public inbox for netdev@vger.kernel.org
 help / color / mirror / Atom feed
From: Daniel Borkmann <daniel@iogearbox.net>
To: netdev@vger.kernel.org
Cc: bpf@vger.kernel.org, kuba@kernel.org, davem@davemloft.net,
	razor@blackwall.org, pabeni@redhat.com, willemb@google.com,
	sdf@fomichev.me, john.fastabend@gmail.com, martin.lau@kernel.org,
	jordan@jrife.io, maciej.fijalkowski@intel.com,
	magnus.karlsson@intel.com, dw@davidwei.uk, toke@redhat.com,
	yangzhenze@bytedance.com, wangdongdong.6@bytedance.com
Subject: [PATCH net-next v8 06/16] net: Proxy net_mp_{open,close}_rxq for leased queues
Date: Thu, 29 Jan 2026 23:28:20 +0100	[thread overview]
Message-ID: <20260129222830.439687-7-daniel@iogearbox.net> (raw)
In-Reply-To: <20260129222830.439687-1-daniel@iogearbox.net>

From: David Wei <dw@davidwei.uk>

When a process in a container wants to setup a memory provider, it will
use the virtual netdev and a leased rxq, and call net_mp_{open,close}_rxq
to try and restart the queue. At this point, proxy the queue restart on
the real rxq in the physical netdev.

For memory providers (io_uring zero-copy rx and devmem), it causes the
real rxq in the physical netdev to be filled from a memory provider that
has DMA mapped memory from a process within a container.

Signed-off-by: David Wei <dw@davidwei.uk>
Co-developed-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
 include/net/netdev_rx_queue.h |   4 ++
 net/core/netdev_rx_queue.c    | 100 +++++++++++++++++++++++++++-------
 2 files changed, 83 insertions(+), 21 deletions(-)

diff --git a/include/net/netdev_rx_queue.h b/include/net/netdev_rx_queue.h
index 589192f67d97..0286c81fb156 100644
--- a/include/net/netdev_rx_queue.h
+++ b/include/net/netdev_rx_queue.h
@@ -75,4 +75,8 @@ enum netif_lease_dir {
 struct netdev_rx_queue *
 __netif_get_rx_queue_lease(struct net_device **dev, unsigned int *rxq,
 			   enum netif_lease_dir dir);
+struct netdev_rx_queue *
+netif_get_rx_queue_lease_locked(struct net_device **dev, unsigned int *rxq);
+void netif_put_rx_queue_lease_locked(struct net_device *orig_dev,
+				     struct net_device *dev);
 #endif /* _LINUX_NETDEV_RX_QUEUE_H */
diff --git a/net/core/netdev_rx_queue.c b/net/core/netdev_rx_queue.c
index f30fe7550455..34e523f1e926 100644
--- a/net/core/netdev_rx_queue.c
+++ b/net/core/netdev_rx_queue.c
@@ -68,6 +68,29 @@ __netif_get_rx_queue_lease(struct net_device **dev, unsigned int *rxq_idx,
 	return rxq;
 }
 
+struct netdev_rx_queue *
+netif_get_rx_queue_lease_locked(struct net_device **dev, unsigned int *rxq_idx)
+{
+	struct net_device *orig_dev = *dev;
+	struct netdev_rx_queue *rxq;
+
+	/* Locking order is always from the virtual to the physical device
+	 * see netdev_nl_queue_create_doit().
+	 */
+	netdev_ops_assert_locked(orig_dev);
+	rxq = __netif_get_rx_queue_lease(dev, rxq_idx, NETIF_VIRT_TO_PHYS);
+	if (rxq && orig_dev != *dev)
+		netdev_lock(*dev);
+	return rxq;
+}
+
+void netif_put_rx_queue_lease_locked(struct net_device *orig_dev,
+				     struct net_device *dev)
+{
+	if (orig_dev != dev)
+		netdev_unlock(dev);
+}
+
 bool netif_rx_queue_lease_get_owner(struct net_device **dev,
 				    unsigned int *rxq_idx)
 {
@@ -191,24 +214,15 @@ int netdev_rx_queue_restart(struct net_device *dev, unsigned int rxq_idx)
 }
 EXPORT_SYMBOL_NS_GPL(netdev_rx_queue_restart, "NETDEV_INTERNAL");
 
