mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Dragos Tatulea <dtatulea@nvidia.com>
To: "Jason Wang" <jasowang@redhat.com>,
	"Eugenio Pérez" <eperezma@redhat.com>,
	"Si-Wei Liu" <si-wei.liu@oracle.com>,
	"Michael S . Tsirkin" <mst@redhat.com>,
	"Saeed Mahameed" <saeedm@nvidia.com>,
	virtualization@lists.linux-foundation.org,
	linux-kernel@vger.kernel.org
Cc: Dragos Tatulea <dtatulea@nvidia.com>,
	Parav Pandit <parav@nvidia.com>,
	Xuan Zhuo <xuanzhuo@linux.alibaba.com>, <kvm@vger.kernel.org>
Subject: [PATCH 10/16] vdpa/mlx5: Move mr mutex out of mr struct
Date: Tue, 12 Sep 2023 16:01:20 +0300	[thread overview]
Message-ID: <20230912130132.561193-11-dtatulea@nvidia.com> (raw)
In-Reply-To: <20230912130132.561193-1-dtatulea@nvidia.com>

The mutex is named like it is supposed to protect only the mkey but in
reality it is a global lock for all mr resources.

Shift the mutex to it's rightful location (struct mlx5_vdpa_dev) and
give it a more appropriate name.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
---
 drivers/vdpa/mlx5/core/mlx5_vdpa.h |  4 ++--
 drivers/vdpa/mlx5/core/mr.c        | 13 +++++++------
 drivers/vdpa/mlx5/core/resources.c |  6 +++---
 3 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/drivers/vdpa/mlx5/core/mlx5_vdpa.h b/drivers/vdpa/mlx5/core/mlx5_vdpa.h
index 01d4ee58ccb1..9c6ac42c21e1 100644
--- a/drivers/vdpa/mlx5/core/mlx5_vdpa.h
+++ b/drivers/vdpa/mlx5/core/mlx5_vdpa.h
@@ -34,8 +34,6 @@ struct mlx5_vdpa_mr {
 	/* state of dvq mr */
 	bool initialized;
 
-	/* serialize mkey creation and destruction */
-	struct mutex mkey_mtx;
 	bool user_mr;
 };
 
@@ -94,6 +92,8 @@ struct mlx5_vdpa_dev {
 	u32 generation;
 
 	struct mlx5_vdpa_mr mr;
+	/* serialize mr access */
+	struct mutex mr_mtx;
 	struct mlx5_control_vq cvq;
 	struct workqueue_struct *wq;
 	unsigned int group2asid[MLX5_VDPA_NUMVQ_GROUPS];
diff --git a/drivers/vdpa/mlx5/core/mr.c b/drivers/vdpa/mlx5/core/mr.c
index 6f29e8eaabb1..abd6a6fb122f 100644
--- a/drivers/vdpa/mlx5/core/mr.c
+++ b/drivers/vdpa/mlx5/core/mr.c
@@ -509,11 +509,11 @@ static void _mlx5_vdpa_destroy_mr(struct mlx5_vdpa_dev *mvdev, struct mlx5_vdpa_
 void mlx5_vdpa_destroy_mr(struct mlx5_vdpa_dev *mvdev,
 			  struct mlx5_vdpa_mr *mr)
 {
-	mutex_lock(&mr->mkey_mtx);
+	mutex_lock(&mvdev->mr_mtx);
 
 	_mlx5_vdpa_destroy_mr(mvdev, mr);
 
-	mutex_unlock(&mr->mkey_mtx);
+	mutex_unlock(&mvdev->mr_mtx);
 }
 
 void mlx5_vdpa_destroy_mr_resources(struct mlx5_vdpa_dev *mvdev)
@@ -550,9 +550,10 @@ int mlx5_vdpa_create_mr(struct mlx5_vdpa_dev *mvdev,
 {
 	int err;
 
-	mutex_lock(&mvdev->mr.mkey_mtx);
+	mutex_lock(&mvdev->mr_mtx);
 	err = _mlx5_vdpa_create_mr(mvdev, mr, iotlb);
-	mutex_unlock(&mvdev->mr.mkey_mtx);
+	mutex_unlock(&mvdev->mr_mtx);
+
 	return err;
 }
 
@@ -563,14 +564,14 @@ int mlx5_vdpa_handle_set_map(struct mlx5_vdpa_dev *mvdev, struct vhost_iotlb *io
 	int err = 0;
 
