linux-nvme.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: "hch@lst.de" <hch@lst.de>
To: Logan Gunthorpe <logang@deltatee.com>
Cc: "sagi@grimberg.me" <sagi@grimberg.me>,
	"linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
	Chaitanya Kulkarni <Chaitanya.Kulkarni@wdc.com>,
	"hch@lst.de" <hch@lst.de>
Subject: Re: [PATCH 01/14] nvmet: for pt I/O cmds annotate req->sg_cnt likely
Date: Fri, 14 Aug 2020 12:33:56 +0200	[thread overview]
Message-ID: <20200814103356.GA20513@lst.de> (raw)
In-Reply-To: <e432874f-6a22-e56a-777f-990e283114e7@deltatee.com>

On Mon, Aug 10, 2020 at 04:55:54PM -0600, Logan Gunthorpe wrote:
> 
> 
> On 2020-08-10 4:54 p.m., Chaitanya Kulkarni wrote:
> > On 8/10/20 14:54, Logan Gunthorpe wrote:
> >> Hey,
> >>
> >> You forgot to CC me again....
> >>
> > Ugghhh..
> >> On 2020-08-10 12:54 p.m., Chaitanya Kulkarni wrote:
> >>> I/O Commands (nvme_cmd_read, nvme_cmd_write) are most command commands
> >>> when accessing passthru controller. For I/O commands data is represented
> >>> in SG list, so sg_cnt != 0 in most cases.
> >>>
> >>> Annotate req->sg_cnt check with likely.
> >> Does adding these likely annotations actually produce any benefit? Why
> >> can't we rely on the branch predictor in these instances?
> >>
> > 
> > They are present in the target code. I don't see any problem having it 
> > in a passthru code unless it shows severe performance degradation.
> 
> Isn't that backwards? Don't we only add performance optimizations (which
> likely/unlikely are) when they show demonstrable increases in performance?

Yes, this looks like pretty premature optimization.

_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

  parent reply	other threads:[~2020-08-14 10:34 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-10 18:54 [PATCH 00/14] nvmet: passthru few fixes and improvements Chaitanya Kulkarni
2020-08-10 18:54 ` [PATCH 01/14] nvmet: for pt I/O cmds annotate req->sg_cnt likely Chaitanya Kulkarni
2020-08-10 21:54   ` Logan Gunthorpe
2020-08-10 22:54     ` Chaitanya Kulkarni
2020-08-10 22:55       ` Logan Gunthorpe
2020-08-10 23:16         ` Chaitanya Kulkarni
2020-08-14 10:33         ` hch [this message]
2020-08-10 18:54 ` [PATCH 02/14] nvmet: for pt I/O commands use likely for ns check Chaitanya Kulkarni
2020-08-10 18:54 ` [PATCH 03/14] nvmet: use consistent type with id->nlbaf Chaitanya Kulkarni
2020-08-10 21:55   ` Logan Gunthorpe
2020-08-10 22:56     ` Chaitanya Kulkarni
2020-08-10 22:57       ` Logan Gunthorpe
2020-08-10 23:20         ` Chaitanya Kulkarni
2020-08-10 18:54 ` [PATCH 04/14] nvmet: use consistent type for op_flag Chaitanya Kulkarni
2020-08-10 21:56   ` Logan Gunthorpe
2020-08-10 22:56     ` Chaitanya Kulkarni
2020-08-10 18:54 ` [PATCH 05/14] nvmet: get rid of the extra variable Chaitanya Kulkarni
2020-08-10 21:58   ` Logan Gunthorpe
2020-08-10 23:00     ` Chaitanya Kulkarni
2020-08-10 23:13       ` Keith Busch
2020-08-10 23:25         ` Chaitanya Kulkarni
2020-08-10 23:32         ` Logan Gunthorpe
2020-08-10 18:54 ` [PATCH 06/14] nvmet: use unlikely for uncommon commands Chaitanya Kulkarni
2020-08-10 20:46   ` Keith Busch
2020-08-10 22:01     ` Logan Gunthorpe
2020-08-10 18:54 ` [PATCH 07/14] nvmet: reuse pt-req-done for rq completion Chaitanya Kulkarni
2020-08-10 22:10   ` Logan Gunthorpe
2020-08-10 23:02     ` Chaitanya Kulkarni
2020-08-14 10:37       ` hch
2020-08-10 18:54 ` [PATCH 08/14] nvmet: use nvme write cmd group setting op_flag Chaitanya Kulkarni
2020-08-10 21:11   ` Keith Busch
2020-08-10 23:04     ` Chaitanya Kulkarni
2020-08-10 23:10       ` Keith Busch
2020-08-10 23:23         ` Chaitanya Kulkarni
2020-08-10 23:26           ` Keith Busch
2020-08-10 23:33             ` Chaitanya Kulkarni
2020-08-10 22:14   ` Logan Gunthorpe
2020-08-10 23:08     ` Chaitanya Kulkarni
2020-08-10 23:23       ` Logan Gunthorpe
2020-08-10 23:32         ` Chaitanya Kulkarni
2020-08-10 23:34           ` Logan Gunthorpe
2020-08-10 18:54 ` [PATCH 09/14] nvmet: based on passthru req control use REQ_FUA Chaitanya Kulkarni
2020-08-14 10:37   ` Christoph Hellwig
2020-08-10 18:54 ` [PATCH 10/14] nvmet: prep patch for pt ctrl put wrapper Chaitanya Kulkarni
2020-08-10 18:54 ` [PATCH 11/14] nvmet: fix nvme module ref count Oops Chaitanya Kulkarni
2020-08-10 22:39   ` Logan Gunthorpe
2020-08-10 23:11     ` Chaitanya Kulkarni
2020-08-10 18:54 ` [PATCH 12/14] block: move blk_rq_bio_prep() to linux/blk-mq.h Chaitanya Kulkarni
2020-08-10 18:54 ` [PATCH 13/14] nvmet: use minimized version of blk_rq_append_bio Chaitanya Kulkarni
2020-08-10 22:41   ` Logan Gunthorpe
2020-08-10 18:54 ` [PATCH 14/14] nvmet: use inline bio for passthru fast path Chaitanya Kulkarni
2020-08-10 22:47   ` Logan Gunthorpe
2020-08-10 23:14     ` 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=20200814103356.GA20513@lst.de \
    --to=hch@lst.de \
    --cc=Chaitanya.Kulkarni@wdc.com \
    --cc=linux-nvme@lists.infradead.org \
    --cc=logang@deltatee.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).