> On May 1, 2022, at 10:18 AM, Bernard Metzler <bmt@xxxxxxxxxxxxxx> wrote:
>
>
>
>> -----Original Message-----
>> From: Chuck Lever <chuck.lever@xxxxxxxxxx>
>> Sent: Thursday, 28 April 2022 19:49
>> To: Bernard Metzler <BMT@xxxxxxxxxxxxxx>
>> Cc: linux-rdma@xxxxxxxxxxxxxxx
>> Subject: [EXTERNAL] [PATCH] RDMA/siw: Enable siw on tunnel devices
>>
>> From: Bernard Metzler <bmt@xxxxxxxxxxxxxx>
>>
>> Signed-off-by: Bernard Metzler <bmt@xxxxxxxxxxxxxx>
>> ---
>> drivers/infiniband/sw/siw/siw_main.c | 5 +++--
>> 1 file changed, 3 insertions(+), 2 deletions(-)
>>
>>
>> Hi Bernard!
>>
>> How come this change isn't in the upstream siw driver yet?
>>
>>
>
> Hi Chuck,
> Good question! Did I ever send the patch to linux-rdma, or was
> that conversation off-list? Sorry for asking, I can't find it in my
> or linux-rdma history...
I don't recall seeing this patch on the public mailing list.
The conversation was between Ben Coddington, you, and me, as
I recall.
> Thank you!
> Bernard
>
>
>> diff --git a/drivers/infiniband/sw/siw/siw_main.c
>> b/drivers/infiniband/sw/siw/siw_main.c
>> index e5c586913d0b..dacc174604bf 100644
>> --- a/drivers/infiniband/sw/siw/siw_main.c
>> +++ b/drivers/infiniband/sw/siw/siw_main.c
>> @@ -119,6 +119,7 @@ static int siw_dev_qualified(struct net_device
>> *netdev)
>> * <linux/if_arp.h> for type identifiers.
>> */
>> if (netdev->type == ARPHRD_ETHER || netdev->type ==
>> ARPHRD_IEEE802 ||
>> + netdev->type == ARPHRD_NONE ||
>> (netdev->type == ARPHRD_LOOPBACK && loopback_enabled))
>> return 1;
>>
>> @@ -315,12 +316,12 @@ static struct siw_device *siw_device_create(struct
>> net_device *netdev)
>>
>> sdev->netdev = netdev;
>>
>> - if (netdev->type != ARPHRD_LOOPBACK) {
>> + if (netdev->type != ARPHRD_LOOPBACK && netdev->type !=
>> ARPHRD_NONE) {
>> addrconf_addr_eui48((unsigned char *)&base_dev-
>>> node_guid,
>> netdev->dev_addr);
>> } else {
>> /*
>> - * The loopback device does not have a HW address,
>> + * 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);
--
Chuck Lever
