From: Yishai Hadas <yishaih@nvidia.com>
To: <jgg@ziepe.ca>, <leon@kernel.org>
Cc: <linux-rdma@vger.kernel.org>, <selvin.xavier@broadcom.com>,
<kalesh-anakkur.purayil@broadcom.com>,
<chengyou@linux.alibaba.com>, <kaishen@linux.alibaba.com>,
<tangchengchang@huawei.com>, <huangjunxian6@hisilicon.com>,
<abhijit.gangurde@amd.com>, <allen.hubbe@amd.com>,
<longli@microsoft.com>, <kotaranov@microsoft.com>,
<mkalderon@marvell.com>, <bryan-bt.tan@broadcom.com>,
<vishnu.dasa@broadcom.com>, <yishaih@nvidia.com>,
<maorg@nvidia.com>
Subject: [PATCH V1 rdma-next 07/15] RDMA/umem: Derive DMA direction from IB access flags
Date: Tue, 15 Sep 2026 17:09:25 +0300 [thread overview]
Message-ID: <20260915140933.40580-8-yishaih@nvidia.com> (raw)
In-Reply-To: <20260915140933.40580-1-yishaih@nvidia.com>
Map non-writable user memory DMA_TO_DEVICE: when no write access is
granted, the NIC only reads the pages and DMA_BIDIRECTIONAL is
unnecessarily broad. Where the IOMMU enforces direction this prevents
the device writing to read-only buffers; where it does not the mapping
is still semantically correct.
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
---
drivers/infiniband/core/umem.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
index f1da70c4e352..b1ab4f0359b0 100644
--- a/drivers/infiniband/core/umem.c
+++ b/drivers/infiniband/core/umem.c
@@ -159,6 +159,11 @@ unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem,
}
EXPORT_SYMBOL(ib_umem_find_best_pgsz);
+static inline enum dma_data_direction ib_access_dma_dir(int access)
+{
+ return ib_access_writable(access) ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE;
+}
+
static struct ib_umem *__ib_umem_get_va(struct ib_device *device,
unsigned long addr, size_t size,
int access,
@@ -329,7 +334,8 @@ struct ib_umem *ib_umem_get_desc(struct ib_device *device,
const struct ib_uverbs_buffer_desc *desc,
int access)
{
- return __ib_umem_get_desc_dir(device, desc, access, DMA_BIDIRECTIONAL);
+ return __ib_umem_get_desc_dir(device, desc, access,
+ ib_access_dma_dir(access));
}
EXPORT_SYMBOL(ib_umem_get_desc);
@@ -482,7 +488,7 @@ struct ib_umem *ib_umem_get_attr(struct ib_device *device,
u16 attr_id, size_t size, int access)
{
return ib_umem_get_from_attrs(device, attrs, attr_id, NULL, size,
- access, DMA_BIDIRECTIONAL);
+ access, ib_access_dma_dir(access));
}
EXPORT_SYMBOL(ib_umem_get_attr);
@@ -521,7 +527,8 @@ struct ib_umem *ib_umem_get_attr_or_va(struct ib_device *device,
int access)
{
return ib_umem_get_from_attrs_or_va(device, attrs, attr_id, NULL, addr,
- size, access, DMA_BIDIRECTIONAL);
+ size, access,
+ ib_access_dma_dir(access));
}
EXPORT_SYMBOL(ib_umem_get_attr_or_va);
--
2.18.1
next prev parent reply other threads:[~2026-09-15 14:11 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-15 14:09 [PATCH V1 rdma-next 00/15] DMA direction and mlx5 driver correctness fixes Yishai Hadas
2026-09-15 14:09 ` [PATCH V1 rdma-next 01/15] RDMA/erdma: Pin CQ buffer writable to match device DMA write access Yishai Hadas
2026-09-15 14:22 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 02/15] RDMA/hns: " Yishai Hadas
2026-09-15 14:20 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 03/15] RDMA/vmw_pvrdma: Pin QP and SRQ rings " Yishai Hadas
2026-09-15 14:21 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 04/15] RDMA/umem: Reuse ib_umem_get_cq_buf_or_va() for VA-only CQ pinning Yishai Hadas
2026-09-15 14:21 ` sashiko-bot
2026-09-20 12:24 ` Leon Romanovsky
2026-09-22 7:25 ` Yishai Hadas
2026-09-15 14:09 ` [PATCH V1 rdma-next 05/15] RDMA/umem: Support an explicit DMA direction other than DMA_BIDIRECTIONAL Yishai Hadas
2026-09-15 14:21 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 06/15] RDMA/umem: Map CQ buffers DMA_FROM_DEVICE Yishai Hadas
2026-09-15 14:25 ` sashiko-bot
2026-09-15 14:09 ` Yishai Hadas [this message]
2026-09-15 14:28 ` [PATCH V1 rdma-next 07/15] RDMA/umem: Derive DMA direction from IB access flags sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 08/15] RDMA/mlx5: Fix mlx5_ib_dev_res_init() failure when XRC cap is absent Yishai Hadas
2026-09-15 14:24 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 09/15] RDMA/mlx5: Put resource reference in mlx5_ib_wq_event() Yishai Hadas
2026-09-15 14:25 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 10/15] RDMA/mlx5: Set WQ event handler before firmware RQ insertion Yishai Hadas
2026-09-15 14:30 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 11/15] RDMA/mlx5: Set SRQ event handler before xarray insertion Yishai Hadas
2026-09-15 14:36 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 12/15] RDMA/mlx5: Set RQ event handler for raw-packet QP Yishai Hadas
2026-09-15 14:32 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 13/15] RDMA/mlx5: Set QP event handler before firmware QPC insertion Yishai Hadas
2026-09-15 14:35 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 14/15] RDMA/mlx5: Initialize QP/RQ/SQ resource refcount before publishing it Yishai Hadas
2026-09-15 14:29 ` sashiko-bot
2026-09-15 14:09 ` [PATCH V1 rdma-next 15/15] RDMA/mlx5: Fix signed integer overflow in EQE qp_srq type shift Yishai Hadas
2026-09-15 14:31 ` sashiko-bot
2026-09-28 12:17 ` [PATCH V1 rdma-next 00/15] DMA direction and mlx5 driver correctness fixes Leon Romanovsky
2026-09-28 12:19 ` 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=20260915140933.40580-8-yishaih@nvidia.com \
--to=yishaih@nvidia.com \
--cc=abhijit.gangurde@amd.com \
--cc=allen.hubbe@amd.com \
--cc=bryan-bt.tan@broadcom.com \
--cc=chengyou@linux.alibaba.com \
--cc=huangjunxian6@hisilicon.com \
--cc=jgg@ziepe.ca \
--cc=kaishen@linux.alibaba.com \
--cc=kalesh-anakkur.purayil@broadcom.com \
--cc=kotaranov@microsoft.com \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=longli@microsoft.com \
--cc=maorg@nvidia.com \
--cc=mkalderon@marvell.com \
--cc=selvin.xavier@broadcom.com \
--cc=tangchengchang@huawei.com \
--cc=vishnu.dasa@broadcom.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