linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA
@ 2018-09-17 10:41 Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 01/18] kvm: arm/arm64: Fix stage2_flush_memslot for 4 level page table Suzuki K Poulose
                   ` (21 more replies)
  0 siblings, 22 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose


The physical address space size for a VM (IPA size) on arm/arm64 is
limited to a static limit of 40bits. This series adds support for
using an IPA size specific to a VM, allowing to use a size supported
by the host (based on the host kernel configuration and CPU support).
The default size is fixed to 40bits. On arm64, we can allow the limit
to be lowered (limiting the number of levels in stage2 to 2, to prevent
splitting the host PMD huge pages at stage2). We also add support for
handling 52bit IPA addresses (where supported) added by Arm v8.2
extensions.

We need to set the IPA limit as early as the VM creation to keep the
code simpler to avoid sprinkling checks everywhere to ensure that the
IPA is configured. We encode the IPA size in the machine_type
argument to KVM_CREATE_VM ioctl. Bits [7-0] of the type are reserved
for the IPA size. The availability of this feature is advertised by a
new cap KVM_CAP_ARM_VM_PHYS_SHIFT. When supported, this capability
returns the maximum IPA shift supported by the host. The supported IPA
size on a host could be different from the system's PARange indicated
by the CPUs (e.g, kernel limit on the PA size).

Supporting different IPA size requires modification to the stage2 page
table code. The arm64 page table level helpers are defined based on the
page table levels used by the host VA. So, the accessors may not work
if the guest uses more number of levels in stage2 than the stage1
of the host.  The previous versions (v1 & v2) of this series refactored
the stage1 page table accessors to reuse the low-level accessors for an
independent stage2 table. However, due to the level folding in the
generic code, the types are redefined as well. i.e, if the PUD is
folded, the pud_t could be defined as :

 typedef struct { pgd_t pgd; } pud_t;

similarly for pmd_t.  So, without stage1 independent page table entry
types for stage2, we could be dealing with a different type for level
 0-2 entries. This is practically fine on arm/arm64 as the entries
have similar format and size and we always use the appropriate
accessors to get the raw value (i.e, pud_val/pmd_val etc). But not
ideal for a solution upstream. So, this version caps the stage2 page
table levels to that of the stage1. This has the following impact on
the IPA support for various pagesize/host-va combinations :


x-----------------------------------------------------x
| host\ipa    | 40bit | 42bit | 44bit | 48bit | 52bit |
-------------------------------------------------------
| 39bit-4K    |  y    |   y   |  n    |   n   |  n/a  |
-------------------------------------------------------
| 48bit-4K    |  y    |   y   |  y    |   y   |  n/a  |
-------------------------------------------------------
| 36bit-16K   |  y    |   n   |  n    |   n   |  n/a  |
-------------------------------------------------------
| 47bit-16K   |  y    |   y   |  y    |   y   |  n/a  |
-------------------------------------------------------
| 48bit-4K    |  y    |   y   |  y    |   y   |  n/a  |
-------------------------------------------------------
| 42bit-64K   |  y    |   y   |  y    |   n   |  n    |
-------------------------------------------------------
| 48bit-64K   |  y    |   y   |  y    |   y   |  y    |
x-----------------------------------------------------x

Or the following list shows what cannot be supported :

 39bit-4K host  | [44 - 48]
 36bit-16K host | [41 - 48]
 42bit-64K host | [47 - 52]

which is not really bad. We can pursue the independent stage2
page table support and lift the restriction once we get there.
Given there is a proposal for new generic page table walker [0],
it would make sense to make our efforts in sync with it to avoid
diverting from a common API.

52bit support is added for VGIC (including ITS emulation) and handling
of PAR, HPFAR registers.

The series applies on 4.19-rc3. A tree is available here:

	 git://linux-arm.org/linux-skp.git ipa52/v5

Tested with
  - Modified kvmtool, which can only be used for (patches included in
    the series for reference / testing):
    * with virtio-pci upto 44bit PA (Due to 4K page size for virtio-pci
      legacy implemented by kvmtool)
    * Upto 48bit PA with virtio-mmio, due to 32bit PFN limitation.
  - Hacked Qemu (boot loader support for highmem, phys-shift support)
    * with virtio-pci GIC-v3 ITS & MSI upto 52bit on Foundation model.
    Also see [1] for Qemu support.

[0] https://lkml.org/lkml/2018/4/24/777
[1] https://lists.gnu.org/archive/html/qemu-devel/2018-06/msg05759.html

Changes since V4:
 - Rebased on v4.19-rc3
 - Dropped virtio patches queued already by mst.
 - Collect Acks from Christoffer
 - Restrict IPA configuration support to arm64 only
 - Use KVM_CAP_ARM_VM_PHYS_SHIFT for detecting the support for
   IPA size configuration along with the limit on the IPA for the host.
 - Update comments on __load_guest_stage2
 - Add comment about the default value for unknown PARange values.
 - Update Documentation of the API

Changes since V3:
 - Use per-VM VTCR instead per-VM private VTCR bits
 - Allow IPA less than 40bits
 - Split the patch adding support for stage2 dynamic page tables
 - Rearrange the series to keep the userspace API at the end, which
   needs further discussion.
 - Collect Reviews/Acks from Eric & Marc

Changes since V2:
 - Drop "refactoring of host page table helpers" and restrict the IPA size
   to make sure stage2 doesn't use more page table levels than that of the host.
 - Load VTCR for TLB operations on behalf of the VM (Pointed-by: James Morse)
 - Split a couple of patches to make them easier to review.
 - Fall back to normal (non-concatenated) entry level page table support if
   possible.
 - Bump the IOCTL number

Changes since V1:
 - Change the userspace API for configuring VM to encode the IPA
   size in the VM type.  (suggested by Christoffer)
 - Expose the IPA limit on the host via ioctl on /dev/kvm
 - Handle 52bit addresses in PAR & HPFAR
 - Drop patch changing the life time of stage2 PGD
 - Rename macros for 48-to-52 bit conversion for GIC ITS BASER.
   (suggested by Christoffer)
 - Split virtio PFN check patches and address comments.

Kristina Martsenko (1):
  vgic: Add support for 52bit guest physical address

Suzuki K Poulose (17):
  kvm: arm/arm64: Fix stage2_flush_memslot for 4 level page table
  kvm: arm/arm64: Remove spurious WARN_ON
  kvm: arm64: Add helper for loading the stage2 setting for a VM
  arm64: Add a helper for PARange to physical shift conversion
  kvm: arm64: Clean up VTCR_EL2 initialisation
  kvm: arm/arm64: Allow arch specific configurations for VM
  kvm: arm64: Configure VTCR_EL2 per VM
  kvm: arm/arm64: Prepare for VM specific stage2 translations
  kvm: arm64: Prepare for dynamic stage2 page table layout
  kvm: arm64: Make stage2 page table layout dynamic
  kvm: arm64: Dynamic configuration of VTTBR mask
  kvm: arm64: Configure VTCR_EL2.SL0 per VM
  kvm: arm64: Switch to per VM IPA limit
  kvm: arm64: Add 52bit support for PAR to HPFAR conversoin
  kvm: arm64: Set a limit on the IPA size
  kvm: arm64: Limit the minimum number of page table levels
  kvm: arm64: Allow tuning the physical address size for VM

 Documentation/virtual/kvm/api.txt             |   8 +
 arch/arm/include/asm/kvm_arm.h                |   3 +-
 arch/arm/include/asm/kvm_host.h               |   7 +
 arch/arm/include/asm/kvm_mmu.h                |  15 +-
 arch/arm/include/asm/stage2_pgtable.h         |  50 ++--
 arch/arm64/include/asm/cpufeature.h           |  20 ++
 arch/arm64/include/asm/kvm_arm.h              | 130 ++++++++--
 arch/arm64/include/asm/kvm_asm.h              |   2 -
 arch/arm64/include/asm/kvm_host.h             |  16 +-
 arch/arm64/include/asm/kvm_hyp.h              |  10 +
 arch/arm64/include/asm/kvm_mmu.h              |  42 +++-
 arch/arm64/include/asm/stage2_pgtable-nopmd.h |  42 ----
 arch/arm64/include/asm/stage2_pgtable-nopud.h |  39 ---
 arch/arm64/include/asm/stage2_pgtable.h       | 237 +++++++++++++-----
 arch/arm64/kvm/hyp/Makefile                   |   1 -
 arch/arm64/kvm/hyp/s2-setup.c                 |  90 -------
 arch/arm64/kvm/hyp/switch.c                   |   4 +-
 arch/arm64/kvm/hyp/tlb.c                      |   4 +-
 arch/arm64/kvm/reset.c                        | 105 ++++++++
 include/linux/irqchip/arm-gic-v3.h            |   5 +
 include/uapi/linux/kvm.h                      |  10 +
 virt/kvm/arm/arm.c                            |   9 +-
 virt/kvm/arm/mmu.c                            | 120 ++++-----
 virt/kvm/arm/vgic/vgic-its.c                  |  36 +--
 virt/kvm/arm/vgic/vgic-kvm-device.c           |   2 +-
 virt/kvm/arm/vgic/vgic-mmio-v3.c              |   2 -
 26 files changed, 617 insertions(+), 392 deletions(-)
 delete mode 100644 arch/arm64/include/asm/stage2_pgtable-nopmd.h
 delete mode 100644 arch/arm64/include/asm/stage2_pgtable-nopud.h
 delete mode 100644 arch/arm64/kvm/hyp/s2-setup.c

kvmtool changes:

Suzuki K Poulose (4):
  kvmtool: Allow backends to run checks on the KVM device fd
  kvmtool: arm64: Add support for guest physical address size
  kvmtool: arm64: Switch memory layout
  kvmtool: arm: Add support for creating VM with PA size

 arm/aarch32/include/kvm/kvm-arch.h        |  6 +++--
 arm/aarch64/include/kvm/kvm-arch.h        | 15 +++++++++---
 arm/aarch64/include/kvm/kvm-config-arch.h |  5 +++-
 arm/include/arm-common/kvm-arch.h         | 17 ++++++++-----
 arm/include/arm-common/kvm-config-arch.h  |  1 +
 arm/kvm.c                                 | 29 ++++++++++++++++++++++-
 include/kvm/kvm.h                         |  4 ++++
 kvm.c                                     |  2 ++
 8 files changed, 66 insertions(+), 13 deletions(-)

-- 
2.19.0


^ permalink raw reply	[flat|nested] 50+ messages in thread

* [PATCH v5 01/18] kvm: arm/arm64: Fix stage2_flush_memslot for 4 level page table
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 02/18] kvm: arm/arm64: Remove spurious WARN_ON Suzuki K Poulose
                   ` (20 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

So far we have only supported 3 level page table with fixed IPA of
40bits, where PUD is folded. With 4 level page tables, we need
to check if the PUD entry is valid or not. Fix stage2_flush_memslot()
to do this check, before walking down the table.

Acked-by: Christoffer Dall <cdall@kernel.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 virt/kvm/arm/mmu.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index ed162a6c57c5..ee7ce8fa4a12 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -412,7 +412,8 @@ static void stage2_flush_memslot(struct kvm *kvm,
 	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
 	do {
 		next = stage2_pgd_addr_end(addr, end);
-		stage2_flush_puds(kvm, pgd, addr, next);
+		if (!stage2_pgd_none(*pgd))
+			stage2_flush_puds(kvm, pgd, addr, next);
 	} while (pgd++, addr = next, addr != end);
 }
 
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 02/18] kvm: arm/arm64: Remove spurious WARN_ON
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 01/18] kvm: arm/arm64: Fix stage2_flush_memslot for 4 level page table Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 03/18] kvm: arm64: Add helper for loading the stage2 setting for a VM Suzuki K Poulose
                   ` (19 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

On a 4-level page table pgd entry can be empty, unlike a 3-level
page table. Remove the spurious WARN_ON() in stage_get_pud().

Acked-by: Christoffer Dall <cdall@kernel.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 virt/kvm/arm/mmu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index ee7ce8fa4a12..4a285d760ce0 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -1005,7 +1005,7 @@ static pud_t *stage2_get_pud(struct kvm *kvm, struct kvm_mmu_memory_cache *cache
 	pud_t *pud;
 
 	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
-	if (WARN_ON(stage2_pgd_none(*pgd))) {
+	if (stage2_pgd_none(*pgd)) {
 		if (!cache)
 			return NULL;
 		pud = mmu_memory_cache_alloc(cache);
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 03/18] kvm: arm64: Add helper for loading the stage2 setting for a VM
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 01/18] kvm: arm/arm64: Fix stage2_flush_memslot for 4 level page table Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 02/18] kvm: arm/arm64: Remove spurious WARN_ON Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 04/18] arm64: Add a helper for PARange to physical shift conversion Suzuki K Poulose
                   ` (18 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

We load the stage2 context of a guest for different operations,
including running the guest and tlb maintenance on behalf of the
guest. As of now only the vttbr is private to the guest, but this
is about to change with IPA per VM. Add a helper to load the stage2
configuration for a VM, which could do the right thing with the
future changes.

Cc: Christoffer Dall <cdall@kernel.org>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since v2:
 - New patch
---
 arch/arm64/include/asm/kvm_hyp.h | 9 +++++++++
 arch/arm64/kvm/hyp/switch.c      | 2 +-
 arch/arm64/kvm/hyp/tlb.c         | 4 ++--
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h
index 384c34397619..d1bd1e0f14d7 100644
--- a/arch/arm64/include/asm/kvm_hyp.h
+++ b/arch/arm64/include/asm/kvm_hyp.h
@@ -155,5 +155,14 @@ void deactivate_traps_vhe_put(void);
 u64 __guest_enter(struct kvm_vcpu *vcpu, struct kvm_cpu_context *host_ctxt);
 void __noreturn __hyp_do_panic(unsigned long, ...);
 
+/*
+ * Must be called from hyp code running at EL2 with an updated VTTBR
+ * and interrupts disabled.
+ */
+static __always_inline void __hyp_text __load_guest_stage2(struct kvm *kvm)
+{
+	write_sysreg(kvm->arch.vttbr, vttbr_el2);
+}
+
 #endif /* __ARM64_KVM_HYP_H__ */
 
diff --git a/arch/arm64/kvm/hyp/switch.c b/arch/arm64/kvm/hyp/switch.c
index ca46153d7915..9d5ce1a3039a 100644
--- a/arch/arm64/kvm/hyp/switch.c
+++ b/arch/arm64/kvm/hyp/switch.c
@@ -198,7 +198,7 @@ void deactivate_traps_vhe_put(void)
 
 static void __hyp_text __activate_vm(struct kvm *kvm)
 {
-	write_sysreg(kvm->arch.vttbr, vttbr_el2);
+	__load_guest_stage2(kvm);
 }
 
 static void __hyp_text __deactivate_vm(struct kvm_vcpu *vcpu)
diff --git a/arch/arm64/kvm/hyp/tlb.c b/arch/arm64/kvm/hyp/tlb.c
index 131c7772703c..4dbd9c69a96d 100644
--- a/arch/arm64/kvm/hyp/tlb.c
+++ b/arch/arm64/kvm/hyp/tlb.c
@@ -30,7 +30,7 @@ static void __hyp_text __tlb_switch_to_guest_vhe(struct kvm *kvm)
 	 * bits. Changing E2H is impossible (goodbye TTBR1_EL2), so
 	 * let's flip TGE before executing the TLB operation.
 	 */
-	write_sysreg(kvm->arch.vttbr, vttbr_el2);
+	__load_guest_stage2(kvm);
 	val = read_sysreg(hcr_el2);
 	val &= ~HCR_TGE;
 	write_sysreg(val, hcr_el2);
@@ -39,7 +39,7 @@ static void __hyp_text __tlb_switch_to_guest_vhe(struct kvm *kvm)
 
 static void __hyp_text __tlb_switch_to_guest_nvhe(struct kvm *kvm)
 {
-	write_sysreg(kvm->arch.vttbr, vttbr_el2);
+	__load_guest_stage2(kvm);
 	isb();
 }
 
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 04/18] arm64: Add a helper for PARange to physical shift conversion
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (2 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 03/18] kvm: arm64: Add helper for loading the stage2 setting for a VM Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 05/18] kvm: arm64: Clean up VTCR_EL2 initialisation Suzuki K Poulose
                   ` (17 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

On arm64, ID_AA64MMFR0_EL1.PARange encodes the maximum Physical
Address range supported by the CPU. Add a helper to decode this
to actual physical shift. If we hit an unallocated value, return
the maximum range supported by the kernel.
This will be used by KVM to set the VTCR_EL2.T0SZ, as it
is about to move its place. Having this helper keeps the code
movement cleaner.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since V4:
 - Add a comment on the result for unknown values.
---
 arch/arm64/include/asm/cpufeature.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h
index 1717ba1db35d..072cc1c970c2 100644
--- a/arch/arm64/include/asm/cpufeature.h
+++ b/arch/arm64/include/asm/cpufeature.h
@@ -530,6 +530,26 @@ void arm64_set_ssbd_mitigation(bool state);
 static inline void arm64_set_ssbd_mitigation(bool state) {}
 #endif
 
+static inline u32 id_aa64mmfr0_parange_to_phys_shift(int parange)
+{
+	switch (parange) {
+	case 0: return 32;
+	case 1: return 36;
+	case 2: return 40;
+	case 3: return 42;
+	case 4: return 44;
+	case 5: return 48;
+	case 6: return 52;
+	/*
+	 * A future PE could use a value unknown to the kernel.
+	 * However, by the "D10.1.4 Principles of the ID scheme
+	 * for fields in ID registers", ARM DDI 0487C.a, any new
+	 * value is guaranteed to be higher than what we know already.
+	 * As a safe limit, we return the limit supported by the kernel.
+	 */
+	default: return CONFIG_ARM64_PA_BITS;
+	}
+}
 #endif /* __ASSEMBLY__ */
 
 #endif
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 05/18] kvm: arm64: Clean up VTCR_EL2 initialisation
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (3 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 04/18] arm64: Add a helper for PARange to physical shift conversion Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 06/18] kvm: arm/arm64: Allow arch specific configurations for VM Suzuki K Poulose
                   ` (16 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Use the new helper for converting the parange to the physical shift.
Also, add the missing definitions for the VTCR_EL2 register fields
and use them instead of hard coding numbers.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changs sinec v3
 - Update comment about reserved bits
 - Added Reviewed-by from Eric
---
 arch/arm64/include/asm/kvm_arm.h |  3 +++
 arch/arm64/kvm/hyp/s2-setup.c    | 34 ++++++++------------------------
 2 files changed, 11 insertions(+), 26 deletions(-)

diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index aa45df752a16..5f807b680a5f 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -107,6 +107,7 @@
 #define VTCR_EL2_RES1		(1 << 31)
 #define VTCR_EL2_HD		(1 << 22)
 #define VTCR_EL2_HA		(1 << 21)
+#define VTCR_EL2_PS_SHIFT	TCR_EL2_PS_SHIFT
 #define VTCR_EL2_PS_MASK	TCR_EL2_PS_MASK
 #define VTCR_EL2_TG0_MASK	TCR_TG0_MASK
 #define VTCR_EL2_TG0_4K		TCR_TG0_4K
@@ -127,6 +128,8 @@
 #define VTCR_EL2_VS_8BIT	(0 << VTCR_EL2_VS_SHIFT)
 #define VTCR_EL2_VS_16BIT	(1 << VTCR_EL2_VS_SHIFT)
 