-int net_mp_open_rxq(struct net_device *dev, unsigned int rxq_idx,
-		    const struct pp_memory_provider_params *p,
-		    struct netlink_ext_ack *extack)
+static int __net_mp_open_rxq(struct net_device *dev, unsigned int rxq_idx,
+			     const struct pp_memory_provider_params *p,
+			     struct netlink_ext_ack *extack)
 {
 	const struct netdev_queue_mgmt_ops *qops = dev->queue_mgmt_ops;
 	struct netdev_queue_config qcfg[2];
 	struct netdev_rx_queue *rxq;
 	int ret;
 
-	if (!netdev_need_ops_lock(dev))
-		return -EOPNOTSUPP;
-
-	if (rxq_idx >= dev->real_num_rx_queues) {
-		NL_SET_ERR_MSG(extack, "rx queue index out of range");
-		return -ERANGE;
-	}
-	rxq_idx = array_index_nospec(rxq_idx, dev->real_num_rx_queues);
-
 	if (dev->cfg->hds_config != ETHTOOL_TCP_DATA_SPLIT_ENABLED) {
 		NL_SET_ERR_MSG(extack, "tcp-data-split is disabled");
 		return -EINVAL;
@@ -255,17 +269,47 @@ int net_mp_open_rxq(struct net_device *dev, unsigned int rxq_idx,
 	return ret;
 }
 
-void net_mp_close_rxq(struct net_device *dev, unsigned int ifq_idx,
-		      const struct pp_memory_provider_params *old_p)
+int net_mp_open_rxq(struct net_device *dev, unsigned int rxq_idx,
+		    const struct pp_memory_provider_params *p,
+		    struct netlink_ext_ack *extack)
+{
+	struct net_device *orig_dev = dev;
+	int ret;
+
+	if (!netdev_need_ops_lock(dev))
+		return -EOPNOTSUPP;
+
+	if (rxq_idx >= dev->real_num_rx_queues) {
+		NL_SET_ERR_MSG(extack, "rx queue index out of range");
+		return -ERANGE;
+	}
+
+	rxq_idx = array_index_nospec(rxq_idx, dev->real_num_rx_queues);
+
+	if (!netif_get_rx_queue_lease_locked(&dev, &rxq_idx)) {
+		NL_SET_ERR_MSG(extack, "rx queue leased to a virtual netdev");
+		return -EBUSY;
+	}
+	if (!dev->dev.parent) {
+		NL_SET_ERR_MSG(extack, "rx queue belongs to a virtual netdev");
+		ret = -EOPNOTSUPP;
+		goto out;
+	}
+
+	ret = __net_mp_open_rxq(dev, rxq_idx, p, extack);
+out:
+	netif_put_rx_queue_lease_locked(orig_dev, dev);
+	return ret;
+}
+
+static void __net_mp_close_rxq(struct net_device *dev, unsigned int rxq_idx,
+			       const struct pp_memory_provider_params *old_p)
 {
 	struct netdev_queue_config qcfg[2];
 	struct netdev_rx_queue *rxq;
 	int err;
 
-	if (WARN_ON_ONCE(ifq_idx >= dev->real_num_rx_queues))
-		return;
-
-	rxq = __netif_get_rx_queue(dev, ifq_idx);
+	rxq = __netif_get_rx_queue(dev, rxq_idx);
 
 	/* Callers holding a netdev ref may get here after we already
 	 * went thru shutdown via dev_memory_provider_uninstall().
@@ -278,10 +322,24 @@ void net_mp_close_rxq(struct net_device *dev, unsigned int ifq_idx,
 			 rxq->mp_params.mp_priv != old_p->mp_priv))
 		return;
 
-	netdev_queue_config(dev, ifq_idx, &qcfg[0]);
+	netdev_queue_config(dev, rxq_idx, &qcfg[0]);
 	memset(&rxq->mp_params, 0, sizeof(rxq->mp_params));
-	netdev_queue_config(dev, ifq_idx, &qcfg[1]);
+	netdev_queue_config(dev, rxq_idx, &qcfg[1]);
 
-	err = netdev_rx_queue_reconfig(dev, ifq_idx, &qcfg[0], &qcfg[1]);
+	err = netdev_rx_queue_reconfig(dev, rxq_idx, &qcfg[0], &qcfg[1]);
 	WARN_ON(err && err != -ENETDOWN);
 }
+
+void net_mp_close_rxq(struct net_device *dev, unsigned int rxq_idx,
+		      const struct pp_memory_provider_params *old_p)
+{
+	struct net_device *orig_dev = dev;
+
+	if (WARN_ON_ONCE(rxq_idx >= dev->real_num_rx_queues))
+		return;
+	if (WARN_ON_ONCE(!netif_get_rx_queue_lease_locked(&dev, &rxq_idx)))
+		return;
+
+	__net_mp_close_rxq(dev, rxq_idx, old_p);
+	netif_put_rx_queue_lease_locked(orig_dev, dev);
+}
-- 
2.43.0


  parent reply	other threads:[~2026-01-29 22:28 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-29 22:28 [PATCH net-next v8 00/16] netkit: Support for io_uring zero-copy and AF_XDP Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 01/16] net: Add queue-create operation Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 02/16] net: Implement netdev_nl_queue_create_doit Daniel Borkmann
2026-01-31 23:45   ` Jakub Kicinski
2026-03-05  4:38     ` Daniel Borkmann
2026-03-06  2:10       ` Jakub Kicinski
2026-03-06  5:49         ` Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 03/16] net: Add lease info to queue-get response Daniel Borkmann
2026-01-31 23:43   ` Jakub Kicinski
2026-02-01  0:15   ` Jakub Kicinski
2026-01-29 22:28 ` [PATCH net-next v8 04/16] net, ethtool: Disallow leased real rxqs to be resized Daniel Borkmann
2026-01-31 23:45   ` Jakub Kicinski
2026-01-29 22:28 ` [PATCH net-next v8 05/16] net: Slightly simplify net_mp_{open,close}_rxq Daniel Borkmann
2026-01-31 23:48   ` Jakub Kicinski
2026-01-29 22:28 ` Daniel Borkmann [this message]
2026-02-01  0:02   ` [PATCH net-next v8 06/16] net: Proxy net_mp_{open,close}_rxq for leased queues Jakub Kicinski
2026-02-01 22:09     ` David Wei
2026-01-29 22:28 ` [PATCH net-next v8 07/16] net: Proxy netdev_queue_get_dma_dev " Daniel Borkmann
2026-02-01  0:04   ` Jakub Kicinski
2026-02-01 22:23     ` David Wei
2026-01-29 22:28 ` [PATCH net-next v8 08/16] xsk: Extend xsk_rcv_check validation Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 09/16] xsk: Proxy pool management for leased queues Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 10/16] netkit: Add single device mode for netkit Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 11/16] netkit: Implement rtnl_link_ops->alloc and ndo_queue_create Daniel Borkmann
2026-02-01  0:19   ` Jakub Kicinski
2026-02-01 22:27     ` David Wei
2026-01-29 22:28 ` [PATCH net-next v8 12/16] netkit: Add netkit notifier to check for unregistering devices Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 13/16] netkit: Add xsk support for af_xdp applications Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 14/16] selftests/net: Add bpf skb forwarding program Daniel Borkmann
2026-01-29 22:28 ` [PATCH net-next v8 15/16] selftests/net: Add env for container based tests Daniel Borkmann
2026-02-01  0:38   ` Jakub Kicinski
2026-02-01 22:53     ` David Wei
2026-02-02 18:41       ` Jakub Kicinski
2026-02-10  0:25         ` David Wei
2026-02-05  2:08       ` Bobby Eshleman
2026-02-05  2:34         ` Bobby Eshleman
2026-02-10 17:30           ` David Wei
2026-01-29 22:28 ` [PATCH net-next v8 16/16] selftests/net: Add netkit container tests Daniel Borkmann
2026-02-01  0:24   ` Jakub Kicinski
2026-02-01 22:54     ` David Wei
2026-02-05  1:44   ` Bobby Eshleman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260129222830.439687-7-daniel@iogearbox.net \
    --to=daniel@iogearbox.net \
    --cc=bpf@vger.kernel.org \
    --cc=davem@davemloft.net \
    --cc=dw@davidwei.uk \
    --cc=john.fastabend@gmail.com \
    --cc=jordan@jrife.io \
    --cc=kuba@kernel.org \
    --cc=maciej.fijalkowski@intel.com \
    --cc=magnus.karlsson@intel.com \
    --cc=martin.lau@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=razor@blackwall.org \
    --cc=sdf@fomichev.me \
    --cc=toke@redhat.com \
    --cc=wangdongdong.6@bytedance.com \
    --cc=willemb@google.com \
    --cc=yangzhenze@bytedance.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox