On Wed, Feb 26, 2025 at 01:03:09PM -0500, Joe Damato wrote:

On Wed, Feb 26, 2025 at 01:48:50PM +0800, Jason Wang wrote:

quoted

On Tue, Feb 25, 2025 at 10:05 AM Joe Damato [off-list ref] wrote:

quoted

Use netif_queue_set_napi to map NAPIs to queue IDs so that the mapping
can be accessed by user apps, taking care to hold RTNL as needed.

I may miss something but I wonder whether letting the caller hold the
lock is better.

Hmm...

Double checking all the paths over again, here's what I see:
  - refill_work, delayed work that needs RTNL so this change seems
    right?

  - virtnet_disable_queue_pair, called from virtnet_open and
    virtnet_close. When called via NDO these are safe and hold RTNL,
    but they can be called from power management and need RTNL.

  - virtnet_enable_queue_pair called from virtnet_open, safe when
    used via NDO but needs RTNL when used via power management.

  - virtnet_rx_pause called in both paths as you mentioned, one
    which needs RTNL and one which doesn't.

Sorry, I missed more paths:

    - virtnet_rx_resume
    - virtnet_tx_pause and virtnet_tx_resume

which are similar to path you mentioned (virtnet_rx_pause) and need
rtnl in one of two different paths.

Let me know if I missed any paths and what your preferred way to fix
this would be?

I think both options below are possible and I have no strong
preference.

I think there are a couple ways to fix this:

  1. Edit this patch to remove the virtnet_queue_set_napi helper,
     and call netif_queue_set_napi from the napi_enable and
     napi_disable helpers directly. Modify code calling into these
     paths to hold rtnl (or not) as described above.

  2. Modify virtnet_enable_queue_pair, virtnet_disable_queue_pair,
     and virtnet_rx_pause to take a "bool need_rtnl" as an a
     function argument and pass that through.

I'm not sure which is cleaner and I do not have a preference.

Can you let me know which you prefer? I am happy to implement either
one for the next revision.

[...]

quoted

quoted

---
 drivers/net/virtio_net.c | 73 ++++++++++++++++++++++++++++------------
 1 file changed, 52 insertions(+), 21 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index e578885c1093..13bb4a563073 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c

@@ -2807,6 +2807,20 @@ static void skb_recv_done(struct virtqueue *rvq)
        virtqueue_napi_schedule(&rq->napi, rvq);
 }

+static void virtnet_queue_set_napi(struct net_device *dev,
+                                  struct napi_struct *napi,
+                                  enum netdev_queue_type q_type, int qidx,
+                                  bool need_rtnl)
+{
+       if (need_rtnl)
+               rtnl_lock();
+
+       netif_queue_set_napi(dev, qidx, q_type, napi);
+
+       if (need_rtnl)
+               rtnl_unlock();
+}
+
 static void virtnet_napi_do_enable(struct virtqueue *vq,
                                   struct napi_struct *napi)
 {

@@ -2821,15 +2835,21 @@ static void virtnet_napi_do_enable(struct virtqueue *vq,
        local_bh_enable();
 }

-static void virtnet_napi_enable(struct receive_queue *rq)
+static void virtnet_napi_enable(struct receive_queue *rq, bool need_rtnl)
 {
+       struct virtnet_info *vi = rq->vq->vdev->priv;
+       int qidx = vq2rxq(rq->vq);
+
        virtnet_napi_do_enable(rq->vq, &rq->napi);
+       virtnet_queue_set_napi(vi->dev, &rq->napi,
+                              NETDEV_QUEUE_TYPE_RX, qidx, need_rtnl);
 }

-static void virtnet_napi_tx_enable(struct send_queue *sq)
+static void virtnet_napi_tx_enable(struct send_queue *sq, bool need_rtnl)
 {
        struct virtnet_info *vi = sq->vq->vdev->priv;
        struct napi_struct *napi = &sq->napi;
+       int qidx = vq2txq(sq->vq);

        if (!napi->weight)
                return;