+#define VTCR_EL2_T0SZ(x)	TCR_T0SZ(x)
+
 /*
  * We configure the Stage-2 page tables to always restrict the IPA space to be
  * 40 bits wide (T0SZ = 24).  Systems with a PARange smaller than 40 bits are
diff --git a/arch/arm64/kvm/hyp/s2-setup.c b/arch/arm64/kvm/hyp/s2-setup.c
index 603e1ee83e89..e1ca672e937a 100644
--- a/arch/arm64/kvm/hyp/s2-setup.c
+++ b/arch/arm64/kvm/hyp/s2-setup.c
@@ -19,45 +19,27 @@
 #include <asm/kvm_arm.h>
 #include <asm/kvm_asm.h>
 #include <asm/kvm_hyp.h>
+#include <asm/cpufeature.h>
 
 u32 __hyp_text __init_stage2_translation(void)
 {
 	u64 val = VTCR_EL2_FLAGS;
 	u64 parange;
+	u32 phys_shift;
 	u64 tmp;
 
 	/*
 	 * Read the PARange bits from ID_AA64MMFR0_EL1 and set the PS
-	 * bits in VTCR_EL2. Amusingly, the PARange is 4 bits, while
-	 * PS is only 3. Fortunately, bit 19 is RES0 in VTCR_EL2...
+	 * bits in VTCR_EL2. Amusingly, the PARange is 4 bits, but the
+	 * allocated values are limited to 3bits.
 	 */
 	parange = read_sysreg(id_aa64mmfr0_el1) & 7;
 	if (parange > ID_AA64MMFR0_PARANGE_MAX)
 		parange = ID_AA64MMFR0_PARANGE_MAX;
-	val |= parange << 16;
+	val |= parange << VTCR_EL2_PS_SHIFT;
 
 	/* Compute the actual PARange... */
-	switch (parange) {
-	case 0:
-		parange = 32;
-		break;
-	case 1:
-		parange = 36;
-		break;
-	case 2:
-		parange = 40;
-		break;
-	case 3:
-		parange = 42;
-		break;
-	case 4:
-		parange = 44;
-		break;
-	case 5:
-	default:
-		parange = 48;
-		break;
-	}
+	phys_shift = id_aa64mmfr0_parange_to_phys_shift(parange);
 
 	/*
 	 * ... and clamp it to 40 bits, unless we have some braindead
@@ -65,7 +47,7 @@ u32 __hyp_text __init_stage2_translation(void)
 	 * return that value for the rest of the kernel to decide what
 	 * to do.
 	 */
-	val |= 64 - (parange > 40 ? 40 : parange);
+	val |= VTCR_EL2_T0SZ(phys_shift > 40 ? 40 : phys_shift);
 
 	/*
 	 * Check the availability of Hardware Access Flag / Dirty Bit
@@ -86,5 +68,5 @@ u32 __hyp_text __init_stage2_translation(void)
 
 	write_sysreg(val, vtcr_el2);
 
-	return parange;
+	return phys_shift;
 }
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 06/18] kvm: arm/arm64: Allow arch specific configurations for VM
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (4 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 05/18] kvm: arm64: Clean up VTCR_EL2 initialisation Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-20 10:22   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 07/18] kvm: arm64: Configure VTCR_EL2 per VM Suzuki K Poulose
                   ` (15 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Allow the arch backends to perform VM specific initialisation.
This will be later used to handle IPA size configuration and per-VM
VTCR configuration on arm64.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arch/arm/include/asm/kvm_host.h   | 7 +++++++
 arch/arm64/include/asm/kvm_host.h | 2 ++
 arch/arm64/kvm/reset.c            | 7 +++++++
 virt/kvm/arm/arm.c                | 5 +++--
 4 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 3ad482d2f1eb..72d46418e1ef 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -354,4 +354,11 @@ static inline void kvm_vcpu_put_sysregs(struct kvm_vcpu *vcpu) {}
 struct kvm *kvm_arch_alloc_vm(void);
 void kvm_arch_free_vm(struct kvm *kvm);
 
+static inline int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
+{
+	if (type)
+		return -EINVAL;
+	return 0;
+}
+
 #endif /* __ARM_KVM_HOST_H__ */
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 3d6d7336f871..b04280ae1be0 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -513,4 +513,6 @@ void kvm_vcpu_put_sysregs(struct kvm_vcpu *vcpu);
 struct kvm *kvm_arch_alloc_vm(void);
 void kvm_arch_free_vm(struct kvm *kvm);
 
+int kvm_arm_config_vm(struct kvm *kvm, unsigned long type);
+
 #endif /* __ARM64_KVM_HOST_H__ */
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index e37c78bbe1ca..b0c07dab5cb3 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -133,3 +133,10 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 	/* Reset timer */
 	return kvm_timer_vcpu_reset(vcpu);
 }
+
+int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
+{
+	if (type)
+		return -EINVAL;
+	return 0;
+}
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index c92053bc3f96..327d0fd28380 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -120,8 +120,9 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
 	int ret, cpu;
 
-	if (type)
-		return -EINVAL;
+	ret = kvm_arm_config_vm(kvm, type);
+	if (ret)
+		return ret;
 
 	kvm->arch.last_vcpu_ran = alloc_percpu(typeof(*kvm->arch.last_vcpu_ran));
 	if (!kvm->arch.last_vcpu_ran)
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 07/18] kvm: arm64: Configure VTCR_EL2 per VM
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (5 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 06/18] kvm: arm/arm64: Allow arch specific configurations for VM Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-20 10:21   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 08/18] kvm: arm/arm64: Prepare for VM specific stage2 translations Suzuki K Poulose
                   ` (14 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Add support for setting the VTCR_EL2 per VM, rather than hard
coding a value at boot time per CPU. This would allow us to tune
the stage2 page table parameters per VM in the later changes.

We compute the VTCR fields based on the system wide sanitised
feature registers, except for the hardware management of Access
Flags (VTCR_EL2.HA). It is fine to run a system with a mix of
CPUs that may or may not update the page table Access Flags.
Since the bit is RES0 on CPUs that don't support it, the bit
should be ignored on them.

Suggested-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arch/arm64/include/asm/kvm_arm.h  |  3 +-
 arch/arm64/include/asm/kvm_asm.h  |  2 -
 arch/arm64/include/asm/kvm_host.h | 12 ++++--
 arch/arm64/include/asm/kvm_hyp.h  |  1 +
 arch/arm64/kvm/hyp/Makefile       |  1 -
 arch/arm64/kvm/hyp/s2-setup.c     | 72 -------------------------------
 arch/arm64/kvm/reset.c            | 30 +++++++++++++
 7 files changed, 40 insertions(+), 81 deletions(-)
 delete mode 100644 arch/arm64/kvm/hyp/s2-setup.c

diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 5f807b680a5f..14317b3a1820 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -135,8 +135,7 @@
  * 40 bits wide (T0SZ = 24).  Systems with a PARange smaller than 40 bits are
  * not known to exist and will break with this configuration.
  *
- * VTCR_EL2.PS is extracted from ID_AA64MMFR0_EL1.PARange at boot time
- * (see hyp-init.S).
+ * The VTCR_EL2 is configured per VM and is initialised in kvm_arm_config_vm().
  *
  * Note that when using 4K pages, we concatenate two first level page tables
  * together. With 16K pages, we concatenate 16 first level page tables.
diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
index 102b5a5c47b6..0b53c72e7591 100644
--- a/arch/arm64/include/asm/kvm_asm.h
+++ b/arch/arm64/include/asm/kvm_asm.h
@@ -72,8 +72,6 @@ extern void __vgic_v3_init_lrs(void);
 
 extern u32 __kvm_get_mdcr_el2(void);
 
-extern u32 __init_stage2_translation(void);
-
 /* Home-grown __this_cpu_{ptr,read} variants that always work at HYP */
 #define __hyp_this_cpu_ptr(sym)						\
 	({								\
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index b04280ae1be0..5ecd457bce7d 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -61,11 +61,13 @@ struct kvm_arch {
 	u64    vmid_gen;
 	u32    vmid;
 
-	/* 1-level 2nd stage table, protected by kvm->mmu_lock */
+	/* stage2 entry level table */
 	pgd_t *pgd;
 
 	/* VTTBR value associated with above pgd and vmid */
 	u64    vttbr;
+	/* VTCR_EL2 value for this VM */
+	u64    vtcr;
 
 	/* The last vcpu id that ran on each physical CPU */
 	int __percpu *last_vcpu_ran;
@@ -442,10 +444,12 @@ int kvm_arm_vcpu_arch_has_attr(struct kvm_vcpu *vcpu,
 
 static inline void __cpu_init_stage2(void)
 {
-	u32 parange = kvm_call_hyp(__init_stage2_translation);
+	u32 ps;
 
-	WARN_ONCE(parange < 40,
-		  "PARange is %d bits, unsupported configuration!", parange);
+	/* Sanity check for minimum IPA size support */
+	ps = id_aa64mmfr0_parange_to_phys_shift(read_sysreg(id_aa64mmfr0_el1) & 0x7);
+	WARN_ONCE(ps < 40,
+		  "PARange is %d bits, unsupported configuration!", ps);
 }
 
 /* Guest/host FPSIMD coordination helpers */
diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h
index d1bd1e0f14d7..23aca66767f9 100644
--- a/arch/arm64/include/asm/kvm_hyp.h
+++ b/arch/arm64/include/asm/kvm_hyp.h
@@ -161,6 +161,7 @@ void __noreturn __hyp_do_panic(unsigned long, ...);
  */
 static __always_inline void __hyp_text __load_guest_stage2(struct kvm *kvm)
 {
+	write_sysreg(kvm->arch.vtcr, vtcr_el2);
 	write_sysreg(kvm->arch.vttbr, vttbr_el2);
 }
 
diff --git a/arch/arm64/kvm/hyp/Makefile b/arch/arm64/kvm/hyp/Makefile
index 2fabc2dc1966..82d1904328ad 100644
--- a/arch/arm64/kvm/hyp/Makefile
+++ b/arch/arm64/kvm/hyp/Makefile
@@ -19,7 +19,6 @@ obj-$(CONFIG_KVM_ARM_HOST) += switch.o
 obj-$(CONFIG_KVM_ARM_HOST) += fpsimd.o
 obj-$(CONFIG_KVM_ARM_HOST) += tlb.o
 obj-$(CONFIG_KVM_ARM_HOST) += hyp-entry.o
-obj-$(CONFIG_KVM_ARM_HOST) += s2-setup.o
 
 # KVM code is run at a different exception code with a different map, so
 # compiler instrumentation that inserts callbacks or checks into the code may
diff --git a/arch/arm64/kvm/hyp/s2-setup.c b/arch/arm64/kvm/hyp/s2-setup.c
deleted file mode 100644
index e1ca672e937a..000000000000
--- a/arch/arm64/kvm/hyp/s2-setup.c
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * Copyright (C) 2016 - ARM Ltd
- * Author: Marc Zyngier <marc.zyngier@arm.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <linux/types.h>
-#include <asm/kvm_arm.h>
-#include <asm/kvm_asm.h>
-#include <asm/kvm_hyp.h>
-#include <asm/cpufeature.h>
-
-u32 __hyp_text __init_stage2_translation(void)
-{
-	u64 val = VTCR_EL2_FLAGS;
-	u64 parange;
-	u32 phys_shift;
-	u64 tmp;
-
-	/*
-	 * Read the PARange bits from ID_AA64MMFR0_EL1 and set the PS
-	 * bits in VTCR_EL2. Amusingly, the PARange is 4 bits, but the
-	 * allocated values are limited to 3bits.
-	 */
-	parange = read_sysreg(id_aa64mmfr0_el1) & 7;
-	if (parange > ID_AA64MMFR0_PARANGE_MAX)
-		parange = ID_AA64MMFR0_PARANGE_MAX;
-	val |= parange << VTCR_EL2_PS_SHIFT;
-
-	/* Compute the actual PARange... */
-	phys_shift = id_aa64mmfr0_parange_to_phys_shift(parange);
-
-	/*
-	 * ... and clamp it to 40 bits, unless we have some braindead
-	 * HW that implements less than that. In all cases, we'll
-	 * return that value for the rest of the kernel to decide what
-	 * to do.
-	 */
-	val |= VTCR_EL2_T0SZ(phys_shift > 40 ? 40 : phys_shift);
-
-	/*
-	 * Check the availability of Hardware Access Flag / Dirty Bit
-	 * Management in ID_AA64MMFR1_EL1 and enable the feature in VTCR_EL2.
-	 */
-	tmp = (read_sysreg(id_aa64mmfr1_el1) >> ID_AA64MMFR1_HADBS_SHIFT) & 0xf;
-	if (tmp)
-		val |= VTCR_EL2_HA;
-
-	/*
-	 * Read the VMIDBits bits from ID_AA64MMFR1_EL1 and set the VS
-	 * bit in VTCR_EL2.
-	 */
-	tmp = (read_sysreg(id_aa64mmfr1_el1) >> ID_AA64MMFR1_VMIDBITS_SHIFT) & 0xf;
-	val |= (tmp == ID_AA64MMFR1_VMIDBITS_16) ?
-			VTCR_EL2_VS_16BIT :
-			VTCR_EL2_VS_8BIT;
-
-	write_sysreg(val, vtcr_el2);
-
-	return phys_shift;
-}
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index b0c07dab5cb3..e0c49377b771 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -26,6 +26,7 @@
 
 #include <kvm/arm_arch_timer.h>
 
+#include <asm/cpufeature.h>
 #include <asm/cputype.h>
 #include <asm/ptrace.h>
 #include <asm/kvm_arm.h>
@@ -134,9 +135,38 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 	return kvm_timer_vcpu_reset(vcpu);
 }
 
+/*
+ * Configure the VTCR_EL2 for this VM. The VTCR value is common
+ * across all the physical CPUs on the system. We use system wide
+ * sanitised values to fill in different fields, except for Hardware
+ * Management of Access Flags. HA Flag is set unconditionally on
+ * all CPUs, as it is safe to run with or without the feature and
+ * the bit is RES0 on CPUs that don't support it.
+ */
 int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
 {
+	u64 vtcr = VTCR_EL2_FLAGS;
+	u64 parange;
+
 	if (type)
 		return -EINVAL;
+
+	parange = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1) & 7;
+	if (parange > ID_AA64MMFR0_PARANGE_MAX)
+		parange = ID_AA64MMFR0_PARANGE_MAX;
+	vtcr |= parange << VTCR_EL2_PS_SHIFT;
+
+	/*
+	 * Enable the Hardware Access Flag management, unconditionally
+	 * on all CPUs. The features is RES0 on CPUs without the support
+	 * and must be ignored by the CPUs.
+	 */
+	vtcr |= VTCR_EL2_HA;
+
+	/* Set the vmid bits */
+	vtcr |= (kvm_get_vmid_bits() == 16) ?
+		VTCR_EL2_VS_16BIT :
+		VTCR_EL2_VS_8BIT;
+	kvm->arch.vtcr = vtcr;
 	return 0;
 }
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 08/18] kvm: arm/arm64: Prepare for VM specific stage2 translations
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (6 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 07/18] kvm: arm64: Configure VTCR_EL2 per VM Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-20 14:07   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 09/18] kvm: arm64: Prepare for dynamic stage2 page table layout Suzuki K Poulose
                   ` (13 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Right now the stage2 page table for a VM is hard coded, assuming
an IPA of 40bits. As we are about to add support for per VM IPA,
prepare the stage2 page table helpers to accept the kvm instance
to make the right decision for the VM. No functional changes.
Adds stage2_pgd_size(kvm) to replace S2_PGD_SIZE. Also, moves
some of the definitions in arm32 to align with the arm64.
Also drop the _AC() specifier constants wherever possible.

Cc: Christoffer Dall <cdall@kernel.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since V3:
 - Improve the comment about kvm_mmu_cache_min_pages()
 - Drop _AC() in arm64 definitions
 - Move kvm_mmu_cache_min_pages() in arm to stage2_pgtable.h in
   line with arm64.
---
 arch/arm/include/asm/kvm_arm.h                |   3 +-
 arch/arm/include/asm/kvm_mmu.h                |  13 +-
 arch/arm/include/asm/stage2_pgtable.h         |  50 +++++---
 arch/arm64/include/asm/kvm_mmu.h              |   7 +-
 arch/arm64/include/asm/stage2_pgtable-nopmd.h |  18 +--
 arch/arm64/include/asm/stage2_pgtable-nopud.h |  16 +--
 arch/arm64/include/asm/stage2_pgtable.h       |  58 +++++----
 virt/kvm/arm/arm.c                            |   2 +-
 virt/kvm/arm/mmu.c                            | 119 +++++++++---------
 virt/kvm/arm/vgic/vgic-kvm-device.c           |   2 +-
 10 files changed, 156 insertions(+), 132 deletions(-)

diff --git a/arch/arm/include/asm/kvm_arm.h b/arch/arm/include/asm/kvm_arm.h
index 3ab8b3781bfe..c3f1f9b304b7 100644
--- a/arch/arm/include/asm/kvm_arm.h
+++ b/arch/arm/include/asm/kvm_arm.h
@@ -133,8 +133,7 @@
  * space.
  */
 #define KVM_PHYS_SHIFT	(40)
-#define KVM_PHYS_SIZE	(_AC(1, ULL) << KVM_PHYS_SHIFT)
-#define KVM_PHYS_MASK	(KVM_PHYS_SIZE - _AC(1, ULL))
+
 #define PTRS_PER_S2_PGD	(_AC(1, ULL) << (KVM_PHYS_SHIFT - 30))
 
 /* Virtualization Translation Control Register (VTCR) bits */
diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
index 265ea9cf7df7..12ae5fbbcf01 100644
--- a/arch/arm/include/asm/kvm_mmu.h
+++ b/arch/arm/include/asm/kvm_mmu.h
@@ -35,16 +35,12 @@
 		addr;							\
 	})
 
-/*
- * KVM_MMU_CACHE_MIN_PAGES is the number of stage2 page table translation levels.
- */
-#define KVM_MMU_CACHE_MIN_PAGES	2
-
 #ifndef __ASSEMBLY__
 
 #include <linux/highmem.h>
 #include <asm/cacheflush.h>
 #include <asm/cputype.h>
+#include <asm/kvm_arm.h>
 #include <asm/kvm_hyp.h>
 #include <asm/pgalloc.h>
 #include <asm/stage2_pgtable.h>
@@ -52,6 +48,13 @@
 /* Ensure compatibility with arm64 */
 #define VA_BITS			32
 
+#define kvm_phys_shift(kvm)		KVM_PHYS_SHIFT
+#define kvm_phys_size(kvm)		(1ULL << kvm_phys_shift(kvm))
+#define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - 1ULL)
+#define kvm_vttbr_baddr_mask(kvm)	VTTBR_BADDR_MASK
+
+#define stage2_pgd_size(kvm)		(PTRS_PER_S2_PGD * sizeof(pgd_t))
+
 int create_hyp_mappings(void *from, void *to, pgprot_t prot);
 int create_hyp_io_mappings(phys_addr_t phys_addr, size_t size,
 			   void __iomem **kaddr,
diff --git a/arch/arm/include/asm/stage2_pgtable.h b/arch/arm/include/asm/stage2_pgtable.h
index 460d616bb2d6..f6a7ea805232 100644
--- a/arch/arm/include/asm/stage2_pgtable.h
+++ b/arch/arm/include/asm/stage2_pgtable.h
@@ -19,43 +19,53 @@
 #ifndef __ARM_S2_PGTABLE_H_
 #define __ARM_S2_PGTABLE_H_
 
