On Wed, Mar 11, 2015 at 04:09:17PM +0800, Fam Zheng wrote:
> Currently shutdown is nop for virtio devices, but the core code could
> remove things behind us such as MSI-X handler etc. For example in the
> case of virtio-scsi-pci, the device may still try to send interupts,
> which will be on IRQ lines seeing MSI-X disabled. Those interrupts will
> be unhandled, and may cause flood.
This sounds very tentative. Do you, in fact, observe some problems
with virtio scsi? How to reproduce them? this needs to go
into the commit messages.
> Remove the device in "shutdown" callback to allow device drivers clean
> up things.
>
> Signed-off-by: Fam Zheng <famz@xxxxxxxxxx>
I'm concerned this will cause more hangs on shutdown: one
of the reasons for reboot is device mal-functioning.
How about we just reset devices instead? Something like
the below (untested).
Signed-off-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c
index 5ce2aa4..0769941 100644
--- a/drivers/virtio/virtio.c
+++ b/drivers/virtio/virtio.c
@@ -269,6 +269,17 @@ static int virtio_dev_remove(struct device *_d)
return 0;
}
+static void virtio_dev_shutdown(struct device *_d)
+{
+ struct virtio_device *dev = dev_to_virtio(_d);
+ /*
+ * Reset the device to make it stop sending interrupts, DMA, etc.
+ * We are shutting down, no need for full cleanup.
+ */
+ dev->config->reset(dev);
+
+}
+
static struct bus_type virtio_bus = {
.name = "virtio",
.match = virtio_dev_match,
@@ -276,6 +288,7 @@ static struct bus_type virtio_bus = {
.uevent = virtio_uevent,
.probe = virtio_dev_probe,
.remove = virtio_dev_remove,
+ .shutdown = virtio_dev_shutdown,
};
bool virtio_device_is_legacy_only(struct virtio_device_id id)
_______________________________________________
Virtualization mailing list
Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
