mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
To: Christoph Hellwig <hch@lst.de>
Cc: "Nicholas A. Bellinger" <nab@daterainc.com>,
	target-devel <target-devel@vger.kernel.org>,
	linux-scsi <linux-scsi@vger.kernel.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Hannes Reinecke <hare@suse.de>,
	Sagi Grimberg <sagig@mellanox.com>
Subject: Re: [RFC 0/2] target: Add TFO->complete_irq queue_work bypass
Date: Thu, 04 Jun 2015 00:06:09 -0700	[thread overview]
Message-ID: <1433401569.18125.112.camel@haakon3.risingtidesystems.com> (raw)
In-Reply-To: <20150603125756.GA19696@lst.de>

On Wed, 2015-06-03 at 14:57 +0200, Christoph Hellwig wrote:
> This makes lockdep very unhappy, rightly so.  If you execute
> one end_io function inside another you basіcally nest every possible
> lock taken in the I/O completion path.  Also adding more work
> to the hardirq path generally isn't a smart idea.  Can you explain
> what issues you were seeing and how much this helps?  Note that
> the workqueue usage in the target core so far is fairly basic, so
> there should some low hanging fruit.

So I've been using tcm_loop + RAMDISK backends for prototyping, but this
patch is intended for vhost-scsi so it can avoid the unnecessary
queue_work() context switch within target_complete_cmd() for all backend
driver types.

This is because vhost_work_queue() is just updating vhost_dev->work_list
and immediately wake_up_process() into a different vhost_worker()
process context.  For heavy small block workloads into fast IBLOCK
backends, avoiding this extra context switch should be a nice efficiency
win.

Also, AFAIK RDMA fabrics are allowed to do ib_post_send() response
callbacks directly from IRQ context as well.

Perhaps tcm_loop LLD code should just be limited to RAMDISK here..?


  reply	other threads:[~2015-06-04  7:06 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-22  7:57 Nicholas A. Bellinger
2015-05-22  7:57 ` [RFC 1/2] target: Add support for fabric IRQ completion Nicholas A. Bellinger
2015-06-09  7:27   ` Christoph Hellwig
2015-06-29  9:51     ` Sagi Grimberg
2015-05-22  7:57 ` [RFC 2/2] loopback: Enable TFO->complete_irq for fast-path ->scsi_done Nicholas A. Bellinger
2015-06-03 12:57 ` [RFC 0/2] target: Add TFO->complete_irq queue_work bypass Christoph Hellwig
2015-06-04  7:06   ` Nicholas A. Bellinger [this message]
2015-06-04 17:01     ` Sagi Grimberg
2015-06-09  7:19     ` Christoph Hellwig
2015-06-10  7:10       ` Nicholas A. Bellinger

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=1433401569.18125.112.camel@haakon3.risingtidesystems.com \
    --to=nab@linux-iscsi.org \
    --cc=hare@suse.de \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=nab@daterainc.com \
    --cc=sagig@mellanox.com \
    --cc=target-devel@vger.kernel.org \
    /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

all inboxes | Powered by JetHome®