-#define stage2_pgd_none(pgd)			pgd_none(pgd)
-#define stage2_pgd_clear(pgd)			pgd_clear(pgd)
-#define stage2_pgd_present(pgd)			pgd_present(pgd)
-#define stage2_pgd_populate(pgd, pud)		pgd_populate(NULL, pgd, pud)
-#define stage2_pud_offset(pgd, address)		pud_offset(pgd, address)
-#define stage2_pud_free(pud)			pud_free(NULL, pud)
+/*
+ * kvm_mmu_cache_min_pages() is the number of pages required
+ * to install a stage-2 translation. We pre-allocate the entry
+ * level table at VM creation. Since we have a 3 level page-table,
+ * we need only two pages to add a new mapping.
+ */
+#define kvm_mmu_cache_min_pages(kvm)	2
 
-#define stage2_pud_none(pud)			pud_none(pud)
-#define stage2_pud_clear(pud)			pud_clear(pud)
-#define stage2_pud_present(pud)			pud_present(pud)
-#define stage2_pud_populate(pud, pmd)		pud_populate(NULL, pud, pmd)
-#define stage2_pmd_offset(pud, address)		pmd_offset(pud, address)
-#define stage2_pmd_free(pmd)			pmd_free(NULL, pmd)
+#define stage2_pgd_none(kvm, pgd)		pgd_none(pgd)
+#define stage2_pgd_clear(kvm, pgd)		pgd_clear(pgd)
+#define stage2_pgd_present(kvm, pgd)		pgd_present(pgd)
+#define stage2_pgd_populate(kvm, pgd, pud)	pgd_populate(NULL, pgd, pud)
+#define stage2_pud_offset(kvm, pgd, address)	pud_offset(pgd, address)
+#define stage2_pud_free(kvm, pud)		pud_free(NULL, pud)
 
-#define stage2_pud_huge(pud)			pud_huge(pud)
+#define stage2_pud_none(kvm, pud)		pud_none(pud)
+#define stage2_pud_clear(kvm, pud)		pud_clear(pud)
+#define stage2_pud_present(kvm, pud)		pud_present(pud)
+#define stage2_pud_populate(kvm, pud, pmd)	pud_populate(NULL, pud, pmd)
+#define stage2_pmd_offset(kvm, pud, address)	pmd_offset(pud, address)
+#define stage2_pmd_free(kvm, pmd)		pmd_free(NULL, pmd)
+
+#define stage2_pud_huge(kvm, pud)		pud_huge(pud)
 
 /* Open coded p*d_addr_end that can deal with 64bit addresses */
-static inline phys_addr_t stage2_pgd_addr_end(phys_addr_t addr, phys_addr_t end)
+static inline phys_addr_t
+stage2_pgd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
 	phys_addr_t boundary = (addr + PGDIR_SIZE) & PGDIR_MASK;
 
 	return (boundary - 1 < end - 1) ? boundary : end;
 }
 
-#define stage2_pud_addr_end(addr, end)		(end)
+#define stage2_pud_addr_end(kvm, addr, end)	(end)
 
-static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
+static inline phys_addr_t
+stage2_pmd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
 	phys_addr_t boundary = (addr + PMD_SIZE) & PMD_MASK;
 
 	return (boundary - 1 < end - 1) ? boundary : end;
 }
 
-#define stage2_pgd_index(addr)				pgd_index(addr)
+#define stage2_pgd_index(kvm, addr)		pgd_index(addr)
 
-#define stage2_pte_table_empty(ptep)			kvm_page_empty(ptep)
-#define stage2_pmd_table_empty(pmdp)			kvm_page_empty(pmdp)
-#define stage2_pud_table_empty(pudp)			false
+#define stage2_pte_table_empty(kvm, ptep)	kvm_page_empty(ptep)
+#define stage2_pmd_table_empty(kvm, pmdp)	kvm_page_empty(pmdp)
+#define stage2_pud_table_empty(kvm, pudp)	false
 
 #endif	/* __ARM_S2_PGTABLE_H_ */
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index d6fff7de5539..3a032066e52c 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -141,8 +141,11 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
  * We currently only support a 40bit IPA.
  */
 #define KVM_PHYS_SHIFT	(40)
-#define KVM_PHYS_SIZE	(1UL << KVM_PHYS_SHIFT)
-#define KVM_PHYS_MASK	(KVM_PHYS_SIZE - 1UL)
+
+#define kvm_phys_shift(kvm)		KVM_PHYS_SHIFT
+#define kvm_phys_size(kvm)		(_AC(1, ULL) << kvm_phys_shift(kvm))
+#define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - _AC(1, ULL))
+#define kvm_vttbr_baddr_mask(kvm)	VTTBR_BADDR_MASK
 
 #include <asm/stage2_pgtable.h>
 
diff --git a/arch/arm64/include/asm/stage2_pgtable-nopmd.h b/arch/arm64/include/asm/stage2_pgtable-nopmd.h
index 2656a0fd05a6..0280dedbf75f 100644
--- a/arch/arm64/include/asm/stage2_pgtable-nopmd.h
+++ b/arch/arm64/include/asm/stage2_pgtable-nopmd.h
@@ -26,17 +26,17 @@
 #define S2_PMD_SIZE		(1UL << S2_PMD_SHIFT)
 #define S2_PMD_MASK		(~(S2_PMD_SIZE-1))
 
-#define stage2_pud_none(pud)			(0)
-#define stage2_pud_present(pud)			(1)
-#define stage2_pud_clear(pud)			do { } while (0)
-#define stage2_pud_populate(pud, pmd)		do { } while (0)
-#define stage2_pmd_offset(pud, address)		((pmd_t *)(pud))
+#define stage2_pud_none(kvm, pud)		(0)
+#define stage2_pud_present(kvm, pud)		(1)
+#define stage2_pud_clear(kvm, pud)		do { } while (0)
+#define stage2_pud_populate(kvm, pud, pmd)	do { } while (0)
+#define stage2_pmd_offset(kvm, pud, address)	((pmd_t *)(pud))
 
-#define stage2_pmd_free(pmd)			do { } while (0)
+#define stage2_pmd_free(kvm, pmd)		do { } while (0)
 
-#define stage2_pmd_addr_end(addr, end)		(end)
+#define stage2_pmd_addr_end(kvm, addr, end)	(end)
 
-#define stage2_pud_huge(pud)			(0)
-#define stage2_pmd_table_empty(pmdp)		(0)
+#define stage2_pud_huge(kvm, pud)		(0)
+#define stage2_pmd_table_empty(kvm, pmdp)	(0)
 
 #endif
diff --git a/arch/arm64/include/asm/stage2_pgtable-nopud.h b/arch/arm64/include/asm/stage2_pgtable-nopud.h
index 5ee87b54ebf3..cd6304e203be 100644
--- a/arch/arm64/include/asm/stage2_pgtable-nopud.h
+++ b/arch/arm64/include/asm/stage2_pgtable-nopud.h
@@ -24,16 +24,16 @@
 #define S2_PUD_SIZE		(_AC(1, UL) << S2_PUD_SHIFT)
 #define S2_PUD_MASK		(~(S2_PUD_SIZE-1))
 
-#define stage2_pgd_none(pgd)			(0)
-#define stage2_pgd_present(pgd)			(1)
-#define stage2_pgd_clear(pgd)			do { } while (0)
-#define stage2_pgd_populate(pgd, pud)	do { } while (0)
+#define stage2_pgd_none(kvm, pgd)		(0)
+#define stage2_pgd_present(kvm, pgd)		(1)
+#define stage2_pgd_clear(kvm, pgd)		do { } while (0)
+#define stage2_pgd_populate(kvm, pgd, pud)	do { } while (0)
 
-#define stage2_pud_offset(pgd, address)		((pud_t *)(pgd))
+#define stage2_pud_offset(kvm, pgd, address)	((pud_t *)(pgd))
 
-#define stage2_pud_free(x)			do { } while (0)
+#define stage2_pud_free(kvm, x)			do { } while (0)
 
-#define stage2_pud_addr_end(addr, end)		(end)
-#define stage2_pud_table_empty(pmdp)		(0)
+#define stage2_pud_addr_end(kvm, addr, end)	(end)
+#define stage2_pud_table_empty(kvm, pmdp)	(0)
 
 #endif
diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
index 8b68099348e5..11891612be14 100644
--- a/arch/arm64/include/asm/stage2_pgtable.h
+++ b/arch/arm64/include/asm/stage2_pgtable.h
@@ -55,7 +55,7 @@
 
 /* S2_PGDIR_SHIFT is the size mapped by top-level stage2 entry */
 #define S2_PGDIR_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(4 - STAGE2_PGTABLE_LEVELS)
-#define S2_PGDIR_SIZE			(_AC(1, UL) << S2_PGDIR_SHIFT)
+#define S2_PGDIR_SIZE			(1UL << S2_PGDIR_SHIFT)
 #define S2_PGDIR_MASK			(~(S2_PGDIR_SIZE - 1))
 
 /*
@@ -65,28 +65,30 @@
 #define PTRS_PER_S2_PGD			(1 << (KVM_PHYS_SHIFT - S2_PGDIR_SHIFT))
 
 /*
- * KVM_MMU_CACHE_MIN_PAGES is the number of stage2 page table translation
- * levels in addition to the PGD.
+ * kvm_mmmu_cache_min_pages() is the number of pages required to install
+ * a stage-2 translation. We pre-allocate the entry level page table at
+ * the VM creation.
  */
-#define KVM_MMU_CACHE_MIN_PAGES		(STAGE2_PGTABLE_LEVELS - 1)
+#define kvm_mmu_cache_min_pages(kvm)	(STAGE2_PGTABLE_LEVELS - 1)
 
 
 #if STAGE2_PGTABLE_LEVELS > 3
 
 #define S2_PUD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(1)
-#define S2_PUD_SIZE			(_AC(1, UL) << S2_PUD_SHIFT)
+#define S2_PUD_SIZE			(1UL << S2_PUD_SHIFT)
 #define S2_PUD_MASK			(~(S2_PUD_SIZE - 1))
 
-#define stage2_pgd_none(pgd)				pgd_none(pgd)
-#define stage2_pgd_clear(pgd)				pgd_clear(pgd)
-#define stage2_pgd_present(pgd)				pgd_present(pgd)
-#define stage2_pgd_populate(pgd, pud)			pgd_populate(NULL, pgd, pud)
-#define stage2_pud_offset(pgd, address)			pud_offset(pgd, address)
-#define stage2_pud_free(pud)				pud_free(NULL, pud)
+#define stage2_pgd_none(kvm, pgd)		pgd_none(pgd)
+#define stage2_pgd_clear(kvm, pgd)		pgd_clear(pgd)
+#define stage2_pgd_present(kvm, pgd)		pgd_present(pgd)
+#define stage2_pgd_populate(kvm, pgd, pud)	pgd_populate(NULL, pgd, pud)
+#define stage2_pud_offset(kvm, pgd, address)	pud_offset(pgd, address)
+#define stage2_pud_free(kvm, pud)		pud_free(NULL, pud)
 
-#define stage2_pud_table_empty(pudp)			kvm_page_empty(pudp)
+#define stage2_pud_table_empty(kvm, pudp)	kvm_page_empty(pudp)
 
-static inline phys_addr_t stage2_pud_addr_end(phys_addr_t addr, phys_addr_t end)
+static inline phys_addr_t
+stage2_pud_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
 	phys_addr_t boundary = (addr + S2_PUD_SIZE) & S2_PUD_MASK;
 
@@ -99,20 +101,21 @@ static inline phys_addr_t stage2_pud_addr_end(phys_addr_t addr, phys_addr_t end)
 #if STAGE2_PGTABLE_LEVELS > 2
 
 #define S2_PMD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(2)
-#define S2_PMD_SIZE			(_AC(1, UL) << S2_PMD_SHIFT)
+#define S2_PMD_SIZE			(1UL << S2_PMD_SHIFT)
 #define S2_PMD_MASK			(~(S2_PMD_SIZE - 1))
 
-#define stage2_pud_none(pud)				pud_none(pud)
-#define stage2_pud_clear(pud)				pud_clear(pud)
-#define stage2_pud_present(pud)				pud_present(pud)
-#define stage2_pud_populate(pud, pmd)			pud_populate(NULL, pud, pmd)
-#define stage2_pmd_offset(pud, address)			pmd_offset(pud, address)
-#define stage2_pmd_free(pmd)				pmd_free(NULL, pmd)
+#define stage2_pud_none(kvm, pud)		pud_none(pud)
+#define stage2_pud_clear(kvm, pud)		pud_clear(pud)
+#define stage2_pud_present(kvm, pud)		pud_present(pud)
+#define stage2_pud_populate(kvm, pud, pmd)	pud_populate(NULL, pud, pmd)
+#define stage2_pmd_offset(kvm, pud, address)	pmd_offset(pud, address)
+#define stage2_pmd_free(kvm, pmd)		pmd_free(NULL, pmd)
 
-#define stage2_pud_huge(pud)				pud_huge(pud)
-#define stage2_pmd_table_empty(pmdp)			kvm_page_empty(pmdp)
+#define stage2_pud_huge(kvm, pud)		pud_huge(pud)
+#define stage2_pmd_table_empty(kvm, pmdp)	kvm_page_empty(pmdp)
 
-static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
+static inline phys_addr_t
+stage2_pmd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
 	phys_addr_t boundary = (addr + S2_PMD_SIZE) & S2_PMD_MASK;
 
@@ -121,7 +124,7 @@ static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
 
 #endif		/* STAGE2_PGTABLE_LEVELS > 2 */
 
-#define stage2_pte_table_empty(ptep)			kvm_page_empty(ptep)
+#define stage2_pte_table_empty(kvm, ptep)	kvm_page_empty(ptep)
 
 #if STAGE2_PGTABLE_LEVELS == 2
 #include <asm/stage2_pgtable-nopmd.h>
@@ -129,10 +132,13 @@ static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
 #include <asm/stage2_pgtable-nopud.h>
 #endif
 
+#define stage2_pgd_size(kvm)	(PTRS_PER_S2_PGD * sizeof(pgd_t))
 
-#define stage2_pgd_index(addr)				(((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1))
+#define stage2_pgd_index(kvm, addr) \
+	(((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1))
 
-static inline phys_addr_t stage2_pgd_addr_end(phys_addr_t addr, phys_addr_t end)
+static inline phys_addr_t
+stage2_pgd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
 	phys_addr_t boundary = (addr + S2_PGDIR_SIZE) & S2_PGDIR_MASK;
 
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 327d0fd28380..43e716bc3f08 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -545,7 +545,7 @@ static void update_vttbr(struct kvm *kvm)
 
 	/* update vttbr to be used with the new vmid */
 	pgd_phys = virt_to_phys(kvm->arch.pgd);
-	BUG_ON(pgd_phys & ~VTTBR_BADDR_MASK);
+	BUG_ON(pgd_phys & ~kvm_vttbr_baddr_mask(kvm));
 	vmid = ((u64)(kvm->arch.vmid) << VTTBR_VMID_SHIFT) & VTTBR_VMID_MASK(kvm_vmid_bits);
 	kvm->arch.vttbr = kvm_phys_to_vttbr(pgd_phys) | vmid;
 
diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index 4a285d760ce0..7e477b3cae5b 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -45,7 +45,6 @@ static phys_addr_t hyp_idmap_vector;
 
 static unsigned long io_map_base;
 
-#define S2_PGD_SIZE	(PTRS_PER_S2_PGD * sizeof(pgd_t))
 #define hyp_pgd_order get_order(PTRS_PER_PGD * sizeof(pgd_t))
 
 #define KVM_S2PTE_FLAG_IS_IOMAP		(1UL << 0)
@@ -150,20 +149,20 @@ static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc)
 
 static void clear_stage2_pgd_entry(struct kvm *kvm, pgd_t *pgd, phys_addr_t addr)
 {
-	pud_t *pud_table __maybe_unused = stage2_pud_offset(pgd, 0UL);
-	stage2_pgd_clear(pgd);
+	pud_t *pud_table __maybe_unused = stage2_pud_offset(kvm, pgd, 0UL);
+	stage2_pgd_clear(kvm, pgd);
 	kvm_tlb_flush_vmid_ipa(kvm, addr);
-	stage2_pud_free(pud_table);
+	stage2_pud_free(kvm, pud_table);
 	put_page(virt_to_page(pgd));
 }
 
 static void clear_stage2_pud_entry(struct kvm *kvm, pud_t *pud, phys_addr_t addr)
 {
-	pmd_t *pmd_table __maybe_unused = stage2_pmd_offset(pud, 0);
-	VM_BUG_ON(stage2_pud_huge(*pud));
-	stage2_pud_clear(pud);
+	pmd_t *pmd_table __maybe_unused = stage2_pmd_offset(kvm, pud, 0);
+	VM_BUG_ON(stage2_pud_huge(kvm, *pud));
+	stage2_pud_clear(kvm, pud);
 	kvm_tlb_flush_vmid_ipa(kvm, addr);
-	stage2_pmd_free(pmd_table);
+	stage2_pmd_free(kvm, pmd_table);
 	put_page(virt_to_page(pud));
 }
 
@@ -252,7 +251,7 @@ static void unmap_stage2_ptes(struct kvm *kvm, pmd_t *pmd,
 		}
 	} while (pte++, addr += PAGE_SIZE, addr != end);
 
-	if (stage2_pte_table_empty(start_pte))
+	if (stage2_pte_table_empty(kvm, start_pte))
 		clear_stage2_pmd_entry(kvm, pmd, start_addr);
 }
 
@@ -262,9 +261,9 @@ static void unmap_stage2_pmds(struct kvm *kvm, pud_t *pud,
 	phys_addr_t next, start_addr = addr;
 	pmd_t *pmd, *start_pmd;
 
-	start_pmd = pmd = stage2_pmd_offset(pud, addr);
+	start_pmd = pmd = stage2_pmd_offset(kvm, pud, addr);
 	do {
-		next = stage2_pmd_addr_end(addr, end);
+		next = stage2_pmd_addr_end(kvm, addr, end);
 		if (!pmd_none(*pmd)) {
 			if (pmd_thp_or_huge(*pmd)) {
 				pmd_t old_pmd = *pmd;
@@ -281,7 +280,7 @@ static void unmap_stage2_pmds(struct kvm *kvm, pud_t *pud,
 		}
 	} while (pmd++, addr = next, addr != end);
 
-	if (stage2_pmd_table_empty(start_pmd))
+	if (stage2_pmd_table_empty(kvm, start_pmd))
 		clear_stage2_pud_entry(kvm, pud, start_addr);
 }
 
@@ -291,14 +290,14 @@ static void unmap_stage2_puds(struct kvm *kvm, pgd_t *pgd,
 	phys_addr_t next, start_addr = addr;
 	pud_t *pud, *start_pud;
 
-	start_pud = pud = stage2_pud_offset(pgd, addr);
+	start_pud = pud = stage2_pud_offset(kvm, pgd, addr);
 	do {
-		next = stage2_pud_addr_end(addr, end);
-		if (!stage2_pud_none(*pud)) {
-			if (stage2_pud_huge(*pud)) {
+		next = stage2_pud_addr_end(kvm, addr, end);
+		if (!stage2_pud_none(kvm, *pud)) {
+			if (stage2_pud_huge(kvm, *pud)) {
 				pud_t old_pud = *pud;
 
-				stage2_pud_clear(pud);
+				stage2_pud_clear(kvm, pud);
 				kvm_tlb_flush_vmid_ipa(kvm, addr);
 				kvm_flush_dcache_pud(old_pud);
 				put_page(virt_to_page(pud));
@@ -308,7 +307,7 @@ static void unmap_stage2_puds(struct kvm *kvm, pgd_t *pgd,
 		}
 	} while (pud++, addr = next, addr != end);
 
