From: Christoph Hellwig <hch@infradead.org>
To: Max Gurtovoy <maxg@mellanox.com>
Cc: Israel Rukshin <israelr@mellanox.com>,
James Smart <jsmart2021@gmail.com>,
Sagi Grimberg <sagi@grimberg.me>,
linux-nvme@lists.infradead.org
Subject: Re: [RFC 3/6] nvme: Introduce max_meta_segments ctrl and ops attribute
Date: Tue, 21 Apr 2020 11:04:20 -0700 [thread overview]
Message-ID: <20200421180420.GC31265@infradead.org> (raw)
In-Reply-To: <7e85bcc2-cba8-4795-5c15-f7a586d339ac@mellanox.com>
On Tue, Feb 25, 2020 at 11:49:36AM +0200, Max Gurtovoy wrote:
>
> On 2/24/2020 8:48 PM, James Smart wrote:
> > This patch was originally posted by Max Gurtovoy <maxg@mellanox.com>
> >
> > This patch doesn't change any logic, and is needed as a preparation
> > for adding PI support for fabrics drivers.
> >
> > This patch parameterizes the number of sgl segments supported for
> > a separate metadata buffer.
> >
> > The parameter is added to the nvme_ctrl_ops struct and the nvme_ctrl_ops
> > struct. nvme_init_ctrl() was modified to initialize the controller
> > value from the ops value. It was done in this manner such that if the
> > transport supports a singular/unchanged value, it can be set in the ops
> > struct and no other code is necessary. However, if the transport must
> > set the value on a per-controller basis, likely due to differents in the
> > host transport hardware, it can directly modify the field in the ctrl
> > struct.
>
> I don't think we need to add a ops for that and override it per ctrl if
> needed later on.
>
> And if a new op is preferred, please add a callback (see bellow).
>
> Btw,
>
> This was already reviewed by Sagi and Marting so it would be nice to leave
> it as-is (or maybe only rename it to max_metadata_segments).
I have to say I prefer your simpler version as well.
_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2020-04-21 18:04 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-24 18:48 [RFC 0/6] nvme: core layer metadata/pi support James Smart
2020-02-24 18:48 ` [RFC 1/6] nvme: Add ns to nvme_request struct James Smart
2020-04-21 18:03 ` Christoph Hellwig
2020-02-24 18:48 ` [RFC 2/6] nvme: make nvme_ns_has_pi accessible to transports James Smart
2020-04-21 18:03 ` Christoph Hellwig
2020-02-24 18:48 ` [RFC 3/6] nvme: Introduce max_meta_segments ctrl and ops attribute James Smart
2020-02-25 9:49 ` Max Gurtovoy
2020-04-21 18:04 ` Christoph Hellwig [this message]
2020-02-24 18:48 ` [RFC 4/6] nvme: Add t_flags and pi_blkszs ctrl and ops attributes James Smart
2020-04-21 18:10 ` Christoph Hellwig
2020-02-24 18:48 ` [RFC 5/6] nvme: Add pi_flags to nvme_request for transport pi support James Smart
2020-04-21 18:11 ` Christoph Hellwig
2020-02-24 18:48 ` [RFC 6/6] nvme: Introduce NVME_INLINE_PROT_SG_CNT James Smart
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=20200421180420.GC31265@infradead.org \
--to=hch@infradead.org \
--cc=israelr@mellanox.com \
--cc=jsmart2021@gmail.com \
--cc=linux-nvme@lists.infradead.org \
--cc=maxg@mellanox.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;
as well as URLs for NNTP newsgroup(s).