Linux RDMA and InfiniBand development
 help / color / mirror / Atom feed
From: Wenpeng Liang <liangwenpeng@huawei.com>
To: <jgg@nvidia.com>, <leon@kernel.org>
Cc: <linux-rdma@vger.kernel.org>, <linuxarm@huawei.com>,
	<liangwenpeng@huawei.com>
Subject: [PATCH v2 for-next 6/7] RDMA/hns: Support MR's restrack ops for hns driver
Date: Mon, 22 Aug 2022 18:44:54 +0800	[thread overview]
Message-ID: <20220822104455.2311053-7-liangwenpeng@huawei.com> (raw)
In-Reply-To: <20220822104455.2311053-1-liangwenpeng@huawei.com>

The MR restrack attributes come from the queue information maintained by
the driver.

For example:

$ rdma res show mr dev hns_0 mrn 6 -dd -jp
[ {
        "ifindex": 4,
        "ifname": "hns_0",
        "mrn": 6,
        "rkey": "300",
        "lkey": "300",
        "mrlen": 131072,
        "pdn": 8,
        "pid": 1524,
        "comm": "ib_send_bw"
    },
    "drv_pbl_hop_num": 2,
    "drv_ba_pg_shift": 14,
    "drv_buf_pg_shift": 12
}

Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
---
 drivers/infiniband/hw/hns/hns_roce_device.h   |  1 +
 drivers/infiniband/hw/hns/hns_roce_main.c     |  1 +
 drivers/infiniband/hw/hns/hns_roce_restrack.c | 30 +++++++++++++++++++
 3 files changed, 32 insertions(+)

diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h
index e0395870b819..30a67bc70f1a 100644
--- a/drivers/infiniband/hw/hns/hns_roce_device.h
+++ b/drivers/infiniband/hw/hns/hns_roce_device.h
@@ -1228,6 +1228,7 @@ int hns_roce_fill_res_cq_entry(struct sk_buff *msg, struct ib_cq *ib_cq);
 int hns_roce_fill_res_cq_entry_raw(struct sk_buff *msg, struct ib_cq *ib_cq);
 int hns_roce_fill_res_qp_entry(struct sk_buff *msg, struct ib_qp *ib_qp);
 int hns_roce_fill_res_qp_entry_raw(struct sk_buff *msg, struct ib_qp *ib_qp);
+int hns_roce_fill_res_mr_entry(struct sk_buff *msg, struct ib_mr *ib_mr);
 struct hns_user_mmap_entry *
 hns_roce_user_mmap_entry_insert(struct ib_ucontext *ucontext, u64 address,
 				size_t length,
diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c
index 17bc73c108f2..ff4386b5c064 100644
--- a/drivers/infiniband/hw/hns/hns_roce_main.c
+++ b/drivers/infiniband/hw/hns/hns_roce_main.c
@@ -570,6 +570,7 @@ static const struct ib_device_ops hns_roce_dev_restrack_ops = {
 	.fill_res_cq_entry_raw = hns_roce_fill_res_cq_entry_raw,
 	.fill_res_qp_entry = hns_roce_fill_res_qp_entry,
 	.fill_res_qp_entry_raw = hns_roce_fill_res_qp_entry_raw,
+	.fill_res_mr_entry = hns_roce_fill_res_mr_entry,
 };
 
 static int hns_roce_register_device(struct hns_roce_dev *hr_dev)
diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c
index 9bafc627864b..84f942e19743 100644
--- a/drivers/infiniband/hw/hns/hns_roce_restrack.c
+++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c
@@ -168,3 +168,33 @@ int hns_roce_fill_res_qp_entry_raw(struct sk_buff *msg, struct ib_qp *ib_qp)
 
 	return ret;
 }
+
+int hns_roce_fill_res_mr_entry(struct sk_buff *msg, struct ib_mr *ib_mr)
+{
+	struct hns_roce_mr *hr_mr = to_hr_mr(ib_mr);
+	struct nlattr *table_attr;
+
+	table_attr = nla_nest_start(msg, RDMA_NLDEV_ATTR_DRIVER);
+	if (!table_attr)
+		return -EMSGSIZE;
+
+	if (rdma_nl_put_driver_u32_hex(msg, "pbl_hop_num", hr_mr->pbl_hop_num))
+		goto err;
+
+	if (rdma_nl_put_driver_u32_hex(msg, "ba_pg_shift",
+				       hr_mr->pbl_mtr.hem_cfg.ba_pg_shift))
+		goto err;
+
+	if (rdma_nl_put_driver_u32_hex(msg, "buf_pg_shift",
+				       hr_mr->pbl_mtr.hem_cfg.buf_pg_shift))
+		goto err;
+
+	nla_nest_end(msg, table_attr);
+
+	return 0;
+
+err:
+	nla_nest_cancel(msg, table_attr);
+
+	return -EMSGSIZE;
+}
-- 
2.30.0


  parent reply	other threads:[~2022-08-22 10:45 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-22 10:44 [PATCH v2 for-next 0/7] RDMA/hns: Add more restrack attributes to hns driver Wenpeng Liang
2022-08-22 10:44 ` [PATCH v2 for-next 1/7] RDMA/hns: Remove redundant DFX file and DFX ops structure Wenpeng Liang
2022-08-22 10:44 ` [PATCH v2 for-next 2/7] RDMA/hns: Add or remove CQ's restrack attributes Wenpeng Liang
2022-08-22 10:44 ` [PATCH v2 for-next 3/7] RDMA/hns: Support CQ's restrack raw ops for hns driver Wenpeng Liang
2022-08-22 10:44 ` [PATCH v2 for-next 4/7] RDMA/hns: Support QP's restrack " Wenpeng Liang
2022-08-22 10:44 ` [PATCH v2 for-next 5/7] RDMA/hns: Support QP's restrack raw " Wenpeng Liang
2022-08-22 10:44 ` Wenpeng Liang [this message]
2022-08-22 10:44 ` [PATCH v2 for-next 7/7] RDMA/hns: Support MR's " Wenpeng Liang
2022-08-23  8:37 ` [PATCH v2 for-next 0/7] RDMA/hns: Add more restrack attributes to " 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=20220822104455.2311053-7-liangwenpeng@huawei.com \
    --to=liangwenpeng@huawei.com \
    --cc=jgg@nvidia.com \
    --cc=leon@kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=linuxarm@huawei.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