Merge branch 'net-restore-the-structure-of-driver-facing-qcfg-api'

Jakub Kicinski says:

====================
net: restore the structure of driver-facing qcfg API

The goal of qcfg objects is to let us seamlessly support new use cases
without modifying all the drivers. We want to pull all the logic of
combining configuration supplied via different interfaces into the core
and present the drivers with a flat queue-by-queue configuration.
Additionally we want to separate the current effective configuration
from the user intent (default vs user setting vs memory provider setting).

Restructure the recently added code to re-introduce the pieces that
are missing compared to the old RFC:
https://lore.kernel.org/20250421222827.283737-1-kuba@kernel.org
Namely:
 - the netdev_queue_config() helper
 - queue config validation callback

I hopefully removed all the more "out there" parts of the RFC.
====================

Link: https://patch.msgid.link/20260122005113.2476634-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

7 files changed, 152 insertions, 49 deletions

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 0e0c88c122f8..8fc0720c3057 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4326,12 +4326,12 @@ static void bnxt_init_ring_struct(struct bnxt *bp)
 
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
+		struct netdev_queue_config qcfg;
 		struct bnxt_ring_mem_info *rmem;
 		struct bnxt_cp_ring_info *cpr;
 		struct bnxt_rx_ring_info *rxr;
 		struct bnxt_tx_ring_info *txr;
 		struct bnxt_ring_struct *ring;
-		struct netdev_rx_queue *rxq;
 
 		if (!bnapi)
 			continue;
@@ -4349,8 +4349,8 @@ static void bnxt_init_ring_struct(struct bnxt *bp)
 		if (!rxr)
 			goto skip_rx;
 
-		rxq = __netif_get_rx_queue(bp->dev, i);
-		rxr->rx_page_size = rxq->qcfg.rx_page_size;
+		netdev_queue_config(bp->dev, i, &qcfg);
+		rxr->rx_page_size = qcfg.rx_page_size;
 
 		ring = &rxr->rx_ring_struct;
 		rmem = &ring->ring_mem;
@@ -15983,8 +15983,12 @@ static void bnxt_queue_default_qcfg(struct net_device *dev,
 	qcfg->rx_page_size = BNXT_RX_PAGE_SIZE;
 }
 
-static int bnxt_validate_qcfg(struct bnxt *bp, struct netdev_queue_config *qcfg)
+static int bnxt_validate_qcfg(struct net_device *dev,
+			      struct netdev_queue_config *qcfg,
+			      struct netlink_ext_ack *extack)
 {
+	struct bnxt *bp = netdev_priv(dev);
+
 	/* Older chips need MSS calc so rx_page_size is not supported */
 	if (!(bp->flags & BNXT_FLAG_CHIP_P5_PLUS) &&
 	    qcfg->rx_page_size != BNXT_RX_PAGE_SIZE)
@@ -16012,10 +16016,6 @@ static int bnxt_queue_mem_alloc(struct net_device *dev,
 	if (!bp->rx_ring)
 		return -ENETDOWN;
 
-	rc = bnxt_validate_qcfg(bp, qcfg);
-	if (rc < 0)
-		return rc;
-
 	rxr = &bp->rx_ring[idx];
 	clone = qmem;
 	memcpy(clone, rxr, sizeof(*rxr));
@@ -16311,9 +16311,13 @@ static const struct netdev_queue_mgmt_ops bnxt_queue_mgmt_ops = {
 	.ndo_queue_start	= bnxt_queue_start,
 	.ndo_queue_stop		= bnxt_queue_stop,
 	.ndo_default_qcfg	= bnxt_queue_default_qcfg,
+	.ndo_validate_qcfg	= bnxt_validate_qcfg,
 	.supported_params	= QCFG_RX_PAGE_SIZE,
 };
 
+static const struct netdev_queue_mgmt_ops bnxt_queue_mgmt_ops_unsupp = {
+};
+
 static void bnxt_remove_one(struct pci_dev *pdev)
 {
 	struct net_device *dev = pci_get_drvdata(pdev);
@@ -16966,9 +16970,10 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	if (BNXT_SUPPORTS_NTUPLE_VNIC(bp))
 		bp->rss_cap |= BNXT_RSS_CAP_MULTI_RSS_CTX;
+
+	dev->queue_mgmt_ops = &bnxt_queue_mgmt_ops_unsupp;
 	if (BNXT_SUPPORTS_QUEUE_API(bp))
 		dev->queue_mgmt_ops = &bnxt_queue_mgmt_ops;
-	dev->request_ops_lock = true;
 	dev->netmem_tx = true;
 
