On Sat, 2013-03-02 at 01:16 -0600, Vijay Mohan Pandarathil wrote:
> - New VFIO_SET_IRQ ioctl option to pass the eventfd that is signaled when
> an error occurs in the vfio_pci_device
>
> - Register pci_error_handler for the vfio_pci driver
>
> - When the device encounters an error, the error handler registered by
> the vfio_pci driver gets invoked by the AER infrastructure
>
> - In the error handler, signal the eventfd registered for the device.
>
> - This results in the qemu eventfd handler getting invoked and
> appropriate action taken for the guest.
>
> Signed-off-by: Vijay Mohan Pandarathil <vijaymohan.pandarathil@xxxxxx>
> ---
> drivers/vfio/pci/vfio_pci.c | 44 ++++++++++++++++++++++++++++++++-
> drivers/vfio/pci/vfio_pci_intrs.c | 49 +++++++++++++++++++++++++++++++++++++
> drivers/vfio/pci/vfio_pci_private.h | 1 +
> include/uapi/linux/vfio.h | 1 +
> 4 files changed, 94 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/vfio/pci/vfio_pci.c
> index 8189cb6..acfcb1a 100644
> --- a/drivers/vfio/pci/vfio_pci.c
> +++ b/drivers/vfio/pci/vfio_pci.c
> @@ -201,7 +201,9 @@ static int vfio_pci_get_irq_count(struct vfio_pci_device *vdev, int irq_type)
>
> return (flags & PCI_MSIX_FLAGS_QSIZE) + 1;
> }
> - }
> + } else if (irq_type == VFIO_PCI_ERR_IRQ_INDEX)
> + if (pci_is_pcie(vdev->pdev))
> + return 1;
>
> return 0;
> }
> @@ -317,6 +319,17 @@ static long vfio_pci_ioctl(void *device_data,
> if (info.argsz < minsz || info.index >= VFIO_PCI_NUM_IRQS)
> return -EINVAL;
>
> + switch (info.index) {
> + case VFIO_PCI_INTX_IRQ_INDEX ... VFIO_PCI_MSIX_IRQ_INDEX:
> + break;
> + case VFIO_PCI_ERR_IRQ_INDEX:
> + if (pci_is_pcie(vdev->pdev))
> + break;
> + /* pass thru to return error */
> + default:
> + return -EINVAL;
> + }
> +
> info.flags = VFIO_IRQ_INFO_EVENTFD;
>
> info.count = vfio_pci_get_irq_count(vdev, info.index);
> @@ -551,11 +564,40 @@ static void vfio_pci_remove(struct pci_dev *pdev)
> kfree(vdev);
> }
>
> +static pci_ers_result_t vfio_pci_aer_err_detected(struct pci_dev *pdev,
> + pci_channel_state_t state)
> +{
> + struct vfio_pci_device *vdev;
> + struct vfio_device *device;
> +
> + device = vfio_device_get_from_dev(&pdev->dev);
> + if (device == NULL)
> + return PCI_ERS_RESULT_DISCONNECT;
> +
> + vdev = vfio_device_data(device);
> + if (vdev == NULL) {
> + vfio_device_put(device);
> + return PCI_ERS_RESULT_DISCONNECT;
> + }
> +
> + if (vdev->err_trigger)
> + eventfd_signal(vdev->err_trigger, 1);
> +
> + vfio_device_put(device);
> +
> + return PCI_ERS_RESULT_CAN_RECOVER;
> +}
> +
> +static struct pci_error_handlers vfio_err_handlers = {
> + .error_detected = vfio_pci_aer_err_detected,
> +};
> +
> static struct pci_driver vfio_pci_driver = {
> .name = "vfio-pci",
> .id_table = NULL, /* only dynamic ids */
> .probe = vfio_pci_probe,
> .remove = vfio_pci_remove,
> + .err_handler = &vfio_err_handlers,
> };
>
> static void __exit vfio_pci_cleanup(void)
> diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
> index 3639371..4a29830 100644
> --- a/drivers/vfio/pci/vfio_pci_intrs.c
> +++ b/drivers/vfio/pci/vfio_pci_intrs.c
> @@ -745,6 +745,48 @@ static int vfio_pci_set_msi_trigger(struct vfio_pci_device *vdev,
