From: Christoffer Dall <cdall@kernel.org>
To: kvmarm@lists.cs.columbia.edu,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Radim Krčmář" <rkrcmar@redhat.com>
Cc: linux-arm-kernel@lists.infradead.org, kvm@vger.kernel.org,
Marc Zyngier <marc.zyngier@arm.com>,
stable@vger.kernel.org,
Kristina Martsenko <kristina.martsenko@arm.com>
Subject: [PULL 08/19] arm64: KVM: fix VTTBR_BADDR_MASK BUG_ON off-by-one
Date: Mon, 4 Dec 2017 15:03:37 +0100 [thread overview]
Message-ID: <20171204140348.21965-6-cdall@kernel.org> (raw)
In-Reply-To: <20171204135637.21620-1-cdall@kernel.org>
From: Kristina Martsenko <kristina.martsenko@arm.com>
VTTBR_BADDR_MASK is used to sanity check the size and alignment of the
VTTBR address. It seems to currently be off by one, thereby only
allowing up to 47-bit addresses (instead of 48-bit) and also
insufficiently checking the alignment. This patch fixes it.
As an example, with 4k pages, before this patch we have:
PHYS_MASK_SHIFT = 48
VTTBR_X = 37 - 24 = 13
VTTBR_BADDR_SHIFT = 13 - 1 = 12
VTTBR_BADDR_MASK = ((1 << 35) - 1) << 12 = 0x00007ffffffff000
Which is wrong, because the mask doesn't allow bit 47 of the VTTBR
address to be set, and only requires the address to be 12-bit (4k)
aligned, while it actually needs to be 13-bit (8k) aligned because we
concatenate two 4k tables.
With this patch, the mask becomes 0x0000ffffffffe000, which is what we
want.
Fixes: 0369f6a34b9f ("arm64: KVM: EL2 register definitions")
Cc: <stable@vger.kernel.org> # 3.11.x
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Kristina Martsenko <kristina.martsenko@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
---
arch/arm64/include/asm/kvm_arm.h | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 7f069ff37f06..715d395ef45b 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -170,8 +170,7 @@
#define VTCR_EL2_FLAGS (VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
#define VTTBR_X (VTTBR_X_TGRAN_MAGIC - VTCR_EL2_T0SZ_IPA)
-#define VTTBR_BADDR_SHIFT (VTTBR_X - 1)
-#define VTTBR_BADDR_MASK (((UL(1) << (PHYS_MASK_SHIFT - VTTBR_X)) - 1) << VTTBR_BADDR_SHIFT)
+#define VTTBR_BADDR_MASK (((UL(1) << (PHYS_MASK_SHIFT - VTTBR_X)) - 1) << VTTBR_X)
#define VTTBR_VMID_SHIFT (UL(48))
#define VTTBR_VMID_MASK(size) (_AT(u64, (1 << size) - 1) << VTTBR_VMID_SHIFT)
--
2.14.2
WARNING: multiple messages have this Message-ID (diff)
From: cdall@kernel.org (Christoffer Dall)
To: linux-arm-kernel@lists.infradead.org
Subject: [PULL 08/19] arm64: KVM: fix VTTBR_BADDR_MASK BUG_ON off-by-one
Date: Mon, 4 Dec 2017 15:03:37 +0100 [thread overview]
Message-ID: <20171204140348.21965-6-cdall@kernel.org> (raw)
In-Reply-To: <20171204135637.21620-1-cdall@kernel.org>
From: Kristina Martsenko <kristina.martsenko@arm.com>
VTTBR_BADDR_MASK is used to sanity check the size and alignment of the
VTTBR address. It seems to currently be off by one, thereby only
allowing up to 47-bit addresses (instead of 48-bit) and also
insufficiently checking the alignment. This patch fixes it.
As an example, with 4k pages, before this patch we have:
PHYS_MASK_SHIFT = 48
VTTBR_X = 37 - 24 = 13
VTTBR_BADDR_SHIFT = 13 - 1 = 12
VTTBR_BADDR_MASK = ((1 << 35) - 1) << 12 = 0x00007ffffffff000
Which is wrong, because the mask doesn't allow bit 47 of the VTTBR
address to be set, and only requires the address to be 12-bit (4k)
aligned, while it actually needs to be 13-bit (8k) aligned because we
concatenate two 4k tables.
With this patch, the mask becomes 0x0000ffffffffe000, which is what we
want.