-	if (stage2_pud_table_empty(start_pud))
+	if (stage2_pud_table_empty(kvm, start_pud))
 		clear_stage2_pgd_entry(kvm, pgd, start_addr);
 }
 
@@ -332,7 +331,7 @@ static void unmap_stage2_range(struct kvm *kvm, phys_addr_t start, u64 size)
 	assert_spin_locked(&kvm->mmu_lock);
 	WARN_ON(size & ~PAGE_MASK);
 
-	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
+	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
 	do {
 		/*
 		 * Make sure the page table is still active, as another thread
@@ -341,8 +340,8 @@ static void unmap_stage2_range(struct kvm *kvm, phys_addr_t start, u64 size)
 		 */
 		if (!READ_ONCE(kvm->arch.pgd))
 			break;
-		next = stage2_pgd_addr_end(addr, end);
-		if (!stage2_pgd_none(*pgd))
+		next = stage2_pgd_addr_end(kvm, addr, end);
+		if (!stage2_pgd_none(kvm, *pgd))
 			unmap_stage2_puds(kvm, pgd, addr, next);
 		/*
 		 * If the range is too large, release the kvm->mmu_lock
@@ -371,9 +370,9 @@ static void stage2_flush_pmds(struct kvm *kvm, pud_t *pud,
 	pmd_t *pmd;
 	phys_addr_t next;
 
-	pmd = stage2_pmd_offset(pud, addr);
+	pmd = stage2_pmd_offset(kvm, pud, addr);
 	do {
-		next = stage2_pmd_addr_end(addr, end);
+		next = stage2_pmd_addr_end(kvm, addr, end);
 		if (!pmd_none(*pmd)) {
 			if (pmd_thp_or_huge(*pmd))
 				kvm_flush_dcache_pmd(*pmd);
@@ -389,11 +388,11 @@ static void stage2_flush_puds(struct kvm *kvm, pgd_t *pgd,
 	pud_t *pud;
 	phys_addr_t next;
 
-	pud = stage2_pud_offset(pgd, addr);
+	pud = stage2_pud_offset(kvm, pgd, addr);
 	do {
-		next = stage2_pud_addr_end(addr, end);
-		if (!stage2_pud_none(*pud)) {
-			if (stage2_pud_huge(*pud))
+		next = stage2_pud_addr_end(kvm, addr, end);
+		if (!stage2_pud_none(kvm, *pud)) {
+			if (stage2_pud_huge(kvm, *pud))
 				kvm_flush_dcache_pud(*pud);
 			else
 				stage2_flush_pmds(kvm, pud, addr, next);
@@ -409,10 +408,10 @@ static void stage2_flush_memslot(struct kvm *kvm,
 	phys_addr_t next;
 	pgd_t *pgd;
 
-	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
+	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
 	do {
-		next = stage2_pgd_addr_end(addr, end);
-		if (!stage2_pgd_none(*pgd))
+		next = stage2_pgd_addr_end(kvm, addr, end);
+		if (!stage2_pgd_none(kvm, *pgd))
 			stage2_flush_puds(kvm, pgd, addr, next);
 	} while (pgd++, addr = next, addr != end);
 }
@@ -898,7 +897,7 @@ int kvm_alloc_stage2_pgd(struct kvm *kvm)
 	}
 
 	/* Allocate the HW PGD, making sure that each page gets its own refcount */
-	pgd = alloc_pages_exact(S2_PGD_SIZE, GFP_KERNEL | __GFP_ZERO);
+	pgd = alloc_pages_exact(stage2_pgd_size(kvm), GFP_KERNEL | __GFP_ZERO);
 	if (!pgd)
 		return -ENOMEM;
 
@@ -987,7 +986,7 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
 
 	spin_lock(&kvm->mmu_lock);
 	if (kvm->arch.pgd) {
-		unmap_stage2_range(kvm, 0, KVM_PHYS_SIZE);
+		unmap_stage2_range(kvm, 0, kvm_phys_size(kvm));
 		pgd = READ_ONCE(kvm->arch.pgd);
 		kvm->arch.pgd = NULL;
 	}
@@ -995,7 +994,7 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
 
 	/* Free the HW pgd, one page at a time */
 	if (pgd)
-		free_pages_exact(pgd, S2_PGD_SIZE);
+		free_pages_exact(pgd, stage2_pgd_size(kvm));
 }
 
 static pud_t *stage2_get_pud(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
@@ -1004,16 +1003,16 @@ static pud_t *stage2_get_pud(struct kvm *kvm, struct kvm_mmu_memory_cache *cache
 	pgd_t *pgd;
 	pud_t *pud;
 
-	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
-	if (stage2_pgd_none(*pgd)) {
+	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
+	if (stage2_pgd_none(kvm, *pgd)) {
 		if (!cache)
 			return NULL;
 		pud = mmu_memory_cache_alloc(cache);
-		stage2_pgd_populate(pgd, pud);
+		stage2_pgd_populate(kvm, pgd, pud);
 		get_page(virt_to_page(pgd));
 	}
 
-	return stage2_pud_offset(pgd, addr);
+	return stage2_pud_offset(kvm, pgd, addr);
 }
 
 static pmd_t *stage2_get_pmd(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
@@ -1026,15 +1025,15 @@ static pmd_t *stage2_get_pmd(struct kvm *kvm, struct kvm_mmu_memory_cache *cache
 	if (!pud)
 		return NULL;
 
-	if (stage2_pud_none(*pud)) {
+	if (stage2_pud_none(kvm, *pud)) {
 		if (!cache)
 			return NULL;
 		pmd = mmu_memory_cache_alloc(cache);
-		stage2_pud_populate(pud, pmd);
+		stage2_pud_populate(kvm, pud, pmd);
 		get_page(virt_to_page(pud));
 	}
 
-	return stage2_pmd_offset(pud, addr);
+	return stage2_pmd_offset(kvm, pud, addr);
 }
 
 static int stage2_set_pmd_huge(struct kvm *kvm, struct kvm_mmu_memory_cache
@@ -1208,8 +1207,9 @@ int kvm_phys_addr_ioremap(struct kvm *kvm, phys_addr_t guest_ipa,
 		if (writable)
 			pte = kvm_s2pte_mkwrite(pte);
 
-		ret = mmu_topup_memory_cache(&cache, KVM_MMU_CACHE_MIN_PAGES,
-						KVM_NR_MEM_OBJS);
+		ret = mmu_topup_memory_cache(&cache,
+					     kvm_mmu_cache_min_pages(kvm),
+					     KVM_NR_MEM_OBJS);
 		if (ret)
 			goto out;
 		spin_lock(&kvm->mmu_lock);
@@ -1297,19 +1297,21 @@ static void stage2_wp_ptes(pmd_t *pmd, phys_addr_t addr, phys_addr_t end)
 
 /**
  * stage2_wp_pmds - write protect PUD range
+ * kvm:		kvm instance for the VM
  * @pud:	pointer to pud entry
  * @addr:	range start address
  * @end:	range end address
  */
-static void stage2_wp_pmds(pud_t *pud, phys_addr_t addr, phys_addr_t end)
+static void stage2_wp_pmds(struct kvm *kvm, pud_t *pud,
+			   phys_addr_t addr, phys_addr_t end)
 {
 	pmd_t *pmd;
 	phys_addr_t next;
 
-	pmd = stage2_pmd_offset(pud, addr);
+	pmd = stage2_pmd_offset(kvm, pud, addr);
 
 	do {
-		next = stage2_pmd_addr_end(addr, end);
+		next = stage2_pmd_addr_end(kvm, addr, end);
 		if (!pmd_none(*pmd)) {
 			if (pmd_thp_or_huge(*pmd)) {
 				if (!kvm_s2pmd_readonly(pmd))
@@ -1329,18 +1331,19 @@ static void stage2_wp_pmds(pud_t *pud, phys_addr_t addr, phys_addr_t end)
   *
   * Process PUD entries, for a huge PUD we cause a panic.
   */
-static void  stage2_wp_puds(pgd_t *pgd, phys_addr_t addr, phys_addr_t end)
+static void  stage2_wp_puds(struct kvm *kvm, pgd_t *pgd,
+			    phys_addr_t addr, phys_addr_t end)
 {
 	pud_t *pud;
 	phys_addr_t next;
 
-	pud = stage2_pud_offset(pgd, addr);
+	pud = stage2_pud_offset(kvm, pgd, addr);
 	do {
-		next = stage2_pud_addr_end(addr, end);
-		if (!stage2_pud_none(*pud)) {
+		next = stage2_pud_addr_end(kvm, addr, end);
+		if (!stage2_pud_none(kvm, *pud)) {
 			/* TODO:PUD not supported, revisit later if supported */
-			BUG_ON(stage2_pud_huge(*pud));
-			stage2_wp_pmds(pud, addr, next);
+			BUG_ON(stage2_pud_huge(kvm, *pud));
+			stage2_wp_pmds(kvm, pud, addr, next);
 		}
 	} while (pud++, addr = next, addr != end);
 }
@@ -1356,7 +1359,7 @@ static void stage2_wp_range(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 	pgd_t *pgd;
 	phys_addr_t next;
 
-	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
+	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
 	do {
 		/*
 		 * Release kvm_mmu_lock periodically if the memory region is
@@ -1370,9 +1373,9 @@ static void stage2_wp_range(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 		cond_resched_lock(&kvm->mmu_lock);
 		if (!READ_ONCE(kvm->arch.pgd))
 			break;
-		next = stage2_pgd_addr_end(addr, end);
-		if (stage2_pgd_present(*pgd))
-			stage2_wp_puds(pgd, addr, next);
+		next = stage2_pgd_addr_end(kvm, addr, end);
+		if (stage2_pgd_present(kvm, *pgd))
+			stage2_wp_puds(kvm, pgd, addr, next);
 	} while (pgd++, addr = next, addr != end);
 }
 
@@ -1521,7 +1524,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 	up_read(&current->mm->mmap_sem);
 
 	/* We need minimum second+third level pages */
-	ret = mmu_topup_memory_cache(memcache, KVM_MMU_CACHE_MIN_PAGES,
+	ret = mmu_topup_memory_cache(memcache, kvm_mmu_cache_min_pages(kvm),
 				     KVM_NR_MEM_OBJS);
 	if (ret)
 		return ret;
@@ -1764,7 +1767,7 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	}
 
 	/* Userspace should not be able to register out-of-bounds IPAs */
-	VM_BUG_ON(fault_ipa >= KVM_PHYS_SIZE);
+	VM_BUG_ON(fault_ipa >= kvm_phys_size(vcpu->kvm));
 
 	if (fault_status == FSC_ACCESS) {
 		handle_access_fault(vcpu, fault_ipa);
@@ -2063,7 +2066,7 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 	 * space addressable by the KVM guest IPA space.
 	 */
 	if (memslot->base_gfn + memslot->npages >=
-	    (KVM_PHYS_SIZE >> PAGE_SHIFT))
+	    (kvm_phys_size(kvm) >> PAGE_SHIFT))
 		return -EFAULT;
 
 	down_read(&current->mm->mmap_sem);
diff --git a/virt/kvm/arm/vgic/vgic-kvm-device.c b/virt/kvm/arm/vgic/vgic-kvm-device.c
index 6ada2432e37c..114dce9f4bf5 100644
--- a/virt/kvm/arm/vgic/vgic-kvm-device.c
+++ b/virt/kvm/arm/vgic/vgic-kvm-device.c
@@ -25,7 +25,7 @@
 int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
 		      phys_addr_t addr, phys_addr_t alignment)
 {
-	if (addr & ~KVM_PHYS_MASK)
+	if (addr & ~kvm_phys_mask(kvm))
 		return -E2BIG;
 
 	if (!IS_ALIGNED(addr, alignment))
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 09/18] kvm: arm64: Prepare for dynamic stage2 page table layout
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (7 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 08/18] kvm: arm/arm64: Prepare for VM specific stage2 translations Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 10/18] kvm: arm64: Make stage2 page table layout dynamic Suzuki K Poulose
                   ` (12 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Our stage2 page table helpers are statically defined based
on the fixed IPA of 40bits and the host page size. As we are
about to add support for configurable IPA size for VMs, we
need to make the page table checks for each VM. This patch
prepares the stage2 helpers to make the transition to a VM
dependent table layout easier. Instead of statically defining
the table helpers based on the page table levels, we now
check the page table levels in the helpers to do the right
thing. In effect, it simply converts the macros to static
inline functions.

Cc: Eric Auger <eric.auger@redhat.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arch/arm64/include/asm/kvm_mmu.h              |  12 +-
 arch/arm64/include/asm/stage2_pgtable-nopmd.h |  42 -----
 arch/arm64/include/asm/stage2_pgtable-nopud.h |  39 ----
 arch/arm64/include/asm/stage2_pgtable.h       | 173 ++++++++++++++----
 4 files changed, 139 insertions(+), 127 deletions(-)
 delete mode 100644 arch/arm64/include/asm/stage2_pgtable-nopmd.h
 delete mode 100644 arch/arm64/include/asm/stage2_pgtable-nopud.h

diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index 3a032066e52c..7342d2c51773 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -147,6 +147,12 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
 #define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - _AC(1, ULL))
 #define kvm_vttbr_baddr_mask(kvm)	VTTBR_BADDR_MASK
 
+static inline bool kvm_page_empty(void *ptr)
+{
+	struct page *ptr_page = virt_to_page(ptr);
+	return page_count(ptr_page) == 1;
+}
+
 #include <asm/stage2_pgtable.h>
 
 int create_hyp_mappings(void *from, void *to, pgprot_t prot);
@@ -241,12 +247,6 @@ static inline bool kvm_s2pmd_exec(pmd_t *pmdp)
 	return !(READ_ONCE(pmd_val(*pmdp)) & PMD_S2_XN);
 }
 
-static inline bool kvm_page_empty(void *ptr)
-{
-	struct page *ptr_page = virt_to_page(ptr);
-	return page_count(ptr_page) == 1;
-}
-
 #define hyp_pte_table_empty(ptep) kvm_page_empty(ptep)
 
 #ifdef __PAGETABLE_PMD_FOLDED
diff --git a/arch/arm64/include/asm/stage2_pgtable-nopmd.h b/arch/arm64/include/asm/stage2_pgtable-nopmd.h
deleted file mode 100644
index 0280dedbf75f..000000000000
--- a/arch/arm64/include/asm/stage2_pgtable-nopmd.h
+++ /dev/null
@@ -1,42 +0,0 @@
-/*
- * Copyright (C) 2016 - ARM Ltd
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#ifndef __ARM64_S2_PGTABLE_NOPMD_H_
-#define __ARM64_S2_PGTABLE_NOPMD_H_
-
-#include <asm/stage2_pgtable-nopud.h>
-
-#define __S2_PGTABLE_PMD_FOLDED
-
-#define S2_PMD_SHIFT		S2_PUD_SHIFT
-#define S2_PTRS_PER_PMD		1
-#define S2_PMD_SIZE		(1UL << S2_PMD_SHIFT)
-#define S2_PMD_MASK		(~(S2_PMD_SIZE-1))
-
-#define stage2_pud_none(kvm, pud)		(0)
-#define stage2_pud_present(kvm, pud)		(1)
-#define stage2_pud_clear(kvm, pud)		do { } while (0)
-#define stage2_pud_populate(kvm, pud, pmd)	do { } while (0)
-#define stage2_pmd_offset(kvm, pud, address)	((pmd_t *)(pud))
-
-#define stage2_pmd_free(kvm, pmd)		do { } while (0)
-
-#define stage2_pmd_addr_end(kvm, addr, end)	(end)
-
-#define stage2_pud_huge(kvm, pud)		(0)
-#define stage2_pmd_table_empty(kvm, pmdp)	(0)
-
-#endif
diff --git a/arch/arm64/include/asm/stage2_pgtable-nopud.h b/arch/arm64/include/asm/stage2_pgtable-nopud.h
deleted file mode 100644
index cd6304e203be..000000000000
--- a/arch/arm64/include/asm/stage2_pgtable-nopud.h
+++ /dev/null
@@ -1,39 +0,0 @@
-/*
- * Copyright (C) 2016 - ARM Ltd
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#ifndef __ARM64_S2_PGTABLE_NOPUD_H_
-#define __ARM64_S2_PGTABLE_NOPUD_H_
-
-#define __S2_PGTABLE_PUD_FOLDED
-
-#define S2_PUD_SHIFT		S2_PGDIR_SHIFT
-#define S2_PTRS_PER_PUD		1
-#define S2_PUD_SIZE		(_AC(1, UL) << S2_PUD_SHIFT)
-#define S2_PUD_MASK		(~(S2_PUD_SIZE-1))
-
-#define stage2_pgd_none(kvm, pgd)		(0)
-#define stage2_pgd_present(kvm, pgd)		(1)
-#define stage2_pgd_clear(kvm, pgd)		do { } while (0)
-#define stage2_pgd_populate(kvm, pgd, pud)	do { } while (0)
-
-#define stage2_pud_offset(kvm, pgd, address)	((pud_t *)(pgd))
-
-#define stage2_pud_free(kvm, x)			do { } while (0)
-
-#define stage2_pud_addr_end(kvm, addr, end)	(end)
-#define stage2_pud_table_empty(kvm, pmdp)	(0)
-
-#endif
diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
index 11891612be14..384f9e982cba 100644
--- a/arch/arm64/include/asm/stage2_pgtable.h
+++ b/arch/arm64/include/asm/stage2_pgtable.h
@@ -19,6 +19,7 @@
 #ifndef __ARM64_S2_PGTABLE_H_
 #define __ARM64_S2_PGTABLE_H_
 
+#include <linux/hugetlb.h>
 #include <asm/pgtable.h>
 
 /*
@@ -71,71 +72,163 @@
  */
 #define kvm_mmu_cache_min_pages(kvm)	(STAGE2_PGTABLE_LEVELS - 1)
 
-
-#if STAGE2_PGTABLE_LEVELS > 3
-
+/* Stage2 PUD definitions when the level is present */
+#define STAGE2_PGTABLE_HAS_PUD		(STAGE2_PGTABLE_LEVELS > 3)
 #define S2_PUD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(1)
 #define S2_PUD_SIZE			(1UL << S2_PUD_SHIFT)
 #define S2_PUD_MASK			(~(S2_PUD_SIZE - 1))
 
-#define stage2_pgd_none(kvm, pgd)		pgd_none(pgd)
-#define stage2_pgd_clear(kvm, pgd)		pgd_clear(pgd)
-#define stage2_pgd_present(kvm, pgd)		pgd_present(pgd)
-#define stage2_pgd_populate(kvm, pgd, pud)	pgd_populate(NULL, pgd, pud)
-#define stage2_pud_offset(kvm, pgd, address)	pud_offset(pgd, address)
-#define stage2_pud_free(kvm, pud)		pud_free(NULL, pud)
-
-#define stage2_pud_table_empty(kvm, pudp)	kvm_page_empty(pudp)
+static inline bool stage2_pgd_none(struct kvm *kvm, pgd_t pgd)
+{
+	if (STAGE2_PGTABLE_HAS_PUD)
+		return pgd_none(pgd);
+	else
+		return 0;
+}
+
+static inline void stage2_pgd_clear(struct kvm *kvm, pgd_t *pgdp)
+{
+	if (STAGE2_PGTABLE_HAS_PUD)
+		pgd_clear(pgdp);
+}
+
+static inline bool stage2_pgd_present(struct kvm *kvm, pgd_t pgd)
+{
+	if (STAGE2_PGTABLE_HAS_PUD)
+		return pgd_present(pgd);
+	else
+		return 1;
+}
+
+static inline void stage2_pgd_populate(struct kvm *kvm, pgd_t *pgd, pud_t *pud)
+{
+	if (STAGE2_PGTABLE_HAS_PUD)
+		pgd_populate(NULL, pgd, pud);
+}
+
+static inline pud_t *stage2_pud_offset(struct kvm *kvm,
+				       pgd_t *pgd, unsigned long address)
+{
+	if (STAGE2_PGTABLE_HAS_PUD)
+		return pud_offset(pgd, address);
+	else
+		return (pud_t *)pgd;
+}
+
+static inline void stage2_pud_free(struct kvm *kvm, pud_t *pud)
+{
+	if (STAGE2_PGTABLE_HAS_PUD)
+		pud_free(NULL, pud);
+}
+
+static inline bool stage2_pud_table_empty(struct kvm *kvm, pud_t *pudp)
+{
+	if (STAGE2_PGTABLE_HAS_PUD)
+		return kvm_page_empty(pudp);
+	else
+		return false;
+}
 
 static inline phys_addr_t
 stage2_pud_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
-	phys_addr_t boundary = (addr + S2_PUD_SIZE) & S2_PUD_MASK;
+	if (STAGE2_PGTABLE_HAS_PUD) {
+		phys_addr_t boundary = (addr + S2_PUD_SIZE) & S2_PUD_MASK;
 
