public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Marcelo Tosatti <mtosatti@redhat.com>
To: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
Cc: gleb@redhat.com, kvm@vger.kernel.org
Subject: Re: [PATCH 8/8] KVM: MMU: Move free_zapped_mmu_pages() out of the protection of mmu_lock
Date: Mon, 4 Feb 2013 11:50:00 -0200	[thread overview]
Message-ID: <20130204135000.GC9005@amt.cnet> (raw)
In-Reply-To: <20130123191811.efba4200.yoshikawa_takuya_b1@lab.ntt.co.jp>

On Wed, Jan 23, 2013 at 07:18:11PM +0900, Takuya Yoshikawa wrote:
> We noticed that kvm_mmu_zap_all() could take hundreds of milliseconds
> for zapping mmu pages with mmu_lock held.
> 
> Although we need to do conditional rescheduling for completely
> fixing this issue, we can reduce the hold time to some extent by moving
> free_zapped_mmu_pages() out of the protection.  Since invalid_list can
> be very long, the effect is not negligible.
> 
> Note: this patch does not treat non-trivial cases.
> 
> Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>

Can you describe the case thats biting? Is it

        /*
         * If memory slot is created, or moved, we need to clear all
         * mmio sptes.
         */
        if (npages && old.base_gfn != mem->guest_phys_addr >> PAGE_SHIFT) {
                kvm_mmu_zap_all(kvm);
                kvm_reload_remote_mmus(kvm);
        }

Because conditional rescheduling for kvm_mmu_zap_all() might not be
desirable: KVM_SET_USER_MEMORY has low latency requirements.



  reply	other threads:[~2013-02-04 14:12 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-01-23 10:12 [PATCH 0/8] KVM: Reduce mmu_lock hold time when zapping mmu pages Takuya Yoshikawa
2013-01-23 10:13 ` [PATCH 1/8] KVM: MMU: Fix and clean up for_each_gfn_* macros Takuya Yoshikawa
2013-01-28 12:24   ` Gleb Natapov
2013-01-28 12:29     ` Takuya Yoshikawa
2013-01-23 10:13 ` [PATCH 2/8] KVM: MMU: Use list_for_each_entry_safe in kvm_mmu_commit_zap_page() Takuya Yoshikawa
2013-01-23 10:14 ` [PATCH 3/8] KVM: MMU: Add a parameter to kvm_mmu_prepare_zap_page() to update the next position Takuya Yoshikawa
2013-01-23 10:15 ` [PATCH 4/8] KVM: MMU: Introduce for_each_gfn_indirect_valid_sp_safe macro Takuya Yoshikawa
2013-01-23 10:16 ` [PATCH 5/8] KVM: MMU: Delete hash_link node in kvm_mmu_prepare_zap_page() Takuya Yoshikawa
2013-01-23 10:16 ` [PATCH 6/8] KVM: MMU: Introduce free_zapped_mmu_pages() for freeing mmu pages in a list Takuya Yoshikawa
2013-01-23 10:17 ` [PATCH 7/8] KVM: MMU: Split out free_zapped_mmu_pages() from kvm_mmu_commit_zap_page() Takuya Yoshikawa
2013-01-23 10:18 ` [PATCH 8/8] KVM: MMU: Move free_zapped_mmu_pages() out of the protection of mmu_lock Takuya Yoshikawa
2013-02-04 13:50   ` Marcelo Tosatti [this message]
2013-02-05  2:21     ` Takuya Yoshikawa
2013-01-23 10:44 ` [PATCH 0/8] KVM: Reduce mmu_lock hold time when zapping mmu pages Xiao Guangrong
2013-01-23 13:28   ` Takuya Yoshikawa
2013-01-23 13:45     ` Xiao Guangrong
2013-01-23 14:49       ` Takuya Yoshikawa
2013-01-23 15:45         ` Xiao Guangrong
2013-02-04 13:42   ` Marcelo Tosatti
2013-02-05  5:30     ` Xiao Guangrong
2013-02-04 13:29 ` Marcelo Tosatti

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=20130204135000.GC9005@amt.cnet \
    --to=mtosatti@redhat.com \
    --cc=gleb@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=yoshikawa_takuya_b1@lab.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox