netdev - [net-next 08/15] net/mlx5: Removing rmap per IRQ

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20210615040123.287101-9-saeed@kernel.org>
Date:   Mon, 14 Jun 2021 21:01:16 -0700
From:   Saeed Mahameed <saeed@...nel.org>
To:     "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>
Cc:     netdev@...r.kernel.org, Leon Romanovsky <leonro@...dia.com>,
        Shay Drory <shayd@...dia.com>,
        Saeed Mahameed <saeedm@...dia.com>
Subject: [net-next 08/15] net/mlx5: Removing rmap per IRQ

From: Shay Drory <shayd@...dia.com>

In next patches, IRQs will be requested according to demand, instead of
statically on driver boot.
Also, currently, rmap is managed by the IRQ layer. rmap management will
move out from the IRQ layer in future patches.

Therefore, we want to remove the IRQ from the rmap, when IRQ is destroyed,
instead of removing all the IRQs from the rmap when irq_table is destroyed.

Signed-off-by: Shay Drory <shayd@...dia.com>
Reviewed-by: Leon Romanovsky <leonro@...dia.com>
Signed-off-by: Saeed Mahameed <saeedm@...dia.com>
---
 .../net/ethernet/mellanox/mlx5/core/pci_irq.c   | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index 81b06b5693cd..6a5a6ec0ddbf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -154,8 +154,14 @@ static void irq_release(struct kref *kref)
 {
 	struct mlx5_irq *irq = container_of(kref, struct mlx5_irq, kref);
 
+	/* free_irq requires that affinity and rmap will be cleared
+	 * before calling it. This is why there is asymmetry with set_rmap
+	 * which should be called after alloc_irq but before request_irq.
+	 */
 	irq_set_affinity_hint(irq->irqn, NULL);
 	free_cpumask_var(irq->mask);
+	/* this line is releasing this irq from the rmap */
+	irq_set_affinity_notifier(irq->irqn, NULL);
 	free_irq(irq->irqn, &irq->nh);
 }
 
@@ -378,6 +384,11 @@ int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 	return err;
 }
 
+static void irq_table_clear_rmap(struct mlx5_irq_table *table)
+{
+	cpu_rmap_put(table->rmap);
+}
+
 void mlx5_irq_table_destroy(struct mlx5_core_dev *dev)
 {
 	struct mlx5_irq_table *table = dev->priv.irq_table;
@@ -386,11 +397,7 @@ void mlx5_irq_table_destroy(struct mlx5_core_dev *dev)
 	if (mlx5_core_is_sf(dev))
 		return;
 
-	/* free_irq requires that affinity and rmap will be cleared
-	 * before calling it. This is why there is asymmetry with set_rmap
-	 * which should be called after alloc_irq but before request_irq.
-	 */
-	irq_clear_rmap(dev);
+	irq_table_clear_rmap(table);
 	for (i = 0; i < table->nvec; i++)
 		irq_release(&mlx5_irq_get(dev, i)->kref);
 	pci_free_irq_vectors(dev->pdev);
-- 
2.31.1