@@ -2843,20 +2863,31 @@ static void virtnet_napi_tx_enable(struct send_queue *sq)
        }

        virtnet_napi_do_enable(sq->vq, napi);
+       virtnet_queue_set_napi(vi->dev, napi, NETDEV_QUEUE_TYPE_TX, qidx,
+                              need_rtnl);
 }

-static void virtnet_napi_tx_disable(struct send_queue *sq)
+static void virtnet_napi_tx_disable(struct send_queue *sq, bool need_rtnl)
 {
+       struct virtnet_info *vi = sq->vq->vdev->priv;
        struct napi_struct *napi = &sq->napi;
+       int qidx = vq2txq(sq->vq);

-       if (napi->weight)
+       if (napi->weight) {
+               virtnet_queue_set_napi(vi->dev, NULL, NETDEV_QUEUE_TYPE_TX,
+                                      qidx, need_rtnl);
                napi_disable(napi);
+       }
 }

-static void virtnet_napi_disable(struct receive_queue *rq)
+static void virtnet_napi_disable(struct receive_queue *rq, bool need_rtnl)
 {
+       struct virtnet_info *vi = rq->vq->vdev->priv;
        struct napi_struct *napi = &rq->napi;
+       int qidx = vq2rxq(rq->vq);

+       virtnet_queue_set_napi(vi->dev, NULL, NETDEV_QUEUE_TYPE_TX, qidx,
+                              need_rtnl);
        napi_disable(napi);
 }

@@ -2870,9 +2901,9 @@ static void refill_work(struct work_struct *work)
        for (i = 0; i < vi->curr_queue_pairs; i++) {
                struct receive_queue *rq = &vi->rq[i];

-               virtnet_napi_disable(rq);
+               virtnet_napi_disable(rq, true);
                still_empty = !try_fill_recv(vi, rq, GFP_KERNEL);
-               virtnet_napi_enable(rq);
+               virtnet_napi_enable(rq, true);

                /* In theory, this can happen: if we don't get any buffers in
                 * we will *never* try to fill again.

@@ -3069,8 +3100,8 @@ static int virtnet_poll(struct napi_struct *napi, int budget)

 static void virtnet_disable_queue_pair(struct virtnet_info *vi, int qp_index)
 {
-       virtnet_napi_tx_disable(&vi->sq[qp_index]);
-       virtnet_napi_disable(&vi->rq[qp_index]);
+       virtnet_napi_tx_disable(&vi->sq[qp_index], false);
+       virtnet_napi_disable(&vi->rq[qp_index], false);
        xdp_rxq_info_unreg(&vi->rq[qp_index].xdp_rxq);
 }

@@ -3089,8 +3120,8 @@ static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index)
        if (err < 0)
                goto err_xdp_reg_mem_model;

-       virtnet_napi_enable(&vi->rq[qp_index]);
-       virtnet_napi_tx_enable(&vi->sq[qp_index]);
+       virtnet_napi_enable(&vi->rq[qp_index], false);
+       virtnet_napi_tx_enable(&vi->sq[qp_index], false);

        return 0;

@@ -3342,7 +3373,7 @@ static void virtnet_rx_pause(struct virtnet_info *vi, struct receive_queue *rq)
        bool running = netif_running(vi->dev);

        if (running) {
-               virtnet_napi_disable(rq);
+               virtnet_napi_disable(rq, true);

During the resize, the rtnl lock has been held on the ethtool path

        rtnl_lock();
        rc = __dev_ethtool(net, ifr, useraddr, ethcmd, state);
        rtnl_unlock();

virtnet_rx_resize()
    virtnet_rx_pause()

and in the case of XSK binding, I see ASSERT_RTNL in xp_assign_dev() at least.

Thanks for catching this. I re-read all the paths and I think I've
outlined a few other issues above.

Please let me know which of the proposed methods above you'd like me
to implement to get this merged.

Thanks.

---
pw-bot: cr