[Qemu-devel] [PATCH v2] virtio-pci: call pci reset variant when guest clears status.

[Qemu-devel] [PATCH v2] virtio-pci: call pci reset variant when guest clears status.

[Gerd Hoffmann]

Actually fixes linux not finding virtio 1.0 device virtqueues after
reboot.  Which is new I think, any chance linux kernel virtio code
became more strict in 4.3?

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
 hw/virtio/virtio-pci.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 94667e6..8213d94 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -47,6 +47,7 @@
 
 static void virtio_pci_bus_new(VirtioBusState *bus, size_t bus_size,
                                VirtIOPCIProxy *dev);
+static void virtio_pci_reset(DeviceState *qdev);
 
 /* virtio device */
 /* DeviceState to VirtIOPCIProxy. For use off data-path. TODO: use QOM. */
@@ -432,8 +433,7 @@ static void virtio_ioport_write(void *opaque, uint32_t addr, uint32_t val)
         }
 
         if (vdev->status == 0) {
-            virtio_reset(vdev);
-            msix_unuse_all_vectors(&proxy->pci_dev);
+            virtio_pci_reset(DEVICE(proxy));
         }
 
         /* Linux before 2.6.34 drives the device without enabling
@@ -1351,8 +1351,7 @@ static void virtio_pci_common_write(void *opaque, hwaddr addr,
         }
 
         if (vdev->status == 0) {
-            virtio_reset(vdev);
-            msix_unuse_all_vectors(&proxy->pci_dev);
+            virtio_pci_reset(DEVICE(proxy));
         }
 
         break;
-- 
1.8.3.1

Re: [Qemu-devel] [PATCH v2] virtio-pci: call pci reset variant when guest clears status.

[Michael S. Tsirkin]

On Wed, Jan 27, 2016 at 03:09:58PM +0100, Gerd Hoffmann wrote:
> Actually fixes linux not finding virtio 1.0 device virtqueues after
> reboot.  Which is new I think, any chance linux kernel virtio code
> became more strict in 4.3?
> 
> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
> ---
>  hw/virtio/virtio-pci.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
> index 94667e6..8213d94 100644
> --- a/hw/virtio/virtio-pci.c
> +++ b/hw/virtio/virtio-pci.c
> @@ -47,6 +47,7 @@
>  
>  static void virtio_pci_bus_new(VirtioBusState *bus, size_t bus_size,
>                                 VirtIOPCIProxy *dev);
> +static void virtio_pci_reset(DeviceState *qdev);
>  
>  /* virtio device */
>  /* DeviceState to VirtIOPCIProxy. For use off data-path. TODO: use QOM. */
> @@ -432,8 +433,7 @@ static void virtio_ioport_write(void *opaque, uint32_t addr, uint32_t val)
>          }
>  
>          if (vdev->status == 0) {
> -            virtio_reset(vdev);
> -            msix_unuse_all_vectors(&proxy->pci_dev);
> +            virtio_pci_reset(DEVICE(proxy));
>          }
>  
>          /* Linux before 2.6.34 drives the device without enabling

Aren't there two call sites in virtio_ioport_write?

> @@ -1351,8 +1351,7 @@ static void virtio_pci_common_write(void *opaque, hwaddr addr,
>          }
>  
>          if (vdev->status == 0) {
> -            virtio_reset(vdev);
> -            msix_unuse_all_vectors(&proxy->pci_dev);
> +            virtio_pci_reset(DEVICE(proxy));
>          }
>  
>          break;
> -- 
> 1.8.3.1

Re: [Qemu-devel] [PATCH v2] virtio-pci: call pci reset variant when guest clears status.

[Gerd Hoffmann]

> >          if (vdev->status == 0) {
> > -            virtio_reset(vdev);
> > -            msix_unuse_all_vectors(&proxy->pci_dev);
> > +            virtio_pci_reset(DEVICE(proxy));
> >          }
> >  
> >          /* Linux before 2.6.34 drives the device without enabling
> 
> Aren't there two call sites in virtio_ioport_write?

/me was specifically fixing clear-status code paths mentioned in the
commit message.

Yes, there is another one, not fully sure what is supposed to happen
there.

cheers,
  Gerd

Re: [Qemu-devel] [PATCH v2] virtio-pci: call pci reset variant when guest clears status.

[Michael S. Tsirkin]

On Thu, Jan 28, 2016 at 10:33:25AM +0100, Gerd Hoffmann wrote:
> > >          if (vdev->status == 0) {
> > > -            virtio_reset(vdev);
> > > -            msix_unuse_all_vectors(&proxy->pci_dev);
> > > +            virtio_pci_reset(DEVICE(proxy));
> > >          }
> > >  
> > >          /* Linux before 2.6.34 drives the device without enabling
> > 
> > Aren't there two call sites in virtio_ioport_write?
> 
> /me was specifically fixing clear-status code paths mentioned in the
> commit message.
> 
> Yes, there is another one, not fully sure what is supposed to happen
> there.
> 
> cheers,
>   Gerd

Same thing really - virtio 0 has two equivalent ways to
reset a device: write 0 into status or write 0 into PFN
register.

-- 
MST