kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Richard W.M. Jones" <rjones@redhat.com>
To: Gleb Natapov <gleb@redhat.com>
Cc: kvm@vger.kernel.org, avi@redhat.com, mtosatti@redhat.com
Subject: Re: [PATCHv5 0/4] improve speed of "rep ins" emulation
Date: Mon, 13 Aug 2012 15:39:05 +0100	[thread overview]
Message-ID: <20120813143905.GA13806@rhmail.home.annexia.org> (raw)
In-Reply-To: <1343659101-24877-1-git-send-email-gleb@redhat.com>

On Mon, Jul 30, 2012 at 05:38:17PM +0300, Gleb Natapov wrote:
> And now for something completely different.
> 
> So this series (or rather the last patch of it) takes different approach
> to "rep ins" optimization. Instead of writing separate fast path for
> it do the fast path inside emulator itself. This way nobody can say the
> code is not reused!
> 
> Patch 1,2 are now, strictly speaking, not needed, but I think this is still
> nice cleanup and, in case of patch 1, eliminates some ifs at each KVM_RUN ioctl.
>  
> Gleb Natapov (4):
>   Provide userspace IO exit completion callback.
>   KVM: emulator: make x86 emulation modes enum instead of defines
>   KVM: emulator: string_addr_inc() cleanup
>   KVM: emulator: optimize "rep ins" handling.
> 
>  arch/x86/include/asm/kvm_emulate.h |   26 +++++-----
>  arch/x86/include/asm/kvm_host.h    |    1 +
>  arch/x86/kvm/emulate.c             |   48 ++++++++++++++-----
>  arch/x86/kvm/x86.c                 |   92 +++++++++++++++++++++---------------
>  4 files changed, 104 insertions(+), 63 deletions(-)

I tested this patch series and it really helped to improve the
performance of loading the libguestfs -kernel and -initrd:

  https://bugzilla.redhat.com/show_bug.cgi?id=847546

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
virt-p2v converts physical machines to virtual machines.  Boot with a
live CD or over the network (PXE) and turn machines into Xen guests.
http://et.redhat.com/~rjones/virt-p2v

      parent reply	other threads:[~2012-08-13 14:39 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-30 14:38 [PATCHv5 0/4] improve speed of "rep ins" emulation Gleb Natapov
2012-07-30 14:38 ` [PATCHv5 1/4] Provide userspace IO exit completion callback Gleb Natapov
2012-08-02 19:26   ` Marcelo Tosatti
2012-08-05 14:49     ` Gleb Natapov
2012-07-30 14:38 ` [PATCHv5 2/4] KVM: emulator: make x86 emulation modes enum instead of defines Gleb Natapov
2012-07-30 14:38 ` [PATCHv5 3/4] KVM: emulator: string_addr_inc() cleanup Gleb Natapov
2012-07-30 14:38 ` [PATCHv5 4/4] KVM: emulator: optimize "rep ins" handling Gleb Natapov
2012-08-05 15:03   ` Avi Kivity
2012-08-05 15:18     ` Gleb Natapov
2012-08-05 15:20       ` Avi Kivity
2012-08-06  8:50   ` Avi Kivity
2012-08-06  8:58     ` Gleb Natapov
2012-08-06  9:28       ` Avi Kivity
2012-08-06 11:05         ` Gleb Natapov
2012-08-06 11:39           ` Avi Kivity
2012-08-06 11:49             ` Gleb Natapov
2012-08-06 12:08               ` Avi Kivity
2012-08-07 12:07                 ` Gleb Natapov
2012-08-13 14:39 ` Richard W.M. Jones [this message]

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=20120813143905.GA13806@rhmail.home.annexia.org \
    --to=rjones@redhat.com \
    --cc=avi@redhat.com \
    --cc=gleb@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=mtosatti@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).