All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Matthew Wilcox <willy@infradead.org>,
	linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org,
	target-devel@vger.kernel.org,
	linux1394-devel@lists.sourceforge.net, linux-usb@vger.kernel.org,
	kvm@vger.kernel.org, virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, Juergen Gross <jgross@suse.com>,
	qla2xxx-upstream@qlogic.com,
	Kent Overstreet <kent.overstreet@gmail.com>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Subject: Re: [PATCH 1/2] Convert target drivers to use sbitmap
Date: Tue, 15 May 2018 10:20:37 -0600	[thread overview]
Message-ID: <6ab7d206-ba46-6b1e-5210-4894ec4dc553@kernel.dk> (raw)
In-Reply-To: <3a56027b-47bc-dcb8-a465-3670031572f1@kernel.dk>

On 5/15/18 10:11 AM, Jens Axboe wrote:
> On 5/15/18 10:00 AM, Matthew Wilcox wrote:
>> From: Matthew Wilcox <mawilcox@microsoft.com>
>>
>> The sbitmap and the percpu_ida perform essentially the same task,
>> allocating tags for commands.  Since the sbitmap is more used than
>> the percpu_ida, convert the percpu_ida users to the sbitmap API.
> 
> It should also be the same performance as percpu_ida in light use, and
> performs much better at > 50% utilization of the tag space. I think
> that's better justification than "more used than".

Had to search long and hard for the perf numbers I did for percpu_ida
on higher utilization, but here it is:

https://lkml.org/lkml/2014/4/22/553

-- 
Jens Axboe


WARNING: multiple messages have this Message-ID
From: Jens Axboe <axboe@kernel.dk>
To: Matthew Wilcox <willy@infradead.org>,
	linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org,
	target-devel@vger.kernel.org,
	linux1394-devel@lists.sourceforge.net, linux-usb@vger.kernel.org,
	kvm@vger.kernel.org, virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, Juergen Gross <jgross@suse.com>,
	qla2xxx-upstream@qlogic.com,
	Kent Overstreet <kent.overstreet@gmail.com>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Subject: Re: [PATCH 1/2] Convert target drivers to use sbitmap
Date: Tue, 15 May 2018 16:20:37 +0000	[thread overview]
Message-ID: <6ab7d206-ba46-6b1e-5210-4894ec4dc553@kernel.dk> (raw)
In-Reply-To: <3a56027b-47bc-dcb8-a465-3670031572f1@kernel.dk>

On 5/15/18 10:11 AM, Jens Axboe wrote:
> On 5/15/18 10:00 AM, Matthew Wilcox wrote:
>> From: Matthew Wilcox <mawilcox@microsoft.com>
>>
>> The sbitmap and the percpu_ida perform essentially the same task,
>> allocating tags for commands.  Since the sbitmap is more used than
>> the percpu_ida, convert the percpu_ida users to the sbitmap API.
> 
> It should also be the same performance as percpu_ida in light use, and
> performs much better at > 50% utilization of the tag space. I think
> that's better justification than "more used than".

Had to search long and hard for the perf numbers I did for percpu_ida
on higher utilization, but here it is:

https://lkml.org/lkml/2014/4/22/553

-- 
Jens Axboe


WARNING: multiple messages have this Message-ID
From: Jens Axboe <axboe@kernel.dk>
To: Matthew Wilcox <willy@infradead.org>,
	linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org,
	target-devel@vger.kernel.org,
	linux1394-devel@lists.sourceforge.net, linux-usb@vger.kernel.org,
	kvm@vger.kernel.org, virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, Juergen Gross <jgross@suse.com>,
	qla2xxx-upstream@qlogic.com,
	Kent Overstreet <kent.overstreet@gmail.com>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Subject: [1/2] Convert target drivers to use sbitmap
Date: Tue, 15 May 2018 10:20:37 -0600	[thread overview]
Message-ID: <6ab7d206-ba46-6b1e-5210-4894ec4dc553@kernel.dk> (raw)

On 5/15/18 10:11 AM, Jens Axboe wrote:
> On 5/15/18 10:00 AM, Matthew Wilcox wrote:
>> From: Matthew Wilcox <mawilcox@microsoft.com>
>>
>> The sbitmap and the percpu_ida perform essentially the same task,
>> allocating tags for commands.  Since the sbitmap is more used than
>> the percpu_ida, convert the percpu_ida users to the sbitmap API.
> 
> It should also be the same performance as percpu_ida in light use, and
> performs much better at > 50% utilization of the tag space. I think
> that's better justification than "more used than".

