Merge branch 'mlx5-misc-fixes-2024-11-07'

Tariq Toukan says:

====================
mlx5 misc fixes 2024-11-07

This patchset provides misc bug fixes from the team to the mlx5 core and
Eth drivers.
====================

Link: https://patch.msgid.link/20241107183527.676877-1-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Author: kuba-moo

drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c

@@ -866,7 +866,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv,
 	return 0;
 
 err_rule:
-	mlx5_tc_ct_entry_destroy_mod_hdr(ct_priv, zone_rule->attr, zone_rule->mh);
+	mlx5_tc_ct_entry_destroy_mod_hdr(ct_priv, attr, zone_rule->mh);
 	mlx5_put_label_mapping(ct_priv, attr->ct_attr.ct_labels_id);
 err_mod_hdr:
 	kfree(attr);

drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c

@@ -660,7 +660,7 @@ tx_sync_info_get(struct mlx5e_ktls_offload_context_tx *priv_tx,
 	while (remaining > 0) {
 		skb_frag_t *frag = &record->frags[i];
 
-		get_page(skb_frag_page(frag));
+		page_ref_inc(skb_frag_page(frag));
 		remaining -= skb_frag_size(frag);
 		info->frags[i++] = *frag;
 	}
@@ -763,7 +763,7 @@ void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq,
 	stats = sq->stats;
 
 	mlx5e_tx_dma_unmap(sq->pdev, dma);
-	put_page(wi->resync_dump_frag_page);
+	page_ref_dec(wi->resync_dump_frag_page);
 	stats->tls_dump_packets++;
 	stats->tls_dump_bytes += wi->num_bytes;
 }
@@ -816,12 +816,12 @@ mlx5e_ktls_tx_handle_ooo(struct mlx5e_ktls_offload_context_tx *priv_tx,
 
 err_out:
 	for (; i < info.nr_frags; i++)
-		/* The put_page() here undoes the page ref obtained in tx_sync_info_get().
+		/* The page_ref_dec() here undoes the page ref obtained in tx_sync_info_get().
 		 * Page refs obtained for the DUMP WQEs above (by page_ref_add) will be
 		 * released only upon their completions (or in mlx5e_free_txqsq_descs,
 		 * if channel closes).
 		 */
-		put_page(skb_frag_page(&info.frags[i]));
+		page_ref_dec(skb_frag_page(&info.frags[i]));
 
 	return MLX5E_KTLS_SYNC_FAIL;
 }

drivers/net/ethernet/mellanox/mlx5/core/en_main.c

@@ -4267,7 +4267,8 @@ void mlx5e_set_xdp_feature(struct net_device *netdev)
 	struct mlx5e_params *params = &priv->channels.params;
 	xdp_features_t val;
 
-	if (params->packet_merge.type != MLX5E_PACKET_MERGE_NONE) {
+	if (!netdev->netdev_ops->ndo_bpf ||
+	    params->packet_merge.type != MLX5E_PACKET_MERGE_NONE) {
 		xdp_clear_features_flag(netdev);
 		return;
 	}

drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c

@@ -36,6 +36,7 @@
 #include "en.h"
 #include "en/port.h"
 #include "eswitch.h"
+#include "lib/mlx5.h"
 
 static int mlx5e_test_health_info(struct mlx5e_priv *priv)
 {
@@ -247,6 +248,9 @@ static int mlx5e_cond_loopback(struct mlx5e_priv *priv)
 	if (is_mdev_switchdev_mode(priv->mdev))
 		return -EOPNOTSUPP;
 
+	if (mlx5_get_sd(priv->mdev))
+		return -EOPNOTSUPP;
+
 	return 0;
 }
 

drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c

@@ -2527,8 +2527,11 @@ static void __esw_offloads_unload_rep(struct mlx5_eswitch *esw,
 				      struct mlx5_eswitch_rep *rep, u8 rep_type)
 {
 	if (atomic_cmpxchg(&rep->rep_data[rep_type].state,
-			   REP_LOADED, REP_REGISTERED) == REP_LOADED)
+			   REP_LOADED, REP_REGISTERED) == REP_LOADED) {
+		if (rep_type == REP_ETH)
+			__esw_offloads_unload_rep(esw, rep, REP_IB);
 		esw->offloads.rep_ops[rep_type]->unload(rep);
+	}
 }
 
 static void __unload_reps_all_vport(struct mlx5_eswitch *esw, u8 rep_type)

drivers/net/ethernet/mellanox/mlx5/core/fs_core.c

@@ -2105,13 +2105,22 @@ lookup_fte_locked(struct mlx5_flow_group *g,
 		fte_tmp = NULL;
 		goto out;
 	}
+
+	nested_down_write_ref_node(&fte_tmp->node, FS_LOCK_CHILD);
+
 	if (!fte_tmp->node.active) {
+		up_write_ref_node(&fte_tmp->node, false);
+
+		if (take_write)
+			up_write_ref_node(&g->node, false);
+		else
+			up_read_ref_node(&g->node);
+
 		tree_put_node(&fte_tmp->node, false);
-		fte_tmp = NULL;
-		goto out;
+
+		return NULL;
 	}
 
-	nested_down_write_ref_node(&fte_tmp->node, FS_LOCK_CHILD);
 out:
 	if (take_write)
 		up_write_ref_node(&g->node, false);

drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c

@@ -593,9 +593,11 @@ static void irq_pool_free(struct mlx5_irq_pool *pool)
 	kvfree(pool);
 }
 
-static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pcif_vec)
+static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pcif_vec,
+			  bool dynamic_vec)
 {
 	struct mlx5_irq_table *table = dev->priv.irq_table;
+	int sf_vec_available = sf_vec;
 	int num_sf_ctrl;
 	int err;
 
@@ -616,6 +618,13 @@ static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pcif_vec)
 	num_sf_ctrl = DIV_ROUND_UP(mlx5_sf_max_functions(dev),
 				   MLX5_SFS_PER_CTRL_IRQ);
 	num_sf_ctrl = min_t(int, MLX5_IRQ_CTRL_SF_MAX, num_sf_ctrl);
+	if (!dynamic_vec && (num_sf_ctrl + 1) > sf_vec_available) {
+		mlx5_core_dbg(dev,
+			      "Not enough IRQs for SFs control and completion pool, required=%d avail=%d\n",
+			      num_sf_ctrl + 1, sf_vec_available);
+		return 0;
+	}
+
 	table->sf_ctrl_pool = irq_pool_alloc(dev, pcif_vec, num_sf_ctrl,
 					     "mlx5_sf_ctrl",
 					     MLX5_EQ_SHARE_IRQ_MIN_CTRL,
@@ -624,9 +633,11 @@ static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pcif_vec)
 		err = PTR_ERR(table->sf_ctrl_pool);
 		goto err_pf;
 	}
-	/* init sf_comp_pool */
+	sf_vec_available -= num_sf_ctrl;
+
+	/* init sf_comp_pool, remaining vectors are for the SF completions */
 	table->sf_comp_pool = irq_pool_alloc(dev, pcif_vec + num_sf_ctrl,
-					     sf_vec - num_sf_ctrl, "mlx5_sf_comp",
+					     sf_vec_available, "mlx5_sf_comp",
 					     MLX5_EQ_SHARE_IRQ_MIN_COMP,
 					     MLX5_EQ_SHARE_IRQ_MAX_COMP);
 	if (IS_ERR(table->sf_comp_pool)) {
@@ -715,6 +726,7 @@ int mlx5_irq_table_get_num_comp(struct mlx5_irq_table *table)
 int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 {
 	int num_eqs = mlx5_max_eq_cap_get(dev);
+	bool dynamic_vec;
 	int total_vec;
 	int pcif_vec;
 	int req_vec;
@@ -724,21 +736,31 @@ int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 	if (mlx5_core_is_sf(dev))
 		return 0;
 
+	/* PCI PF vectors usage is limited by online cpus, device EQs and
+	 * PCI MSI-X capability.
+	 */
 	pcif_vec = MLX5_CAP_GEN(dev, num_ports) * num_online_cpus() + 1;
 	pcif_vec = min_t(int, pcif_vec, num_eqs);
+	pcif_vec = min_t(int, pcif_vec, pci_msix_vec_count(dev->pdev));
 
 	total_vec = pcif_vec;
 	if (mlx5_sf_max_functions(dev))
 		total_vec += MLX5_MAX_MSIX_PER_SF * mlx5_sf_max_functions(dev);
 	total_vec = min_t(int, total_vec, pci_msix_vec_count(dev->pdev));
-	pcif_vec = min_t(int, pcif_vec, pci_msix_vec_count(dev->pdev));
 
 	req_vec = pci_msix_can_alloc_dyn(dev->pdev) ? 1 : total_vec;
 	n = pci_alloc_irq_vectors(dev->pdev, 1, req_vec, PCI_IRQ_MSIX);
 	if (n < 0)
 		return n;
 
-	err = irq_pools_init(dev, total_vec - pcif_vec, pcif_vec);
+	/* Further limit vectors of the pools based on platform for non dynamic case */
+	dynamic_vec = pci_msix_can_alloc_dyn(dev->pdev);
+	if (!dynamic_vec) {
+		pcif_vec = min_t(int, n, pcif_vec);
+		total_vec = min_t(int, n, total_vec);
+	}
+
+	err = irq_pools_init(dev, total_vec - pcif_vec, pcif_vec, dynamic_vec);
 	if (err)
 		pci_free_irq_vectors(dev->pdev);