
    > + continue;
    > +
    > + if (!p->wildcard && p->datamatch != args->datamatch)
    > + continue;
    > +
    > + __kvm_io_bus_unregister_dev(bus, &p->dev);
    > + ioeventfd_release(p);
    > + ret = 0;
    > + break;
    > + }
    > +
    > + up_write(&kvm->slots_lock);
    > +
    > + eventfd_ctx_put(eventfd);
    > +
    > + return ret;
    > +}
    > +
    > +int
    > +kvm_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args)
    > +{
    > + if (args->flags & KVM_IOEVENTFD_FLAG_DEASSIGN)
    > + return kvm_deassign_ioeventfd(kvm, args);
    > +
    > + return kvm_assign_ioeventfd(kvm, args);
    > +}
    > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
    > index dd92b44..14e1f32 100644
    > --- a/virt/kvm/kvm_main.c
    > +++ b/virt/kvm/kvm_main.c
    > @@ -979,7 +979,7 @@ static struct kvm *kvm_create_vm(void)
    > spin_lock_init(&kvm->mmu_lock);
    > spin_lock_init(&kvm->requests_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);
    > @@ -2271,6 +2271,15 @@ static long kvm_vm_ioctl(struct file *filp,
    > r = kvm_irqfd(kvm, data.fd, data.gsi, data.flags);
    > break;
    > }
    > + case KVM_IOEVENTFD: {
    > + struct kvm_ioeventfd data;
    > +
    > + r = -EFAULT;
    > + if (copy_from_user(&data, argp, sizeof data))
    > + goto out;
    > + r = kvm_ioeventfd(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 majordomo@vger.kernel.org
    > More majordomo info at http://vger.kernel.org/majordomo-info.html
    >


    [unhandled content-type:application/pgp-signature]
    \
     
     \ /
      Last update: 2009-07-09 17:05    [W:0.112 / U:149.784 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site