mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Leon Romanovsky <leon@kernel.org>
To: Jason Gunthorpe <jgg@ziepe.ca>, Leon Romanovsky <leon@kernel.org>,
	Yishai Hadas <yishaih@nvidia.com>,
	Chiara Meiohas <cmeiohas@nvidia.com>,
	Michal Kalderon <mkalderon@marvell.com>,
	Selvin Xavier <selvin.xavier@broadcom.com>
Cc: linux-rdma@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH rdma-next 4/6] RDMA/mlx5: Avoid direct access to DMA device pointer
Date: Sun,  4 Jan 2026 15:51:36 +0200	[thread overview]
Message-ID: <20260104-ib-core-misc-v1-4-00367f77f3a8@nvidia.com> (raw)
In-Reply-To: <20260104-ib-core-misc-v1-0-00367f77f3a8@nvidia.com>

From: Leon Romanovsky <leonro@nvidia.com>

The dma_device field is marked as internal and must not be accessed by
drivers or ULPs. Remove all direct mlx5 references to this field.

Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/infiniband/hw/mlx5/mr.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 325fa04cbe8a..a7b37e3df072 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -1646,10 +1646,13 @@ reg_user_mr_dmabuf(struct ib_pd *pd, struct device *dma_device,
 						 offset, length, fd,
 						 access_flags,
 						 &mlx5_ib_dmabuf_attach_ops);
-	else
+	else if (dma_device)
 		umem_dmabuf = ib_umem_dmabuf_get_pinned_with_dma_device(&dev->ib_dev,
 				dma_device, offset, length,
 				fd, access_flags);
+	else
+		umem_dmabuf = ib_umem_dmabuf_get_pinned(
+			&dev->ib_dev, offset, length, fd, access_flags);
 
 	if (IS_ERR(umem_dmabuf)) {
 		mlx5_ib_dbg(dev, "umem_dmabuf get failed (%pe)\n", umem_dmabuf);
@@ -1782,10 +1785,8 @@ struct ib_mr *mlx5_ib_reg_user_mr_dmabuf(struct ib_pd *pd, u64 offset,
 		return reg_user_mr_dmabuf_by_data_direct(pd, offset, length, virt_addr,
 							 fd, access_flags);
 
-	return reg_user_mr_dmabuf(pd, pd->device->dma_device,
-				  offset, length, virt_addr,
-				  fd, access_flags, MLX5_MKC_ACCESS_MODE_MTT,
-				  dmah);
+	return reg_user_mr_dmabuf(pd, NULL, offset, length, virt_addr, fd,
+				  access_flags, MLX5_MKC_ACCESS_MODE_MTT, dmah);
 }
 
 /*

-- 
2.52.0


  parent reply	other threads:[~2026-01-04 13:52 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-04 13:51 [PATCH rdma-next 0/6] Batch of unrelated cleanups in IB Leon Romanovsky
2026-01-04 13:51 ` [PATCH rdma-next 1/6] RDMA/umem: Remove redundant DMABUF ops check Leon Romanovsky
2026-01-05  3:29   ` Kalesh Anakkur Purayil
2026-01-04 13:51 ` [PATCH rdma-next 2/6] RDMA/core: Avoid exporting module local functions and remove not-used ones Leon Romanovsky
2026-01-05  3:45   ` Kalesh Anakkur Purayil
2026-01-05  5:01   ` Zhu Yanjun
2026-01-04 13:51 ` [PATCH rdma-next 3/6] RDMA/mlx5: Fix ucaps init error flow Leon Romanovsky
2026-01-05  3:37   ` Kalesh Anakkur Purayil
2026-01-04 13:51 ` Leon Romanovsky [this message]
2026-01-05  3:45   ` [PATCH rdma-next 4/6] RDMA/mlx5: Avoid direct access to DMA device pointer Kalesh Anakkur Purayil
2026-01-04 13:51 ` [PATCH rdma-next 5/6] RDMA/qedr: Remove unused defines Leon Romanovsky
2026-01-05  3:34   ` Kalesh Anakkur Purayil
2026-01-04 13:51 ` [PATCH rdma-next 6/6] RDMA/ocrdma: Remove unused OCRDMA_UVERBS definition Leon Romanovsky
2026-01-05  3:41   ` Kalesh Anakkur Purayil
2026-01-05  9:07 ` [PATCH rdma-next 0/6] Batch of unrelated cleanups in IB Leon Romanovsky

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=20260104-ib-core-misc-v1-4-00367f77f3a8@nvidia.com \
    --to=leon@kernel.org \
    --cc=cmeiohas@nvidia.com \
    --cc=jgg@ziepe.ca \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=mkalderon@marvell.com \
    --cc=selvin.xavier@broadcom.com \
    --cc=yishaih@nvidia.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®