Skip to contentSkip to conversation

FAILED: patch "[PATCH] KVM: x86/mmu: Fix struct guest_walker arrays for 5-level" failed to apply to 5.4-stable tree

PATCH [email protected] 3 participants 3 messages Feb 17, 2020 – Feb 18, 2020

Subsystems (3)
Stable branch x86 KVM x86
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 5.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From f6ab0107a4942dbf9a5cf0cca3f37e184870a360 Mon Sep 17 00:00:00 2001
From: Sean Christopherson <[email protected]>
Date: Fri, 7 Feb 2020 09:37:42 -0800
Subject: [PATCH] KVM: x86/mmu: Fix struct guest_walker arrays for 5-level
paging

Define PT_MAX_FULL_LEVELS as PT64_ROOT_MAX_LEVEL, i.e. 5, to fix shadow paging for 5-level guest page tables. PT_MAX_FULL_LEVELS is used to size the arrays that track guest pages table information, i.e. using a "max levels" of 4 causes KVM to access garbage beyond the end of an array when querying state for level 5 entries. E.g. FNAME(gpte_changed) will read garbage and most likely return %true for a level 5 entry, soft-hanging the guest because FNAME(fetch) will restart the guest instead of creating SPTEs because it thinks the guest PTE has changed.

Note, KVM doesn't yet support 5-level nested EPT, so PT_MAX_FULL_LEVELS gets to stay "4" for the PTTYPE_EPT case.

Fixes 855feb673640 ("KVM: MMU: Add 5 level EPT & Shadow page table support.")
Cc [email protected]
Signed-off-by Sean Christopherson <[email protected]>
Signed-off-by Paolo Bonzini <[email protected]>
arch/x86/kvm/mmu/paging_tmpl.h +1 -1
index 4e1ef0473663..e4c8a4cbf407 100644
@@ -33,7 +33,7 @@
 	#define PT_GUEST_ACCESSED_SHIFT PT_ACCESSED_SHIFT
 	#define PT_HAVE_ACCESSED_DIRTY(mmu) true
 	#ifdef CONFIG_X86_64
-	#define PT_MAX_FULL_LEVELS 4
+	#define PT_MAX_FULL_LEVELS PT64_ROOT_MAX_LEVEL
 	#define CMPXCHG cmpxchg
 	#else
 	#define CMPXCHG cmpxchg64
maintainer
#1 lore raw
Quoted 27 lines · On Mon, Feb 17, 2020 at 08:14:31PM +0100, [email protected] wrote:

The patch below does not apply to the 5.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From f6ab0107a4942dbf9a5cf0cca3f37e184870a360 Mon Sep 17 00:00:00 2001
From: Sean Christopherson <[email protected]>
Date: Fri, 7 Feb 2020 09:37:42 -0800
Subject: [PATCH] KVM: x86/mmu: Fix struct guest_walker arrays for 5-level paging

Define PT_MAX_FULL_LEVELS as PT64_ROOT_MAX_LEVEL, i.e. 5, to fix shadow paging for 5-level guest page tables. PT_MAX_FULL_LEVELS is used to size the arrays that track guest pages table information, i.e. using a "max levels" of 4 causes KVM to access garbage beyond the end of an array when querying state for level 5 entries. E.g. FNAME(gpte_changed) will read garbage and most likely return %true for a level 5 entry, soft-hanging the guest because FNAME(fetch) will restart the guest instead of creating SPTEs because it thinks the guest PTE has changed.

Quoted 7 lines

Note, KVM doesn't yet support 5-level nested EPT, so PT_MAX_FULL_LEVELS gets to stay "4" for the PTTYPE_EPT case.

Fixes 855feb673640 ("KVM: MMU: Add 5 level EPT & Shadow page table support.")
Cc [email protected]
Signed-off-by Sean Christopherson <[email protected]>
Signed-off-by Paolo Bonzini <[email protected]>

File name is different in 5.4 and 4.19. Fixed and queued up for both.

Signature · 2 lines
Thanks,
Sasha
#2 lore raw
Quoted 31 lines · Sasha Levin
Quoted 30 lines · On Mon, Feb 17, 2020 at 08:14:31PM +0100, [email protected] wrote:

The patch below does not apply to the 5.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From f6ab0107a4942dbf9a5cf0cca3f37e184870a360 Mon Sep 17 00:00:00 2001
From: Sean Christopherson <[email protected]>
Date: Fri, 7 Feb 2020 09:37:42 -0800
Subject: [PATCH] KVM: x86/mmu: Fix struct guest_walker arrays for 5-level paging

Define PT_MAX_FULL_LEVELS as PT64_ROOT_MAX_LEVEL, i.e. 5, to fix shadow paging for 5-level guest page tables. PT_MAX_FULL_LEVELS is used to size the arrays that track guest pages table information, i.e. using a "max levels" of 4 causes KVM to access garbage beyond the end of an array when querying state for level 5 entries. E.g. FNAME(gpte_changed) will read garbage and most likely return %true for a level 5 entry, soft-hanging the guest because FNAME(fetch) will restart the guest instead of creating SPTEs because it thinks the guest PTE has changed.

Note, KVM doesn't yet support 5-level nested EPT, so PT_MAX_FULL_LEVELS gets to stay "4" for the PTTYPE_EPT case.

Quoted 6 lines
Quoted 4 lines
Fixes 855feb673640 ("KVM: MMU: Add 5 level EPT & Shadow page table support.")
Cc [email protected]
Signed-off-by Sean Christopherson <[email protected]>
Signed-off-by Paolo Bonzini <[email protected]>

File name is different in 5.4 and 4.19. Fixed and queued up for both.

Thanks! Backporting the non-trivial conflicts is on my todo list, unfortunately my todo list is rather long right now...