All of lore.kernel.org
 help / color / mirror / Atom feed
From: Felipe Franciosi <felipe.franciosi@citrix.com>
To: Roger Pau Monne <roger.pau@citrix.com>,
	Greg KH <gregkh@linuxfoundation.org>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: "axboe@kernel.dk" <axboe@kernel.dk>,
	"xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>,
	"jerry.snitselaar@oracle.com" <jerry.snitselaar@oracle.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"stable@vger.kernel.org" <stable@vger.kernel.org>
Subject: Re: Backport request to stable of two performance related fixes for xen-blkfront (3.13 fixes to earlier trees)
Date: Mon, 2 Jun 2014 07:13:58 +0000	[thread overview]
Message-ID: <9F2C4E7DFB7839489C89757A66C5AD62641E57__13766.1588182256$1401700128$gmane$org@AMSPEX01CL03.citrite.net> (raw)
In-Reply-To: <537DF36F.5070909@citrix.com>

[-- Attachment #1: Type: text/plain, Size: 4048 bytes --]

> -----Original Message-----
> From: Roger Pau Monné [mailto:roger.pau@citrix.com]
> Sent: 22 May 2014 13:54
> To: Greg KH; Konrad Rzeszutek Wilk
> Cc: linux-kernel@vger.kernel.org; stable@vger.kernel.org; xen-
> devel@lists.xenproject.org; Felipe Franciosi; jerry.snitselaar@oracle.com;
> axboe@kernel.dk
> Subject: Re: Backport request to stable of two performance related fixes for
> xen-blkfront (3.13 fixes to earlier trees)
> 
> On 20/05/14 05:19, Greg KH wrote:
> > On Wed, May 14, 2014 at 03:11:22PM -0400, Konrad Rzeszutek Wilk wrote:
> >> Hey Greg
> >>
> >> This email is in regards to backporting two patches to stable that
> >> fall under the 'performance' rule:
> >>
> >>  bfe11d6de1c416cea4f3f0f35f864162063ce3fa
> >>  fbe363c476afe8ec992d3baf682670a4bd1b6ce6
> >>
> >> I've copied Jerry - the maintainer of the Oracle's kernel. I don't
> >> have the emails of the other distros maintainers but the bugs associated
> with it are:
> >>
> >> https://bugzilla.redhat.com/show_bug.cgi?id=1096909
> >> (RHEL7)
> >> https://bugs.launchpad.net/ubuntu/+bug/1319003
> >> (Ubuntu 13.10)
> >>
> >> The following distros are affected:
> >>
> >> (x) Ubuntu 13.04 and derivatives (3.8)
> >> (v) Ubuntu 13.10 and derivatives (3.11), supported until 2014-07
> >> (x) Fedora 17 (3.8 and 3.9 in updates)
> >> (x) Fedora 18 (3.8, 3.9, 3.10, 3.11 in updates)
> >> (v) Fedora 19 (3.9; 3.10, 3.11, 3.12 in updates; fixed with latest
> >> update to 3.13), supported until TBA
> >> (v) Fedora 20 (3.11; 3.12 in updates; fixed with latest update to
> >> 3.13), supported until TBA
> >> (v) RHEL 7 and derivatives (3.10), expected to be supported until
> >> about 2025
> >> (v) openSUSE 13.1 (3.11), expected to be supported until at least
> >> 2016-08
> >> (v) SLES 12 (3.12), expected to be supported until about 2024
> >> (v) Mageia 3 (3.8), supported until 2014-11-19
> >> (v) Mageia 4 (3.12), supported until 2015-08-01
> >> (v) Oracle Enterprise Linux with Unbreakable Enterprise Kernel
> >> Release 3 (3.8), supported until TBA
> >>
> >> Here is the analysis of the problem and what was put in the RHEL7 bug.
> >> The Oracle bug does not exist (as I just backport them in the kernel
> >> and send a GIT PULL to Jerry) - but if you would like I can certainly
> >> furnish you with one (it would be identical to what is mentioned below).
> >>
> >> If you are OK with the backport, I am volunteering Roger and Felipe
> >> to assist in jamming^H^H^H^Hbackporting the patches into earlier kernels.
> >
> > Sure, can you provide backported patches?  As-is, they don't apply to
> > the 3.10-stable kernel.
> 
> Here are the backported patches to 3.10 stable, I would like however to get
> some testing/benchmarking on them before applying, since it's not a trivial
> backport. Could you please give them a spin Felipe?

Apologies for the delay in remeasuring this.

