On 10/20/23 19:08, Namjae Jeon wrote:
2023-10-21 5:27 GMT+09:00, Tom Talpey <tom@xxxxxxxxxx>:
On 10/20/2023 10:45 AM, Namjae Jeon wrote:
2023-10-20 22:02 GMT+09:00, Kangjing "Chaser" Huang
<huangkangjing@xxxxxxxxx>:
Physical ib_device does not have an underlying net_device, thus its
association with IPoIB net_device cannot be retrieved via
ops.get_netdev() or ib_device_get_by_netdev(). ksmbd reads physical
ib_device port GUID from the lower 16 bytes of the hardware addresses on
IPoIB net_device and match its underlying ib_device using ib_find_gid()
Signed-off-by: Kangjing Huang <huangkangjing@xxxxxxxxx>
Acked-by: Namjae Jeon <linkinjeon@xxxxxxxxxx>
Can you fix the warnings from checkpatch.pl ? You need to run it like
this. ./scripts/checkpatch.pl [patch]
WARNING: Block comments use a trailing */ on a separate line
#148: FILE: fs/smb/server/transport_rdma.c:2256:
+ * check for matching infiniband GUID in hw_addr */
WARNING: From:/Signed-off-by: email name mismatch: 'From: "Kangjing
\Chaser\ Huang" <huangkangjing@xxxxxxxxx>' != 'Signed-off-by: Kangjing
Huang <huangkangjing@xxxxxxxxx>'
total: 0 errors, 2 warnings, 54 lines checked
And have you made this patch on linux 6.6-rc6 kernel ? because I can't
apply this patch with the following error.
checking file fs/smb/server/transport_rdma.c
Hunk #1 FAILED at 2140.
Hunk #2 FAILED at 2241.
2 out of 2 hunks FAILED
Sorry about that. I did do it on v6.6-rc6. It may be my mail client's
issue with plain text emails, I'm sending again together with the
comment format fixed and the review tag using git send-email.
Thanks.
With these glitches addressed, please add my
Reviewed-by: Tom Talpey <tom@xxxxxxxxxx>
Thanks for your review!
Thanks for the update.
v2 -> v1:
* Add more detailed description to comment
---
fs/smb/server/transport_rdma.c | 39
+++++++++++++++++++++++++---------
1 file changed, 29 insertions(+), 10 deletions(-)
diff --git a/fs/smb/server/transport_rdma.c
b/fs/smb/server/transport_rdma.c
index 3b269e1f523a..a623e29b2760 100644
--- a/fs/smb/server/transport_rdma.c
+++ b/fs/smb/server/transport_rdma.c
@@ -2140,8 +2140,7 @@ static int smb_direct_ib_client_add(struct
ib_device
*ib_dev)
if (ib_dev->node_type != RDMA_NODE_IB_CA)
smb_direct_port = SMB_DIRECT_PORT_IWARP;
- if (!ib_dev->ops.get_netdev ||
- !rdma_frwr_is_supported(&ib_dev->attrs))
+ if (!rdma_frwr_is_supported(&ib_dev->attrs))
return 0;
smb_dev = kzalloc(sizeof(*smb_dev), GFP_KERNEL);
@@ -2241,17 +2240,37 @@ bool ksmbd_rdma_capable_netdev(struct net_device
*netdev)
for (i = 0; i < smb_dev->ib_dev->phys_port_cnt; i++) {
struct net_device *ndev;
- ndev = smb_dev->ib_dev->ops.get_netdev(smb_dev->ib_dev,
- i + 1);
- if (!ndev)
- continue;
+ if (smb_dev->ib_dev->ops.get_netdev) {
+ ndev = smb_dev->ib_dev->ops.get_netdev(
+ smb_dev->ib_dev, i + 1);
+ if (!ndev)
+ continue;
- if (ndev == netdev) {
+ if (ndev == netdev) {
+ dev_put(ndev);
+ rdma_capable = true;
+ goto out;
+ }
dev_put(ndev);
- rdma_capable = true;
- goto out;
+ /* if ib_dev does not implement ops.get_netdev
+ * check for matching infiniband GUID in hw_addr */
+ } else if (netdev->type == ARPHRD_INFINIBAND) {
+ struct netdev_hw_addr *ha;
+ union ib_gid gid;
+ u32 port_num;
+ int ret;
+
+ netdev_hw_addr_list_for_each(
+ ha, &netdev->dev_addrs) {
+ memcpy(&gid, ha->addr + 4, sizeof(gid));
+ ret = ib_find_gid(smb_dev->ib_dev, &gid,
+ &port_num, NULL);
+ if (!ret) {
+ rdma_capable = true;
+ goto out;
+ }
+ }
}
- dev_put(ndev);
}
}
out:
--
2.42.0
