All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Roy Shterman <roy.shterman@gmail.com>, qemu-devel@nongnu.org
Cc: famz@redhat.com, stefanha@redhat.com
Subject: Re: [Qemu-devel] virtio-scsi/blk dataplane and guest memory allocation
Date: Thu, 21 Jan 2016 10:01:05 +0100	[thread overview]
Message-ID: <56A09E51.4020600@redhat.com> (raw)
In-Reply-To: <CAOBqJ55v7zZN-g7ezAhBgmnDf6x6ksw2_W=6Xxi3FH_Vf_timg@mail.gmail.com>



On 20/01/2016 21:12, Roy Shterman wrote:
> Hi,
> 
> I have two questions,
> 
> First, I'm developing for Libiscsi and trying to work with virtio-scsi
> dataplane or even virtio-blk dataplane and it doesn't works well.
> 
> I'm working with latest qemu and latest Libiscsi in RedHat 7 libvirt
> package.
> 
> my iscsi xml part is : 
> 
> virtio-blk - 
> 
>     <disk type='network' device='lun'>
>       <driver name='qemu' type='raw'/>
>       <source protocol='iscsi' name='iqn.2001-04.com.r-dcs03-tgt-3/1'>
>         <host name='11.212.32.52' port='3260'/>
>       </source>
>       <target dev='vdb' bus='virtio'/>
>       <address type='pci' domain='0x0000' bus='0x00' slot='0x07'
> function='0x0'/>
>     </disk>
> 
> virtio-scsi - 
> 
>     <disk type='network' device='lun'>
>       <driver name='qemu' type='raw'/>
>       <source protocol='iscsi' name='iqn.2001-04.com.r-dcs03-tgt-1/1>
>         <host name='11.212.32.52' port='3260'/>
>       </source>
>       <target dev='sda' bus='scsi'/>
>       <address type='drive' controller='0' bus='0' target='0' unit='0'/>
>     </disk>
>     <controller type='scsi' index='0' model='virtio-scsi'>
>       <address type='pci' domain='0x0000' bus='0x00' slot='0x06'
> function='0x0'/>
>     </controller>

There is now support for dataplane in libvirt.  See
https://libvirt.org/formatdomain.html#elementsIOThreadsAllocation and
then you can add an iothread='NN' (NN is a number) to the <driver
name='qemu' type='raw'/> element.

> second thing, I'm trying to look for the code where QEMU allocate all
> guest memory (2 GB) in my case.

Start at memory_allocate_system_memory; ultimately you'll reach
qemu_anon_ram_alloc which is basically an mmap.

paolo

  parent reply	other threads:[~2016-01-21  9:01 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-20 20:12 [Qemu-devel] virtio-scsi/blk dataplane and guest memory allocation Roy Shterman
2016-01-21  3:53 ` Fam Zheng
2016-01-21  9:01 ` Paolo Bonzini [this message]
2016-01-27 19:03   ` Roy Shterman
2016-01-28  1:24     ` Fam Zheng
2016-01-28  7:28     ` Roy Shterman
2016-01-28  7:38       ` Fam Zheng
2016-01-28  8:46       ` Paolo Bonzini
2016-01-28  8:53         ` Daniel P. Berrange
2016-01-30  8:29           ` Roy Shterman
2016-01-30  8:33             ` Roy Shterman

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=56A09E51.4020600@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=famz@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=roy.shterman@gmail.com \
    --cc=stefanha@redhat.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 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.