All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gleb Natapov <gleb@redhat.com>
To: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Cc: avi.kivity@gmail.com, mtosatti@redhat.com, pbonzini@redhat.com,
	linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Subject: Re: [PATCH v6 2/7] KVM: MMU: delete shadow page from hash list in kvm_mmu_prepare_zap_page
Date: Sun, 19 May 2013 13:47:13 +0300	[thread overview]
Message-ID: <20130519104713.GE4725@redhat.com> (raw)
In-Reply-To: <1368738782-18649-3-git-send-email-xiaoguangrong@linux.vnet.ibm.com>

On Fri, May 17, 2013 at 05:12:57AM +0800, Xiao Guangrong wrote:
> Move deletion shadow page from the hash list from kvm_mmu_commit_zap_page to
> kvm_mmu_prepare_zap_page so that we can call kvm_mmu_commit_zap_page
> once for multiple kvm_mmu_prepare_zap_page that can help us to avoid
> unnecessary TLB flush
> 
Don't we call kvm_mmu_commit_zap_page() once for multiple
kvm_mmu_prepare_zap_page() now when possible? kvm_mmu_commit_zap_page()
gets a list as a parameter. I am not against the change, but wish to
understand it better.

> Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
> ---
>  arch/x86/kvm/mmu.c |    8 ++++++--
>  1 files changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
> index 40d7b2d..682ecb4 100644
> --- a/arch/x86/kvm/mmu.c
> +++ b/arch/x86/kvm/mmu.c
> @@ -1466,7 +1466,7 @@ static inline void kvm_mod_used_mmu_pages(struct kvm *kvm, int nr)
>  static void kvm_mmu_free_page(struct kvm_mmu_page *sp)
>  {
>  	ASSERT(is_empty_shadow_page(sp->spt));
> -	hlist_del(&sp->hash_link);
> +
>  	list_del(&sp->link);
>  	free_page((unsigned long)sp->spt);
>  	if (!sp->role.direct)
> @@ -1655,7 +1655,8 @@ static void kvm_mmu_commit_zap_page(struct kvm *kvm,
>  
>  #define for_each_gfn_indirect_valid_sp(_kvm, _sp, _gfn)			\
>  	for_each_gfn_sp(_kvm, _sp, _gfn)				\
> -		if ((_sp)->role.direct || (_sp)->role.invalid) {} else
> +		if ((_sp)->role.direct ||				\
> +		      ((_sp)->role.invalid && WARN_ON(1))) {} else
>  
>  /* @sp->gfn should be write-protected at the call site */
>  static int __kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp,
> @@ -2074,6 +2075,9 @@ static int kvm_mmu_prepare_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp,
>  		unaccount_shadowed(kvm, sp->gfn);
>  	if (sp->unsync)
>  		kvm_unlink_unsync_page(kvm, sp);
> +
> +	hlist_del_init(&sp->hash_link);
> +
What about moving this inside if() bellow and making it hlist_del()?
Leave the page on the hash if root_count is non zero.

>  	if (!sp->root_count) {
>  		/* Count self */
>  		ret++;
> -- 
> 1.7.7.6

--
			Gleb.

  reply	other threads:[~2013-05-19 10:47 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-16 21:12 [PATCH v6 0/7] KVM: MMU: fast zap all shadow pages Xiao Guangrong
2013-05-16 21:12 ` [PATCH v6 1/7] KVM: MMU: drop unnecessary kvm_reload_remote_mmus Xiao Guangrong
2013-05-16 21:12 ` [PATCH v6 2/7] KVM: MMU: delete shadow page from hash list in kvm_mmu_prepare_zap_page Xiao Guangrong
2013-05-19 10:47   ` Gleb Natapov [this message]
2013-05-20  9:19     ` Xiao Guangrong
2013-05-20  9:42       ` Gleb Natapov
2013-05-16 21:12 ` [PATCH v6 3/7] KVM: MMU: fast invalidate all pages Xiao Guangrong
2013-05-19 10:04   ` Gleb Natapov
2013-05-20  9:12     ` Xiao Guangrong
2013-05-20 19:46   ` Marcelo Tosatti
2013-05-20 20:15     ` Gleb Natapov
2013-05-20 20:40       ` Marcelo Tosatti
2013-05-21  3:36         ` Xiao Guangrong
2013-05-21  8:45           ` Gleb Natapov
2013-05-22  1:41           ` Marcelo Tosatti
2013-05-21  8:39         ` Gleb Natapov
2013-05-22  1:33           ` Marcelo Tosatti
2013-05-22  6:34             ` Gleb Natapov
2013-05-22  8:46               ` Xiao Guangrong
2013-05-22  8:54                 ` Gleb Natapov
2013-05-22  9:41                   ` Xiao Guangrong
2013-05-22 13:17                     ` Gleb Natapov
2013-05-22 15:25                       ` Xiao Guangrong
2013-05-22 15:42                         ` Gleb Natapov
2013-05-22 15:06               ` Marcelo Tosatti
2013-05-16 21:12 ` [PATCH v6 4/7] KVM: MMU: zap pages in batch Xiao Guangrong
2013-05-16 21:13 ` [PATCH v6 5/7] KVM: x86: use the fast way to invalidate all pages Xiao Guangrong
2013-05-16 21:13 ` [PATCH v6 6/7] KVM: MMU: show mmu_valid_gen in shadow page related tracepoints Xiao Guangrong
2013-05-16 21:13 ` [PATCH v6 7/7] KVM: MMU: add tracepoint for kvm_mmu_invalidate_all_pages Xiao Guangrong
2013-05-19 10:49 ` [PATCH v6 0/7] KVM: MMU: fast zap all shadow pages Gleb Natapov

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=20130519104713.GE4725@redhat.com \
    --to=gleb@redhat.com \
    --cc=avi.kivity@gmail.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=xiaoguangrong@linux.vnet.ibm.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.