 	rc = register_netdev(dev);
diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h
index 2ab3eae8e8c3..95ed28212f4e 100644
--- a/include/net/netdev_queues.h
+++ b/include/net/netdev_queues.h
@@ -139,7 +139,16 @@ enum {
  * @ndo_queue_get_dma_dev: Get dma device for zero-copy operations to be used
  *			   for this queue. Return NULL on error.
  *
- * @ndo_default_qcfg:	Populate queue config struct with defaults. Optional.
+ * @ndo_default_qcfg:	(Optional) Populate queue config struct with defaults.
+ *			Queue config structs are passed to this helper before
+ *			the user-requested settings are applied.
+ *
+ * @ndo_validate_qcfg: (Optional) Check if queue config is supported.
+ *			Called when configuration affecting a queue may be
+ *			changing, either due to NIC-wide config, or config
+ *			scoped to the queue at a specified index.
+ *			When NIC-wide config is changed the callback will
+ *			be invoked for all queues.
  *
  * @supported_params:	Bitmask of supported parameters, see QCFG_*.
  *
@@ -164,12 +173,18 @@ struct netdev_queue_mgmt_ops {
 				  int idx);
 	void	(*ndo_default_qcfg)(struct net_device *dev,
 				    struct netdev_queue_config *qcfg);
+	int	(*ndo_validate_qcfg)(struct net_device *dev,
+				     struct netdev_queue_config *qcfg,
+				     struct netlink_ext_ack *extack);
 	struct device *	(*ndo_queue_get_dma_dev)(struct net_device *dev,
 						 int idx);
 
 	unsigned int supported_params;
 };
 
+void netdev_queue_config(struct net_device *dev, int rxq,
+			 struct netdev_queue_config *qcfg);
+
 bool netif_rxq_has_unreadable_mp(struct net_device *dev, int idx);
 
 /**
diff --git a/net/core/Makefile b/net/core/Makefile
index 9ef2099c5426..d643a5a7fd18 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_NETDEV_ADDR_LIST_TEST) += dev_addr_lists_test.o
 
 obj-y += net-sysfs.o
 obj-y += hotdata.o
+obj-y += netdev_config.o
 obj-y += netdev_rx_queue.o
 obj-y += netdev_queues.o
 obj-$(CONFIG_PAGE_POOL) += page_pool.o page_pool_user.o
diff --git a/net/core/dev.c b/net/core/dev.c
index ec0d0cdfc078..43de5af0d6ec 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -11282,21 +11282,6 @@ static void netdev_free_phy_link_topology(struct net_device *dev)
 	}
 }
 
-static void init_rx_queue_cfgs(struct net_device *dev)
-{
-	const struct netdev_queue_mgmt_ops *qops = dev->queue_mgmt_ops;
-	struct netdev_rx_queue *rxq;
-	int i;
-
-	if (!qops || !qops->ndo_default_qcfg)
-		return;
-
-	for (i = 0; i < dev->num_rx_queues; i++) {
-		rxq = __netif_get_rx_queue(dev, i);
-		qops->ndo_default_qcfg(dev, &rxq->qcfg);
-	}
-}
-
 /**
  * register_netdevice() - register a network device
  * @dev: device to register
@@ -11342,8 +11327,6 @@ int register_netdevice(struct net_device *dev)
 	if (!dev->name_node)
 		goto out;
 
-	init_rx_queue_cfgs(dev);
-
 	/* Init, if this function is available */
 	if (dev->netdev_ops->ndo_init) {
 		ret = dev->netdev_ops->ndo_init(dev);
diff --git a/net/core/dev.h b/net/core/dev.h
index da18536cbd35..98793a738f43 100644
--- a/net/core/dev.h
+++ b/net/core/dev.h
@@ -10,6 +10,7 @@
 
 struct net;
 struct netlink_ext_ack;
+struct netdev_queue_config;
 struct cpumask;
 
 /* Random bits of netdevice that don't need to be exposed */
@@ -91,6 +92,10 @@ extern struct rw_semaphore dev_addr_sem;
 extern struct list_head net_todo_list;
 void netdev_run_todo(void);
 