> return 0;
> }
>
> +static int vfio_pci_set_err_trigger(struct vfio_pci_device *vdev,
> + unsigned index, unsigned start,
> + unsigned count, uint32_t flags, void *data)
> +{
> + int32_t fd = *(int32_t *)data;
> +
> + if ((index != VFIO_PCI_ERR_IRQ_INDEX) ||
> + !(flags & VFIO_IRQ_SET_DATA_TYPE_MASK))
> + return -EINVAL;
> +
> + /* DATA_NONE/DATA_BOOL enables loopback testing */
> +
> + if (flags & VFIO_IRQ_SET_DATA_NONE) {
> + if (vdev->err_trigger)
> + eventfd_signal(vdev->err_trigger, 1);
> + return 0;
> + } else if (flags & VFIO_IRQ_SET_DATA_BOOL) {
> + uint8_t trigger = *(uint8_t *)data;
> + if (trigger && vdev->err_trigger)
> + eventfd_signal(vdev->err_trigger, 1);
> + return 0;
> + }
> +
> + /* Handle SET_DATA_EVENTFD */
> +
> + if (fd == -1) {
> + if (vdev->err_trigger)
> + eventfd_ctx_put(vdev->err_trigger);
I mentioned ordering/locking issues back on v3 and I don't think they've
been addressed yet.
What happens if error_detected is called here?
> + vdev->err_trigger = NULL;
> + return 0;
> + } else if (fd >= 0) {
> + struct eventfd_ctx *efdctx;
> + efdctx = eventfd_ctx_fdget(fd);
> + if (IS_ERR(efdctx))
> + return PTR_ERR(efdctx);
> + if (vdev->err_trigger)
> + eventfd_ctx_put(vdev->err_trigger);
Or here?
Both are brief windows where vdev->err_trigger is neither NULL nor
valid. The other trigger setup functions do a 1) disable, 2) re-enable
where the disable is synchronous and avoids this race. I don't know if
you have that capability, so we have to assume that error_detected can
be called at any time. I notice that report_error_detected() wraps the
callback in a device_lock(), so you could potentially use
device_lock/unlock here to avoid racing it. Thanks,
Alex
> + vdev->err_trigger = efdctx;
> + return 0;
> + } else
> + return -EINVAL;
> +}
> int vfio_pci_set_irqs_ioctl(struct vfio_pci_device *vdev, uint32_t flags,
> unsigned index, unsigned start, unsigned count,
> void *data)
> @@ -779,6 +821,13 @@ int vfio_pci_set_irqs_ioctl(struct vfio_pci_device *vdev, uint32_t flags,
> break;
> }
> break;
> + case VFIO_PCI_ERR_IRQ_INDEX:
> + switch (flags & VFIO_IRQ_SET_ACTION_TYPE_MASK) {
> + case VFIO_IRQ_SET_ACTION_TRIGGER:
> + if (pci_is_pcie(vdev->pdev))
> + func = vfio_pci_set_err_trigger;
> + break;
> + }
> }
>
> if (!func)
> diff --git a/drivers/vfio/pci/vfio_pci_private.h b/drivers/vfio/pci/vfio_pci_private.h
> index d7e55d0..9c6d5d0 100644
> --- a/drivers/vfio/pci/vfio_pci_private.h
> +++ b/drivers/vfio/pci/vfio_pci_private.h
> @@ -56,6 +56,7 @@ struct vfio_pci_device {
> bool has_vga;
> struct pci_saved_state *pci_saved_state;
> atomic_t refcnt;
> + struct eventfd_ctx *err_trigger;
> };
>
> #define is_intx(vdev) (vdev->irq_type == VFIO_PCI_INTX_IRQ_INDEX)
> diff --git a/include/uapi/linux/vfio.h b/include/uapi/linux/vfio.h
> index 4f41f30..284ff24 100644
> --- a/include/uapi/linux/vfio.h
> +++ b/include/uapi/linux/vfio.h
> @@ -319,6 +319,7 @@ enum {
> VFIO_PCI_INTX_IRQ_INDEX,
> VFIO_PCI_MSI_IRQ_INDEX,
> VFIO_PCI_MSIX_IRQ_INDEX,
> + VFIO_PCI_ERR_IRQ_INDEX,
> VFIO_PCI_NUM_IRQS
> };
>
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
