public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Avi Kivity <avi@redhat.com>
To: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	kvm@vger.kernel.org
Subject: Re: [PATCH 7/7] KVM: MMU: Validate all gptes during fetch, not just those used for new pages
Date: Mon, 12 Jul 2010 11:58:55 +0300	[thread overview]
Message-ID: <4C3AD94F.8030809@redhat.com> (raw)
In-Reply-To: <1278862955-6890-8-git-send-email-avi@redhat.com>

On 07/11/2010 06:42 PM, Avi Kivity wrote:
> Currently, when we fetch an spte, we only verify that gptes match those that
> the walker saw if we build new shadow pages for them.
>
> However, this misses the following race:
>
>    vcpu1            vcpu2
>
>    walk
>                    change gpte
>                    walk
>                    instantiate sp
>
>    fetch existing sp
>
> Fix by validating every gpte, regardless of whether it is used for building
> a new sp or not.
>
> Signed-off-by: Avi Kivity<avi@redhat.com>
> ---
>   arch/x86/kvm/paging_tmpl.h |   18 ++++++++++--------
>   1 files changed, 10 insertions(+), 8 deletions(-)
>
> diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
> index a7f8295..4bbd0c7 100644
> --- a/arch/x86/kvm/paging_tmpl.h
> +++ b/arch/x86/kvm/paging_tmpl.h
> @@ -325,7 +325,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
>   			 int *ptwrite, pfn_t pfn)
>   {
>   	unsigned access = gw->pt_access;
> -	struct kvm_mmu_page *sp;
> +	struct kvm_mmu_page *uninitialized_var(sp);
>   	u64 *sptep = NULL;
>   	int uninitialized_var(level);
>   	bool dirty = is_dirty_gpte(gw->ptes[gw->level - 1]);
> @@ -343,18 +343,19 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
>   	     shadow_walk_okay(&iterator)&&  iterator.level>  gw->level;
>   	     shadow_walk_next(&iterator)) {
>   		gfn_t table_gfn;
> +		bool new_page = false;
>
>   		level = iterator.level;
>   		sptep = iterator.sptep;
>
>   		drop_spte_if_large(vcpu, sptep);
>
> -		if (is_shadow_present_pte(*sptep))
> -			continue;
>    

See, this gets dropped.

> -
> -		table_gfn = gw->table_gfn[level - 2];
> -		sp = kvm_mmu_get_page(vcpu, table_gfn, addr, level-1,
> -				      false, access, sptep);
> +		if (!is_shadow_present_pte(*sptep)) {
> +			table_gfn = gw->table_gfn[level - 2];
> +			sp = kvm_mmu_get_page(vcpu, table_gfn, addr, level-1,
> +					      false, access, sptep);
> +			new_page = true;
> +		}
>
>   		if (!FNAME(validate_indirect_spte)(vcpu, sptep, sp,
>   						   gw, level)) {
>    

Now we need to change validate_indirect_spte() to account for all levels.



-- 
error compiling committee.c: too many arguments to function


      parent reply	other threads:[~2010-07-12  8:58 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-11 15:42 [PATCH 0/7] Simplify and fix fetch() Avi Kivity
2010-07-11 15:42 ` [PATCH 1/7] KVM: MMU: Add link_shadow_page() helper Avi Kivity
2010-07-12  4:58   ` Xiao Guangrong
2010-07-11 15:42 ` [PATCH 2/7] KVM: MMU: Use __set_spte to link shadow pages Avi Kivity
2010-07-12  4:58   ` Xiao Guangrong
2010-07-11 15:42 ` [PATCH 3/7] KVM: MMU: Add drop_spte_if_large() helper Avi Kivity
2010-07-12  4:59   ` Xiao Guangrong
2010-07-11 15:42 ` [PATCH 4/7] KVM: MMU: Add validate_direct_spte() helper Avi Kivity
2010-07-12  5:00   ` Xiao Guangrong
2010-07-11 15:42 ` [PATCH 5/7] KVM: MMU: Add validate_indirect_spte() helper Avi Kivity
2010-07-12  5:01   ` Xiao Guangrong
2010-07-12  5:12   ` Xiao Guangrong
2010-07-12  8:50     ` Avi Kivity
2010-07-11 15:42 ` [PATCH 6/7] KVM: MMU: Simplify spte fetch() function Avi Kivity
2010-07-11 16:08   ` Avi Kivity
2010-07-12  5:10     ` Xiao Guangrong
2010-07-12  8:52       ` Avi Kivity
2010-07-12  5:08   ` Xiao Guangrong
2010-07-12  8:53     ` Avi Kivity
2010-07-11 15:42 ` [PATCH 7/7] KVM: MMU: Validate all gptes during fetch, not just those used for new pages Avi Kivity
2010-07-12  5:10   ` Xiao Guangrong
2010-07-12  8:58   ` Avi Kivity [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=4C3AD94F.8030809@redhat.com \
    --to=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    --cc=xiaoguangrong@cn.fujitsu.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