+int netdev_queue_config_validate(struct net_device *dev, int rxq_idx,
+				 struct netdev_queue_config *qcfg,
+				 struct netlink_ext_ack *extack);
+
 /* netdev management, shared between various uAPI entry points */
 struct netdev_name_node {
 	struct hlist_node hlist;
diff --git a/net/core/netdev_config.c b/net/core/netdev_config.c
new file mode 100644
index 000000000000..f14af365d5cd
--- /dev/null
+++ b/net/core/netdev_config.c
@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/netdevice.h>
+#include <net/netdev_queues.h>
+#include <net/netdev_rx_queue.h>
+
+#include "dev.h"
+
+static int netdev_nop_validate_qcfg(struct net_device *dev,
+				    struct netdev_queue_config *qcfg,
+				    struct netlink_ext_ack *extack)
+{
+	return 0;
+}
+
+static int __netdev_queue_config(struct net_device *dev, int rxq_idx,
+				 struct netdev_queue_config *qcfg,
+				 struct netlink_ext_ack *extack,
+				 bool validate)
+{
+	int (*validate_cb)(struct net_device *dev,
+			   struct netdev_queue_config *qcfg,
+			   struct netlink_ext_ack *extack);
+	struct pp_memory_provider_params *mpp;
+	int err;
+
+	validate_cb = netdev_nop_validate_qcfg;
+	if (validate && dev->queue_mgmt_ops->ndo_validate_qcfg)
+		validate_cb = dev->queue_mgmt_ops->ndo_validate_qcfg;
+
+	memset(qcfg, 0, sizeof(*qcfg));
+
+	/* Get defaults from the driver, in case user config not set */
+	if (dev->queue_mgmt_ops->ndo_default_qcfg)
+		dev->queue_mgmt_ops->ndo_default_qcfg(dev, qcfg);
+	err = validate_cb(dev, qcfg, extack);
+	if (err)
+		return err;
+
+	/* Apply MP overrides */
+	mpp = &__netif_get_rx_queue(dev, rxq_idx)->mp_params;
+	if (mpp->rx_page_size)
+		qcfg->rx_page_size = mpp->rx_page_size;
+	err = validate_cb(dev, qcfg, extack);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+/**
+ * netdev_queue_config() - get configuration for a given queue
+ * @dev:      net_device instance
+ * @rxq_idx:  index of the queue of interest
+ * @qcfg: queue configuration struct (output)
+ *
+ * Render the configuration for a given queue. This helper should be used
+ * by drivers which support queue configuration to retrieve config for
+ * a particular queue.
+ *
+ * @qcfg is an output parameter and is always fully initialized by this
+ * function. Some values may not be set by the user, drivers may either
+ * deal with the "unset" values in @qcfg, or provide the callback
+ * to populate defaults in queue_management_ops.
+ */
+void netdev_queue_config(struct net_device *dev, int rxq_idx,
+			 struct netdev_queue_config *qcfg)
+{
+	__netdev_queue_config(dev, rxq_idx, qcfg, NULL, false);
+}
+EXPORT_SYMBOL(netdev_queue_config);
+
+int netdev_queue_config_validate(struct net_device *dev, int rxq_idx,
+				 struct netdev_queue_config *qcfg,
+				 struct netlink_ext_ack *extack)
+{
+	return __netdev_queue_config(dev, rxq_idx, qcfg, extack, true);
+}
diff --git a/net/core/netdev_rx_queue.c b/net/core/netdev_rx_queue.c
index b81cad90ba2f..668a90658f25 100644
--- a/net/core/netdev_rx_queue.c
+++ b/net/core/netdev_rx_queue.c
@@ -7,6 +7,7 @@
 #include <net/netdev_rx_queue.h>
 #include <net/page_pool/memory_provider.h>
 
+#include "dev.h"
 #include "page_pool_priv.h"
 
 /* See also page_pool_is_unreadable() */
@@ -18,11 +19,13 @@ bool netif_rxq_has_unreadable_mp(struct net_device *dev, int idx)
 }
 EXPORT_SYMBOL(netif_rxq_has_unreadable_mp);
 
-int netdev_rx_queue_restart(struct net_device *dev, unsigned int rxq_idx)
+static int netdev_rx_queue_reconfig(struct net_device *dev,
+				    unsigned int rxq_idx,
+				    struct netdev_queue_config *qcfg_old,
+				    struct netdev_queue_config *qcfg_new)
 {
 	struct netdev_rx_queue *rxq = __netif_get_rx_queue(dev, rxq_idx);
 	const struct netdev_queue_mgmt_ops *qops = dev->queue_mgmt_ops;
-	struct netdev_queue_config qcfg;
 	void *new_mem, *old_mem;
 	int err;
 
@@ -30,21 +33,8 @@ int netdev_rx_queue_restart(struct net_device *dev, unsigned int rxq_idx)
 	    !qops->ndo_queue_mem_alloc || !qops->ndo_queue_start)
 		return -EOPNOTSUPP;
 
-	if (WARN_ON_ONCE(qops->supported_params && !qops->ndo_default_qcfg))
-		return -EINVAL;
-
 	netdev_assert_locked(dev);
 
-	memset(&qcfg, 0, sizeof(qcfg));
-	if (qops->ndo_default_qcfg)
-		qops->ndo_default_qcfg(dev, &qcfg);
-
-	if (rxq->mp_params.rx_page_size) {
-		if (!(qops->supported_params & QCFG_RX_PAGE_SIZE))
-			return -EOPNOTSUPP;
-		qcfg.rx_page_size = rxq->mp_params.rx_page_size;
-	}
-
 	new_mem = kvzalloc(qops->ndo_queue_mem_size, GFP_KERNEL);
 	if (!new_mem)
 		return -ENOMEM;
@@ -55,7 +45,7 @@ int netdev_rx_queue_restart(struct net_device *dev, unsigned int rxq_idx)
 		goto err_free_new_mem;
 	}
 
