From: Chuck Lever <cel@kernel.org>
To: netdev@vger.kernel.org, linux-rdma@vger.kernel.org
Cc: BMT@zurich.ibm.com, tom@talpey.com
Subject: [PATCH RFC 3/3] RDMA/siw: Require non-zero 6-byte MACs for soft iWARP
Date: Fri, 05 May 2023 11:43:11 -0400 [thread overview]
Message-ID: <168330138101.5953.12575990094340826016.stgit@oracle-102.nfsv4bat.org> (raw)
In-Reply-To: <168330051600.5953.11366152375575299483.stgit@oracle-102.nfsv4bat.org>
From: Chuck Lever <chuck.lever@oracle.com>
In the past, LOOPBACK and NONE (tunnel) devices had all-zero MAC
addresses. siw_device_create() would fall back to copying the
device's name in those cases, because an all-zero MAC address breaks
the RDMA core IP-to-device lookup mechanism.
However, in some cases, the net_device::name field is also empty.
So we're back at square one.
Rather than checking the device type, look at the
net_device::addr_len field. If it's got the right number of octets
and it is not all zeroes, use that.
Then, to enable siw support for that device/address type, change
the device driver to ensure such devices have a valid 6-octet MAC
address. For virtual devices, using eth_hw_addr_random() is
sufficient.
Fixes: a2d36b02c15d ("RDMA/siw: Enable siw on tunnel devices")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
drivers/infiniband/sw/siw/siw_main.c | 22 +++++++---------------
1 file changed, 7 insertions(+), 15 deletions(-)
diff --git a/drivers/infiniband/sw/siw/siw_main.c b/drivers/infiniband/sw/siw/siw_main.c
index 65b5cda5457b..2c31bf397993 100644
--- a/drivers/infiniband/sw/siw/siw_main.c
+++ b/drivers/infiniband/sw/siw/siw_main.c
@@ -304,10 +304,15 @@ static const struct ib_device_ops siw_device_ops = {
static struct siw_device *siw_device_create(struct net_device *netdev)
{
+ static const u8 zeromac[ETH_ALEN] = { 0 };
struct siw_device *sdev = NULL;
struct ib_device *base_dev;
int rv;
+ if ((netdev->addr_len != ETH_ALEN) ||
+ (memcmp(netdev->dev_addr, zeromac, ETH_ALEN) == 0))
+ return NULL;
+
sdev = ib_alloc_device(siw_device, base_dev);
if (!sdev)
return NULL;
@@ -316,21 +321,8 @@ static struct siw_device *siw_device_create(struct net_device *netdev)
sdev->netdev = netdev;
- if (netdev->type != ARPHRD_LOOPBACK && netdev->type != ARPHRD_NONE) {
- addrconf_addr_eui48((unsigned char *)&base_dev->node_guid,
- netdev->dev_addr);
- } else {
- /*
- * This device does not have a HW address,
- * but connection mangagement lib expects gid != 0
- */
- size_t len = min_t(size_t, strlen(base_dev->name), 6);
- char addr[6] = { };
-
- memcpy(addr, base_dev->name, len);
- addrconf_addr_eui48((unsigned char *)&base_dev->node_guid,
- addr);
- }
+ addrconf_addr_eui48((unsigned char *)&base_dev->node_guid,
+ netdev->dev_addr);
base_dev->uverbs_cmd_mask |= BIT_ULL(IB_USER_VERBS_CMD_POST_SEND);
next prev parent reply other threads:[~2023-05-05 15:43 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-05 15:41 [PATCH RFC 0/3] siw on tunnel devices Chuck Lever
2023-05-05 15:42 ` [PATCH RFC 1/3] net/tun: Ensure tun devices have a MAC address Chuck Lever
2023-05-05 16:59 ` Stephen Hemminger
2023-05-05 17:09 ` Chuck Lever III
2023-05-05 15:42 ` [PATCH RFC 2/3] net/lo: Ensure lo " Chuck Lever
2023-05-05 16:57 ` Stephen Hemminger
2023-05-05 15:43 ` Chuck Lever [this message]
2023-05-05 19:58 ` [PATCH RFC 3/3] RDMA/siw: Require non-zero 6-byte MACs for soft iWARP Jason Gunthorpe
2023-05-05 20:03 ` Chuck Lever III
2023-05-06 18:05 ` Chuck Lever III
2023-05-23 19:18 ` Chuck Lever III
2023-05-23 19:44 ` Tom Talpey
2023-05-23 22:50 ` Chuck Lever III
2023-05-31 19:04 ` Jason Gunthorpe
2023-05-31 19:11 ` Chuck Lever III
2023-05-31 20:09 ` Jason Gunthorpe
2023-05-31 20:19 ` Chuck Lever III
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=168330138101.5953.12575990094340826016.stgit@oracle-102.nfsv4bat.org \
--to=cel@kernel.org \
--cc=BMT@zurich.ibm.com \
--cc=linux-rdma@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=tom@talpey.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;
as well as URLs for NNTP newsgroup(s).