WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

Re: [Xen-devel] why do I get bad disk write performance in the kernel 3.

To: Ian Campbell <Ian.Campbell@xxxxxxxxxx>
Subject: Re: [Xen-devel] why do I get bad disk write performance in the kernel 3.1?
From: Roberto Scudeller <beto.rvs@xxxxxxxxx>
Date: Thu, 10 Nov 2011 12:25:02 -0200
Cc: "xen-devel@xxxxxxxxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxxxxxxxx>, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
Delivery-date: Thu, 10 Nov 2011 06:27:43 -0800
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=lIXCYlJ+VXt+xdMhVLA/AjeztqVRy7TuHwCdVt3oIdI=; b=D0BmHZX0ZENzkBYf4MvSWnG/0vT+DA6Wn/vRjl3RPwkRuGs3iMMqtUm5Rk+yTOx7BH r8b1iDc7Vp0tvoySdnkRvSkewNt988QBM/q0fQnkRNlYB0jp4t/1DftYSf1HFQ7q6fsw 41tOKtFCvdi9LevHBj5JIZq5QDmE+k0c945rM=
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <1320917040.955.197.camel@xxxxxxxxxxxxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <CAOdhoheOhKGWp3f-tvXDX5SuGMXmsyY4fP==EXL-CWfiF52NzQ@xxxxxxxxxxxxxx> <20111109151838.GA26277@xxxxxxxxxxxxxxxxxxx> <CAOdhohc+UoicOGTW2W_rzdf7d4RR48QyOb4dr=f2g-LAWKB-dA@xxxxxxxxxxxxxx> <20111110034237.GA3156@xxxxxxxxxxxxxxxxxxx> <1320917040.955.197.camel@xxxxxxxxxxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Ian,


That depends on whether Roberto is using xm/xend or xl, I think.
Roberto?

I'm using "xl create windows.cfg"
 

I'm not sure what xend does for 'tap2:aio' if blktap isn't available. I
suspect it doesn't fallback to qdisk. Roberto can you report the content
of xenstore under /local/domain/0/backend/ (use xenstore-ls <path>).

# xenstore-ls /local/domain/0/backend
qdisk = ""
 10 = ""
  51712 = ""
   frontend = "/local/domain/10/device/vbd/51712"
   params = "aio:/local-disk/benchCM-windows-2003-64b-std/xvda"
   frontend-id = "10"
   >   removable = "0"
   bootable = "1"
   state = "4"
   dev = "xvda"
   type = "tap"
   mode = "w"
   feature-barrier = "1"
   info = "0"
   sector-size = "512"
   sectors = "67108864"
   hotplug-status = "connected"
  51728 = ""
   frontend = "/local/domain/10/device/vbd/51728"
   params = "aio:/local-disk/benchCM-windows-2003-64b-std/xvdb"
   frontend-id = "10"
   >   removable = "0"
   bootable = "1"
   state = "4"
   dev = "xvdb"
   type = "tap"
   mode = "w"
   feature-barrier = "1"
   info = "0"
   sector-size = "512"
   sectors = "614400"
   hotplug-status = "connected"
console = ""
 10 = ""
  0 = ""
   frontend = "/local/domain/10/console"
   frontend-id = "10"
   >   state = "1"
   domain = "benchCM-windows-2003-64b-std-test"
   protocol = "vt100"


 
> You could also try the 'file:/local-disk' which will setup a loopback device
> and use that. Try that as well.

This is true with xm/xend but with xl you will need to manually setup
the looback and use phy:/dev/loop in order to perform this experiment.
For xl file: turns into qdisk unless blktap is available.

I'm execute with file:/, but in xenstore-list, continue with qdisk. Look:
device = ""
 suspend = ""
  event-channel = "6"
 vbd = ""
  51712 = ""
   backend = "/local/domain/0/backend/qdisk/10/51712"
   backend-id = "0"
   state = "4"
   virtual-device = "51712"
   device-type = "disk"
   ring-ref = "16383"
   event-channel = "7"
  51728 = ""
   backend = "/local/domain/0/backend/qdisk/10/51728"
   backend-id = "0"
   state = "4"
   virtual-device = "51728"
   device-type = "disk"
   ring-ref = "16371"
   event-channel = "8"

Konrad,
I try with "file://" and result was qdisk again. Do you think is LVM test need?

I testing my windows.cfg in xen 4.1.2-rc1-pre with kernel 2.6.32.43, the result of xenstore-ls:
# xenstore-ls /local/domain/0/backend
vbd = ""
 2 = ""
  51712 = ""
   frontend = "/local/domain/2/device/vbd/51712"
   tapdisk-params = "aio:/disk-local/benchCM-windows-2003-64b-std/xvda"
   params = "/dev/xen/blktap-2/tapdev2"
   physical-device = "fd:2"
   frontend-id = "2"
   >    removable = "0"
   bootable = "1"
   state = "4"
   dev = "xvda"
   type = "phy"
   mode = "w"
   feature-barrier = "1"
   sectors = "67108864"
   info = "0"
   sector-size = "512"
  51728 = ""
   frontend = "/local/domain/2/device/vbd/51728"
   tapdisk-params = "aio:/local-disk/benchCM-windows-2003-64b-std/xvdb"
   params = "/dev/xen/blktap-2/tapdev3"
   physical-device = "fd:3"
   frontend-id = "2"
   >   removable = "0"
   bootable = "1"
   state = "4"
   dev = "xvdb"
   type = "phy"
   mode = "w"
   feature-barrier = "1"
   sectors = "614400"
   info = "0"
   sector-size = "512"
console = ""
 2 = ""
  0 = ""
   frontend = "/local/domain/2/console"
   frontend-id = "2"
   >   state = "1"
   domain = "benchCM-windows-2003-64b-std-test"
   protocol = "vt100"

Why did kernel 3.1 use the qdisk? I want to use blktap backend ('tap2:aio:/disk-local/xvda,xvda,w').


Thanks for help.

--
Roberto Scudeller


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel