[linux] 02/02: [x86] KVM: MMU: always terminate page walks at level 1 (CVE-2017-12188)

[debian-kernel at lists.debian.org]

This is an automated email from the git hooks/post-receive script.

carnil pushed a commit to branch sid
in repository linux.

commit ccefd718c8b67a6697b516af82de4abbdf0d4102
Author: Salvatore Bonaccorso <carnil at debian.org>
Date:   Fri Oct 13 18:09:11 2017 +0200

    [x86] KVM: MMU: always terminate page walks at level 1 (CVE-2017-12188)
---
 debian/changelog                                   |  1 +
 ...MU-always-terminate-page-walks-at-level-1.patch | 83 ++++++++++++++++++++++
 debian/patches/series                              |  1 +
 3 files changed, 85 insertions(+)

diff --git a/debian/changelog b/debian/changelog
index aef371e..b53f324 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -16,6 +16,7 @@ linux (4.13.4-2) UNRELEASED; urgency=medium
   * ALSA: seq: Fix use-after-free at creating a port (CVE-2017-15265)
   * [x86] KVM: nVMX: update last_nonleaf_level when initializing nested EPT
     (CVE-2017-12188)
+  * [x86] KVM: MMU: always terminate page walks at level 1 (CVE-2017-12188)
 
  -- Ben Hutchings <ben at decadent.org.uk>  Wed, 04 Oct 2017 23:14:54 +0100
 
diff --git a/debian/patches/bugfix/x86/KVM-MMU-always-terminate-page-walks-at-level-1.patch b/debian/patches/bugfix/x86/KVM-MMU-always-terminate-page-walks-at-level-1.patch
new file mode 100644
index 0000000..47cbead
--- /dev/null
+++ b/debian/patches/bugfix/x86/KVM-MMU-always-terminate-page-walks-at-level-1.patch
@@ -0,0 +1,83 @@
+From: Ladi Prosek <lprosek at redhat.com>
+Date: Thu, 5 Oct 2017 11:10:23 +0200
+Subject: KVM: MMU: always terminate page walks at level 1
+Origin: https://git.kernel.org/linus/829ee279aed43faa5cb1e4d65c0cad52f2426c53
+Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-12188
+
+is_last_gpte() is not equivalent to the pseudo-code given in commit
+6bb69c9b69c31 ("KVM: MMU: simplify last_pte_bitmap") because an incorrect
+value of last_nonleaf_level may override the result even if level == 1.
+
+It is critical for is_last_gpte() to return true on level == 1 to
+terminate page walks. Otherwise memory corruption may occur as level
+is used as an index to various data structures throughout the page
+walking code.  Even though the actual bug would be wherever the MMU is
+initialized (as in the previous patch), be defensive and ensure here
+that is_last_gpte() returns the correct value.
+
+This patch is also enough to fix CVE-2017-12188.
+
+Fixes: 6bb69c9b69c315200ddc2bc79aee14c0184cf5b2
+Cc: stable at vger.kernel.org
+Cc: Andy Honig <ahonig at google.com>
+Signed-off-by: Ladi Prosek <lprosek at redhat.com>
+[Panic if walk_addr_generic gets an incorrect level; this is a serious
+ bug and it's not worth a WARN_ON where the recovery path might hide
+ further exploitable issues; suggested by Andrew Honig. - Paolo]
+Signed-off-by: Paolo Bonzini <pbonzini at redhat.com>
+---
+ arch/x86/kvm/mmu.c         | 14 +++++++-------
+ arch/x86/kvm/paging_tmpl.h |  3 ++-
+ 2 files changed, 9 insertions(+), 8 deletions(-)
+
+diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
+index 3c25f20115bc..7a69cf053711 100644
+--- a/arch/x86/kvm/mmu.c
++++ b/arch/x86/kvm/mmu.c
+@@ -3974,19 +3974,19 @@ static inline bool is_last_gpte(struct kvm_mmu *mmu,
+ 				unsigned level, unsigned gpte)
+ {
+ 	/*
+-	 * PT_PAGE_TABLE_LEVEL always terminates.  The RHS has bit 7 set
+-	 * iff level <= PT_PAGE_TABLE_LEVEL, which for our purpose means
+-	 * level == PT_PAGE_TABLE_LEVEL; set PT_PAGE_SIZE_MASK in gpte then.
+-	 */
+-	gpte |= level - PT_PAGE_TABLE_LEVEL - 1;
+-
+-	/*
+ 	 * The RHS has bit 7 set iff level < mmu->last_nonleaf_level.
+ 	 * If it is clear, there are no large pages at this level, so clear
+ 	 * PT_PAGE_SIZE_MASK in gpte if that is the case.
+ 	 */
+ 	gpte &= level - mmu->last_nonleaf_level;
+ 
++	/*
++	 * PT_PAGE_TABLE_LEVEL always terminates.  The RHS has bit 7 set
++	 * iff level <= PT_PAGE_TABLE_LEVEL, which for our purpose means
++	 * level == PT_PAGE_TABLE_LEVEL; set PT_PAGE_SIZE_MASK in gpte then.
++	 */
++	gpte |= level - PT_PAGE_TABLE_LEVEL - 1;
++
+ 	return gpte & PT_PAGE_SIZE_MASK;
+ }
+ 
+diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
+index 86b68dc5a649..f18d1f8d332b 100644
+--- a/arch/x86/kvm/paging_tmpl.h
++++ b/arch/x86/kvm/paging_tmpl.h
+@@ -334,10 +334,11 @@ static int FNAME(walk_addr_generic)(struct guest_walker *walker,
+ 		--walker->level;
+ 
+ 		index = PT_INDEX(addr, walker->level);
+-
+ 		table_gfn = gpte_to_gfn(pte);
+ 		offset    = index * sizeof(pt_element_t);
+ 		pte_gpa   = gfn_to_gpa(table_gfn) + offset;
++
++		BUG_ON(walker->level < 1);
+ 		walker->table_gfn[walker->level - 1] = table_gfn;
+ 		walker->pte_gpa[walker->level - 1] = pte_gpa;
+ 
+-- 
+2.11.0
+
diff --git a/debian/patches/series b/debian/patches/series
index 09ca221..408d183 100644
--- a/debian/patches/series
+++ b/debian/patches/series
@@ -127,6 +127,7 @@ bugfix/all/KEYS-prevent-KEYCTL_READ-on-negative-key.patch
 bugfix/all/waitid-Add-missing-access_ok-checks.patch
 bugfix/all/ALSA-seq-Fix-use-after-free-at-creating-a-port.patch
 bugfix/x86/KVM-nVMX-update-last_nonleaf_level-when-initializing.patch
+bugfix/x86/KVM-MMU-always-terminate-page-walks-at-level-1.patch
 
 # Fix exported symbol versions
 bugfix/alpha/alpha-restore-symbol-versions-for-symbols-exported-f.patch

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/kernel/linux.git