I can confirm the backport drastically improves aggregate throughput (at least) when the backend does not support persistent grants. This is very visible with moderate-sized requests and more than one VM (on my graphs, above four guests). Naturally, the magnitude of the regression will inevitably vary depending on the host's characteristics.

The attached graphs were measured on XenServer Creedence #85466 which uses tapdisk3 (grant copy). They are the result of a single run, so some variation is expected.

I used Ubuntu 13.10 64 bit guests and installed 3.10.40. I used 5 SSDs (4 Microns and 1 Fusion-io). Each SSD had 10 LVM Logical Volumes, one assigned to each VM (therefore each VM had 5 extra virtual disks). The benchmark was sequential reads using the specified block size directly on the virtual block device with O_DIRECT for a duration of 10s each, reporting the total throughput. It was synchronised so that the specified number of guests executed the IO operations at the same time.

The only difference between the two tests are Roger's patches. Please see attached.

> 
> Roger.

Cheers,
Felipe

[-- Attachment #2: backport_eval-3.10.40.png --]
[-- Type: image/png, Size: 281535 bytes --]

[-- Attachment #3: backport_eval-3.10.40-roger.png --]
[-- Type: image/png, Size: 282454 bytes --]

[-- Attachment #4: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

  reply	other threads:[~2014-06-02  7:14 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-05-14 19:11 Backport request to stable of two performance related fixes for xen-blkfront (3.13 fixes to earlier trees) Konrad Rzeszutek Wilk
2014-05-14 19:21 ` Josh Boyer
2014-05-14 19:21 ` Josh Boyer
2014-05-20  3:19 ` Greg KH
2014-05-20  3:19   ` Greg KH
2014-05-22 12:54   ` Roger Pau Monné
2014-06-02  7:13     ` Felipe Franciosi [this message]
2014-05-22 12:54   ` Roger Pau Monné
2014-05-20  9:32 ` [Xen-devel] " Vitaly Kuznetsov
2014-05-20  9:54   ` Vitaly Kuznetsov
2014-05-20  9:54   ` [Xen-devel] " Vitaly Kuznetsov
2014-05-20 10:32     ` Roger Pau Monné
2014-05-20 11:41       ` Vitaly Kuznetsov
2014-05-20 11:41       ` [Xen-devel] " Vitaly Kuznetsov
2014-05-20 13:59         ` Felipe Franciosi
2014-05-20 13:59           ` Felipe Franciosi
2014-05-20 13:59           ` [Xen-devel] " Felipe Franciosi
2014-05-22  8:52           ` Vitaly Kuznetsov
2014-05-20 10:32     ` Roger Pau Monné
2014-05-20  9:32 ` Vitaly Kuznetsov
2014-06-04  5:48 ` Greg KH
2014-06-04  5:48 ` Greg KH
2014-06-06 10:47   ` Jiri Slaby
2014-06-06 10:58     ` Vitaly Kuznetsov
2014-06-10 13:19       ` [Xen-devel] " Vitaly Kuznetsov
2014-06-10 16:55         ` Roger Pau Monné
2014-06-12 12:00           ` Vitaly Kuznetsov
2014-06-12 12:32             ` Felipe Franciosi
2014-06-12 12:32               ` Felipe Franciosi
2014-06-12 15:32               ` Vitaly Kuznetsov
2014-06-20 17:06                 ` Felipe Franciosi
2014-06-20 17:06                   ` Felipe Franciosi
2014-06-10 17:26         ` Felipe Franciosi
2014-06-06 10:58     ` Vitaly Kuznetsov
2014-06-06 13:56     ` Greg KH
2014-06-06 13:56     ` Greg KH
2014-06-06 14:02       ` Konrad Rzeszutek Wilk
2014-06-06 14:02       ` Konrad Rzeszutek Wilk
2014-06-06 14:03         ` Jiri Slaby
2014-06-06 14:03         ` Jiri Slaby
2014-06-06 10:47   ` Jiri Slaby
2014-05-14 19:11 Konrad Rzeszutek Wilk

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='9F2C4E7DFB7839489C89757A66C5AD62641E57__13766.1588182256$1401700128$gmane$org@AMSPEX01CL03.citrite.net' \
    --to=felipe.franciosi@citrix.com \
    --cc=axboe@kernel.dk \
    --cc=gregkh@linuxfoundation.org \
    --cc=jerry.snitselaar@oracle.com \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=roger.pau@citrix.com \
    --cc=stable@vger.kernel.org \
    --cc=xen-devel@lists.xenproject.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 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.