Had to search long and hard for the perf numbers I did for percpu_ida
on higher utilization, but here it is:

https://lkml.org/lkml/2014/4/22/553

  parent reply	other threads:[~2018-05-15 16:20 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-15 16:00 [PATCH 0/2] Use sbitmap instead of percpu_ida Matthew Wilcox
2018-05-15 16:00 ` Matthew Wilcox
2018-05-15 16:00 ` Matthew Wilcox
2018-05-15 16:00 ` [PATCH 1/2] Convert target drivers to use sbitmap Matthew Wilcox
2018-05-15 16:00 ` Matthew Wilcox
2018-05-15 16:00   ` [1/2] " Matthew Wilcox
2018-05-15 16:00   ` [PATCH 1/2] " Matthew Wilcox
2018-05-15 16:00   ` Matthew Wilcox
2018-05-15 16:11   ` Jens Axboe
2018-05-15 16:11     ` [1/2] " Jens Axboe
2018-05-15 16:11     ` [PATCH 1/2] " Jens Axboe
2018-05-15 16:20     ` Jens Axboe
2018-05-15 16:20     ` Jens Axboe [this message]
2018-05-15 16:20       ` [1/2] " Jens Axboe
2018-05-15 16:20       ` [PATCH 1/2] " Jens Axboe
2018-06-12  1:18     ` Jens Axboe
2018-06-12  1:18       ` [1/2] " Jens Axboe
2018-06-12  1:18       ` [PATCH 1/2] " Jens Axboe
2018-06-12  3:06       ` Matthew Wilcox
2018-06-12  3:06         ` [1/2] " Matthew Wilcox
2018-06-12  3:06         ` [PATCH 1/2] " Matthew Wilcox
2018-06-12  3:06         ` Matthew Wilcox
2018-05-15 16:11   ` Jens Axboe
2018-05-16  5:54   ` Felipe Balbi
2018-05-16  5:54     ` [1/2] " Felipe Balbi
2018-05-16  5:54     ` [PATCH 1/2] " Felipe Balbi
2018-05-16 12:47   ` kbuild test robot
2018-05-16 12:47   ` kbuild test robot
2018-05-16 12:47     ` [1/2] " kbuild test robot
2018-05-16 12:47     ` [PATCH 1/2] " kbuild test robot
2018-05-16 12:47   ` [RFC PATCH] iscsit_wait_for_tag() can be static kbuild test robot
2018-05-16 12:47   ` kbuild test robot
2018-05-16 12:47     ` [RFC] " kbuild test robot
2018-05-16 12:47     ` [RFC PATCH] " kbuild test robot
2018-05-16 12:47     ` kbuild test robot
2018-06-12 15:22   ` [PATCH 1/2] Convert target drivers to use sbitmap Bart Van Assche
2018-06-12 15:22     ` [1/2] " Bart Van Assche
2018-06-12 15:22     ` [PATCH 1/2] " Bart Van Assche
2018-06-12 15:22     ` Bart Van Assche
2018-06-12 16:15     ` Matthew Wilcox
2018-06-12 16:15     ` Matthew Wilcox
2018-06-12 16:15       ` [1/2] " Matthew Wilcox
2018-06-12 16:15       ` [PATCH 1/2] " Matthew Wilcox
2018-06-12 16:15       ` Matthew Wilcox
2018-06-12 16:32       ` Bart Van Assche
2018-06-12 16:32         ` [1/2] " Bart Van Assche
2018-06-12 16:32         ` [PATCH 1/2] " Bart Van Assche
2018-06-12 16:32         ` Bart Van Assche
2018-06-12 18:08         ` Matthew Wilcox
2018-06-12 18:08           ` [1/2] " Matthew Wilcox
2018-06-12 18:08           ` [PATCH 1/2] " Matthew Wilcox
2018-06-12 18:08           ` Matthew Wilcox
2018-05-15 16:00 ` [PATCH 2/2] Remove percpu_ida Matthew Wilcox
2018-05-15 16:00 ` Matthew Wilcox
2018-05-15 16:00   ` [2/2] " Matthew Wilcox
2018-05-15 16:00   ` [PATCH 2/2] " Matthew Wilcox

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=6ab7d206-ba46-6b1e-5210-4894ec4dc553@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=jgross@suse.com \
    --cc=kent.overstreet@gmail.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=linux1394-devel@lists.sourceforge.net \
    --cc=mawilcox@microsoft.com \
    --cc=netdev@vger.kernel.org \
    --cc=qla2xxx-upstream@qlogic.com \
    --cc=target-devel@vger.kernel.org \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=willy@infradead.org \
    --subject='Re: [PATCH 1/2] Convert target drivers to use sbitmap' \
    /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

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.