All of lore.kernel.org
 help / color / mirror / Atom feed
From: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
To: Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Jason Gunthorpe <jgg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
	Majd Dibbiny <majd-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
	Mark Bloch <markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
	Moni Shoua <monis-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
	Yishai Hadas <yishaih-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>,
	stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: [PATCH rdma-rc 2/4] IB/mlx5: Serialize access to the VMA list
Date: Sun, 24 Dec 2017 13:54:56 +0200	[thread overview]
Message-ID: <20171224115458.27577-3-leon@kernel.org> (raw)
In-Reply-To: <20171224115458.27577-1-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>

From: Majd Dibbiny <majd-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>

User-space applications can do mmap and munmap directly.

Since the VMA list is not protected with a mutex, concurrent
accesses to the VMA list from the mmap and munmap can cause
data corruption.

Cc: <stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> # v4.7
Fixes: 7c2344c3bbf9 ("IB/mlx5: Implements disassociate_ucontext API")
Reviewed-by: Yishai Hadas <yishaih-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Majd Dibbiny <majd-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Signed-off-by: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
 drivers/infiniband/hw/mlx5/main.c    | 8 ++++++++
 drivers/infiniband/hw/mlx5/mlx5_ib.h | 6 ++++++
 2 files changed, 14 insertions(+)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index b4ef4d9b6ce5..8ac50de2b242 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -1463,6 +1463,7 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	}
 
 	INIT_LIST_HEAD(&context->vma_private_list);
+	mutex_init(&context->vma_private_list_mutex);
 	INIT_LIST_HEAD(&context->db_page_list);
 	mutex_init(&context->db_page_mutex);
 
@@ -1624,7 +1625,9 @@ static void  mlx5_ib_vma_close(struct vm_area_struct *area)
 	 * mlx5_ib_disassociate_ucontext().
 	 */
 	mlx5_ib_vma_priv_data->vma = NULL;
+	mutex_lock(mlx5_ib_vma_priv_data->vma_private_list_mutex);
 	list_del(&mlx5_ib_vma_priv_data->list);
+	mutex_unlock(mlx5_ib_vma_priv_data->vma_private_list_mutex);
 	kfree(mlx5_ib_vma_priv_data);
 }
 
@@ -1644,10 +1647,13 @@ static int mlx5_ib_set_vma_data(struct vm_area_struct *vma,
 		return -ENOMEM;
 
 	vma_prv->vma = vma;
+	vma_prv->vma_private_list_mutex = &ctx->vma_private_list_mutex;
 	vma->vm_private_data = vma_prv;
 	vma->vm_ops =  &mlx5_ib_vm_ops;
 
+	mutex_lock(&ctx->vma_private_list_mutex);
 	list_add(&vma_prv->list, vma_head);
+	mutex_unlock(&ctx->vma_private_list_mutex);
 
 	return 0;
 }
@@ -1690,6 +1696,7 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext)
 	 * mlx5_ib_vma_close.
 	 */
 	down_write(&owning_mm->mmap_sem);
+	mutex_lock(&context->vma_private_list_mutex);
 	list_for_each_entry_safe(vma_private, n, &context->vma_private_list,
 				 list) {
 		vma = vma_private->vma;
@@ -1704,6 +1711,7 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext)
 		list_del(&vma_private->list);
 		kfree(vma_private);
 	}
+	mutex_unlock(&context->vma_private_list_mutex);
 	up_write(&owning_mm->mmap_sem);
 	mmput(owning_mm);
 	put_task_struct(owning_process);
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 6dd8cac78de2..dcf30e18409e 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -115,6 +115,9 @@ enum {
 struct mlx5_ib_vma_private_data {
 	struct list_head list;
 	struct vm_area_struct *vma;
+	/* protect vma_private_list add/del
+	 */
+	struct mutex *vma_private_list_mutex;
 };
 
 struct mlx5_ib_ucontext {
@@ -129,6 +132,9 @@ struct mlx5_ib_ucontext {
 	/* Transport Domain number */
 	u32			tdn;
 	struct list_head	vma_private_list;
+	/* protect vma_private_list add/del
+	 */
+	struct mutex		vma_private_list_mutex;
 
 	unsigned long		upd_xlt_page;
 	/* protect ODP/KSM */
-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

WARNING: multiple messages have this Message-ID (diff)
From: Leon Romanovsky <leon@kernel.org>
To: Doug Ledford <dledford@redhat.com>, Jason Gunthorpe <jgg@mellanox.com>
Cc: linux-rdma@vger.kernel.org, Daniel Jurgens <danielj@mellanox.com>,
	Majd Dibbiny <majd@mellanox.com>, Mark Bloch <markb@mellanox.com>,
	Moni Shoua <monis@mellanox.com>,
	Yishai Hadas <yishaih@mellanox.com>,
	stable@vger.kernel.org
Subject: [PATCH rdma-rc 2/4] IB/mlx5: Serialize access to the VMA list
Date: Sun, 24 Dec 2017 13:54:56 +0200	[thread overview]
Message-ID: <20171224115458.27577-3-leon@kernel.org> (raw)
In-Reply-To: <20171224115458.27577-1-leon@kernel.org>

From: Majd Dibbiny <majd@mellanox.com>

User-space applications can do mmap and munmap directly.

