mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
To: linux-scsi <linux-scsi@vger.kernel.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Joe Eykholt <jeykholt@cisco.com>
Cc: Christoph Hellwig <hch@lst.de>,
	FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>,
	Mike Christie <michaelc@cs.wisc.edu>,
	Hannes Reinecke <hare@suse.de>,
	James Bottomley <James.Bottomley@suse.de>,
	Konrad Rzeszutek Wilk <konrad@darnok.org>,
	Boaz Harrosh <bharrosh@panasas.com>,
	Richard Sharpe <realrichardsharpe@gmail.com>,
	Nicholas Bellinger <nab@linux-iscsi.org>
Subject: [PATCH 0/4] tcm: Add pre-allocated struct se_cmd descriptor support
Date: Fri, 10 Sep 2010 02:26:24 -0700	[thread overview]
Message-ID: <1284110784-32763-1-git-send-email-nab@linux-iscsi.org> (raw)

From: Nicholas Bellinger <nab@linux-iscsi.org>

Greetings Joe and Co,

This patch series adds support to TCM Core for allowing TCM fabric modules to use
pre-allocated struct se_cmd descriptors and sense data buffers living within
fabric module dependent data structures.  This patch series was orginally inspiried
as a future TCM v4.1 optimization by Joe Eykholt in a thread here:

http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/88a6434fa9cad163

The first patch adds the necessary TCM Core infrastructure to support the
initialization and proper release of struct se_cmd living within TCM fabric
module dependent data structures.  This includes the conversion of existing logic
and the addition of the new transport_init_se_cmd() EXPORT_SYMBOL() code.

The second patch converts the TCM_Loop fabric module to use
struct tcm_loop_cmd->tl_se_cmd and struct tcm_loop_cmd->tl_sense_buf.

The third patch converts the LIO-Target fabric module to use
struct iscsi_cmd->se_cmd and struct iscsi_cmd->sense_buffer.

The forth patch converts the Open-FCoE.org / TCM_FC fabric module to
use struct ft_cmd->se_cmd and struct ft_cmd->ft_sense_buffer.

So far this has been lightly tested with TCM_Loop and LIO-Target fabric
modules on v2.6.36-rc3 HVM x86 guest, and will require more testing
to properly ensure the patch series is functionally correct, and that
other new TCM modules are still 100% functional using the existing
*transport_alloc_se_cmd() callers.

Many thanks again to Joe for noticing the oppourtunity for improvement
within the per I/O descriptor path and making this astute suggestion!!

Signed-off-by: Nicholas A. Bellinger <nab@linux-iscsi.org>

Nicholas Bellinger (4):
  tcm: Add support for fabric module provided struct se_cmd descriptors
  tcm_loop: Convert to pre-allocated struct se_cmd descriptors
  lio-target: Convert to use pre-allocated struct se_cmd descriptors
  tcm_fc: Convert to use pre-allocated struct se_cmd descriptors

 drivers/target/lio-target/iscsi_target.c       |   27 +++--
 drivers/target/lio-target/iscsi_target_core.h  |   10 ++-
 drivers/target/lio-target/iscsi_target_erl1.c  |    8 +-
 drivers/target/lio-target/iscsi_target_tmr.c   |   23 ++---
 drivers/target/lio-target/iscsi_target_util.c  |   44 ++++----
 drivers/target/target_core_transport.c         |  131 +++++++++++++++++++-----
 drivers/target/tcm_fc/tcm_fc.h                 |    3 +-
 drivers/target/tcm_fc/tfc_cmd.c                |   59 +++++------
 drivers/target/tcm_fc/tfc_io.c                 |    4 +-
 drivers/target/tcm_loop/tcm_loop_core.h        |    6 +-
 drivers/target/tcm_loop/tcm_loop_fabric.c      |   12 +-
 drivers/target/tcm_loop/tcm_loop_fabric_scsi.c |   37 +++----
 include/target/target_core_base.h              |    3 +
 include/target/target_core_transport.h         |    4 +
 14 files changed, 226 insertions(+), 145 deletions(-)


                 reply	other threads:[~2010-09-10  9:26 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=1284110784-32763-1-git-send-email-nab@linux-iscsi.org \
    --to=nab@linux-iscsi.org \
    --cc=James.Bottomley@suse.de \
    --cc=bharrosh@panasas.com \
    --cc=fujita.tomonori@lab.ntt.co.jp \
    --cc=hare@suse.de \
    --cc=hch@lst.de \
    --cc=jeykholt@cisco.com \
    --cc=konrad@darnok.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=michaelc@cs.wisc.edu \
    --cc=realrichardsharpe@gmail.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

all inboxes | Powered by JetHome®