From: Ming Lei <ming.lei@redhat.com>
To: Christoph Hellwig <hch@lst.de>
Cc: John Garry <john.garry@huawei.com>,
James Bottomley <James.Bottomley@hansenpartnership.com>,
linux-scsi@vger.kernel.org,
"Martin K . Petersen" <martin.petersen@oracle.com>,
Sumanesh Samanta <sumanesh.samanta@broadcom.com>,
"Ewan D . Milne" <emilne@redhat.com>,
Hannes Reinecke <hare@suse.de>,
Bart Van Assche <bart.vanassche@wdc.com>
Subject: Re: [PATCH] scsi: avoid to fetch scsi host template instance in IO path
Date: Tue, 3 Mar 2020 06:30:40 +0800 [thread overview]
Message-ID: <20200302223040.GD13940@ming.t460p> (raw)
In-Reply-To: <20200302131541.GA13265@lst.de>
On Mon, Mar 02, 2020 at 02:15:41PM +0100, Christoph Hellwig wrote:
> On Mon, Mar 02, 2020 at 10:58:55AM +0000, John Garry wrote:
> > On 28/02/2020 09:33, Ming Lei wrote:
> >> scsi host template struct is quite big, and the following three
> >> fields are needed in SCSI IO path:
> >>
> >> - queuecommand
> >> - commit_rqs
> >> - cmd_size
> >
> > Would it have been nearly as good to reorganise Scsi host template
> > structure to ensure that these are adjacent?
> >
> > I say nearly, as it avoids the shost->hostt read.
>
> That would be worth trying. Replicating function pointers out of
> read-only data structures generally isn't a very good idea.
>
OK, I will try to re-organize host template and see if it can reach
same performance, but it still introduces one extra fetch in IO path.
BTW, we replicate function pointer in blk-mq too, such as q->mq_ops.
Thanks,
Ming
prev parent reply other threads:[~2020-03-02 22:31 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-28 9:33 [PATCH] scsi: avoid to fetch scsi host template instance in IO path Ming Lei
2020-03-02 4:32 ` Bart Van Assche
2020-03-02 10:58 ` John Garry
2020-03-02 13:15 ` Christoph Hellwig
2020-03-02 22:30 ` Ming Lei [this message]
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=20200302223040.GD13940@ming.t460p \
--to=ming.lei@redhat.com \
--cc=James.Bottomley@hansenpartnership.com \
--cc=bart.vanassche@wdc.com \
--cc=emilne@redhat.com \
--cc=hare@suse.de \
--cc=hch@lst.de \
--cc=john.garry@huawei.com \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=sumanesh.samanta@broadcom.com \
/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