Since the VMA list is not protected with a mutex, concurrent
accesses to the VMA list from the mmap and munmap can cause
data corruption.

Cc: <stable@vger.kernel.org> # v4.7
Fixes: 7c2344c3bbf9 ("IB/mlx5: Implements disassociate_ucontext API")
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 drivers/infiniband/hw/mlx5/main.c    | 8 ++++++++
 drivers/infiniband/hw/mlx5/mlx5_ib.h | 6 ++++++
 2 files changed, 14 insertions(+)

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index b4ef4d9b6ce5..8ac50de2b242 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -1463,6 +1463,7 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	}
 
 	INIT_LIST_HEAD(&context->vma_private_list);
+	mutex_init(&context->vma_private_list_mutex);
 	INIT_LIST_HEAD(&context->db_page_list);
 	mutex_init(&context->db_page_mutex);
 
@@ -1624,7 +1625,9 @@ static void  mlx5_ib_vma_close(struct vm_area_struct *area)
 	 * mlx5_ib_disassociate_ucontext().
 	 */
 	mlx5_ib_vma_priv_data->vma = NULL;
+	mutex_lock(mlx5_ib_vma_priv_data->vma_private_list_mutex);
 	list_del(&mlx5_ib_vma_priv_data->list);
+	mutex_unlock(mlx5_ib_vma_priv_data->vma_private_list_mutex);
 	kfree(mlx5_ib_vma_priv_data);
 }
 
@@ -1644,10 +1647,13 @@ static int mlx5_ib_set_vma_data(struct vm_area_struct *vma,
 		return -ENOMEM;
 
 	vma_prv->vma = vma;
+	vma_prv->vma_private_list_mutex = &ctx->vma_private_list_mutex;
 	vma->vm_private_data = vma_prv;
 	vma->vm_ops =  &mlx5_ib_vm_ops;
 
+	mutex_lock(&ctx->vma_private_list_mutex);
 	list_add(&vma_prv->list, vma_head);
+	mutex_unlock(&ctx->vma_private_list_mutex);
 
 	return 0;
 }
@@ -1690,6 +1696,7 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext)
 	 * mlx5_ib_vma_close.
 	 */
 	down_write(&owning_mm->mmap_sem);
+	mutex_lock(&context->vma_private_list_mutex);
 	list_for_each_entry_safe(vma_private, n, &context->vma_private_list,
 				 list) {
 		vma = vma_private->vma;
@@ -1704,6 +1711,7 @@ static void mlx5_ib_disassociate_ucontext(struct ib_ucontext *ibcontext)
 		list_del(&vma_private->list);
 		kfree(vma_private);
 	}
+	mutex_unlock(&context->vma_private_list_mutex);
 	up_write(&owning_mm->mmap_sem);
 	mmput(owning_mm);
 	put_task_struct(owning_process);
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 6dd8cac78de2..dcf30e18409e 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -115,6 +115,9 @@ enum {
 struct mlx5_ib_vma_private_data {
 	struct list_head list;
 	struct vm_area_struct *vma;
+	/* protect vma_private_list add/del
+	 */
+	struct mutex *vma_private_list_mutex;
 };
 
 struct mlx5_ib_ucontext {
@@ -129,6 +132,9 @@ struct mlx5_ib_ucontext {
 	/* Transport Domain number */
 	u32			tdn;
 	struct list_head	vma_private_list;
+	/* protect vma_private_list add/del
+	 */
+	struct mutex		vma_private_list_mutex;
 
 	unsigned long		upd_xlt_page;
 	/* protect ODP/KSM */
-- 
2.15.1

  parent reply	other threads:[~2017-12-24 11:54 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-24 11:54 [PATCH rdma-rc 0/4] RDMA fixes 2017-12-24 Leon Romanovsky
2017-12-24 11:54 ` [PATCH rdma-rc 1/4] RDMA/core: Enforce requirement to hold lists_rwsem semaphore Leon Romanovsky
     [not found]   ` <20171224115458.27577-2-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2017-12-27 21:49     ` Jason Gunthorpe
2017-12-27 21:49       ` Jason Gunthorpe
2017-12-24 11:54 ` [PATCH rdma-rc 3/4] IB/uverbs: Fix command checking as part of ib_uverbs_ex_modify_qp() Leon Romanovsky
     [not found] ` <20171224115458.27577-1-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2017-12-24 11:54   ` Leon Romanovsky [this message]
2017-12-24 11:54     ` [PATCH rdma-rc 2/4] IB/mlx5: Serialize access to the VMA list Leon Romanovsky
2017-12-24 11:54   ` [PATCH rdma-rc 4/4] IB/core: Verify that QP is security enabled in create and destroy Leon Romanovsky
2017-12-24 11:54     ` Leon Romanovsky
2017-12-27 22:32   ` [PATCH rdma-rc 0/4] RDMA fixes 2017-12-24 Jason Gunthorpe

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=20171224115458.27577-3-leon@kernel.org \
    --to=leon-dgejt+ai2ygdnm+yrofe0a@public.gmane.org \
    --cc=danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
    --cc=dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=jgg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
    --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=majd-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
    --cc=markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
    --cc=monis-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
    --cc=stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=yishaih-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.