On 9/14/2018 4:46 PM, Jason Gunthorpe wrote:
> On Fri, Sep 14, 2018 at 04:37:21PM -0500, Steve Wise wrote:
>>
>>
>> On 9/14/2018 2:35 PM, Jason Gunthorpe wrote:
>>> On Fri, Sep 14, 2018 at 01:45:05PM -0500, Steve Wise wrote:
>>>
>>>>>> Since the rdma core currently assigned device name strings, I'm thinking
>>>>>> to just follow that scheme. This leads to slightly different command
>>>>>> syntax:
>>>>>>
>>>>>> rdma link add TYPE dev DEV
>>>>>>
>>>>>> EG: rdma link add rxe dev eth0
>>>>>>
>>>>>> Then the kernel would return the resulting ib_device name and index.
>>>>>
>>>>> I would rather we specify the ib_device name from userspace from the
>>>>> start..
>>>>>
>>>>
>>>> Why? This is not current practice for rdma devices: the core names them.
>>>
>>> That is a bad practice.. Leon has been working on device renaming like
>>> netdev has, so it would make no sense to have no name here and have
>>> renaming down the road.
>>>
>>> Jason
>>>
>>
>> This causes a slit kink in arbitrarily naming rxe rdma devices. From
>> providers/rxe/rxe.c:
>>
>> static const struct verbs_match_ent hca_table[] = {
>> /* FIXME: rxe needs a more reliable way to detect the rxe device */
>> VERBS_NAME_MATCH("rxe", NULL),
>> {},
>> };
>>
>>
>> So when I add a new rxe device named "foo0", it gets added fine in the
>> kernel, but the user space side skips it. It doesn't show up in
>> ibv_devices or ibv_devinfo for example. But the sysfs entries are there...
>
> Yes, this needs fixing too :(
>
> I was hoping we'd get to netlink discovery before this became an
> issue.. The right solution is to use the new driver id number, I
> think.
>
> Jason
>
Do you mean the rdma_driver_id enum? That would I methinks...
