From: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
To: Vasu Dev <vasu.dev@intel.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>,
linux-scsi@vger.kernel.org, Joe Eykholt <jeykholt@cisco.com>,
James Bottomley <James.Bottomley@suse.de>,
devel@open-fcoe.org, Robert Love <robert.w.love@intel.com>
Subject: Re: [PATCH] scsi, fcoe, libfc: drop scsi host_lock use from fc_queuecommand
Date: Fri, 03 Sep 2010 18:20:38 -0700 [thread overview]
Message-ID: <1283563238.20549.113.camel@haakon2.linux-iscsi.org> (raw)
In-Reply-To: <20100903222715.6237.75737.stgit@localhost.localdomain>
On Fri, 2010-09-03 at 15:27 -0700, Vasu Dev wrote:
> Currently fc_queuecommand drops this lock very early on and then re-acquires
> this lock just before return, this re-acquired lock gets dropped immediately
> by its fast path caller scsi_dispatch_cmd, this re-acquire and then immediate
> drop on each IO hits performance especially with small size IOs on multi-core
> systems, this hit is significant about 25% with 512 bytes size IOs.
>
> This lock is not needed in fc_queuecommand and calling fc_queuecommand
> without this lock held removes above described performance hit.
>
> So this patch adds unlocked_qcmds flag to drop host_lock before
> calling only fc_queuecommand and removes re-acquire & then drop
> for each IO. Added flag, drops this lock only if LLD wants such
> as fcoe.ko does here for fc_queuecommand. This change won't affect
> any existing LLD since added unlocked_qcmds flag will be zero
> in those cases and their host lock uses would effectively remain
> same after this patch.
>
> As per Mike and Nab comments, restored host lock around SHOST_DEL state
> in scsi_dispatch_cmd.
>
> Signed-off-by: Vasu Dev <vasu.dev@intel.com>
> ---
> drivers/scsi/fcoe/fcoe.c | 1 +
> drivers/scsi/libfc/fc_fcp.c | 14 +++++---------
> drivers/scsi/scsi.c | 7 ++++++-
> include/scsi/scsi_host.h | 3 +++
> 4 files changed, 15 insertions(+), 10 deletions(-)
>
<SNIP>
> index b7bdecb..1814c51 100644
> --- a/include/scsi/scsi_host.h
> +++ b/include/scsi/scsi_host.h
> @@ -636,6 +636,9 @@ struct Scsi_Host {
> /* Asynchronous scan in progress */
> unsigned async_scan:1;
>
> + /* call queuecommand without Scsi_Host lock held */
> + unsigned unlocked_qcmds:1;
> +
I am not sure why this structure member (or any of the other ones) need
to be using :1 at this point, considering drivers/scsi/hosts.c:scsi_host_alloc()
is doing a kzalloc() on struct Scsi_Host to begin with..
Other than that minor nit, I have merged this patch into branch at
lio-core-2.6.git/scsi-unlocked-qcmds available here:
git://git.kernel.org/pub/scm/linux/kernel/git/nab/lio-core-2.6.git scsi-unlocked-qcmds
and via gitweb here:
http://git.kernel.org/?p=linux/kernel/git/nab/lio-core-2.6.git;a=shortlog;h=refs/heads/scsi-unlocked_qcmds
I will have a look at running some modified TCM_Loop and Linux/iSCSI Initiators tests
using with SHT->unlocked-qcmds=1 in the very near future on this code. 8-)
Many thanks again for your very worthwhile efforts Vasu!
Best,
--nab
next prev parent reply other threads:[~2010-09-04 1:24 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-03 22:27 [PATCH] scsi, fcoe, libfc: drop scsi host_lock use from fc_queuecommand Vasu Dev
2010-09-04 1:20 ` Nicholas A. Bellinger [this message]
2010-09-12 16:37 ` Bart Van Assche
2010-09-15 17:31 ` Vasu Dev
2010-09-15 18:01 ` Bart Van Assche
2010-09-24 6:41 ` Jens Axboe
2010-09-25 16:55 ` Bart Van Assche
2010-09-26 3:19 ` Jens Axboe
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=1283563238.20549.113.camel@haakon2.linux-iscsi.org \
--to=nab@linux-iscsi.org \
--cc=James.Bottomley@suse.de \
--cc=devel@open-fcoe.org \
--cc=jeykholt@cisco.com \
--cc=linux-scsi@vger.kernel.org \
--cc=michaelc@cs.wisc.edu \
--cc=robert.w.love@intel.com \
--cc=vasu.dev@intel.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