-	err = qops->ndo_queue_mem_alloc(dev, &qcfg, new_mem, rxq_idx);
+	err = qops->ndo_queue_mem_alloc(dev, qcfg_new, new_mem, rxq_idx);
 	if (err)
 		goto err_free_old_mem;
 
@@ -68,7 +58,7 @@ int netdev_rx_queue_restart(struct net_device *dev, unsigned int rxq_idx)
 		if (err)
 			goto err_free_new_queue_mem;
 
-		err = qops->ndo_queue_start(dev, &qcfg, new_mem, rxq_idx);
+		err = qops->ndo_queue_start(dev, qcfg_new, new_mem, rxq_idx);
 		if (err)
 			goto err_start_queue;
 	} else {
@@ -80,7 +70,6 @@ int netdev_rx_queue_restart(struct net_device *dev, unsigned int rxq_idx)
 	kvfree(old_mem);
 	kvfree(new_mem);
 
-	rxq->qcfg = qcfg;
 	return 0;
 
 err_start_queue:
@@ -91,7 +80,7 @@ err_start_queue:
 	 * WARN if we fail to recover the old rx queue, and at least free
 	 * old_mem so we don't also leak that.
 	 */
-	if (qops->ndo_queue_start(dev, &rxq->qcfg, old_mem, rxq_idx)) {
+	if (qops->ndo_queue_start(dev, qcfg_old, old_mem, rxq_idx)) {
 		WARN(1,
 		     "Failed to restart old queue in error path. RX queue %d may be unhealthy.",
 		     rxq_idx);
@@ -109,12 +98,22 @@ err_free_new_mem:
 
 	return err;
 }
+
+int netdev_rx_queue_restart(struct net_device *dev, unsigned int rxq_idx)
+{
+	struct netdev_queue_config qcfg;
+
+	netdev_queue_config(dev, rxq_idx, &qcfg);
+	return netdev_rx_queue_reconfig(dev, rxq_idx, &qcfg, &qcfg);
+}
 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)
 {
+	const struct netdev_queue_mgmt_ops *qops = dev->queue_mgmt_ops;
+	struct netdev_queue_config qcfg[2];
 	struct netdev_rx_queue *rxq;
 	int ret;
 
@@ -139,6 +138,10 @@ int __net_mp_open_rxq(struct net_device *dev, unsigned int rxq_idx,
 		NL_SET_ERR_MSG(extack, "unable to custom memory provider to device with XDP program attached");
 		return -EEXIST;
 	}
+	if (p->rx_page_size && !(qops->supported_params & QCFG_RX_PAGE_SIZE)) {
+		NL_SET_ERR_MSG(extack, "device does not support: rx_page_size");
+		return -EOPNOTSUPP;
+	}
 
 	rxq = __netif_get_rx_queue(dev, rxq_idx);
 	if (rxq->mp_params.mp_ops) {
@@ -152,11 +155,20 @@ int __net_mp_open_rxq(struct net_device *dev, unsigned int rxq_idx,
 	}
 #endif
 
+	netdev_queue_config(dev, rxq_idx, &qcfg[0]);
 	rxq->mp_params = *p;
-	ret = netdev_rx_queue_restart(dev, rxq_idx);
+	ret = netdev_queue_config_validate(dev, rxq_idx, &qcfg[1], extack);
 	if (ret)
-		memset(&rxq->mp_params, 0, sizeof(rxq->mp_params));
+		goto err_clear_mp;
 
+	ret = netdev_rx_queue_reconfig(dev, rxq_idx, &qcfg[0], &qcfg[1]);
+	if (ret)
+		goto err_clear_mp;
+
+	return 0;
+
+err_clear_mp:
+	memset(&rxq->mp_params, 0, sizeof(rxq->mp_params));
 	return ret;
 }
 
@@ -174,6 +186,7 @@ int net_mp_open_rxq(struct net_device *dev, unsigned int rxq_idx,
 void __net_mp_close_rxq(struct net_device *dev, unsigned int ifq_idx,
 			const struct pp_memory_provider_params *old_p)
 {
+	struct netdev_queue_config qcfg[2];
 	struct netdev_rx_queue *rxq;
 	int err;
 
@@ -193,8 +206,11 @@ 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]);
 	memset(&rxq->mp_params, 0, sizeof(rxq->mp_params));
-	err = netdev_rx_queue_restart(dev, ifq_idx);
+	netdev_queue_config(dev, ifq_idx, &qcfg[1]);
+
+	err = netdev_rx_queue_reconfig(dev, ifq_idx, &qcfg[0], &qcfg[1]);
 	WARN_ON(err && err != -ENETDOWN);
 }