Fixes: 0369f6a34b9f ("arm64: KVM: EL2 register definitions")
Cc: <stable@vger.kernel.org> # 3.11.x
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Kristina Martsenko <kristina.martsenko@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
---
arch/arm64/include/asm/kvm_arm.h | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 7f069ff37f06..715d395ef45b 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -170,8 +170,7 @@
#define VTCR_EL2_FLAGS (VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
#define VTTBR_X (VTTBR_X_TGRAN_MAGIC - VTCR_EL2_T0SZ_IPA)
-#define VTTBR_BADDR_SHIFT (VTTBR_X - 1)
-#define VTTBR_BADDR_MASK (((UL(1) << (PHYS_MASK_SHIFT - VTTBR_X)) - 1) << VTTBR_BADDR_SHIFT)
+#define VTTBR_BADDR_MASK (((UL(1) << (PHYS_MASK_SHIFT - VTTBR_X)) - 1) << VTTBR_X)
#define VTTBR_VMID_SHIFT (UL(48))
#define VTTBR_VMID_MASK(size) (_AT(u64, (1 << size) - 1) << VTTBR_VMID_SHIFT)
--
2.14.2
WARNING: multiple messages have this Message-ID (diff)
From: Christoffer Dall <cdall@kernel.org>
To: kvmarm@lists.cs.columbia.edu,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Radim Krčmář" <rkrcmar@redhat.com>
Cc: linux-arm-kernel@lists.infradead.org, kvm@vger.kernel.org,
Marc Zyngier <marc.zyngier@arm.com>,
Kristina Martsenko <kristina.martsenko@arm.com>,
stable@vger.kernel.org,
Christoffer Dall <christoffer.dall@linaro.org>
Subject: [PULL 08/19] arm64: KVM: fix VTTBR_BADDR_MASK BUG_ON off-by-one
Date: Mon, 4 Dec 2017 15:03:37 +0100 [thread overview]
Message-ID: <20171204140348.21965-6-cdall@kernel.org> (raw)
In-Reply-To: <20171204135637.21620-1-cdall@kernel.org>
From: Kristina Martsenko <kristina.martsenko@arm.com>
VTTBR_BADDR_MASK is used to sanity check the size and alignment of the
VTTBR address. It seems to currently be off by one, thereby only
allowing up to 47-bit addresses (instead of 48-bit) and also
insufficiently checking the alignment. This patch fixes it.
As an example, with 4k pages, before this patch we have:
PHYS_MASK_SHIFT = 48
VTTBR_X = 37 - 24 = 13
VTTBR_BADDR_SHIFT = 13 - 1 = 12
VTTBR_BADDR_MASK = ((1 << 35) - 1) << 12 = 0x00007ffffffff000
Which is wrong, because the mask doesn't allow bit 47 of the VTTBR
address to be set, and only requires the address to be 12-bit (4k)
aligned, while it actually needs to be 13-bit (8k) aligned because we
concatenate two 4k tables.
With this patch, the mask becomes 0x0000ffffffffe000, which is what we
want.
Fixes: 0369f6a34b9f ("arm64: KVM: EL2 register definitions")
Cc: <stable@vger.kernel.org> # 3.11.x
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Kristina Martsenko <kristina.martsenko@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
---
arch/arm64/include/asm/kvm_arm.h | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 7f069ff37f06..715d395ef45b 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -170,8 +170,7 @@
#define VTCR_EL2_FLAGS (VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
#define VTTBR_X (VTTBR_X_TGRAN_MAGIC - VTCR_EL2_T0SZ_IPA)
-#define VTTBR_BADDR_SHIFT (VTTBR_X - 1)
-#define VTTBR_BADDR_MASK (((UL(1) << (PHYS_MASK_SHIFT - VTTBR_X)) - 1) << VTTBR_BADDR_SHIFT)
+#define VTTBR_BADDR_MASK (((UL(1) << (PHYS_MASK_SHIFT - VTTBR_X)) - 1) << VTTBR_X)
#define VTTBR_VMID_SHIFT (UL(48))
#define VTTBR_VMID_MASK(size) (_AT(u64, (1 << size) - 1) << VTTBR_VMID_SHIFT)
--
2.14.2
next prev parent reply other threads:[~2017-12-04 14:00 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-04 13:56 [PULL 00/19] KVM/ARM Fixes for v4.15 Christoffer Dall
2017-12-04 13:56 ` Christoffer Dall
2017-12-04 13:56 ` [PULL 01/19] KVM: arm/arm64: Don't enable/disable physical timer access on VHE Christoffer Dall
2017-12-04 13:56 ` Christoffer Dall
2017-12-04 13:56 ` [PULL 02/19] KVM: arm/arm64: VGIC: extend !vgic_is_initialized guard Christoffer Dall
2017-12-04 13:56 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 03/19] KVM: arm/arm64: vgic-irqfd: Fix MSI entry allocation Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 04/19] KVM: arm/arm64: vgic: Preserve the revious read from the pending table Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 05/19] KVM: arm/arm64: vgic-its: " Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 06/19] KVM: arm/arm64: vgic-its: Check result of allocation before use Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 07/19] KVM: arm/arm64: vgic-v4: Only perform an unmap for valid vLPIs Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall [this message]
2017-12-04 14:03 ` [PULL 08/19] arm64: KVM: fix VTTBR_BADDR_MASK BUG_ON off-by-one Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 09/19] arm: KVM: Fix " Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 10/19] KVM: arm/arm64: debug: Introduce helper for single-step Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 11/19] kvm: arm64: handle single-stepping trapped instructions Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 12/19] kvm: arm64: handle single-step of userspace mmio instructions Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 13/19] kvm: arm64: handle single-step during SError exceptions Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 14/19] kvm: arm64: handle single-step of hyp emulated mmio instructions Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 15/19] KVM: arm/arm64: Avoid attempting to load timer vgic state without a vgic Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 16/19] kvm: arm: don't treat unavailable HYP mode as an error Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 17/19] KVM: arm/arm64: Fix spinlock acquisition in vgic_set_owner Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 18/19] KVM: arm/arm64: kvm_arch_destroy_vm cleanups Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` [PULL 19/19] KVM: arm/arm64: Fix broken GICH_ELRSR big endian conversion Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-04 14:03 ` Christoffer Dall
2017-12-05 17:59 ` [PULL 00/19] KVM/ARM Fixes for v4.15 Radim Krčmář
2017-12-05 17:59 ` Radim Krčmář
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=20171204140348.21965-6-cdall@kernel.org \
--to=cdall@kernel.org \
--cc=kristina.martsenko@arm.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=marc.zyngier@arm.com \
--cc=pbonzini@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=stable@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.