-	return (boundary - 1 < end - 1) ? boundary : end;
+		return (boundary - 1 < end - 1) ? boundary : end;
+	} else {
+		return end;
+	}
 }
 
-#endif		/* STAGE2_PGTABLE_LEVELS > 3 */
-
-
-#if STAGE2_PGTABLE_LEVELS > 2
-
+/* Stage2 PMD definitions when the level is present */
+#define STAGE2_PGTABLE_HAS_PMD		(STAGE2_PGTABLE_LEVELS > 2)
 #define S2_PMD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(2)
 #define S2_PMD_SIZE			(1UL << S2_PMD_SHIFT)
 #define S2_PMD_MASK			(~(S2_PMD_SIZE - 1))
 
-#define stage2_pud_none(kvm, pud)		pud_none(pud)
-#define stage2_pud_clear(kvm, pud)		pud_clear(pud)
-#define stage2_pud_present(kvm, pud)		pud_present(pud)
-#define stage2_pud_populate(kvm, pud, pmd)	pud_populate(NULL, pud, pmd)
-#define stage2_pmd_offset(kvm, pud, address)	pmd_offset(pud, address)
-#define stage2_pmd_free(kvm, pmd)		pmd_free(NULL, pmd)
-
-#define stage2_pud_huge(kvm, pud)		pud_huge(pud)
-#define stage2_pmd_table_empty(kvm, pmdp)	kvm_page_empty(pmdp)
+static inline bool stage2_pud_none(struct kvm *kvm, pud_t pud)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		return pud_none(pud);
+	else
+		return 0;
+}
+
+static inline void stage2_pud_clear(struct kvm *kvm, pud_t *pud)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		pud_clear(pud);
+}
+
+static inline bool stage2_pud_present(struct kvm *kvm, pud_t pud)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		return pud_present(pud);
+	else
+		return 1;
+}
+
+static inline void stage2_pud_populate(struct kvm *kvm, pud_t *pud, pmd_t *pmd)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		pud_populate(NULL, pud, pmd);
+}
+
+static inline pmd_t *stage2_pmd_offset(struct kvm *kvm,
+				       pud_t *pud, unsigned long address)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		return pmd_offset(pud, address);
+	else
+		return (pmd_t *)pud;
+}
+
+static inline void stage2_pmd_free(struct kvm *kvm, pmd_t *pmd)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		pmd_free(NULL, pmd);
+}
+
+static inline bool stage2_pud_huge(struct kvm *kvm, pud_t pud)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		return pud_huge(pud);
+	else
+		return 0;
+}
+
+static inline bool stage2_pmd_table_empty(struct kvm *kvm, pmd_t *pmdp)
+{
+	if (STAGE2_PGTABLE_HAS_PMD)
+		return kvm_page_empty(pmdp);
+	else
+		return 0;
+}
 
 static inline phys_addr_t
 stage2_pmd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
-	phys_addr_t boundary = (addr + S2_PMD_SIZE) & S2_PMD_MASK;
+	if (STAGE2_PGTABLE_HAS_PMD) {
+		phys_addr_t boundary = (addr + S2_PMD_SIZE) & S2_PMD_MASK;
 
-	return (boundary - 1 < end - 1) ? boundary : end;
+		return (boundary - 1 < end - 1) ? boundary : end;
+	} else {
+		return end;
+	}
 }
 
-#endif		/* STAGE2_PGTABLE_LEVELS > 2 */
-
-#define stage2_pte_table_empty(kvm, ptep)	kvm_page_empty(ptep)
-
-#if STAGE2_PGTABLE_LEVELS == 2
-#include <asm/stage2_pgtable-nopmd.h>
-#elif STAGE2_PGTABLE_LEVELS == 3
-#include <asm/stage2_pgtable-nopud.h>
-#endif
+static inline bool stage2_pte_table_empty(struct kvm *kvm, pte_t *ptep)
+{
+	return kvm_page_empty(ptep);
+}
 
 #define stage2_pgd_size(kvm)	(PTRS_PER_S2_PGD * sizeof(pgd_t))
 
-#define stage2_pgd_index(kvm, addr) \
-	(((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1))
+static inline unsigned long stage2_pgd_index(struct kvm *kvm, phys_addr_t addr)
+{
+	return (((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1));
+}
 
 static inline phys_addr_t
 stage2_pgd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 10/18] kvm: arm64: Make stage2 page table layout dynamic
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (8 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 09/18] kvm: arm64: Prepare for dynamic stage2 page table layout Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-20 14:07   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask Suzuki K Poulose
                   ` (11 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Switch to dynamic stage2 page table layout based on the given
VM. So far we had a common stage2 table layout determined at
compile time. Make decision based on the VM instance depending
on the IPA limit for the VM. Adds helpers to compute the stage2
parameters based on the guest's IPA and uses them to make the decisions.

The IPA limit is still fixed to 40bits and the build time check
to ensure the stage2 doesn't exceed the host kernels page table
levels is retained. Also make sure that the host has pud/pmd helpers
are used only when they are available at host.

Cc: Christoffer Dall <cdall@kernel.org>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arch/arm64/include/asm/stage2_pgtable.h | 84 +++++++++++++++----------
 1 file changed, 52 insertions(+), 32 deletions(-)

diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
index 384f9e982cba..e5acda8e2e31 100644
--- a/arch/arm64/include/asm/stage2_pgtable.h
+++ b/arch/arm64/include/asm/stage2_pgtable.h
@@ -22,6 +22,13 @@
 #include <linux/hugetlb.h>
 #include <asm/pgtable.h>
 
+/*
+ * PGDIR_SHIFT determines the size a top-level page table entry can map
+ * and depends on the number of levels in the page table. Compute the
+ * PGDIR_SHIFT for a given number of levels.
+ */
+#define pt_levels_pgdir_shift(n)	ARM64_HW_PGTABLE_LEVEL_SHIFT(4 - (n))
+
 /*
  * The hardware supports concatenation of up to 16 tables at stage2 entry level
  * and we use the feature whenever possible.
@@ -30,11 +37,13 @@
  * On arm64, the smallest PAGE_SIZE supported is 4k, which means
  *             (PAGE_SHIFT - 3) > 4 holds for all page sizes.
  * This implies, the total number of page table levels at stage2 expected
- * by the hardware is actually the number of levels required for (KVM_PHYS_SHIFT - 4)
+ * by the hardware is actually the number of levels required for (IPA_SHIFT - 4)
  * in normal translations(e.g, stage1), since we cannot have another level in
- * the range (KVM_PHYS_SHIFT, KVM_PHYS_SHIFT - 4).
+ * the range (IPA_SHIFT, IPA_SHIFT - 4).
  */
-#define STAGE2_PGTABLE_LEVELS		ARM64_HW_PGTABLE_LEVELS(KVM_PHYS_SHIFT - 4)
+#define stage2_pgtable_levels(ipa)	ARM64_HW_PGTABLE_LEVELS((ipa) - 4)
+#define STAGE2_PGTABLE_LEVELS		stage2_pgtable_levels(KVM_PHYS_SHIFT)
+#define kvm_stage2_levels(kvm)		stage2_pgtable_levels(kvm_phys_shift(kvm))
 
 /*
  * With all the supported VA_BITs and 40bit guest IPA, the following condition
@@ -54,33 +63,42 @@
 #error "Unsupported combination of guest IPA and host VA_BITS."
 #endif
 
-/* S2_PGDIR_SHIFT is the size mapped by top-level stage2 entry */
-#define S2_PGDIR_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(4 - STAGE2_PGTABLE_LEVELS)
-#define S2_PGDIR_SIZE			(1UL << S2_PGDIR_SHIFT)
-#define S2_PGDIR_MASK			(~(S2_PGDIR_SIZE - 1))
+
+/* stage2_pgdir_shift() is the size mapped by top-level stage2 entry for the VM */
+#define stage2_pgdir_shift(kvm)		pt_levels_pgdir_shift(kvm_stage2_levels(kvm))
+#define stage2_pgdir_size(kvm)		(1ULL << stage2_pgdir_shift(kvm))
+#define stage2_pgdir_mask(kvm)		~(stage2_pgdir_size(kvm) - 1)
 
 /*
  * The number of PTRS across all concatenated stage2 tables given by the
  * number of bits resolved at the initial level.
  */
-#define PTRS_PER_S2_PGD			(1 << (KVM_PHYS_SHIFT - S2_PGDIR_SHIFT))
+#define __s2_pgd_ptrs(ipa, lvls)	(1 << ((ipa) - pt_levels_pgdir_shift((lvls))))
+#define __s2_pgd_size(ipa, lvls)	(__s2_pgd_ptrs((ipa), (lvls)) * sizeof(pgd_t))
+
+#define stage2_pgd_ptrs(kvm)		__s2_pgd_ptrs(kvm_phys_shift(kvm), kvm_stage2_levels(kvm))
+#define stage2_pgd_size(kvm)		__s2_pgd_size(kvm_phys_shift(kvm), kvm_stage2_levels(kvm))
 
 /*
  * kvm_mmmu_cache_min_pages() is the number of pages required to install
  * a stage-2 translation. We pre-allocate the entry level page table at
  * the VM creation.
  */
-#define kvm_mmu_cache_min_pages(kvm)	(STAGE2_PGTABLE_LEVELS - 1)
+#define kvm_mmu_cache_min_pages(kvm)	(kvm_stage2_levels(kvm) - 1)
 
 /* Stage2 PUD definitions when the level is present */
-#define STAGE2_PGTABLE_HAS_PUD		(STAGE2_PGTABLE_LEVELS > 3)
+static inline bool kvm_stage2_has_pud(struct kvm *kvm)
+{
+	return (CONFIG_PGTABLE_LEVELS > 3) && (kvm_stage2_levels(kvm) > 3);
+}
+
 #define S2_PUD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(1)
 #define S2_PUD_SIZE			(1UL << S2_PUD_SHIFT)
 #define S2_PUD_MASK			(~(S2_PUD_SIZE - 1))
 
 static inline bool stage2_pgd_none(struct kvm *kvm, pgd_t pgd)
 {
-	if (STAGE2_PGTABLE_HAS_PUD)
+	if (kvm_stage2_has_pud(kvm))
 		return pgd_none(pgd);
 	else
 		return 0;
@@ -88,13 +106,13 @@ static inline bool stage2_pgd_none(struct kvm *kvm, pgd_t pgd)
 
 static inline void stage2_pgd_clear(struct kvm *kvm, pgd_t *pgdp)
 {
-	if (STAGE2_PGTABLE_HAS_PUD)
+	if (kvm_stage2_has_pud(kvm))
 		pgd_clear(pgdp);
 }
 
 static inline bool stage2_pgd_present(struct kvm *kvm, pgd_t pgd)
 {
-	if (STAGE2_PGTABLE_HAS_PUD)
+	if (kvm_stage2_has_pud(kvm))
 		return pgd_present(pgd);
 	else
 		return 1;
@@ -102,14 +120,14 @@ static inline bool stage2_pgd_present(struct kvm *kvm, pgd_t pgd)
 
 static inline void stage2_pgd_populate(struct kvm *kvm, pgd_t *pgd, pud_t *pud)
 {
-	if (STAGE2_PGTABLE_HAS_PUD)
+	if (kvm_stage2_has_pud(kvm))
 		pgd_populate(NULL, pgd, pud);
 }
 
 static inline pud_t *stage2_pud_offset(struct kvm *kvm,
 				       pgd_t *pgd, unsigned long address)
 {
-	if (STAGE2_PGTABLE_HAS_PUD)
+	if (kvm_stage2_has_pud(kvm))
 		return pud_offset(pgd, address);
 	else
 		return (pud_t *)pgd;
@@ -117,13 +135,13 @@ static inline pud_t *stage2_pud_offset(struct kvm *kvm,
 
 static inline void stage2_pud_free(struct kvm *kvm, pud_t *pud)
 {
-	if (STAGE2_PGTABLE_HAS_PUD)
+	if (kvm_stage2_has_pud(kvm))
 		pud_free(NULL, pud);
 }
 
 static inline bool stage2_pud_table_empty(struct kvm *kvm, pud_t *pudp)
 {
-	if (STAGE2_PGTABLE_HAS_PUD)
+	if (kvm_stage2_has_pud(kvm))
 		return kvm_page_empty(pudp);
 	else
 		return false;
@@ -132,7 +150,7 @@ static inline bool stage2_pud_table_empty(struct kvm *kvm, pud_t *pudp)
 static inline phys_addr_t
 stage2_pud_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
-	if (STAGE2_PGTABLE_HAS_PUD) {
+	if (kvm_stage2_has_pud(kvm)) {
 		phys_addr_t boundary = (addr + S2_PUD_SIZE) & S2_PUD_MASK;
 
 		return (boundary - 1 < end - 1) ? boundary : end;
@@ -142,14 +160,18 @@ stage2_pud_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 }
 
 /* Stage2 PMD definitions when the level is present */
-#define STAGE2_PGTABLE_HAS_PMD		(STAGE2_PGTABLE_LEVELS > 2)
+static inline bool kvm_stage2_has_pmd(struct kvm *kvm)
+{
+	return (CONFIG_PGTABLE_LEVELS > 2) && (kvm_stage2_levels(kvm) > 2);
+}
+
 #define S2_PMD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(2)
 #define S2_PMD_SIZE			(1UL << S2_PMD_SHIFT)
 #define S2_PMD_MASK			(~(S2_PMD_SIZE - 1))
 
 static inline bool stage2_pud_none(struct kvm *kvm, pud_t pud)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		return pud_none(pud);
 	else
 		return 0;
@@ -157,13 +179,13 @@ static inline bool stage2_pud_none(struct kvm *kvm, pud_t pud)
 
 static inline void stage2_pud_clear(struct kvm *kvm, pud_t *pud)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		pud_clear(pud);
 }
 
 static inline bool stage2_pud_present(struct kvm *kvm, pud_t pud)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		return pud_present(pud);
 	else
 		return 1;
@@ -171,14 +193,14 @@ static inline bool stage2_pud_present(struct kvm *kvm, pud_t pud)
 
 static inline void stage2_pud_populate(struct kvm *kvm, pud_t *pud, pmd_t *pmd)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		pud_populate(NULL, pud, pmd);
 }
 
 static inline pmd_t *stage2_pmd_offset(struct kvm *kvm,
 				       pud_t *pud, unsigned long address)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		return pmd_offset(pud, address);
 	else
 		return (pmd_t *)pud;
@@ -186,13 +208,13 @@ static inline pmd_t *stage2_pmd_offset(struct kvm *kvm,
 
 static inline void stage2_pmd_free(struct kvm *kvm, pmd_t *pmd)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		pmd_free(NULL, pmd);
 }
 
 static inline bool stage2_pud_huge(struct kvm *kvm, pud_t pud)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		return pud_huge(pud);
 	else
 		return 0;
@@ -200,7 +222,7 @@ static inline bool stage2_pud_huge(struct kvm *kvm, pud_t pud)
 
 static inline bool stage2_pmd_table_empty(struct kvm *kvm, pmd_t *pmdp)
 {
-	if (STAGE2_PGTABLE_HAS_PMD)
+	if (kvm_stage2_has_pmd(kvm))
 		return kvm_page_empty(pmdp);
 	else
 		return 0;
@@ -209,7 +231,7 @@ static inline bool stage2_pmd_table_empty(struct kvm *kvm, pmd_t *pmdp)
 static inline phys_addr_t
 stage2_pmd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
-	if (STAGE2_PGTABLE_HAS_PMD) {
+	if (kvm_stage2_has_pmd(kvm)) {
 		phys_addr_t boundary = (addr + S2_PMD_SIZE) & S2_PMD_MASK;
 
 		return (boundary - 1 < end - 1) ? boundary : end;
@@ -223,17 +245,15 @@ static inline bool stage2_pte_table_empty(struct kvm *kvm, pte_t *ptep)
 	return kvm_page_empty(ptep);
 }
 
-#define stage2_pgd_size(kvm)	(PTRS_PER_S2_PGD * sizeof(pgd_t))
-
 static inline unsigned long stage2_pgd_index(struct kvm *kvm, phys_addr_t addr)
 {
-	return (((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1));
+	return (((addr) >> stage2_pgdir_shift(kvm)) & (stage2_pgd_ptrs(kvm) - 1));
 }
 
 static inline phys_addr_t
 stage2_pgd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
 {
-	phys_addr_t boundary = (addr + S2_PGDIR_SIZE) & S2_PGDIR_MASK;
+	phys_addr_t boundary = (addr + stage2_pgdir_size(kvm)) & stage2_pgdir_mask(kvm);
 
 	return (boundary - 1 < end - 1) ? boundary : end;
 }
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (9 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 10/18] kvm: arm64: Make stage2 page table layout dynamic Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-20 14:07   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 12/18] kvm: arm64: Configure VTCR_EL2.SL0 per VM Suzuki K Poulose
                   ` (10 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

On arm64 VTTBR_EL2:BADDR holds the base address for the stage2
translation table. The Arm ARM mandates that the bits BADDR[x-1:0]
should be 0, where 'x' is defined for a given IPA Size and the
number of levels for a translation granule size. It is defined
using some magical constants. This patch is a reverse engineered
implementation to calculate the 'x' at runtime for a given ipa and
number of page table levels. See patch for more details.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since V3:
 - Update reference to latest ARM ARM and improve commentary
---
 arch/arm64/include/asm/kvm_arm.h | 63 +++++++++++++++++++++++++++++---
 arch/arm64/include/asm/kvm_mmu.h | 25 ++++++++++++-
 2 files changed, 81 insertions(+), 7 deletions(-)

diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 14317b3a1820..3fb1d440be6e 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -123,7 +123,6 @@
 #define VTCR_EL2_SL0_MASK	(3 << VTCR_EL2_SL0_SHIFT)
 #define VTCR_EL2_SL0_LVL1	(1 << VTCR_EL2_SL0_SHIFT)
 #define VTCR_EL2_T0SZ_MASK	0x3f
-#define VTCR_EL2_T0SZ_40B	24
 #define VTCR_EL2_VS_SHIFT	19
 #define VTCR_EL2_VS_8BIT	(0 << VTCR_EL2_VS_SHIFT)
 #define VTCR_EL2_VS_16BIT	(1 << VTCR_EL2_VS_SHIFT)
@@ -140,11 +139,8 @@
  * Note that when using 4K pages, we concatenate two first level page tables
  * together. With 16K pages, we concatenate 16 first level page tables.
  *
- * The magic numbers used for VTTBR_X in this patch can be found in Tables
- * D4-23 and D4-25 in ARM DDI 0487A.b.
  */
 
