Linux-NVME Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Max Gurtovoy <mgurtovoy@nvidia.com>
To: Christoph Hellwig <hch@lst.de>, Chao Leng <lengchao@huawei.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>,
	linux-nvme@lists.infradead.org, sagi@grimberg.me,
	kbusch@kernel.org
Subject: Re: [PATCH] nvme: add DIX support for nvme-rdma
Date: Tue, 6 Sep 2022 12:34:13 +0300	[thread overview]
Message-ID: <42216010-70a0-e0ef-2b4f-b25072792f36@nvidia.com> (raw)
In-Reply-To: <20220906065903.GA2345@lst.de>


On 9/6/2022 9:59 AM, Christoph Hellwig wrote:
> On Tue, Sep 06, 2022 at 10:13:42AM +0800, Chao Leng wrote:
>> Although the NVMe protocol does not define in detail how DIX should
>> be supported.
> NVMe describes how protection information works very well.  It does not
> use "DIX" at all to reference those features.  And remember that even
> for SCSI DIX is not a normative on the wire protocol, but just an
> interface for HBAs on how to allow the host to interact with the actual
> wire protocol.

Maybe we can extend the spec and say that for Fabrics only we can use 
IP_CSUM as protection for "host_buf-------->host HBA".

On the wire we'll use extended LBA with CRC as defined today.

>
>> But The NVMe base spec says:
>> Additionally, support has been added for many Enterprise capabilities like
>> end-to-end data protection (compatible with SCSI Protection Information,
>> commonly known as T10 DIF, and SNIA DIX standards), enhanced error reporting,
>> and virtualization.
> None of which is normative language.


  reply	other threads:[~2022-09-06  9:34 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-29  8:12 [PATCH] nvme: add DIX support for nvme-rdma Chao Leng
2022-08-29  9:02 ` Sagi Grimberg
2022-08-29 10:43 ` Max Gurtovoy
2022-08-29 13:16   ` Chao Leng
2022-08-29 14:56     ` Max Gurtovoy
2022-08-29 15:10       ` Keith Busch
2022-08-29 23:47         ` Max Gurtovoy
2022-08-30 12:18         ` Chao Leng
2022-08-30 14:49           ` Keith Busch
2022-08-30  2:38       ` Martin K. Petersen
2022-08-30 12:21         ` Chao Leng
2022-09-05  6:37           ` Christoph Hellwig
2022-09-06  2:13             ` Chao Leng
2022-09-06  6:59               ` Christoph Hellwig
2022-09-06  9:34                 ` Max Gurtovoy [this message]
2022-09-06  9:35                   ` Christoph Hellwig
2022-09-06 10:05                     ` Max Gurtovoy
2022-09-06 10:13                 ` Chao Leng
2022-08-30 12:15       ` Chao Leng

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=42216010-70a0-e0ef-2b4f-b25072792f36@nvidia.com \
    --to=mgurtovoy@nvidia.com \
    --cc=hch@lst.de \
    --cc=kbusch@kernel.org \
    --cc=lengchao@huawei.com \
    --cc=linux-nvme@lists.infradead.org \
    --cc=martin.petersen@oracle.com \
    --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