All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marcelo Tosatti <mtosatti@redhat.com>
To: kvm@vger.kernel.org
Cc: avi@redhat.com, aarcange@redhat.com,
	Marcelo Tosatti <mtosatti@redhat.com>
Subject: [patch 3/4] KVM: MMU: flush TLBs on writable -> read-only spte overwrite
Date: Fri, 22 Oct 2010 14:18:17 -0200	[thread overview]
Message-ID: <20101022162031.499103636@redhat.com> (raw)
In-Reply-To: 20101022161814.098714474@redhat.com

[-- Attachment #1: mmu-flush-tlb-on-overwrite --]
[-- Type: text/plain, Size: 1074 bytes --]

This can happen in the following scenario:

vcpu0			vcpu1
read fault
gup(.write=0)		
			gup(.write=1)
			reuse swap cache, no COW
			set writable spte
			use writable spte
set read-only spte

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Index: kvm/arch/x86/kvm/mmu.c
===================================================================
--- kvm.orig/arch/x86/kvm/mmu.c
+++ kvm/arch/x86/kvm/mmu.c
@@ -2068,6 +2068,16 @@ static void mmu_set_spte(struct kvm_vcpu
 				 spte_to_pfn(*sptep), pfn);
 			drop_spte(vcpu->kvm, sptep, shadow_trap_nonpresent_pte);
 			kvm_flush_remote_tlbs(vcpu->kvm);
+		/*
+ 		 * If we overwrite a writable spte with a read-only one,
+ 		 * drop it and flush remote TLBs. Otherwise rmap_write_protect
+ 		 * will find a read-only spte, even though the writable spte
+ 		 * might be cached on a CPU's TLB.
+ 		 */
+		} else if (is_writable_pte(*sptep) &&
+			  (!(pte_access & ACC_WRITE_MASK) || !dirty)) {
+			drop_spte(vcpu->kvm, sptep, shadow_trap_nonpresent_pte);
+			kvm_flush_remote_tlbs(vcpu->kvm);
 		} else
 			was_rmapped = 1;
 	}



  parent reply	other threads:[~2010-10-22 16:23 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-10-19 16:26 [patch 0/3] [RFC] support read-only mappings Marcelo Tosatti
2010-10-19 16:26 ` [patch 1/3] KVM: VMX: remove setting of shadow_base_ptes for EPT Marcelo Tosatti
2010-10-20 10:24   ` Avi Kivity
2010-10-19 16:26 ` [patch 2/3] KVM: MMU: flush TLBs on writable -> read-only spte overwrite Marcelo Tosatti
2010-10-19 16:26 ` [patch 3/3] KVM: propagate fault r/w information to gup(), allow read-only mappings Marcelo Tosatti
2010-10-20 10:36   ` Avi Kivity
2010-10-22 16:18 ` [patch 0/4] support read-only mappings (v2) Marcelo Tosatti
2010-10-22 16:18   ` [patch 1/4] KVM: VMX: remove setting of shadow_base_ptes for EPT Marcelo Tosatti
2010-10-22 16:18   ` [patch 2/4] KVM: MMU: remove kvm_mmu_set_base_ptes Marcelo Tosatti
2010-10-22 16:18   ` Marcelo Tosatti [this message]
2010-10-22 16:18   ` [patch 4/4] KVM: propagate fault r/w information to gup(), allow read-only memory Marcelo Tosatti
2010-10-27  9:20   ` [patch 0/4] support read-only mappings (v2) 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=20101022162031.499103636@redhat.com \
    --to=mtosatti@redhat.com \
    --cc=aarcange@redhat.com \
    --cc=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    /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.