ioccsz and iorcsz check failing

[Sagi Grimberg]

On 12/19/23 19:24, Max Gurtovoy wrote:
>
>
> On 18/12/2023 18:04, Sagi Grimberg wrote:
>>
>>>> Is anything else in the identify wrong, or is it just these
>>>> fabrics fields?
>>>
>>> So what I am seeing on the wire are a few fabrics commands (connect, 
>>> get
>>> property) followed by the nvme id ctrl command (opcode 0x6).
>>>
>>>   nvmet: nvmet_req_init:962
>>>   nvmet: nvmet_parse_admin_cmd:1011
>>>   nvmet: nvmet_parse_discovery_cmd:359 opcode 6
>>>
>>> This calls then nvmet_execute_disc_identify, so adding
>>>
>>> --- a/drivers/nvme/target/discovery.c
>>> +++ b/drivers/nvme/target/discovery.c
>>> @@ -249,6 +249,7 @@ static void nvmet_execute_disc_identify(struct 
>>> nvmet_req *req)
>>>   {
>>>          struct nvmet_ctrl *ctrl = req->sq->ctrl;
>>>          struct nvme_id_ctrl *id;
>>> +       u32 cmd_capsule_size;
>>>          u16 status = 0;
>>>
>>>          if (!nvmet_check_transfer_len(req, NVME_IDENTIFY_DATA_SIZE))
>>> @@ -289,6 +290,17 @@ static void nvmet_execute_disc_identify(struct 
>>> nvmet_req *req)
>>>                  id->sgls |= cpu_to_le32(1 << 2);
>>>          if (req->port->inline_data_size)
>>>                  id->sgls |= cpu_to_le32(1 << 20);
>>> +       /*
>>> +        * Max command capsule size is sqe + in-capsule data size.
>>> +        * Disable in-capsule data for Metadata capable controllers.
>>> +        */
>>> +       cmd_capsule_size = sizeof(struct nvme_command);
>>> +       if (!ctrl->pi_support)
>>> +               cmd_capsule_size += req->port->inline_data_size;
>>> +       id->ioccsz = cpu_to_le32(cmd_capsule_size / 16);
>>
>> Yes, this is the culprit. Nice that it exposed a bug.
>>
>> There is no in-capsule data for discovery controllers afaict.
>
> Also, the discovery controllers can't support PI AFAIK.

Indeed, we should simply set it to sizeof(struct nvme_command) / 16.

It is true that for tcp transport we should add additional 8192 which
is mandatory for nvme-tcp. But that can be done in an incremental patch.