> On 31 Mar 2021, at 15:35, Jason Gunthorpe <jgg@xxxxxxxxxx> wrote:
>
> On Wed, Mar 31, 2021 at 01:34:06PM +0000, Haakon Bugge wrote:
>
>>> Actually I bet you could do this same thing entirely in userspace by
>>> adjusting rdma_init_qp_attr() to copy the data that would be stored in
>>> the cm_id.. ??
>>
>> This will definitely not solve the issue for kernel ULP, e.g., RDS.
>
> Sure, that makes sense to have some rdmacm api in-kernel only
Let me send a v2 doing only that.
>> Further, why do we have rdma_set_option() with option RDMA_OPTION_ID_ACK_TIMEOUT ?
>
> It may have been a mistake to do it like that
I see what you are saying. If id_priv in user space was augmented with timeout_set, timeout, rnr_retry_timer_set, and rnr_retry_time and corresponding rdma_set_{ack_timeout,rnr_retry_timer} functions, use said information in ucma_modify_qp_rtr() and ucma_modify_qp_rts(), it will be a user-space game only.
Håkon
