> > > Can't this problem be solved by adding "offset" field in
> > > virtio_video_mem_entry?
> > >
> > > struct virtio_video_mem_entry {
> > > le64 addr;
> > > le32 length;
> > > le32 offset;
> > > u8 padding[4];
> > > };
> > >
> > > Here, "addr" must be the same in every mem_entry for (1)
All the queues notifications use the same register on MMIO transport
layer. Add a feature bit (39) for enhancing the notification capability.
The detailed mechanism would be in next patch.
Co-developed-by: Chao Peng
Signed-off-by: Chao Peng
Co-developed-by: Liu Jiang
Signed-off-by: Liu Jiang
C
Bit 1 msi_sharing reported in the MsiState register indicates the mapping mode
device uses.
Bit 1 is 0 - device uses MSI non-sharing mode. This indicates vector per event
and
fixed static vectors and events relationship. This fits for devices with a high
interrupt
rate and best performance;
Bit
The current MMIO transport layer uses a single, dedicated interrupt
signal, which brings performance penalty. Add a feature bit (40)
for introducing MSI capability.
Co-developed-by: Chao Peng
Signed-off-by: Chao Peng
Co-developed-by: Liu Jiang
Signed-off-by: Liu Jiang
Co-developed-by: Zha Bin
The current virtio over MMIO has some limitations that impact the performance.
It only supports single legacy, dedicated interrupt and one virtqueue
notification register for all virtqueues which cause performance penalties.
To address such limitations, we proposed to update virtio-mmio spec with
With VIRTIO_F_MMIO_MSI feature bit offered, the Message Signal
Interrupts (MSI) is supported as first priority. For any reason it
fails to use MSI, it need use the single dedicated interrupt as before.
For MSI vectors and events mapping relationship, introduce in next patch.
Co-developed-by: Chao
With VIRTIO_F_MMIO_NOTIFICATION feature bit offered, the notification
mechanism is enhanced. Driver reads QueueNotify register to get
notification structure and calculate notification addresses of
each virtqueue.
Co-developed-by: Chao Peng
Signed-off-by: Chao Peng
Co-developed-by: Liu Jiang
Sig
On 1/5/2020 7:04 PM, Michael S. Tsirkin wrote:
struct virtio_mmio_vq_info {
@@ -101,6 +107,8 @@ struct virtio_mmio_vq_info {
};
+static void vm_free_msi_irqs(struct virtio_device *vdev);
+static int vm_request_msi_vectors(struct virtio_device *vdev, int nirqs);
/* Configura
Hi,
On Mon, Jan 20, 2020 at 7:48 PM Gerd Hoffmann wrote:
>
> Hi,
>
> > > Hmm, using (ii) the API, then check whenever your three plane buffers
> > > happen to have the correct layout for (1) hardware looks somewhat
> > > backwards to me.
> >
> > Can't this problem be solved by adding "offset" f
Hi,
> > Hmm, using (ii) the API, then check whenever your three plane buffers
> > happen to have the correct layout for (1) hardware looks somewhat
> > backwards to me.
>
> Can't this problem be solved by adding "offset" field in
> virtio_video_mem_entry?
>
> struct virtio_video_mem_entry {
>
10 matches
Mail list logo
