2013-05-22 06:14:13

by Jason Wang

Subject: [PATCH V2] virtio_net: enable napi for all possible queues during open

Commit 55257d72bd1c51f25106350f4983ec19f62ed1fa (virtio-net: fill only rx
queues which are being used) only does the napi enabling during open for
curr_queue_pairs. This will break multiqueue receiving since napi of new queues
were still disabled after changing the number of queues.

This patch fixes this by enabling napi for all possible queues during open.

Cc: Sasha Levin <[email protected]>
Signed-off-by: Jason Wang <[email protected]>
---
Changes from V1:
- enable napi in open instead of virtnet_set_queues() to avoid kernel BUG() when
trying to do napi_enable() for a already enabled queue
---
drivers/net/virtio_net.c | 9 +++++----
1 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 2d6abac..241527d 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -636,10 +636,11 @@ static int virtnet_open(struct net_device *dev)
struct virtnet_info *vi = netdev_priv(dev);
int i;

- for (i = 0; i < vi->curr_queue_pairs; i++) {
- /* Make sure we have some buffers: if oom use wq. */
- if (!try_fill_recv(&vi->rq[i], GFP_KERNEL))
- schedule_delayed_work(&vi->refill, 0);
+ for (i = 0; i < vi->max_queue_pairs; i++) {
+ if (i < vi->curr_queue_pairs)
+ /* Make sure we have some buffers: if oom use wq. */
+ if (!try_fill_recv(&vi->rq[i], GFP_KERNEL))
+ schedule_delayed_work(&vi->refill, 0);
virtnet_napi_enable(&vi->rq[i]);
}

--
1.7.1

2013-05-23 03:44:48

by Rusty Russell

Subject: Re: [PATCH V2] virtio_net: enable napi for all possible queues during open

Jason Wang <[email protected]> writes:
> Commit 55257d72bd1c51f25106350f4983ec19f62ed1fa (virtio-net: fill only rx
> queues which are being used) only does the napi enabling during open for
> curr_queue_pairs. This will break multiqueue receiving since napi of new queues
> were still disabled after changing the number of queues.
>
> This patch fixes this by enabling napi for all possible queues during open.
>
> Cc: Sasha Levin <[email protected]>
> Signed-off-by: Jason Wang <[email protected]>

Acked-by: Rusty Russell <[email protected]>

Dave, please send to Linus (fixes master).

Cheers,
Rusty.

> ---
> Changes from V1:
> - enable napi in open instead of virtnet_set_queues() to avoid kernel BUG() when
> trying to do napi_enable() for a already enabled queue
> ---
> drivers/net/virtio_net.c | 9 +++++----
> 1 files changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 2d6abac..241527d 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -636,10 +636,11 @@ static int virtnet_open(struct net_device *dev)
> struct virtnet_info *vi = netdev_priv(dev);
> int i;
>
> - for (i = 0; i < vi->curr_queue_pairs; i++) {
> - /* Make sure we have some buffers: if oom use wq. */
> - if (!try_fill_recv(&vi->rq[i], GFP_KERNEL))
> - schedule_delayed_work(&vi->refill, 0);
> + for (i = 0; i < vi->max_queue_pairs; i++) {
> + if (i < vi->curr_queue_pairs)
> + /* Make sure we have some buffers: if oom use wq. */
> + if (!try_fill_recv(&vi->rq[i], GFP_KERNEL))
> + schedule_delayed_work(&vi->refill, 0);
> virtnet_napi_enable(&vi->rq[i]);
> }
>
> --
> 1.7.1

2013-05-23 05:59:42

by Michael S. Tsirkin

Subject: Re: [PATCH V2] virtio_net: enable napi for all possible queues during open

On Wed, May 22, 2013 at 02:03:58PM +0800, Jason Wang wrote:
> Commit 55257d72bd1c51f25106350f4983ec19f62ed1fa (virtio-net: fill only rx
> queues which are being used) only does the napi enabling during open for
> curr_queue_pairs. This will break multiqueue receiving since napi of new queues
> were still disabled after changing the number of queues.
>
> This patch fixes this by enabling napi for all possible queues during open.
>
> Cc: Sasha Levin <[email protected]>
> Signed-off-by: Jason Wang <[email protected]>

Acked-by: Michael S. Tsirkin <[email protected]>

> ---
> Changes from V1:
> - enable napi in open instead of virtnet_set_queues() to avoid kernel BUG() when
> trying to do napi_enable() for a already enabled queue
> ---
> drivers/net/virtio_net.c | 9 +++++----
> 1 files changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 2d6abac..241527d 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -636,10 +636,11 @@ static int virtnet_open(struct net_device *dev)
> struct virtnet_info *vi = netdev_priv(dev);
> int i;
>
> - for (i = 0; i < vi->curr_queue_pairs; i++) {
> - /* Make sure we have some buffers: if oom use wq. */
> - if (!try_fill_recv(&vi->rq[i], GFP_KERNEL))
> - schedule_delayed_work(&vi->refill, 0);
> + for (i = 0; i < vi->max_queue_pairs; i++) {
> + if (i < vi->curr_queue_pairs)
> + /* Make sure we have some buffers: if oom use wq. */
> + if (!try_fill_recv(&vi->rq[i], GFP_KERNEL))
> + schedule_delayed_work(&vi->refill, 0);
> virtnet_napi_enable(&vi->rq[i]);
> }
>
> --
> 1.7.1

2013-05-23 07:11:26

by David Miller

Subject: Re: [PATCH V2] virtio_net: enable napi for all possible queues during open

From: Jason Wang <[email protected]>
Date: Wed, 22 May 2013 14:03:58 +0800

> Commit 55257d72bd1c51f25106350f4983ec19f62ed1fa (virtio-net: fill only rx
> queues which are being used) only does the napi enabling during open for
> curr_queue_pairs. This will break multiqueue receiving since napi of new queues
> were still disabled after changing the number of queues.
>
> This patch fixes this by enabling napi for all possible queues during open.
>
> Cc: Sasha Levin <[email protected]>
> Signed-off-by: Jason Wang <[email protected]>

Applied, thanks.