linux-kernel - [PATCH 5.13 047/127] vdpa/mlx5: Fix queue type selection logic

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: <20210824165607.709387-48-sashal@kernel.org>
Date:   Tue, 24 Aug 2021 12:54:47 -0400
From:   Sasha Levin <sashal@...nel.org>
To:     linux-kernel@...r.kernel.org, stable@...r.kernel.org
Cc:     Eli Cohen <elic@...dia.com>,
        "Michael S . Tsirkin" <mst@...hat.com>,
        Sasha Levin <sashal@...nel.org>
Subject: [PATCH 5.13 047/127] vdpa/mlx5: Fix queue type selection logic

From: Eli Cohen <elic@...dia.com>

[ Upstream commit 879753c816dbbdb2a9a395aa4448d29feee92d1a ]

get_queue_type() comments that splict virtqueue is preferred, however,
the actual logic preferred packed virtqueues. Since firmware has not
supported packed virtqueues we ended up using split virtqueues as was
desired.

Since we do not advertise support for packed virtqueues, we add a check
to verify split virtqueues are indeed supported.

Fixes: 1a86b377aa21 ("vdpa/mlx5: Add VDPA driver for supported mlx5 devices")
Signed-off-by: Eli Cohen <elic@...dia.com>
Link: https://lore.kernel.org/r/20210811053759.66752-1-elic@nvidia.com
Signed-off-by: Michael S. Tsirkin <mst@...hat.com>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
 drivers/vdpa/mlx5/net/mlx5_vnet.c  | 14 ++++++++++----
 include/linux/mlx5/mlx5_ifc_vdpa.h | 10 ++++++----
 2 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
index f3495386698a..103d8f70df8e 100644
--- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
@@ -752,12 +752,12 @@ static int get_queue_type(struct mlx5_vdpa_net *ndev)
 	type_mask = MLX5_CAP_DEV_VDPA_EMULATION(ndev->mvdev.mdev, virtio_queue_type);
 
 	/* prefer split queue */
-	if (type_mask & MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_PACKED)
-		return MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_PACKED;
+	if (type_mask & MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_SPLIT)
+		return MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_SPLIT;
 
-	WARN_ON(!(type_mask & MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_SPLIT));
+	WARN_ON(!(type_mask & MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_PACKED));
 
-	return MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_SPLIT;
+	return MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_PACKED;
 }
 
 static bool vq_is_tx(u16 idx)
@@ -2010,6 +2010,12 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name)
 		return -ENOSPC;
 
 	mdev = mgtdev->madev->mdev;
+	if (!(MLX5_CAP_DEV_VDPA_EMULATION(mdev, virtio_queue_type) &
+	    MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_SPLIT)) {
+		dev_warn(mdev->device, "missing support for split virtqueues\n");
+		return -EOPNOTSUPP;
+	}
+
 	/* we save one virtqueue for control virtqueue should we require it */
 	max_vqs = MLX5_CAP_DEV_VDPA_EMULATION(mdev, max_num_virtio_queues);
 	max_vqs = min_t(u32, max_vqs, MLX5_MAX_SUPPORTED_VQS);
diff --git a/include/linux/mlx5/mlx5_ifc_vdpa.h b/include/linux/mlx5/mlx5_ifc_vdpa.h
index 98b56b75c625..1a9c9d94cb59 100644
--- a/include/linux/mlx5/mlx5_ifc_vdpa.h
+++ b/include/linux/mlx5/mlx5_ifc_vdpa.h
@@ -11,13 +11,15 @@ enum {
 };
 
 enum {
-	MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_SPLIT   = 0x1, // do I check this caps?
-	MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_PACKED  = 0x2,
+	MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_SPLIT   = 0,
+	MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_PACKED  = 1,
 };
 
 enum {
-	MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_SPLIT   = 0,
-	MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_PACKED  = 1,
+	MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_SPLIT =
+		BIT(MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_SPLIT),
+	MLX5_VIRTIO_EMULATION_CAP_VIRTIO_QUEUE_TYPE_PACKED =
+		BIT(MLX5_VIRTIO_EMULATION_VIRTIO_QUEUE_TYPE_PACKED),
 };
 
 struct mlx5_ifc_virtio_q_bits {
-- 
2.30.2