From: Michael Gur <michaelgur@nvidia.com>
To: jgg@ziepe.ca, leon@kernel.org
Cc: linux-rdma@vger.kernel.org, netdev@vger.kernel.org,
Edward Srouji <edwards@nvidia.com>,
Yishai Hadas <yishaih@nvidia.com>,
Michael Gur <michaelgur@nvidia.com>,
linux-kernel@vger.kernel.org (open list)
Subject: [PATCH rdma-next 14/15] RDMA/mlx5: Support new Unordered access flag
Date: Sun, 26 Jul 2026 12:29:28 +0300 [thread overview]
Message-ID: <20260726092943.2880176-15-michaelgur@nvidia.com> (raw)
In-Reply-To: <20260726092943.2880176-1-michaelgur@nvidia.com>
Add mlx5 driver support for IB_ACCESS_UNORDERED, the new uverbs access
flag that enables relaxing the order of read after write on mkeys.
Extend the current enablement of ATS over DMABUF with RO access to also
enable for DMABUF with Unordered access.
Mark IB_ACCESS_UNORDERED as non-UMR-modifiable: there is no UMR
mkey-mask bit for order_read_after_write, so UMR cannot drive it.
Reviewed-by: Yishai Hadas <yishaih@nvidia.com>
Signed-off-by: Michael Gur <michaelgur@nvidia.com>
---
drivers/infiniband/hw/mlx5/mlx5_ib.h | 2 +-
drivers/infiniband/hw/mlx5/mr.c | 7 ++++++-
drivers/infiniband/hw/mlx5/umr.c | 2 +-
drivers/infiniband/hw/mlx5/umr.h | 2 +-
4 files changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 427e7f706c8a..ab245ec1af6b 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -1679,7 +1679,7 @@ static inline bool mlx5_umem_needs_ats(struct mlx5_ib_dev *dev,
{
if (!MLX5_CAP_GEN(dev->mdev, ats) || !umem->is_dmabuf)
return false;
- return access_flags & IB_ACCESS_RELAXED_ORDERING;
+ return access_flags & (IB_ACCESS_RELAXED_ORDERING | IB_ACCESS_UNORDERED);
}
int set_roce_addr(struct mlx5_ib_dev *dev, u32 port_num,
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 5f6d91c47483..31879c0ecfb3 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -71,9 +71,14 @@ static void set_mkc_access_pd_addr_fields(void *mkc, int acc, u64 start_addr,
MLX5_SET(mkc, mkc, lw, !!(acc & IB_ACCESS_LOCAL_WRITE));
MLX5_SET(mkc, mkc, lr, 1);
- if (acc & IB_ACCESS_RELAXED_ORDERING)
+ if (acc & (IB_ACCESS_RELAXED_ORDERING | IB_ACCESS_UNORDERED))
mlx5_core_mkey_set_relaxed_ordering(dev->mdev, mkc);
+ if ((acc & IB_ACCESS_UNORDERED) &&
+ MLX5_CAP_GEN(dev->mdev, mkc_order_read_after_write))
+ MLX5_SET(mkc, mkc, order_read_after_write,
+ MLX5_MKC_ORDER_READ_AFTER_WRITE_RO);
+
MLX5_SET(mkc, mkc, pd, to_mpd(pd)->pdn);
MLX5_SET(mkc, mkc, qpn, 0xffffff);
MLX5_SET64(mkc, mkc, start_addr, start_addr);
diff --git a/drivers/infiniband/hw/mlx5/umr.c b/drivers/infiniband/hw/mlx5/umr.c
index b0799f86994d..689d4542d1ad 100644
--- a/drivers/infiniband/hw/mlx5/umr.c
+++ b/drivers/infiniband/hw/mlx5/umr.c
@@ -451,7 +451,7 @@ static void mlx5r_umr_set_access_flags(struct mlx5_ib_dev *dev,
MLX5_SET(mkc, seg, lw, !!(access_flags & IB_ACCESS_LOCAL_WRITE));
MLX5_SET(mkc, seg, lr, 1);
- if (access_flags & IB_ACCESS_RELAXED_ORDERING)
+ if (access_flags & (IB_ACCESS_RELAXED_ORDERING | IB_ACCESS_UNORDERED))
mlx5_core_mkey_set_relaxed_ordering(dev->mdev, seg);
}
diff --git a/drivers/infiniband/hw/mlx5/umr.h b/drivers/infiniband/hw/mlx5/umr.h
index a6c4f35b75e6..78b91d614a17 100644
--- a/drivers/infiniband/hw/mlx5/umr.h
+++ b/drivers/infiniband/hw/mlx5/umr.h
@@ -49,7 +49,7 @@ static inline unsigned int
mlx5r_umr_get_unchangeable_access_flags(struct mlx5_ib_dev *dev,
unsigned int access_flags)
{
- unsigned int ret = 0;
+ unsigned int ret = access_flags & IB_ACCESS_UNORDERED;
if ((access_flags & IB_ACCESS_REMOTE_ATOMIC) &&
MLX5_CAP_GEN(dev->mdev, atomic) &&
--
2.52.0
next prev parent reply other threads:[~2026-07-26 9:31 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-26 9:29 [PATCH rdma-next 00/15] RDMA: Support HW requiring relaxed ordering and add Unordered access flag Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 01/15] RDMA/mlx5: Allow optional access flags in DM registration Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 02/15] RDMA/mlx5: Allow optional access flags in DEVX UMEM registration Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 03/15] RDMA/core: Allow optional access flags in dmabuf reg ioctl Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 04/15] RDMA/core: Allow optional access flags in reg mr ioctl Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 05/15] mlx5: Rename IFC bits of relaxed ordering fields Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 06/15] net/mlx5: Add IFC bits for new ordering caps Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 07/15] mlx5: Move RO setting helper to core and consolidate mlx5 consumers Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 08/15] net/mlx5: Enable relaxed ordering on resource dump mkey Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 09/15] vfio/mlx5: Enable relaxed ordering on the live migration data mkey Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 10/15] RDMA/mlx5: Enable relaxed ordering on ODP null mkey Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 11/15] RDMA/mlx5: Enable relaxed ordering on Memory Window mkey on strict-RO HW Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 12/15] RDMA/mlx5: Converge UMR access-flag cap checks Michael Gur
2026-07-26 9:29 ` [PATCH rdma-next 13/15] RDMA/uverbs: Add new Unordered MR access flag Michael Gur
2026-07-26 9:29 ` Michael Gur [this message]
2026-07-26 9:29 ` [PATCH rdma-next 15/15] RDMA/mlx5: Enforce relaxed ordering when HW requires it Michael Gur
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=20260726092943.2880176-15-michaelgur@nvidia.com \
--to=michaelgur@nvidia.com \
--cc=edwards@nvidia.com \
--cc=jgg@ziepe.ca \
--cc=leon@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--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 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.