From: Leon Romanovsky <leon@kernel.org>
To: Doug Ledford <dledford@redhat.com>, Jason Gunthorpe <jgg@mellanox.com>
Cc: Leon Romanovsky <leonro@mellanox.com>, linux-rdma@vger.kernel.org
Subject: [PATCH rdma-next] RDMA/ucma: Return stable IB device index as identifier
Date: Thu, 30 Apr 2020 18:29:39 +0300 [thread overview]
Message-ID: <20200430152939.77967-1-leon@kernel.org> (raw)
From: Leon Romanovsky <leonro@mellanox.com>
The librdmacm uses node_guid as identifier to correlate between
IB devices and CMA devices. However FW resets cause to such
"connection" to be lost and require from the user to restart
its application.
Extend UCMA to return IB device index, which is stable identifier.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
---
drivers/infiniband/core/ucma.c | 16 +++++++++-------
include/uapi/rdma/rdma_user_cm.h | 2 ++
2 files changed, 11 insertions(+), 7 deletions(-)
diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c
index 99482dc5934b..d5723465478e 100644
--- a/drivers/infiniband/core/ucma.c
+++ b/drivers/infiniband/core/ucma.c
@@ -848,7 +848,7 @@ static ssize_t ucma_query_route(struct ucma_file *file,
struct sockaddr *addr;
int ret = 0;
- if (out_len < sizeof(resp))
+ if (out_len < offsetof(struct rdma_ucm_query_route_resp, ibdev_index))
return -ENOSPC;
if (copy_from_user(&cmd, inbuf, sizeof(cmd)))
@@ -872,6 +872,7 @@ static ssize_t ucma_query_route(struct ucma_file *file,
goto out;
resp.node_guid = (__force __u64) ctx->cm_id->device->node_guid;
+ resp.ibdev_index = ctx->cm_id->device->index;
resp.port_num = ctx->cm_id->port_num;
if (rdma_cap_ib_sa(ctx->cm_id->device, ctx->cm_id->port_num))
@@ -883,8 +884,8 @@ static ssize_t ucma_query_route(struct ucma_file *file,
out:
mutex_unlock(&ctx->mutex);
- if (copy_to_user(u64_to_user_ptr(cmd.response),
- &resp, sizeof(resp)))
+ if (copy_to_user(u64_to_user_ptr(cmd.response), &resp,
+ min_t(size_t, out_len, sizeof(resp))))
ret = -EFAULT;
ucma_put_ctx(ctx);
@@ -898,6 +899,7 @@ static void ucma_query_device_addr(struct rdma_cm_id *cm_id,
return;
resp->node_guid = (__force __u64) cm_id->device->node_guid;
+ resp->ibdev_index = cm_id->device->index;
resp->port_num = cm_id->port_num;
resp->pkey = (__force __u16) cpu_to_be16(
ib_addr_get_pkey(&cm_id->route.addr.dev_addr));
@@ -910,7 +912,7 @@ static ssize_t ucma_query_addr(struct ucma_context *ctx,
struct sockaddr *addr;
int ret = 0;
- if (out_len < sizeof(resp))
+ if (out_len < offsetof(struct rdma_ucm_query_addr_resp, ibdev_index))
return -ENOSPC;
memset(&resp, 0, sizeof resp);
@@ -925,7 +927,7 @@ static ssize_t ucma_query_addr(struct ucma_context *ctx,
ucma_query_device_addr(ctx->cm_id, &resp);
- if (copy_to_user(response, &resp, sizeof(resp)))
+ if (copy_to_user(response, &resp, min_t(size_t, out_len, sizeof(resp))))
ret = -EFAULT;
return ret;
@@ -977,7 +979,7 @@ static ssize_t ucma_query_gid(struct ucma_context *ctx,
struct sockaddr_ib *addr;
int ret = 0;
- if (out_len < sizeof(resp))
+ if (out_len < offsetof(struct rdma_ucm_query_addr_resp, ibdev_index))
return -ENOSPC;
memset(&resp, 0, sizeof resp);
@@ -1010,7 +1012,7 @@ static ssize_t ucma_query_gid(struct ucma_context *ctx,
&ctx->cm_id->route.addr.dst_addr);
}
- if (copy_to_user(response, &resp, sizeof(resp)))
+ if (copy_to_user(response, &resp, min_t(size_t, out_len, sizeof(resp))))
ret = -EFAULT;
return ret;
diff --git a/include/uapi/rdma/rdma_user_cm.h b/include/uapi/rdma/rdma_user_cm.h
index e545f2de1e13..14d48b462d91 100644
--- a/include/uapi/rdma/rdma_user_cm.h
+++ b/include/uapi/rdma/rdma_user_cm.h
@@ -168,6 +168,7 @@ struct rdma_ucm_query_route_resp {
__u32 num_paths;
__u8 port_num;
__u8 reserved[3];
+ __u32 ibdev_index;
};
struct rdma_ucm_query_addr_resp {
@@ -179,6 +180,7 @@ struct rdma_ucm_query_addr_resp {
__u16 dst_size;
struct __kernel_sockaddr_storage src_addr;
struct __kernel_sockaddr_storage dst_addr;
+ __u32 ibdev_index;
};
struct rdma_ucm_query_path_resp {
--
2.26.2
next reply other threads:[~2020-04-30 15:29 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-30 15:29 Leon Romanovsky [this message]
2020-05-04 11:01 ` [PATCH rdma-next] RDMA/ucma: Return stable IB device index as identifier Gal Pressman
2020-05-04 11:09 ` Leon Romanovsky
2020-05-04 11:13 ` 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=20200430152939.77967-1-leon@kernel.org \
--to=leon@kernel.org \
--cc=dledford@redhat.com \
--cc=jgg@mellanox.com \
--cc=leonro@mellanox.com \
--cc=linux-rdma@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).