linux-nvme.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Hannes Reinecke <hare@suse.de>
To: Christoph Hellwig <hch@lst.de>, Hannes Reinecke <hare@kernel.org>
Cc: Keith Busch <kbusch@kernel.org>, Sagi Grimberg <sagi@grimberg.me>,
	linux-nvme@lists.infradead.org
Subject: Re: [PATCH 1/4] nvme: authentication error are always non-retryable
Date: Fri, 1 Mar 2024 16:26:38 +0100	[thread overview]
Message-ID: <f511975a-822e-470a-a3a2-d7ed652e3cfc@suse.de> (raw)
In-Reply-To: <20240301131208.GA3586@lst.de>

On 3/1/24 14:12, Christoph Hellwig wrote:
>> -	if ((nvme_req(req)->status & 0x7ff) == NVME_SC_AUTH_REQUIRED)
>> -		return AUTHENTICATE;
>> -
>>   	if (blk_noretry_request(req) ||
>>   	    (nvme_req(req)->status & NVME_SC_DNR) ||
>>   	    nvme_req(req)->retries >= nvme_max_retries)
>>   		return COMPLETE;
>>   
>> +	if ((nvme_req(req)->status & 0x7ff) == NVME_SC_AUTH_REQUIRED)
>> +		return AUTHENTICATE;
> 
> This part looks fine (although I'd word the commit message
> differently for it).
> 
>> @@ -467,7 +467,7 @@ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl)
>>   		if (result & NVME_CONNECT_AUTHREQ_ASCR) {
>>   			dev_warn(ctrl->device,
>>   				 "qid 0: secure concatenation is not supported\n");
>> -			ret = NVME_SC_AUTH_REQUIRED;
>> +			ret = NVME_SC_AUTH_REQUIRED | NVME_SC_DNR;
>>   			goto out_free_data;
>>   		}
>>   		/* Authentication required */
>> @@ -475,14 +475,16 @@ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl)
>>   		if (ret) {
>>   			dev_warn(ctrl->device,
>>   				 "qid 0: authentication setup failed\n");
>> -			ret = NVME_SC_AUTH_REQUIRED;
>> +			ret = NVME_SC_AUTH_REQUIRED | NVME_SC_DNR;
> 
> 
> .. but the others should never use nvme status codes as they never
> go out onto the wire.
> 
Would love to, but to my knowledge we have the convention that NVMe 
status codes less than 0 should _not_ be retried. NVMe transport errors,
OTOH, _should_ be retried. 'connect_admin_queue' and 'connect_io_queue' 
now straddles the boundary: technically it's an NVMe command, but in
practice it's a transport command as there are quite a lot of steps
before we even can send the 'connect' command.

So, how do we handle the return codes from 'connect_admin_queue' ?
The NVMe core style (with not retrying negative errors) or the NVMe
transport style (with always retrying until we run out of retries)?

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                  Kernel Storage Architect
hare@suse.de                                +49 911 74053 688
SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg
HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich



  reply	other threads:[~2024-03-01 15:26 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-01 11:28 [PATCH 0/4] nvme: fixes for authentication errors Hannes Reinecke
2024-03-01 11:28 ` [PATCH 1/4] nvme: authentication error are always non-retryable Hannes Reinecke
2024-03-01 13:12   ` Christoph Hellwig
2024-03-01 15:26     ` Hannes Reinecke [this message]
2024-03-07  8:51       ` Sagi Grimberg
2024-03-07 10:32         ` Hannes Reinecke
2024-03-07 11:37           ` Sagi Grimberg
2024-03-01 11:28 ` [PATCH 2/4] nvmet: lock config semaphore when accessing DH-HMAC-CHAP key Hannes Reinecke
2024-03-01 13:13   ` Christoph Hellwig
2024-03-07  8:53   ` Sagi Grimberg
2024-03-01 11:28 ` [PATCH 3/4] nvmet: return DHCHAP status codes from nvmet_setup_auth() Hannes Reinecke
2024-03-01 13:13   ` Christoph Hellwig
2024-03-07  8:56   ` Sagi Grimberg
2024-03-07 11:19     ` Hannes Reinecke
2024-03-07 12:03       ` Sagi Grimberg
2024-03-01 11:28 ` [PATCH 4/4] nvmet-loop: do not call nvme_ctrl_put() after nvme_ctrl_uninit() Hannes Reinecke
2024-03-01 13:14   ` Christoph Hellwig
2024-03-07  8:58   ` Sagi Grimberg
2024-03-01 12:24 ` [PATCH 0/4] nvme: fixes for authentication errors Daniel Wagner
2024-03-03  2:58 ` Chaitanya Kulkarni

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=f511975a-822e-470a-a3a2-d7ed652e3cfc@suse.de \
    --to=hare@suse.de \
    --cc=hare@kernel.org \
    --cc=hch@lst.de \
    --cc=kbusch@kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=sagi@grimberg.me \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).