From: James Bottomley <James.Bottomley@HansenPartnership.com>
To: Bart Van Assche <bvanassche@acm.org>
Cc: Christoph Hellwig <hch@infradead.org>,
Jens Axboe <axboe@kernel.dk>,
linux-scsi@vger.kernel.org
Subject: Re: [PATCH 13/17] scsi: push host_lock down into scsi_{host,target}_queue_ready
Date: Thu, 06 Feb 2014 10:41:16 -0800 [thread overview]
Message-ID: <1391712076.22335.13.camel@dabdike> (raw)
In-Reply-To: <52F3C21F.70409@acm.org>
On Thu, 2014-02-06 at 18:10 +0100, Bart Van Assche wrote:
> On 02/06/14 17:56, James Bottomley wrote:
> > Could you benchmark this lot and show what the actual improvement is
> > just for this series, if any?
>
> I see a performance improvement of 12% with the SRP protocol for the
> SCSI core optimizations alone (I am still busy measuring the impact of
> the blk-mq conversion but I can already see that it is really
> significant). Please note that the performance impact depends a lot on
> the workload (number of LUNs per SCSI host e.g.) so maybe the workload I
> chose is not doing justice to Christoph's work. And it's also important
> to mention that with the workload I ran I was saturating the target
> system CPU (a quad core Intel i5). In other words, results might be
> better with a more powerful target system.
On what? Just the patches I indicated or the whole series? My specific
concern is that swapping a critical section for atomics may not buy us
anything even on x86 and may slow down non-x86. That's the bit I'd like
benchmarks to explore.
James
next prev parent reply other threads:[~2014-02-06 18:41 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-05 12:39 [PATCH 00/17] SCSI data path micro-optimizations Christoph Hellwig
2014-02-05 12:39 ` [PATCH 01/17] scsi: handle command allocation failure in scsi_reset_provider Christoph Hellwig
2014-02-05 12:39 ` [PATCH 02/17] megaraid: simplify internal command handling Christoph Hellwig
2014-02-06 16:40 ` Christoph Hellwig
2014-02-05 12:39 ` [PATCH 03/17] scsi: remove scsi_allocate_command/scsi_free_command Christoph Hellwig
2014-02-05 12:39 ` [PATCH 04/17] scsi: avoid useless free_list lock roundtrips Christoph Hellwig
2014-02-05 23:44 ` James Bottomley
2014-02-06 16:22 ` Christoph Hellwig
2014-02-07 9:05 ` Paolo Bonzini
2014-02-05 12:39 ` [PATCH 05/17] scsi: simplify command allocation and freeing a bit Christoph Hellwig
2014-02-05 23:51 ` James Bottomley
2014-02-06 16:21 ` Christoph Hellwig
2014-02-05 12:39 ` [PATCH 06/17] scsi: add support for per-host cmd pools Christoph Hellwig
2014-02-07 9:13 ` Paolo Bonzini
2014-02-07 12:44 ` Christoph Hellwig
2014-02-07 9:35 ` Mike Christie
2014-02-07 12:46 ` Christoph Hellwig
2014-02-07 21:43 ` Mike Christie
2014-02-10 12:20 ` Christoph Hellwig
2014-02-05 12:39 ` [PATCH 07/17] virtio_scsi: use cmd_size Christoph Hellwig
2014-02-07 9:13 ` Paolo Bonzini
2014-02-05 12:39 ` [PATCH 08/17] scsi: do not manipulate device reference counts in scsi_get/put_command Christoph Hellwig
2014-02-05 12:39 ` [PATCH 09/17] scsi: micro-optimize scsi_request_fn() Christoph Hellwig
2014-02-05 12:39 ` [PATCH 10/17] scsi: micro-optimize scsi_next_command() Christoph Hellwig
2014-02-05 12:39 ` [PATCH 11/17] scsi: micro-optimize scsi_requeue_command() Christoph Hellwig
2014-02-05 12:39 ` [PATCH 12/17] scsi: avoid taking host_lock in scsi_run_queue unless nessecary Christoph Hellwig
2014-02-05 23:54 ` James Bottomley
2014-02-06 16:19 ` Christoph Hellwig
2014-02-05 12:39 ` [PATCH 13/17] scsi: push host_lock down into scsi_{host,target}_queue_ready Christoph Hellwig
2014-02-06 16:56 ` James Bottomley
2014-02-06 17:10 ` Bart Van Assche
2014-02-06 18:41 ` James Bottomley [this message]
2014-02-07 10:42 ` Bart Van Assche
2014-02-06 21:58 ` Nicholas A. Bellinger
2014-02-07 10:32 ` Bart Van Assche
2014-02-07 19:30 ` Nicholas A. Bellinger
2014-02-08 11:00 ` Bart Van Assche
2014-02-09 8:26 ` Nicholas A. Bellinger
2014-02-10 12:09 ` Christoph Hellwig
2014-02-10 19:53 ` Nicholas A. Bellinger
2014-02-10 11:39 ` Christoph Hellwig
2014-02-10 20:09 ` Jens Axboe
2014-02-17 9:36 ` Bart Van Assche
2014-02-17 22:00 ` Christoph Hellwig
2014-02-26 15:39 ` Bart Van Assche
2014-02-10 21:10 ` James Bottomley
2014-02-05 12:39 ` [PATCH 14/17] scsi: convert target_busy to an atomic_t Christoph Hellwig
2014-02-05 12:39 ` [PATCH 15/17] scsi: convert host_busy to atomic_t Christoph Hellwig
2014-02-05 12:39 ` [PATCH 16/17] scsi: convert device_busy " Christoph Hellwig
2014-02-05 12:39 ` [PATCH 17/17] scsi: fix the {host,target,device}_blocked counter mess Christoph Hellwig
2014-02-05 23:41 ` [PATCH 00/17] SCSI data path micro-optimizations James Bottomley
2014-02-06 16:29 ` Christoph Hellwig
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=1391712076.22335.13.camel@dabdike \
--to=james.bottomley@hansenpartnership.com \
--cc=axboe@kernel.dk \
--cc=bvanassche@acm.org \
--cc=hch@infradead.org \
--cc=linux-scsi@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