[Qemu-devel] Re: [PATCHv2] qemu-kvm/vhost: fix up irqfd support

[Alex Williamson <alex.williamson@redhat.com>]

On Wed, 2010-10-06 at 16:56 +0200, Michael S. Tsirkin wrote:
> vhost irqfd support: case where many vqs are
> mapped to a single msix vector is currently broken.
> Fix it up.
> 
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> ---
> 
> This is on top of the qemu patchset, which is unchanged.
> Fixes from v1:
> 	correct error handling 
> 
>  hw/msix.c       |   68 ++++++++++++++++++++++++++++++++++++++-----------------
>  hw/msix.h       |    4 +-
>  hw/pci.h        |    3 +-
>  hw/virtio-pci.c |   56 ++++++++++++++++++++++++++++++++++++++-------
>  4 files changed, 97 insertions(+), 34 deletions(-)
> 
> diff --git a/hw/msix.c b/hw/msix.c
> index 3dd0456..3d4dd61 100644
> --- a/hw/msix.c
> +++ b/hw/msix.c
> @@ -300,10 +300,8 @@ static void msix_mmio_writel(void *opaque, target_phys_addr_t addr,
>      if (kvm_enabled() && kvm_irqchip_in_kernel()) {
>          kvm_msix_update(dev, vector, was_masked, msix_is_masked(dev, vector));
>      }
> -    if (was_masked != msix_is_masked(dev, vector) &&
> -        dev->msix_mask_notifier && dev->msix_mask_notifier_opaque[vector]) {
> +    if (was_masked != msix_is_masked(dev, vector) && dev->msix_mask_notifier) {
>          int r = dev->msix_mask_notifier(dev, vector,
> -					dev->msix_mask_notifier_opaque[vector],
>  					msix_is_masked(dev, vector));
>          assert(r >= 0);
>      }
> @@ -351,9 +349,8 @@ static void msix_mask_all(struct PCIDevice *dev, unsigned nentries)
>          int was_masked = msix_is_masked(dev, vector);
>          dev->msix_table_page[offset] |= MSIX_VECTOR_MASK;
>          if (was_masked != msix_is_masked(dev, vector) &&
> -            dev->msix_mask_notifier && dev->msix_mask_notifier_opaque[vector]) {
> +            dev->msix_mask_notifier) {
>              r = dev->msix_mask_notifier(dev, vector,
> -                                        dev->msix_mask_notifier_opaque[vector],
>                                          msix_is_masked(dev, vector));
>              assert(r >= 0);
>          }
> @@ -379,8 +376,6 @@ int msix_init(struct PCIDevice *dev, unsigned short nentries,
>                                              sizeof *dev->msix_irq_entries);
>      }
>  #endif
> -    dev->msix_mask_notifier_opaque =
> -        qemu_mallocz(nentries * sizeof *dev->msix_mask_notifier_opaque);
>      dev->msix_mask_notifier = NULL;
>      dev->msix_entry_used = qemu_mallocz(MSIX_MAX_ENTRIES *
>                                          sizeof *dev->msix_entry_used);
> @@ -444,8 +439,6 @@ int msix_uninit(PCIDevice *dev)
>      dev->msix_entry_used = NULL;
>      qemu_free(dev->msix_irq_entries);
>      dev->msix_irq_entries = NULL;
> -    qemu_free(dev->msix_mask_notifier_opaque);
> -    dev->msix_mask_notifier_opaque = NULL;
>      dev->cap_present &= ~QEMU_PCI_CAP_MSIX;
>      return 0;
>  }
> @@ -590,46 +583,79 @@ void msix_unuse_all_vectors(PCIDevice *dev)
>      msix_free_irq_entries(dev);
>  }
>  
> -int msix_set_mask_notifier(PCIDevice *dev, unsigned vector, void *opaque)
> +static int msix_set_mask_notifier_for_vector(PCIDevice *dev, unsigned vector)
>  {
>      int r = 0;
>      if (vector >= dev->msix_entries_nr || !dev->msix_entry_used[vector])
>          return 0;
>  
>      assert(dev->msix_mask_notifier);
> -    assert(opaque);
> -    assert(!dev->msix_mask_notifier_opaque[vector]);
>  
>      /* Unmask the new notifier unless vector is masked. */
>      if (!msix_is_masked(dev, vector)) {
> -        r = dev->msix_mask_notifier(dev, vector, opaque, false);
> +        r = dev->msix_mask_notifier(dev, vector, false);
>          if (r < 0) {
>              return r;
>          }
>      }
> -    dev->msix_mask_notifier_opaque[vector] = opaque;
>      return r;
>  }
>  
> -int msix_unset_mask_notifier(PCIDevice *dev, unsigned vector)
> +static int msix_unset_mask_notifier_for_vector(PCIDevice *dev, unsigned vector)
>  {
>      int r = 0;
> -    void *opaque;
>      if (vector >= dev->msix_entries_nr || !dev->msix_entry_used[vector])
>          return 0;
>  
> -    opaque = dev->msix_mask_notifier_opaque[vector];
> -
>      assert(dev->msix_mask_notifier);
> -    assert(opaque);
>  
>      /* Mask the old notifier unless it is already masked. */
>      if (!msix_is_masked(dev, vector)) {
> -        r = dev->msix_mask_notifier(dev, vector, opaque, true);
> +        r = dev->msix_mask_notifier(dev, vector, true);
>          if (r < 0) {
>              return r;
>          }
>      }
> -    dev->msix_mask_notifier_opaque[vector] = NULL;
> +    return r;
> +}

The above need to be combined to a single function now since the only
difference is s/true/false.

Alex

> +
> +int msix_set_mask_notifier(PCIDevice *dev, msix_mask_notifier_func f)
> +{
> +    int r, n;
> +    assert(!dev->msix_mask_notifier);
> +    dev->msix_mask_notifier = f;
> +    for (n = 0; n < dev->msix_entries_nr; ++n) {
> +        r = msix_set_mask_notifier_for_vector(dev, n);
> +        if (r < 0) {
> +            goto undo;
> +        }
> +    }
> +    return 0;
> +
> +undo:
> +    while (--n >= 0) {
> +        msix_unset_mask_notifier_for_vector(dev, n);
> +    }
> +    dev->msix_mask_notifier = NULL;
> +    return r;
> +}
> +
> +int msix_unset_mask_notifier(PCIDevice *dev)
> +{
> +    int r, n;
> +    assert(dev->msix_mask_notifier);
> +    for (n = 0; n < dev->msix_entries_nr; ++n) {
> +        r = msix_unset_mask_notifier_for_vector(dev, n);
> +        if (r < 0) {
> +            goto undo;
> +        }
> +    }
> +    dev->msix_mask_notifier = NULL;
> +    return 0;
> +
> +undo:
> +    while (--n >= 0) {
> +        msix_set_mask_notifier_for_vector(dev, n);
> +    }
>      return r;
>  }
> diff --git a/hw/msix.h b/hw/msix.h
> index 6b21ffb..5a81df5 100644
> --- a/hw/msix.h
> +++ b/hw/msix.h
> @@ -33,6 +33,6 @@ void msix_reset(PCIDevice *dev);
>  
>  extern int msix_supported;
>  
> -int msix_set_mask_notifier(PCIDevice *dev, unsigned vector, void *opaque);
> -int msix_unset_mask_notifier(PCIDevice *dev, unsigned vector);
> +int msix_set_mask_notifier(PCIDevice *dev, msix_mask_notifier_func);
> +int msix_unset_mask_notifier(PCIDevice *dev);
>  #endif
> diff --git a/hw/pci.h b/hw/pci.h
> index ccb99d0..a40dc14 100644
> --- a/hw/pci.h
> +++ b/hw/pci.h
> @@ -131,7 +131,7 @@ enum {
>  #define PCI_CAPABILITY_CONFIG_MSIX_LENGTH 0x10
>  
>  typedef int (*msix_mask_notifier_func)(PCIDevice *, unsigned vector,
> -				       void *opaque, int masked);
> +				       int masked);
>  
>  struct PCIDevice {
>      DeviceState qdev;
> @@ -198,7 +198,6 @@ struct PCIDevice {
>  
>      struct kvm_irq_routing_entry *msix_irq_entries;
>  
> -    void **msix_mask_notifier_opaque;
>      msix_mask_notifier_func msix_mask_notifier;
>  
>      /* Device capability configuration space */
> diff --git a/hw/virtio-pci.c b/hw/virtio-pci.c
> index 232f943..c1204ce 100644
> --- a/hw/virtio-pci.c
> +++ b/hw/virtio-pci.c
> @@ -428,11 +428,10 @@ static void virtio_pci_guest_notifier_read(void *opaque)
>      }
>  }
>  
> -static int virtio_pci_mask_notifier(PCIDevice *dev, unsigned vector,
> -                                    void *opaque, int masked)
> +static int virtio_pci_mask_vq(PCIDevice *dev, unsigned vector,
> +                              VirtQueue *vq, int masked)
>  {
>  #ifdef CONFIG_KVM
> -    VirtQueue *vq = opaque;
>      EventNotifier *notifier = virtio_queue_get_guest_notifier(vq);
>      int r = kvm_set_irqfd(dev->msix_irq_entries[vector].gsi,
>                            event_notifier_get_fd(notifier),