 	*change_map = false;
-	mutex_lock(&mr->mkey_mtx);
+	mutex_lock(&mvdev->mr_mtx);
 	if (mr->initialized) {
 		mlx5_vdpa_info(mvdev, "memory map update\n");
 		*change_map = true;
 	}
 	if (!*change_map)
 		err = _mlx5_vdpa_create_mr(mvdev, mr, iotlb);
-	mutex_unlock(&mr->mkey_mtx);
+	mutex_unlock(&mvdev->mr_mtx);
 
 	return err;
 }
diff --git a/drivers/vdpa/mlx5/core/resources.c b/drivers/vdpa/mlx5/core/resources.c
index d5a59c9035fb..5c5a41b64bfc 100644
--- a/drivers/vdpa/mlx5/core/resources.c
+++ b/drivers/vdpa/mlx5/core/resources.c
@@ -256,7 +256,7 @@ int mlx5_vdpa_alloc_resources(struct mlx5_vdpa_dev *mvdev)
 		mlx5_vdpa_warn(mvdev, "resources already allocated\n");
 		return -EINVAL;
 	}
-	mutex_init(&mvdev->mr.mkey_mtx);
+	mutex_init(&mvdev->mr_mtx);
 	res->uar = mlx5_get_uars_page(mdev);
 	if (IS_ERR(res->uar)) {
 		err = PTR_ERR(res->uar);
@@ -301,7 +301,7 @@ int mlx5_vdpa_alloc_resources(struct mlx5_vdpa_dev *mvdev)
 err_uctx:
 	mlx5_put_uars_page(mdev, res->uar);
 err_uars:
-	mutex_destroy(&mvdev->mr.mkey_mtx);
+	mutex_destroy(&mvdev->mr_mtx);
 	return err;
 }
 
@@ -318,6 +318,6 @@ void mlx5_vdpa_free_resources(struct mlx5_vdpa_dev *mvdev)
 	dealloc_pd(mvdev, res->pdn, res->uid);
 	destroy_uctx(mvdev, res->uid);
 	mlx5_put_uars_page(mvdev->mdev, res->uar);
-	mutex_destroy(&mvdev->mr.mkey_mtx);
+	mutex_destroy(&mvdev->mr_mtx);
 	res->valid = false;
 }
-- 
2.41.0


  parent reply	other threads:[~2023-09-12 13:03 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-12 13:01 [PATCH 00/16] vdpa: Add support for vq descriptor mappings Dragos Tatulea
2023-09-12 13:01 ` [PATCH 01/16] vdpa: introduce dedicated descriptor group for virtqueue Dragos Tatulea
2023-09-12 13:01 ` [PATCH 02/16] vhost-vdpa: introduce descriptor group backend feature Dragos Tatulea
2023-09-12 13:01 ` [PATCH 03/16] vhost-vdpa: uAPI to get dedicated descriptor group id Dragos Tatulea
2023-09-12 13:01 ` [PATCH 04/16] vdpa/mlx5: Create helper function for dma mappings Dragos Tatulea
2023-09-26  3:11   ` Jason Wang
2023-09-12 13:01 ` [PATCH 05/16] vdpa/mlx5: Decouple cvq iotlb handling from hw mapping code Dragos Tatulea
2023-09-26  3:28   ` Jason Wang
2023-09-12 13:01 ` [PATCH 06/16] vdpa/mlx5: Take cvq iotlb lock during refresh Dragos Tatulea
2023-09-26  3:33   ` Jason Wang
2023-09-12 13:01 ` [PATCH 07/16] vdpa/mlx5: Collapse "dvq" mr add/delete functions Dragos Tatulea
2023-09-26  3:33   ` Jason Wang
2023-09-12 13:01 ` [PATCH 08/16] vdpa/mlx5: Rename mr destroy functions Dragos Tatulea
2023-09-26  3:34   ` Jason Wang
2023-09-12 13:01 ` [PATCH 09/16] vdpa/mlx5: Allow creation/deletion of any given mr struct Dragos Tatulea
2023-09-26  4:44   ` Jason Wang
2023-09-26  7:21     ` Dragos Tatulea
2023-10-08  4:25       ` Jason Wang
2023-10-08 12:04         ` Dragos Tatulea
2023-10-09  6:39           ` Jason Wang
2023-10-09  7:06             ` Dragos Tatulea
2023-09-12 13:01 ` Dragos Tatulea [this message]
2023-09-12 13:01 ` [PATCH 11/16] vdpa/mlx5: Improve mr update flow Dragos Tatulea
2023-09-12 13:01 ` [PATCH 12/16] vdpa/mlx5: Introduce mr for vq descriptor Dragos Tatulea
2023-09-12 13:01 ` [PATCH 13/16] vdpa/mlx5: Enable hw support for vq descriptor mapping Dragos Tatulea
2023-09-12 13:01 ` [PATCH 14/16] vdpa/mlx5: Make iotlb helper functions more generic Dragos Tatulea
2023-09-12 13:01 ` [PATCH 15/16] vdpa/mlx5: Update cvq iotlb mapping on ASID change Dragos Tatulea
2023-09-13  1:03 ` [PATCH 00/16] vdpa: Add support for vq descriptor mappings Lei Yang
2023-09-13 16:08   ` Eugenio Perez Martin
2023-09-15  6:34     ` Si-Wei Liu
2023-09-20  1:14       ` Lei Yang
2023-09-25  7:59 ` Dragos Tatulea
2023-09-26 12:22   ` Si-Wei Liu
2023-09-26 12:37     ` Dragos Tatulea

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=20230912130132.561193-11-dtatulea@nvidia.com \
    --to=dtatulea@nvidia.com \
    --cc=eperezma@redhat.com \
    --cc=jasowang@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=parav@nvidia.com \
    --cc=saeedm@nvidia.com \
    --cc=si-wei.liu@oracle.com \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=xuanzhuo@linux.alibaba.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

all inboxes | Powered by JetHome®