From: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
To: Avi Kivity <avi@redhat.com>
Cc: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>,
mtosatti@redhat.com, agraf@suse.de, paulus@samba.org,
aarcange@redhat.com, kvm@vger.kernel.org,
kvm-ppc@vger.kernel.org, linux-kernel@vger.kernel.org,
takuya.yoshikawa@gmail.com
Subject: Re: [PATCH 5/6] KVM: Separate rmap_pde from kvm_lpage_info->write_count
Date: Mon, 02 Jul 2012 02:22:08 +0000 [thread overview]
Message-ID: <4FF105D0.8010400@linux.vnet.ibm.com> (raw)
In-Reply-To: <4FEC96EB.1050307@redhat.com>
On 06/29/2012 01:39 AM, Avi Kivity wrote:
> But I still think it's the right thing since it simplifies the code.
> Maybe we should add a prefetch() on write_count do mitigate the
> overhead, if it starts showing up in profiles.
>
Long time ago, there was a discussion about dropping prefetch in the
operation of list walking:
http://lwn.net/Articles/444336/
IIRC, the conclusion is that it is better to let CPU prefetch memory
by itself.
Actually, when i developed lockless spte walking, i measure the performance if
prefetch was used, but i did not see the improvement.
WARNING: multiple messages have this Message-ID (diff)
From: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
To: Avi Kivity <avi@redhat.com>
Cc: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>,
mtosatti@redhat.com, agraf@suse.de, paulus@samba.org,
aarcange@redhat.com, kvm@vger.kernel.org,
kvm-ppc@vger.kernel.org, linux-kernel@vger.kernel.org,
takuya.yoshikawa@gmail.com
Subject: Re: [PATCH 5/6] KVM: Separate rmap_pde from kvm_lpage_info->write_count
Date: Mon, 02 Jul 2012 10:22:08 +0800 [thread overview]
Message-ID: <4FF105D0.8010400@linux.vnet.ibm.com> (raw)
In-Reply-To: <4FEC96EB.1050307@redhat.com>
On 06/29/2012 01:39 AM, Avi Kivity wrote:
> But I still think it's the right thing since it simplifies the code.
> Maybe we should add a prefetch() on write_count do mitigate the
> overhead, if it starts showing up in profiles.
>
Long time ago, there was a discussion about dropping prefetch in the
operation of list walking:
http://lwn.net/Articles/444336/
IIRC, the conclusion is that it is better to let CPU prefetch memory
by itself.
Actually, when i developed lockless spte walking, i measure the performance if
prefetch was used, but i did not see the improvement.
next prev parent reply other threads:[~2012-07-02 2:22 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-21 8:48 [PATCH 0/6] KVM: Optimize MMU notifier's THP page invalidation -v2 Takuya Yoshikawa
2012-06-21 8:48 ` Takuya Yoshikawa
2012-06-21 8:49 ` [PATCH 1/6] KVM: MMU: Use __gfn_to_rmap() to clean up kvm_handle_hva() Takuya Yoshikawa
2012-06-21 8:49 ` Takuya Yoshikawa
2012-06-21 8:49 ` [PATCH 2/6] KVM: Introduce hva_to_gfn_memslot() for kvm_handle_hva() Takuya Yoshikawa
2012-06-21 8:49 ` Takuya Yoshikawa