> @@ -453,6 +452,34 @@ static int virtio_pci_mask_notifier(PCIDevice *dev, unsigned vector,
>  #endif
>  }
>  
> +static int virtio_pci_mask_notifier(PCIDevice *dev, unsigned vector,
> +                                    int masked)
> +{
> +    VirtIOPCIProxy *proxy = container_of(dev, VirtIOPCIProxy, pci_dev);
> +    VirtIODevice *vdev = proxy->vdev;
> +    int r, n;
> +
> +    for (n = 0; n < VIRTIO_PCI_QUEUE_MAX; n++) {
> +        if (!virtio_queue_get_num(vdev, n)) {
> +            break;
> +        }
> +        if (virtio_queue_vector(vdev, n) != vector) {
> +            continue;
> +        }
> +        r = virtio_pci_mask_vq(dev, vector, virtio_get_queue(vdev, n), masked);
> +        if (r < 0) {
> +            goto undo;
> +        }
> +    }
> +    return 0;
> +undo:
> +    while (--n >= 0) {
> +        virtio_pci_mask_vq(dev, vector, virtio_get_queue(vdev, n), !masked);
> +    }
> +    return r;
> +}
> +
> +
>  static int virtio_pci_set_guest_notifier(void *opaque, int n, bool assign)
>  {
>      VirtIOPCIProxy *proxy = opaque;
> @@ -466,11 +493,7 @@ static int virtio_pci_set_guest_notifier(void *opaque, int n, bool assign)
>          }
>          qemu_set_fd_handler(event_notifier_get_fd(notifier),
>                              virtio_pci_guest_notifier_read, NULL, vq);
> -        msix_set_mask_notifier(&proxy->pci_dev,
> -                               virtio_queue_vector(proxy->vdev, n), vq);
>      } else {
> -        msix_unset_mask_notifier(&proxy->pci_dev,
> -				 virtio_queue_vector(proxy->vdev, n));
>          qemu_set_fd_handler(event_notifier_get_fd(notifier),
>                              NULL, NULL, NULL);
>          /* Test and clear notifier before closing it,
> @@ -488,6 +511,13 @@ static int virtio_pci_set_guest_notifiers(void *opaque, bool assign)
>      VirtIODevice *vdev = proxy->vdev;
>      int r, n;
>  
> +    /* Must unset mask notifier while guest notifier
> +     * is still assigned */
> +    if (!assign) {
> +	    r = msix_unset_mask_notifier(&proxy->pci_dev);
> +            assert(r >= 0);
> +    }
> +
>      for (n = 0; n < VIRTIO_PCI_QUEUE_MAX; n++) {
>          if (!virtio_queue_get_num(vdev, n)) {
>              break;
> @@ -499,6 +529,16 @@ static int virtio_pci_set_guest_notifiers(void *opaque, bool assign)
>          }
>      }
>  
> +    /* Must set mask notifier after guest notifier
> +     * has been assigned */
> +    if (assign) {
> +        r = msix_set_mask_notifier(&proxy->pci_dev,
> +                                   virtio_pci_mask_notifier);
> +        if (r < 0) {
> +            goto assign_error;
> +        }
> +    }
> +
>      return 0;
>  
>  assign_error:
> @@ -584,8 +624,6 @@ static void virtio_init_pci(VirtIOPCIProxy *proxy, VirtIODevice *vdev,
>  
>      proxy->pci_dev.config_write = virtio_write_config;
>  
> -    proxy->pci_dev.msix_mask_notifier = virtio_pci_mask_notifier;
> -
>      size = VIRTIO_PCI_REGION_SIZE(&proxy->pci_dev) + vdev->config_len;
>      if (size & (size-1))
>          size = 1 << qemu_fls(size);