-#define VTCR_EL2_T0SZ_IPA	VTCR_EL2_T0SZ_40B
 #define VTCR_EL2_COMMON_BITS	(VTCR_EL2_SH0_INNER | VTCR_EL2_ORGN0_WBWA | \
 				 VTCR_EL2_IRGN0_WBWA | VTCR_EL2_RES1)
 
@@ -175,9 +171,64 @@
 #endif
 
 #define VTCR_EL2_FLAGS			(VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
-#define VTTBR_X				(VTTBR_X_TGRAN_MAGIC - VTCR_EL2_T0SZ_IPA)
+/*
+ * ARM VMSAv8-64 defines an algorithm for finding the translation table
+ * descriptors in section D4.2.8 in ARM DDI 0487C.a.
+ *
+ * The algorithm defines the expectations on the BaseAddress (for the page
+ * table) bits resolved at each level based on the page size, entry level
+ * and T0SZ. The variable "x" in the algorithm also affects the VTTBR:BADDR
+ * for stage2 page table.
+ *
+ * The value of "x" is calculated as :
+ *	x = Magic_N - T0SZ
+ *
+ * where Magic_N is an integer depending on the page size and the entry
+ * level of the page table as below:
+ *
+ *	--------------------------------------------
+ *	| Entry level		|  4K    16K   64K |
+ *	--------------------------------------------
+ *	| Level: 0 (4 levels)	| 28   |  -  |  -  |
+ *	--------------------------------------------
+ *	| Level: 1 (3 levels)	| 37   | 31  | 25  |
+ *	--------------------------------------------
+ *	| Level: 2 (2 levels)	| 46   | 42  | 38  |
+ *	--------------------------------------------
+ *	| Level: 3 (1 level)	| -    | 53  | 51  |
+ *	--------------------------------------------
+ *
+ * We have a magic formula for the Magic_N below:
+ *
+ *  Magic_N(PAGE_SIZE, Level) = 64 - ((PAGE_SHIFT - 3) * Number_of_levels)
+ *
+ * where Number_of_levels = (4 - Level). We are only interested in the
+ * value for Entry_Level for the stage2 page table.
+ *
+ * So, given that T0SZ = (64 - IPA_SHIFT), we can compute 'x' as follows:
+ *
+ *	x = (64 - ((PAGE_SHIFT - 3) * Number_of_levels)) - (64 - IPA_SHIFT)
+ *	  = IPA_SHIFT - ((PAGE_SHIFT - 3) * Number of levels)
+ *
+ * Here is one way to explain the Magic Formula:
+ *
+ *  x = log2(Size_of_Entry_Level_Table)
+ *
+ * Since, we can resolve (PAGE_SHIFT - 3) bits at each level, and another
+ * PAGE_SHIFT bits in the PTE, we have :
+ *
+ *  Bits_Entry_level = IPA_SHIFT - ((PAGE_SHIFT - 3) * (n - 1) + PAGE_SHIFT)
+ *		     = IPA_SHIFT - (PAGE_SHIFT - 3) * n - 3
+ *  where n = number of levels, and since each pointer is 8bytes, we have:
+ *
+ *  x = Bits_Entry_Level + 3
+ *    = IPA_SHIFT - (PAGE_SHIFT - 3) * n
+ *
+ * The only constraint here is that, we have to find the number of page table
+ * levels for a given IPA size (which we do, see stage2_pt_levels())
+ */
+#define ARM64_VTTBR_X(ipa, levels)	((ipa) - ((levels) * (PAGE_SHIFT - 3)))
 
-#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)
 
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index 7342d2c51773..ac3ca9690bad 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -145,7 +145,6 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
 #define kvm_phys_shift(kvm)		KVM_PHYS_SHIFT
 #define kvm_phys_size(kvm)		(_AC(1, ULL) << kvm_phys_shift(kvm))
 #define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - _AC(1, ULL))
-#define kvm_vttbr_baddr_mask(kvm)	VTTBR_BADDR_MASK
 
 static inline bool kvm_page_empty(void *ptr)
 {
@@ -520,5 +519,29 @@ static inline int hyp_map_aux_data(void)
 
 #define kvm_phys_to_vttbr(addr)		phys_to_ttbr(addr)
 
+/*
+ * Get the magic number 'x' for VTTBR:BADDR of this KVM instance.
+ * With v8.2 LVA extensions, 'x' should be a minimum of 6 with
+ * 52bit IPS.
+ */
+static inline int arm64_vttbr_x(u32 ipa_shift, u32 levels)
+{
+	int x = ARM64_VTTBR_X(ipa_shift, levels);
+
+	return (IS_ENABLED(CONFIG_ARM64_PA_BITS_52) && x < 6) ? 6 : x;
+}
+
+static inline u64 vttbr_baddr_mask(u32 ipa_shift, u32 levels)
+{
+	unsigned int x = arm64_vttbr_x(ipa_shift, levels);
+
+	return GENMASK_ULL(PHYS_MASK_SHIFT - 1, x);
+}
+
+static inline u64 kvm_vttbr_baddr_mask(struct kvm *kvm)
+{
+	return vttbr_baddr_mask(kvm_phys_shift(kvm), kvm_stage2_levels(kvm));
+}
+
 #endif /* __ASSEMBLY__ */
 #endif /* __ARM64_KVM_MMU_H__ */
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 12/18] kvm: arm64: Configure VTCR_EL2.SL0 per VM
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (10 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-20 14:25   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 13/18] kvm: arm64: Switch to per VM IPA limit Suzuki K Poulose
                   ` (9 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

VTCR_EL2 holds the following key stage2 translation table
parameters:
  SL0  - Entry level in the page table lookup.
  T0SZ - Denotes the size of the memory addressed by the table.

We have been using fixed values for the SL0 depending on the
page size as we have a fixed IPA size. But since we are about
to make it dynamic, we need to calculate the SL0 at runtime
per VM. This patch adds a helper to compute the value of SL0
for a VM based on the IPA size.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Cc: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since v3:
 - Update reference to latest ARM ARM.
 - Update per-vm VTCR value of SL0.
 - Add helpers to decode levels from SL0.
 - Didn't pick up Reviewed-by tag from Eric, as there
   are some new changes in this version
---
 arch/arm64/include/asm/kvm_arm.h | 51 +++++++++++++++++++++++++-------
 arch/arm64/kvm/reset.c           |  1 +
 2 files changed, 41 insertions(+), 11 deletions(-)

diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 3fb1d440be6e..5c1487dc5dca 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -121,7 +121,6 @@
 #define VTCR_EL2_IRGN0_WBWA	TCR_IRGN0_WBWA
 #define VTCR_EL2_SL0_SHIFT	6
 #define VTCR_EL2_SL0_MASK	(3 << VTCR_EL2_SL0_SHIFT)
-#define VTCR_EL2_SL0_LVL1	(1 << VTCR_EL2_SL0_SHIFT)
 #define VTCR_EL2_T0SZ_MASK	0x3f
 #define VTCR_EL2_VS_SHIFT	19
 #define VTCR_EL2_VS_8BIT	(0 << VTCR_EL2_VS_SHIFT)
@@ -148,29 +147,59 @@
 /*
  * Stage2 translation configuration:
  * 64kB pages (TG0 = 1)
- * 2 level page tables (SL = 1)
  */
-#define VTCR_EL2_TGRAN_FLAGS		(VTCR_EL2_TG0_64K | VTCR_EL2_SL0_LVL1)
-#define VTTBR_X_TGRAN_MAGIC		38
+#define VTCR_EL2_TGRAN			VTCR_EL2_TG0_64K
+#define VTCR_EL2_TGRAN_SL0_BASE		3UL
+
 #elif defined(CONFIG_ARM64_16K_PAGES)
 /*
  * Stage2 translation configuration:
  * 16kB pages (TG0 = 2)
- * 2 level page tables (SL = 1)
  */
-#define VTCR_EL2_TGRAN_FLAGS		(VTCR_EL2_TG0_16K | VTCR_EL2_SL0_LVL1)
-#define VTTBR_X_TGRAN_MAGIC		42
+#define VTCR_EL2_TGRAN			VTCR_EL2_TG0_16K
+#define VTCR_EL2_TGRAN_SL0_BASE		3UL
 #else	/* 4K */
 /*
  * Stage2 translation configuration:
  * 4kB pages (TG0 = 0)
- * 3 level page tables (SL = 1)
  */
-#define VTCR_EL2_TGRAN_FLAGS		(VTCR_EL2_TG0_4K | VTCR_EL2_SL0_LVL1)
-#define VTTBR_X_TGRAN_MAGIC		37
+#define VTCR_EL2_TGRAN			VTCR_EL2_TG0_4K
+#define VTCR_EL2_TGRAN_SL0_BASE		2UL
 #endif
 
-#define VTCR_EL2_FLAGS			(VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
+#define VTCR_EL2_FLAGS			(VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN)
+/*
+ * VTCR_EL2:SL0 indicates the entry level for Stage2 translation.
+ * Interestingly, it depends on the page size.
+ * See D.10.2.121, VTCR_EL2, in ARM DDI 0487C.a
+ *
+ *	-----------------------------------------
+ *	| Entry level		|  4K  | 16K/64K |
+ *	------------------------------------------
+ *	| Level: 0		|  2   |   -     |
+ *	------------------------------------------
+ *	| Level: 1		|  1   |   2     |
+ *	------------------------------------------
+ *	| Level: 2		|  0   |   1     |
+ *	------------------------------------------
+ *	| Level: 3		|  -   |   0     |
+ *	------------------------------------------
+ *
+ * That table roughly translates to :
+ *
+ *	SL0(PAGE_SIZE, Entry_level) = SL0_BASE(PAGE_SIZE) - Entry_Level
+ *
+ * Where SL0_BASE(4K) = 2 and SL0_BASE(16K) = 3, SL0_BASE(64K) = 3, provided
+ * we take care of ruling out the unsupported cases and
+ * Entry_Level = 4 - Number_of_levels.
+ *
+ */
+#define VTCR_EL2_LVLS_TO_SL0(levels)	\
+	((VTCR_EL2_TGRAN_SL0_BASE - (4 - (levels))) << VTCR_EL2_SL0_SHIFT)
+#define VTCR_EL2_SL0_TO_LVLS(sl0)	\
+	((sl0) + 4 - VTCR_EL2_TGRAN_SL0_BASE)
+#define VTCR_EL2_LVLS(vtcr)		\
+	VTCR_EL2_SL0_TO_LVLS(((vtcr) & VTCR_EL2_SL0_MASK) >> VTCR_EL2_SL0_SHIFT)
 /*
  * ARM VMSAv8-64 defines an algorithm for finding the translation table
  * descriptors in section D4.2.8 in ARM DDI 0487C.a.
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index e0c49377b771..d9b7a00993b6 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -167,6 +167,7 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
 	vtcr |= (kvm_get_vmid_bits() == 16) ?
 		VTCR_EL2_VS_16BIT :
 		VTCR_EL2_VS_8BIT;
+	vtcr |= VTCR_EL2_LVLS_TO_SL0(kvm_stage2_levels(kvm));
 	kvm->arch.vtcr = vtcr;
 	return 0;
 }
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 13/18] kvm: arm64: Switch to per VM IPA limit
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (11 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 12/18] kvm: arm64: Configure VTCR_EL2.SL0 per VM Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [PATCH v5 14/18] vgic: Add support for 52bit guest physical address Suzuki K Poulose
                   ` (8 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Now that we can manage the stage2 page table per VM, switch the
configuration details to per VM instance. The VTCR is updated
with the values specific to the VM based on the configuration.
We store the IPA size and the number of stage2 page table levels
for the guest already in VTCR. Decode it back from the vtcr
field wherever we need it.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arch/arm64/include/asm/kvm_arm.h        | 3 +++
 arch/arm64/include/asm/kvm_mmu.h        | 2 +-
 arch/arm64/include/asm/stage2_pgtable.h | 2 +-
 arch/arm64/kvm/reset.c                  | 4 +++-
 4 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 5c1487dc5dca..0a37c0513ede 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -200,6 +200,9 @@
 	((sl0) + 4 - VTCR_EL2_TGRAN_SL0_BASE)
 #define VTCR_EL2_LVLS(vtcr)		\
 	VTCR_EL2_SL0_TO_LVLS(((vtcr) & VTCR_EL2_SL0_MASK) >> VTCR_EL2_SL0_SHIFT)
+
+#define VTCR_EL2_IPA(vtcr)	(64 - ((vtcr) & VTCR_EL2_T0SZ_MASK))
+
 /*
  * ARM VMSAv8-64 defines an algorithm for finding the translation table
  * descriptors in section D4.2.8 in ARM DDI 0487C.a.
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index ac3ca9690bad..77b1af9e64db 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -142,7 +142,7 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
  */
 #define KVM_PHYS_SHIFT	(40)
 
-#define kvm_phys_shift(kvm)		KVM_PHYS_SHIFT
+#define kvm_phys_shift(kvm)		VTCR_EL2_IPA(kvm->arch.vtcr)
 #define kvm_phys_size(kvm)		(_AC(1, ULL) << kvm_phys_shift(kvm))
 #define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - _AC(1, ULL))
 
diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
index e5acda8e2e31..352ec4158fdf 100644
--- a/arch/arm64/include/asm/stage2_pgtable.h
+++ b/arch/arm64/include/asm/stage2_pgtable.h
@@ -43,7 +43,7 @@
  */
 #define stage2_pgtable_levels(ipa)	ARM64_HW_PGTABLE_LEVELS((ipa) - 4)
 #define STAGE2_PGTABLE_LEVELS		stage2_pgtable_levels(KVM_PHYS_SHIFT)
-#define kvm_stage2_levels(kvm)		stage2_pgtable_levels(kvm_phys_shift(kvm))
+#define kvm_stage2_levels(kvm)		VTCR_EL2_LVLS(kvm->arch.vtcr)
 
 /*
  * With all the supported VA_BITs and 40bit guest IPA, the following condition
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index d9b7a00993b6..51ecf0f7c912 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -167,7 +167,9 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
 	vtcr |= (kvm_get_vmid_bits() == 16) ?
 		VTCR_EL2_VS_16BIT :
 		VTCR_EL2_VS_8BIT;
-	vtcr |= VTCR_EL2_LVLS_TO_SL0(kvm_stage2_levels(kvm));
+	vtcr |= VTCR_EL2_LVLS_TO_SL0(stage2_pgtable_levels(KVM_PHYS_SHIFT));
+	vtcr |= VTCR_EL2_T0SZ(KVM_PHYS_SHIFT);
+
 	kvm->arch.vtcr = vtcr;
 	return 0;
 }
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 14/18] vgic: Add support for 52bit guest physical address
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (12 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 13/18] kvm: arm64: Switch to per VM IPA limit Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-21 14:57   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 15/18] kvm: arm64: Add 52bit support for PAR to HPFAR conversoin Suzuki K Poulose
                   ` (7 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose, Kristina Martsenko

From: Kristina Martsenko <kristina.martsenko@arm.com>

Add support for handling 52bit guest physical address to the
VGIC layer. So far we have limited the guest physical address
to 48bits, by explicitly masking the upper bits. This patch
removes the restriction. We do not have to check if the host
supports 52bit as the gpa is always validated during an access.
(e.g, kvm_{read/write}_guest, kvm_is_visible_gfn()).
Also, the ITS table save-restore is also not affected with
the enhancement. The DTE entries already store the bits[51:8]
of the ITT_addr (with a 256byte alignment).

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Kristina Martsenko <kristina.martsenko@arm.com>
[ Macro clean ups, fix PROPBASER and PENDBASER accesses ]
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 include/linux/irqchip/arm-gic-v3.h |  5 +++++
 virt/kvm/arm/vgic/vgic-its.c       | 36 +++++++++---------------------
 virt/kvm/arm/vgic/vgic-mmio-v3.c   |  2 --
 3 files changed, 15 insertions(+), 28 deletions(-)

diff --git a/include/linux/irqchip/arm-gic-v3.h b/include/linux/irqchip/arm-gic-v3.h
index 8bdbb5f29494..e961f40992d7 100644
--- a/include/linux/irqchip/arm-gic-v3.h
+++ b/include/linux/irqchip/arm-gic-v3.h
@@ -357,6 +357,8 @@
 #define GITS_CBASER_RaWaWt	GIC_BASER_CACHEABILITY(GITS_CBASER, INNER, RaWaWt)
 #define GITS_CBASER_RaWaWb	GIC_BASER_CACHEABILITY(GITS_CBASER, INNER, RaWaWb)
 
+#define GITS_CBASER_ADDRESS(cbaser)	((cbaser) & GENMASK_ULL(52, 12))
+
 #define GITS_BASER_NR_REGS		8
 
 #define GITS_BASER_VALID			(1ULL << 63)
@@ -388,6 +390,9 @@
 #define GITS_BASER_ENTRY_SIZE_MASK	GENMASK_ULL(52, 48)
 #define GITS_BASER_PHYS_52_to_48(phys)					\
 	(((phys) & GENMASK_ULL(47, 16)) | (((phys) >> 48) & 0xf) << 12)
+#define GITS_BASER_ADDR_48_to_52(baser)					\
+	(((baser) & GENMASK_ULL(47, 16)) | (((baser) >> 12) & 0xf) << 48)
+
 #define GITS_BASER_SHAREABILITY_SHIFT	(10)
 #define GITS_BASER_InnerShareable					\
 	GIC_BASER_SHAREABILITY(GITS_BASER, InnerShareable)
diff --git a/virt/kvm/arm/vgic/vgic-its.c b/virt/kvm/arm/vgic/vgic-its.c
index 12502251727e..eb2a390a6c86 100644
--- a/virt/kvm/arm/vgic/vgic-its.c
+++ b/virt/kvm/arm/vgic/vgic-its.c
@@ -241,13 +241,6 @@ static struct its_ite *find_ite(struct vgic_its *its, u32 device_id,
 	list_for_each_entry(dev, &(its)->device_list, dev_list) \
 		list_for_each_entry(ite, &(dev)->itt_head, ite_list)
 
-/*
- * We only implement 48 bits of PA at the moment, although the ITS
- * supports more. Let's be restrictive here.
- */
-#define BASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 16))
-#define CBASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 12))
-
 #define GIC_LPI_OFFSET 8192
 
 #define VITS_TYPER_IDBITS 16
@@ -759,6 +752,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
 {
 	int l1_tbl_size = GITS_BASER_NR_PAGES(baser) * SZ_64K;
 	u64 indirect_ptr, type = GITS_BASER_TYPE(baser);
+	phys_addr_t base = GITS_BASER_ADDR_48_to_52(baser);
 	int esz = GITS_BASER_ENTRY_SIZE(baser);
 	int index;
 	gfn_t gfn;
@@ -783,7 +777,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
 		if (id >= (l1_tbl_size / esz))
 			return false;
 
-		addr = BASER_ADDRESS(baser) + id * esz;
+		addr = base + id * esz;
 		gfn = addr >> PAGE_SHIFT;
 
 		if (eaddr)
@@ -798,7 +792,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
 
 	/* Each 1st level entry is represented by a 64-bit value. */
 	if (kvm_read_guest_lock(its->dev->kvm,
-			   BASER_ADDRESS(baser) + index * sizeof(indirect_ptr),
+			   base + index * sizeof(indirect_ptr),
 			   &indirect_ptr, sizeof(indirect_ptr)))
 		return false;
 
@@ -808,11 +802,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
 	if (!(indirect_ptr & BIT_ULL(63)))
 		return false;
 
-	/*
-	 * Mask the guest physical address and calculate the frame number.
-	 * Any address beyond our supported 48 bits of PA will be caught
-	 * by the actual check in the final step.
-	 */
+	/* Mask the guest physical address and calculate the frame number. */
 	indirect_ptr &= GENMASK_ULL(51, 16);
 
 	/* Find the address of the actual entry */
@@ -1304,9 +1294,6 @@ static u64 vgic_sanitise_its_baser(u64 reg)
 				  GITS_BASER_OUTER_CACHEABILITY_SHIFT,
 				  vgic_sanitise_outer_cacheability);
 
-	/* Bits 15:12 contain bits 51:48 of the PA, which we don't support. */
-	reg &= ~GENMASK_ULL(15, 12);
-
 	/* We support only one (ITS) page size: 64K */
 	reg = (reg & ~GITS_BASER_PAGE_SIZE_MASK) | GITS_BASER_PAGE_SIZE_64K;
 
@@ -1325,11 +1312,8 @@ static u64 vgic_sanitise_its_cbaser(u64 reg)
 				  GITS_CBASER_OUTER_CACHEABILITY_SHIFT,
 				  vgic_sanitise_outer_cacheability);
 
-	/*
-	 * Sanitise the physical address to be 64k aligned.
-	 * Also limit the physical addresses to 48 bits.
-	 */
-	reg &= ~(GENMASK_ULL(51, 48) | GENMASK_ULL(15, 12));
+	/* Sanitise the physical address to be 64k aligned. */
+	reg &= ~GENMASK_ULL(15, 12);
 
 	return reg;
 }
@@ -1375,7 +1359,7 @@ static void vgic_its_process_commands(struct kvm *kvm, struct vgic_its *its)
 	if (!its->enabled)
 		return;
 
-	cbaser = CBASER_ADDRESS(its->cbaser);
+	cbaser = GITS_CBASER_ADDRESS(its->cbaser);
 
 	while (its->cwriter != its->creadr) {
 		int ret = kvm_read_guest_lock(kvm, cbaser + its->creadr,
@@ -2233,7 +2217,7 @@ static int vgic_its_restore_device_tables(struct vgic_its *its)
 	if (!(baser & GITS_BASER_VALID))
 		return 0;
 
-	l1_gpa = BASER_ADDRESS(baser);
+	l1_gpa = GITS_BASER_ADDR_48_to_52(baser);
 
 	if (baser & GITS_BASER_INDIRECT) {
 		l1_esz = GITS_LVL1_ENTRY_SIZE;
@@ -2305,7 +2289,7 @@ static int vgic_its_save_collection_table(struct vgic_its *its)
 {
 	const struct vgic_its_abi *abi = vgic_its_get_abi(its);
 	u64 baser = its->baser_coll_table;
-	gpa_t gpa = BASER_ADDRESS(baser);
+	gpa_t gpa = GITS_BASER_ADDR_48_to_52(baser);
 	struct its_collection *collection;
 	u64 val;
 	size_t max_size, filled = 0;
@@ -2354,7 +2338,7 @@ static int vgic_its_restore_collection_table(struct vgic_its *its)
 	if (!(baser & GITS_BASER_VALID))
 		return 0;
 
-	gpa = BASER_ADDRESS(baser);
+	gpa = GITS_BASER_ADDR_48_to_52(baser);
 
 	max_size = GITS_BASER_NR_PAGES(baser) * SZ_64K;
 
diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c b/virt/kvm/arm/vgic/vgic-mmio-v3.c
index a2a175b08b17..b3d1f0985117 100644
--- a/virt/kvm/arm/vgic/vgic-mmio-v3.c
+++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c
@@ -364,7 +364,6 @@ static u64 vgic_sanitise_pendbaser(u64 reg)
 				  vgic_sanitise_outer_cacheability);
 
 	reg &= ~PENDBASER_RES0_MASK;
-	reg &= ~GENMASK_ULL(51, 48);
 
 	return reg;
 }
@@ -382,7 +381,6 @@ static u64 vgic_sanitise_propbaser(u64 reg)
 				  vgic_sanitise_outer_cacheability);
 
 	reg &= ~PROPBASER_RES0_MASK;
-	reg &= ~GENMASK_ULL(51, 48);
 	return reg;
 }
 
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 15/18] kvm: arm64: Add 52bit support for PAR to HPFAR conversoin
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (13 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 14/18] vgic: Add support for 52bit guest physical address Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-25  9:59   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 16/18] kvm: arm64: Set a limit on the IPA size Suzuki K Poulose
                   ` (6 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Add support for handling 52bit addresses in PAR to HPFAR
conversion. Instead of hardcoding the address limits, we
now use PHYS_MASK_SHIFT.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arch/arm64/include/asm/kvm_arm.h | 7 +++++++
 arch/arm64/kvm/hyp/switch.c      | 2 +-
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 0a37c0513ede..241d1622fa19 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -308,6 +308,13 @@
 
 /* Hyp Prefetch Fault Address Register (HPFAR/HDFAR) */
 #define HPFAR_MASK	(~UL(0xf))
+/*
+ * We have
+ *	PAR	[PA_Shift - 1	: 12] = PA	[PA_Shift - 1 : 12]
+ *	HPFAR	[PA_Shift - 9	: 4]  = FIPA	[PA_Shift - 1 : 12]
+ */
+#define PAR_TO_HPFAR(par)		\
+	(((par) & GENMASK_ULL(PHYS_MASK_SHIFT - 1, 12)) >> 8)
 
 #define kvm_arm_exception_type	\
 	{0, "IRQ" }, 		\
diff --git a/arch/arm64/kvm/hyp/switch.c b/arch/arm64/kvm/hyp/switch.c
index 9d5ce1a3039a..7cc175c88a37 100644
--- a/arch/arm64/kvm/hyp/switch.c
+++ b/arch/arm64/kvm/hyp/switch.c
@@ -263,7 +263,7 @@ static bool __hyp_text __translate_far_to_hpfar(u64 far, u64 *hpfar)
 		return false; /* Translation failed, back to guest */
 
 	/* Convert PAR to HPFAR format */
-	*hpfar = ((tmp >> 12) & ((1UL << 36) - 1)) << 4;
+	*hpfar = PAR_TO_HPFAR(tmp);
 	return true;
 }
 
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 16/18] kvm: arm64: Set a limit on the IPA size
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (14 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 15/18] kvm: arm64: Add 52bit support for PAR to HPFAR conversoin Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-25  9:59   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 17/18] kvm: arm64: Limit the minimum number of page table levels Suzuki K Poulose
                   ` (5 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

So far we have restricted the IPA size of the VM to the default
value (40bits). Now that we can manage the IPA size per VM and
support dynamic stage2 page tables, we can allow VMs to have
larger IPA. This patch introduces a the maximum IPA size
supported on the host. This is decided by the following factors :

 1) Maximum PARange supported by the CPUs - This can be inferred
    from the system wide safe value.
 2) Maximum PA size supported by the host kernel (48 vs 52)
 3) Number of levels in the host page table (as we base our
    stage2 tables on the host table helpers).

Since the stage2 page table code is dependent on the stage1
page table, we always ensure that :

  Number of Levels at Stage1 >= Number of Levels at Stage2

So we limit the IPA to make sure that the above condition
is satisfied. This will affect the following combinations
of VA_BITS and IPA for different page sizes.

  Host configuration | Unsupported IPA ranges
  39bit VA, 4K       | [44, 48]
  36bit VA, 16K      | [41, 48]
  42bit VA, 64K      | [47, 52]

Supporting the above combinations need independent stage2
page table manipulation code, which would need substantial
changes. We could purse the solution independently and
switch the page table code once we have it ready.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since V2:
 - Restrict the IPA size to limit the number of page table
   levels in stage2 to that of stage1 or less.
---
 arch/arm/include/asm/kvm_mmu.h    |  2 ++
 arch/arm64/include/asm/kvm_host.h |  2 ++
 arch/arm64/kvm/reset.c            | 43 +++++++++++++++++++++++++++++++
 virt/kvm/arm/arm.c                |  2 ++
 4 files changed, 49 insertions(+)

diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
index 12ae5fbbcf01..5ad1a54f98dc 100644
--- a/arch/arm/include/asm/kvm_mmu.h
+++ b/arch/arm/include/asm/kvm_mmu.h
@@ -358,6 +358,8 @@ static inline int hyp_map_aux_data(void)
 
 #define kvm_phys_to_vttbr(addr)		(addr)
 
+static inline void kvm_set_ipa_limit(void) {}
+
 #endif	/* !__ASSEMBLY__ */
 
 #endif /* __ARM_KVM_MMU_H__ */
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 5ecd457bce7d..f0474061851d 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -513,6 +513,8 @@ static inline int kvm_arm_have_ssbd(void)
 void kvm_vcpu_load_sysregs(struct kvm_vcpu *vcpu);
 void kvm_vcpu_put_sysregs(struct kvm_vcpu *vcpu);
 
+void kvm_set_ipa_limit(void);
+
 #define __KVM_HAVE_ARCH_VM_ALLOC
 struct kvm *kvm_arch_alloc_vm(void);
 void kvm_arch_free_vm(struct kvm *kvm);
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index 51ecf0f7c912..76972b19bdd7 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -34,6 +34,9 @@
 #include <asm/kvm_coproc.h>
 #include <asm/kvm_mmu.h>
 
+/* Maximum phys_shift supported for any VM on this host */
+static u32 kvm_ipa_limit;
+
 /*
  * ARMv8 Reset Values
  */
@@ -135,6 +138,46 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
 	return kvm_timer_vcpu_reset(vcpu);
 }
 
+void kvm_set_ipa_limit(void)
+{
+	unsigned int ipa_max, va_max, parange;
+
+	parange = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1) & 0x7;
+	ipa_max = id_aa64mmfr0_parange_to_phys_shift(parange);
+
+	/* Raise the limit to the default size for backward compatibility */
+	if (ipa_max < KVM_PHYS_SHIFT) {
+		WARN_ONCE(1,
+			  "PARange is %d bits, unsupported configuration!",
+			  ipa_max);
+		ipa_max = KVM_PHYS_SHIFT;
+	}
+
+	/* Clamp it to the PA size supported by the kernel */
+	ipa_max = (ipa_max > PHYS_MASK_SHIFT) ? PHYS_MASK_SHIFT : ipa_max;
+	/*
+	 * Since our stage2 table is dependent on the stage1 page table code,
+	 * we must always honor the following condition:
+	 *
+	 *  Number of levels in Stage1 >= Number of levels in Stage2.
+	 *
+	 * So clamp the ipa limit further down to limit the number of levels.
+	 * Since we can concatenate upto 16 tables at entry level, we could
+	 * go upto 4bits above the maximum VA addressible with the current
+	 * number of levels.
+	 */
+	va_max = PGDIR_SHIFT + PAGE_SHIFT - 3;
+	va_max += 4;
+
+	if (va_max < ipa_max) {
+		kvm_info("Limiting IPA limit to %dbytes due to host VA bits limitation\n",
+			 va_max);
+		ipa_max = va_max;
+	}
+
+	kvm_ipa_limit = ipa_max;
+}
+
 /*
  * Configure the VTCR_EL2 for this VM. The VTCR value is common
  * across all the physical CPUs on the system. We use system wide
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 43e716bc3f08..631f9a3ad99a 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -1413,6 +1413,8 @@ static int init_common_resources(void)
 	kvm_vmid_bits = kvm_get_vmid_bits();
 	kvm_info("%d-bit VMID\n", kvm_vmid_bits);
 
+	kvm_set_ipa_limit();
+
 	return 0;
 }
 
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 17/18] kvm: arm64: Limit the minimum number of page table levels
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (15 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 16/18] kvm: arm64: Set a limit on the IPA size Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-25 10:00   ` Auger Eric
  2018-09-17 10:41 ` [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM Suzuki K Poulose
                   ` (4 subsequent siblings)
  21 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Since we are about to remove the lower limit on the IPA size,
make sure that we do not go to 1 level page table (e.g, with
32bit IPA on 64K host with concatenation) to avoid splitting
the host PMD huge pages at stage2.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arch/arm64/include/asm/stage2_pgtable.h |  8 +++++++-
 arch/arm64/kvm/reset.c                  | 12 +++++++++++-
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
index 352ec4158fdf..6a56fdff0823 100644
--- a/arch/arm64/include/asm/stage2_pgtable.h
+++ b/arch/arm64/include/asm/stage2_pgtable.h
@@ -72,8 +72,14 @@
 /*
  * The number of PTRS across all concatenated stage2 tables given by the
  * number of bits resolved at the initial level.
+ * If we force more number of levels than necessary, we may have
+ * stage2_pgdir_shift > IPA, in which case, stage2_pgd_ptrs will have
+ * one entry.
  */
-#define __s2_pgd_ptrs(ipa, lvls)	(1 << ((ipa) - pt_levels_pgdir_shift((lvls))))
+#define pgd_ptrs_shift(ipa, pgdir_shift)	\
+	((ipa) > (pgdir_shift) ? ((ipa) - (pgdir_shift)) : 0)
+#define __s2_pgd_ptrs(ipa, lvls)		\
+	(1 << (pgd_ptrs_shift((ipa), pt_levels_pgdir_shift(lvls))))
 #define __s2_pgd_size(ipa, lvls)	(__s2_pgd_ptrs((ipa), (lvls)) * sizeof(pgd_t))
 
 #define stage2_pgd_ptrs(kvm)		__s2_pgd_ptrs(kvm_phys_shift(kvm), kvm_stage2_levels(kvm))
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index 76972b19bdd7..0393bb974b23 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -190,10 +190,19 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
 {
 	u64 vtcr = VTCR_EL2_FLAGS;
 	u64 parange;
+	u8 lvls;
 
 	if (type)
 		return -EINVAL;
 
+	/*
+	 * Use a minimum 2 level page table to prevent splitting
+	 * host PMD huge pages at stage2.
+	 */
+	lvls = stage2_pgtable_levels(KVM_PHYS_SHIFT);
+	if (lvls < 2)
+		lvls = 2;
+
 	parange = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1) & 7;
 	if (parange > ID_AA64MMFR0_PARANGE_MAX)
 		parange = ID_AA64MMFR0_PARANGE_MAX;
@@ -210,7 +219,8 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
 	vtcr |= (kvm_get_vmid_bits() == 16) ?
 		VTCR_EL2_VS_16BIT :
 		VTCR_EL2_VS_8BIT;
-	vtcr |= VTCR_EL2_LVLS_TO_SL0(stage2_pgtable_levels(KVM_PHYS_SHIFT));
+
+	vtcr |= VTCR_EL2_LVLS_TO_SL0(lvls);
 	vtcr |= VTCR_EL2_T0SZ(KVM_PHYS_SHIFT);
 
 	kvm->arch.vtcr = vtcr;
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (16 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 17/18] kvm: arm64: Limit the minimum number of page table levels Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 14:20   ` Peter Maydell
                     ` (2 more replies)
  2018-09-17 10:41 ` [kvmtool PATCH v5 19/18] kvmtool: Allow backends to run checks on the KVM device fd Suzuki K Poulose
                   ` (3 subsequent siblings)
  21 siblings, 3 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose, Peter Maydel

Allow specifying the physical address size limit for a new
VM via the kvm_type argument for the KVM_CREATE_VM ioctl. This
allows us to finalise the stage2 page table as early as possible
and hence perform the right checks on the memory slots
without complication. The size is ecnoded as Log2(PA_Size) in
bits[7:0] of the type field. For backward compatibility the
value 0 is reserved and implies 40bits. Also, lift the limit
of the IPA to host limit and allow lower IPA sizes (e.g, 32).

The userspace could check the extension KVM_CAP_ARM_VM_PHYS_SHIFT
for the availability of this feature. The cap check returns the
maximum limit for the physical address shift supported by the host.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Christoffer Dall <cdall@kernel.org>
Cc: Peter Maydel <peter.maydell@linaro.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
Changes since v4:
 - Fold the introduction of the KVM_CAP_ARM_VM_PHYS_SHIFT to this
   patch to allow detection of the availability of the feature for
   userspace.
 - Document the API
 - Restrict the feature only to arm64.
Changes since V3:
 - Switch to a CAP, that can be checkd via EXTENSIONS on KVM device
   fd, rather than a dedicated ioctl.
---
 Documentation/virtual/kvm/api.txt       |  8 ++++++++
 arch/arm64/include/asm/stage2_pgtable.h | 20 --------------------
 arch/arm64/kvm/reset.c                  | 20 ++++++++++++++++----
 include/uapi/linux/kvm.h                | 10 ++++++++++
 4 files changed, 34 insertions(+), 24 deletions(-)

diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index c664064f76fb..f860251ff27c 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -122,6 +122,14 @@ the default trap & emulate implementation (which changes the virtual
 memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
 flag KVM_VM_MIPS_VZ.
 