2012-06-21 8:50 ` [PATCH 3/6] KVM: MMU: Make kvm_handle_hva() handle range of addresses Takuya Yoshikawa
2012-06-21 8:50 ` Takuya Yoshikawa
2012-06-21 8:51 ` [PATCH 4/6] KVM: Introduce kvm_unmap_hva_range() for kvm_mmu_notifier_invalidate_range_start() Takuya Yoshikawa
2012-06-21 8:51 ` Takuya Yoshikawa
2012-06-21 8:51 ` [PATCH 5/6] KVM: Separate rmap_pde from kvm_lpage_info->write_count Takuya Yoshikawa
2012-06-21 8:51 ` Takuya Yoshikawa
2012-06-21 8:52 ` [PATCH 6/6] KVM: MMU: Avoid handling same rmap_pde in kvm_handle_hva_range() Takuya Yoshikawa
2012-06-21 8:52 ` Takuya Yoshikawa
2012-06-22 14:33 ` Takuya Yoshikawa
2012-06-22 14:33 ` Takuya Yoshikawa
2012-06-21 12:51 ` [PATCH 0/6] KVM: Optimize MMU notifier's THP page invalidation -v2 Alexander Graf
2012-06-21 12:51 ` Alexander Graf
2012-06-28 1:57 ` [PATCH 0/6] KVM: Optimize MMU notifier's THP page invalidation -v3 Takuya Yoshikawa
2012-06-28 1:57 ` Takuya Yoshikawa
2012-06-28 1:58 ` [PATCH 1/6] KVM: MMU: Use __gfn_to_rmap() to clean up kvm_handle_hva() Takuya Yoshikawa
2012-06-28 1:58 ` Takuya Yoshikawa
2012-06-28 1:59 ` [PATCH 2/6] KVM: Introduce hva_to_gfn_memslot() for kvm_handle_hva() Takuya Yoshikawa
2012-06-28 1:59 ` Takuya Yoshikawa
2012-06-28 2:00 ` [PATCH 3/6] KVM: MMU: Make kvm_handle_hva() handle range of addresses Takuya Yoshikawa
2012-06-28 2:00 ` Takuya Yoshikawa
2012-06-28 2:00 ` [PATCH 4/6] KVM: Introduce kvm_unmap_hva_range() for kvm_mmu_notifier_invalidate_range_start() Takuya Yoshikawa
2012-06-28 2:00 ` Takuya Yoshikawa
2012-06-28 2:01 ` [PATCH 5/6] KVM: Separate rmap_pde from kvm_lpage_info->write_count Takuya Yoshikawa
2012-06-28 2:01 ` Takuya Yoshikawa
2012-06-28 3:12 ` Xiao Guangrong
2012-06-28 3:12 ` Xiao Guangrong
2012-06-28 3:45 ` Takuya Yoshikawa
2012-06-28 3:45 ` Takuya Yoshikawa
2012-06-28 17:39 ` Avi Kivity
2012-06-28 17:39 ` Avi Kivity
2012-06-29 1:44 ` Takuya Yoshikawa
2012-06-29 1:44 ` Takuya Yoshikawa
2012-07-02 2:22 ` Xiao Guangrong [this message]
2012-07-02 2:22 ` Xiao Guangrong
2012-06-28 2:02 ` [PATCH 6/6] KVM: MMU: Avoid handling same rmap_pde in kvm_handle_hva_range() Takuya Yoshikawa
2012-06-28 2:02 ` Takuya Yoshikawa
2012-06-28 17:53 ` Avi Kivity
2012-06-28 17:53 ` Avi Kivity
2012-06-29 1:46 ` Takuya Yoshikawa
2012-06-29 1:46 ` Takuya Yoshikawa
2012-07-01 7:41 ` Avi Kivity
2012-07-01 7:41 ` Avi Kivity
2012-07-01 13:18 ` Takuya Yoshikawa
2012-07-01 13:18 ` Takuya Yoshikawa
2012-07-01 13:31 ` Avi Kivity
2012-07-01 13:31 ` Avi Kivity
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=4FF105D0.8010400@linux.vnet.ibm.com \
--to=xiaoguangrong@linux.vnet.ibm.com \
--cc=aarcange@redhat.com \
--cc=agraf@suse.de \
--cc=avi@redhat.com \
--cc=kvm-ppc@vger.kernel.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mtosatti@redhat.com \
--cc=paulus@samba.org \
--cc=takuya.yoshikawa@gmail.com \
--cc=yoshikawa.takuya@oss.ntt.co.jp \
/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.