+To configure the physical address space size for a VM (IPA size) on arm64,
+check KVM_CAP_ARM_VM_PHYS_SHIFT (which returns the maximum limit for the
+IPA shift) and use KVM_VM_TYPE_ARM_PHYS_SHIFT(PHYS_SHIFT). Bits[7-0] of the
+machine type has been reserved for specifying the PHYS_SHIFT.
+The supported range is [32...IPA_LIMIT], where IPA_LIMIT could be
+identified by checking KVM_CAP_ARM_VM_PHYS_SHIFT. For backward compatibility
+a value of 0 selects 40bits.
+
 
 4.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST
 
diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
index 6a56fdff0823..0b339f5a4a7c 100644
--- a/arch/arm64/include/asm/stage2_pgtable.h
+++ b/arch/arm64/include/asm/stage2_pgtable.h
@@ -42,28 +42,8 @@
  * the range (IPA_SHIFT, IPA_SHIFT - 4).
  */
 #define stage2_pgtable_levels(ipa)	ARM64_HW_PGTABLE_LEVELS((ipa) - 4)
-#define STAGE2_PGTABLE_LEVELS		stage2_pgtable_levels(KVM_PHYS_SHIFT)
 #define kvm_stage2_levels(kvm)		VTCR_EL2_LVLS(kvm->arch.vtcr)
 
-/*
- * With all the supported VA_BITs and 40bit guest IPA, the following condition
- * is always true:
- *
- *       STAGE2_PGTABLE_LEVELS <= CONFIG_PGTABLE_LEVELS
- *
- * We base our stage-2 page table walker helpers on this assumption and
- * fall back to using the host version of the helper wherever possible.
- * i.e, if a particular level is not folded (e.g, PUD) at stage2, we fall back
- * to using the host version, since it is guaranteed it is not folded at host.
- *
- * If the condition breaks in the future, we can rearrange the host level
- * definitions and reuse them for stage2. Till then...
- */
-#if STAGE2_PGTABLE_LEVELS > CONFIG_PGTABLE_LEVELS
-#error "Unsupported combination of guest IPA and host VA_BITS."
-#endif
-
-
 /* stage2_pgdir_shift() is the size mapped by top-level stage2 entry for the VM */
 #define stage2_pgdir_shift(kvm)		pt_levels_pgdir_shift(kvm_stage2_levels(kvm))
 #define stage2_pgdir_size(kvm)		(1ULL << stage2_pgdir_shift(kvm))
diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index 0393bb974b23..c9640159e11f 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -89,6 +89,9 @@ int kvm_arch_dev_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_VCPU_EVENTS:
 		r = 1;
 		break;
+	case KVM_CAP_ARM_VM_PHYS_SHIFT:
+		r = kvm_ipa_limit;
+		break;
 	default:
 		r = 0;
 	}
@@ -190,16 +193,25 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
 {
 	u64 vtcr = VTCR_EL2_FLAGS;
 	u64 parange;
-	u8 lvls;
+	u8 lvls, ipa_shift;
 
-	if (type)
+	if (type & ~KVM_VM_TYPE_ARM_PHYS_SHIFT_MASK)
 		return -EINVAL;
 
+	ipa_shift = KVM_VM_TYPE_ARM_PHYS_SHIFT(type);
+	if (ipa_shift) {
+		if (ipa_shift > kvm_ipa_limit ||
+		    ipa_shift < 32)
+			return -EINVAL;
+	} else {
+		ipa_shift = KVM_PHYS_SHIFT;
+	}
+
 	/*
 	 * Use a minimum 2 level page table to prevent splitting
 	 * host PMD huge pages at stage2.
 	 */
-	lvls = stage2_pgtable_levels(KVM_PHYS_SHIFT);
+	lvls = stage2_pgtable_levels(ipa_shift);
 	if (lvls < 2)
 		lvls = 2;
 
@@ -221,7 +233,7 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
 		VTCR_EL2_VS_8BIT;
 
 	vtcr |= VTCR_EL2_LVLS_TO_SL0(lvls);
-	vtcr |= VTCR_EL2_T0SZ(KVM_PHYS_SHIFT);
+	vtcr |= VTCR_EL2_T0SZ(ipa_shift);
 
 	kvm->arch.vtcr = vtcr;
 	return 0;
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 07548de5c988..2a6b29c446db 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -750,6 +750,15 @@ struct kvm_ppc_resize_hpt {
 
 #define KVM_S390_SIE_PAGE_OFFSET 1
 
+/*
+ * On arm64, machine type can be used to request the physical
+ * address size for the VM. Bits[7-0] has been reserved for the PA
+ * size shift (i.e, log2(PA_Size)). For backward compatibility,
+ * value 0 implies the default IPA size, 40bits.
+ */
+#define KVM_VM_TYPE_ARM_PHYS_SHIFT_MASK	0xffULL
+#define KVM_VM_TYPE_ARM_PHYS_SHIFT(x)		\
+	((x) & KVM_VM_TYPE_ARM_PHYS_SHIFT_MASK)
 /*
  * ioctls for /dev/kvm fds:
  */
@@ -952,6 +961,7 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_S390_HPAGE_1M 156
 #define KVM_CAP_NESTED_STATE 157
 #define KVM_CAP_ARM_INJECT_SERROR_ESR 158
+#define KVM_CAP_ARM_VM_PHYS_SHIFT 159 /* returns maximum PA shift for a VM */
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [kvmtool PATCH v5 19/18] kvmtool: Allow backends to run checks on the KVM device fd
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (17 preceding siblings ...)
  2018-09-17 10:41 ` [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [kvmtool PATCH v5 20/18] kvmtool: arm64: Add support for guest physical address size Suzuki K Poulose
                   ` (2 subsequent siblings)
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Allow architectures to perform initialisation based on the
KVM device fd ioctls, even before the VM is created.

Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 include/kvm/kvm.h | 4 ++++
 kvm.c             | 2 ++
 2 files changed, 6 insertions(+)

diff --git a/include/kvm/kvm.h b/include/kvm/kvm.h
index 1edacfd..dc3f0dd 100644
--- a/include/kvm/kvm.h
+++ b/include/kvm/kvm.h
@@ -135,6 +135,10 @@ int kvm__get_sock_by_instance(const char *name);
 int kvm__enumerate_instances(int (*callback)(const char *name, int pid));
 void kvm__remove_socket(const char *name);
 
+#ifndef kvm__arch_init_hyp
+static inline void kvm__arch_init_hyp(struct kvm *kvm) {}
+#endif
+
 void kvm__arch_set_cmdline(char *cmdline, bool video);
 void kvm__arch_init(struct kvm *kvm, const char *hugetlbfs_path, u64 ram_size);
 void kvm__arch_delete_ram(struct kvm *kvm);
diff --git a/kvm.c b/kvm.c
index 7de825a..0894778 100644
--- a/kvm.c
+++ b/kvm.c
@@ -373,6 +373,8 @@ int kvm__init(struct kvm *kvm)
 		goto err_sys_fd;
 	}
 
+	kvm__arch_init_hyp(kvm);
+
 	kvm->vm_fd = ioctl(kvm->sys_fd, KVM_CREATE_VM, KVM_VM_TYPE);
 	if (kvm->vm_fd < 0) {
 		pr_err("KVM_CREATE_VM ioctl");
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [kvmtool PATCH v5 20/18] kvmtool: arm64: Add support for guest physical address size
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (18 preceding siblings ...)
  2018-09-17 10:41 ` [kvmtool PATCH v5 19/18] kvmtool: Allow backends to run checks on the KVM device fd Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [kvmtool PATCH v5 21/18] kvmtool: arm64: Switch memory layout Suzuki K Poulose
  2018-09-17 10:41 ` [kvmtool PATCH v5 22/18] kvmtool: arm: Add support for creating VM with PA size Suzuki K Poulose
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Add an option to specify the physical address size used by this
VM.

Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arm/aarch64/include/kvm/kvm-config-arch.h | 5 ++++-
 arm/include/arm-common/kvm-config-arch.h  | 1 +
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/arm/aarch64/include/kvm/kvm-config-arch.h b/arm/aarch64/include/kvm/kvm-config-arch.h
index 04be43d..dabd22c 100644
--- a/arm/aarch64/include/kvm/kvm-config-arch.h
+++ b/arm/aarch64/include/kvm/kvm-config-arch.h
@@ -8,7 +8,10 @@
 			"Create PMUv3 device"),				\
 	OPT_U64('\0', "kaslr-seed", &(cfg)->kaslr_seed,			\
 			"Specify random seed for Kernel Address Space "	\
-			"Layout Randomization (KASLR)"),
+			"Layout Randomization (KASLR)"),		\
+	OPT_INTEGER('\0', "phys-shift", &(cfg)->phys_shift,		\
+			"Specify maximum physical address size (not "	\
+			"the amount of memory)"),
 
 #include "arm-common/kvm-config-arch.h"
 
diff --git a/arm/include/arm-common/kvm-config-arch.h b/arm/include/arm-common/kvm-config-arch.h
index 6a196f1..e0b531e 100644
--- a/arm/include/arm-common/kvm-config-arch.h
+++ b/arm/include/arm-common/kvm-config-arch.h
@@ -11,6 +11,7 @@ struct kvm_config_arch {
 	bool		has_pmuv3;
 	u64		kaslr_seed;
 	enum irqchip_type irqchip;
+	int		phys_shift;
 };
 
 int irqchip_parser(const struct option *opt, const char *arg, int unset);
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [kvmtool PATCH v5 21/18] kvmtool: arm64: Switch memory layout
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (19 preceding siblings ...)
  2018-09-17 10:41 ` [kvmtool PATCH v5 20/18] kvmtool: arm64: Add support for guest physical address size Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  2018-09-17 10:41 ` [kvmtool PATCH v5 22/18] kvmtool: arm: Add support for creating VM with PA size Suzuki K Poulose
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

If the guest wants to use a larger physical address space place
the RAM at upper half of the address space. Otherwise, it uses the
default layout.

Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arm/aarch32/include/kvm/kvm-arch.h |  6 ++++--
 arm/aarch64/include/kvm/kvm-arch.h | 15 ++++++++++++---
 arm/include/arm-common/kvm-arch.h  | 11 ++++++-----
 arm/kvm.c                          |  2 +-
 4 files changed, 23 insertions(+), 11 deletions(-)

diff --git a/arm/aarch32/include/kvm/kvm-arch.h b/arm/aarch32/include/kvm/kvm-arch.h
index cd31e72..bcd382b 100644
--- a/arm/aarch32/include/kvm/kvm-arch.h
+++ b/arm/aarch32/include/kvm/kvm-arch.h
@@ -3,8 +3,10 @@
 
 #define ARM_KERN_OFFSET(...)	0x8000
 
-#define ARM_MAX_MEMORY(...)	ARM_LOMAP_MAX_MEMORY
-
 #include "arm-common/kvm-arch.h"
 
+#define ARM_MAX_MEMORY(...)	ARM32_MAX_MEMORY
+#define ARM_MEMORY_AREA(...)	ARM32_MEMORY_AREA
+
+
 #endif /* KVM__KVM_ARCH_H */
diff --git a/arm/aarch64/include/kvm/kvm-arch.h b/arm/aarch64/include/kvm/kvm-arch.h
index 9de623a..bad35b9 100644
--- a/arm/aarch64/include/kvm/kvm-arch.h
+++ b/arm/aarch64/include/kvm/kvm-arch.h
@@ -1,14 +1,23 @@
 #ifndef KVM__KVM_ARCH_H
 #define KVM__KVM_ARCH_H
 
+#include "arm-common/kvm-arch.h"
+
+#define ARM64_MEMORY_AREA(phys_shift)	(1UL << (phys_shift - 1))
+#define ARM64_MAX_MEMORY(phys_shift)	\
+	((1ULL << (phys_shift)) - ARM64_MEMORY_AREA(phys_shift))
+
+#define ARM_MEMORY_AREA(kvm)	((kvm)->cfg.arch.aarch32_guest ?	\
+				 ARM32_MEMORY_AREA		:	\
+				 ARM64_MEMORY_AREA(kvm->cfg.arch.phys_shift))
+
 #define ARM_KERN_OFFSET(kvm)	((kvm)->cfg.arch.aarch32_guest	?	\
 				0x8000				:	\
 				0x80000)
 
 #define ARM_MAX_MEMORY(kvm)	((kvm)->cfg.arch.aarch32_guest	?	\
-				ARM_LOMAP_MAX_MEMORY		:	\
-				ARM_HIMAP_MAX_MEMORY)
+				ARM32_MAX_MEMORY		:	\
+				ARM64_MAX_MEMORY(kvm->cfg.arch.phys_shift))
 
-#include "arm-common/kvm-arch.h"
 
 #endif /* KVM__KVM_ARCH_H */
diff --git a/arm/include/arm-common/kvm-arch.h b/arm/include/arm-common/kvm-arch.h
index b9d486d..b29b4b1 100644
--- a/arm/include/arm-common/kvm-arch.h
+++ b/arm/include/arm-common/kvm-arch.h
@@ -6,14 +6,15 @@
 #include <linux/types.h>
 
 #include "arm-common/gic.h"
-
 #define ARM_IOPORT_AREA		_AC(0x0000000000000000, UL)
 #define ARM_MMIO_AREA		_AC(0x0000000000010000, UL)
 #define ARM_AXI_AREA		_AC(0x0000000040000000, UL)
-#define ARM_MEMORY_AREA		_AC(0x0000000080000000, UL)
 
-#define ARM_LOMAP_MAX_MEMORY	((1ULL << 32) - ARM_MEMORY_AREA)
-#define ARM_HIMAP_MAX_MEMORY	((1ULL << 40) - ARM_MEMORY_AREA)
+#define ARM32_MEMORY_AREA	_AC(0x0000000080000000, UL)
+#define ARM32_MAX_MEMORY	((1ULL << 32) - ARM32_MEMORY_AREA)
+
+#define ARM_IOMEM_AREA_END	ARM32_MEMORY_AREA
+
 
 #define ARM_GIC_DIST_BASE	(ARM_AXI_AREA - ARM_GIC_DIST_SIZE)
 #define ARM_GIC_CPUI_BASE	(ARM_GIC_DIST_BASE - ARM_GIC_CPUI_SIZE)
@@ -24,7 +25,7 @@
 #define ARM_IOPORT_SIZE		(ARM_MMIO_AREA - ARM_IOPORT_AREA)
 #define ARM_VIRTIO_MMIO_SIZE	(ARM_AXI_AREA - (ARM_MMIO_AREA + ARM_GIC_SIZE))
 #define ARM_PCI_CFG_SIZE	(1ULL << 24)
-#define ARM_PCI_MMIO_SIZE	(ARM_MEMORY_AREA - \
+#define ARM_PCI_MMIO_SIZE	(ARM_IOMEM_AREA_END - \
 				(ARM_AXI_AREA + ARM_PCI_CFG_SIZE))
 
 #define KVM_IOPORT_AREA		ARM_IOPORT_AREA
diff --git a/arm/kvm.c b/arm/kvm.c
index b824f63..eac2ad2 100644
--- a/arm/kvm.c
+++ b/arm/kvm.c
@@ -30,7 +30,7 @@ void kvm__init_ram(struct kvm *kvm)
 	u64 phys_start, phys_size;
 	void *host_mem;
 
-	phys_start	= ARM_MEMORY_AREA;
+	phys_start	= ARM_MEMORY_AREA(kvm);
 	phys_size	= kvm->ram_size;
 	host_mem	= kvm->ram_start;
 
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* [kvmtool PATCH v5 22/18] kvmtool: arm: Add support for creating VM with PA size
  2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
                   ` (20 preceding siblings ...)
  2018-09-17 10:41 ` [kvmtool PATCH v5 21/18] kvmtool: arm64: Switch memory layout Suzuki K Poulose
@ 2018-09-17 10:41 ` Suzuki K Poulose
  21 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 10:41 UTC (permalink / raw)
  To: linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, eric.auger, pbonzini, rkrcmar,
	will.deacon, catalin.marinas, james.morse, dave.martin,
	julien.grall, linux-kernel, suzuki.poulose

Specify the physical size for the VM encoded in the vm type.

Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
 arm/include/arm-common/kvm-arch.h |  6 +++++-
 arm/kvm.c                         | 27 +++++++++++++++++++++++++++
 2 files changed, 32 insertions(+), 1 deletion(-)

diff --git a/arm/include/arm-common/kvm-arch.h b/arm/include/arm-common/kvm-arch.h
index b29b4b1..d77f3ac 100644
--- a/arm/include/arm-common/kvm-arch.h
+++ b/arm/include/arm-common/kvm-arch.h
@@ -44,7 +44,11 @@
 
 #define KVM_IRQ_OFFSET		GIC_SPI_IRQ_BASE
 
-#define KVM_VM_TYPE		0
+extern unsigned long		kvm_arm_type;
+extern void kvm__arch_init_hyp(struct kvm *kvm);
+
+#define KVM_VM_TYPE		kvm_arm_type
+#define kvm__arch_init_hyp	kvm__arch_init_hyp
 
 #define VIRTIO_DEFAULT_TRANS(kvm)	\
 	((kvm)->cfg.arch.virtio_trans_pci ? VIRTIO_PCI : VIRTIO_MMIO)
diff --git a/arm/kvm.c b/arm/kvm.c
index eac2ad2..6c19dd0 100644
--- a/arm/kvm.c
+++ b/arm/kvm.c
@@ -11,6 +11,8 @@
 #include <linux/kvm.h>
 #include <linux/sizes.h>
 
+unsigned long kvm_arm_type;
+
 struct kvm_ext kvm_req_ext[] = {
 	{ DEFINE_KVM_EXT(KVM_CAP_IRQCHIP) },
 	{ DEFINE_KVM_EXT(KVM_CAP_ONE_REG) },
@@ -18,6 +20,31 @@ struct kvm_ext kvm_req_ext[] = {
 	{ 0, 0 },
 };
 
+#ifndef KVM_CAP_ARM_VM_PHYS_SHIFT
+#define KVM_CAP_ARM_VM_PHYS_SHIFT	159
+#endif
+
+void kvm__arch_init_hyp(struct kvm *kvm)
+{
+	int max_ipa;
+
+	if (!kvm->cfg.arch.phys_shift)
+		kvm->cfg.arch.phys_shift = 40;
+	if (kvm->cfg.arch.phys_shift == 40)
+		return;
+	max_ipa = ioctl(kvm->sys_fd,
+			KVM_CHECK_EXTENSION, KVM_CAP_ARM_VM_PHYS_SHIFT);
+	if (!max_ipa)
+		die("Kernel doesn't support IPA size configuration\n");
+	if ((kvm->cfg.arch.phys_shift > max_ipa) ||
+	    (kvm->cfg.arch.phys_shift < 32))
+		die("Requested PA size (%u) is not supported by the host"
+		    " [32 - %u]bit\n",
+		    kvm->cfg.arch.phys_shift, max_ipa);
+	if (kvm->cfg.arch.phys_shift < 32)
+	kvm_arm_type = kvm->cfg.arch.phys_shift;
+}
+
 bool kvm__arch_cpu_supports_vm(void)
 {
 	/* The KVM capability check is enough. */
-- 
2.19.0


^ permalink raw reply related	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-17 10:41 ` [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM Suzuki K Poulose
@ 2018-09-17 14:20   ` Peter Maydell
  2018-09-17 14:43     ` Suzuki K Poulose
  2018-09-18  1:55   ` Peter Maydell
  2018-09-25 10:00   ` Auger Eric
  2 siblings, 1 reply; 50+ messages in thread
From: Peter Maydell @ 2018-09-17 14:20 UTC (permalink / raw)
  To: Suzuki K Poulose
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

On 17 September 2018 at 11:41, Suzuki K Poulose <suzuki.poulose@arm.com> wrote:
> Allow specifying the physical address size limit for a new
> VM via the kvm_type argument for the KVM_CREATE_VM ioctl. This
> allows us to finalise the stage2 page table as early as possible
> and hence perform the right checks on the memory slots
> without complication. The size is ecnoded as Log2(PA_Size) in
> bits[7:0] of the type field. For backward compatibility the
> value 0 is reserved and implies 40bits. Also, lift the limit
> of the IPA to host limit and allow lower IPA sizes (e.g, 32).
>
> The userspace could check the extension KVM_CAP_ARM_VM_PHYS_SHIFT
> for the availability of this feature. The cap check returns the
> maximum limit for the physical address shift supported by the host.
>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Cc: Peter Maydel <peter.maydell@linaro.org>

Typo: my surname has two "l"s in it.

thanks
-- PMM

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-17 14:20   ` Peter Maydell
@ 2018-09-17 14:43     ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-17 14:43 UTC (permalink / raw)
  To: Peter Maydell
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

Hi Peter,

On 17/09/2018 15:20, Peter Maydell wrote:
> On 17 September 2018 at 11:41, Suzuki K Poulose <suzuki.poulose@arm.com> wrote:
>> Allow specifying the physical address size limit for a new
>> VM via the kvm_type argument for the KVM_CREATE_VM ioctl. This
>> allows us to finalise the stage2 page table as early as possible
>> and hence perform the right checks on the memory slots
>> without complication. The size is ecnoded as Log2(PA_Size) in
>> bits[7:0] of the type field. For backward compatibility the
>> value 0 is reserved and implies 40bits. Also, lift the limit
>> of the IPA to host limit and allow lower IPA sizes (e.g, 32).
>>
>> The userspace could check the extension KVM_CAP_ARM_VM_PHYS_SHIFT
>> for the availability of this feature. The cap check returns the
>> maximum limit for the physical address shift supported by the host.
>>
>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>> Cc: Christoffer Dall <cdall@kernel.org>
>> Cc: Peter Maydel <peter.maydell@linaro.org>
> 
> Typo: my surname has two "l"s in it.

Sorry about that. I have fixed this locally. Btw, did you get
a chance to look at the patch itself ? If you did, are you fine
with the API ?

Suzuki


^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-17 10:41 ` [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM Suzuki K Poulose
  2018-09-17 14:20   ` Peter Maydell
@ 2018-09-18  1:55   ` Peter Maydell
  2018-09-18 15:16     ` Suzuki K Poulose
  2018-09-25 10:00   ` Auger Eric
  2 siblings, 1 reply; 50+ messages in thread
From: Peter Maydell @ 2018-09-18  1:55 UTC (permalink / raw)
  To: Suzuki K Poulose
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

On 17 September 2018 at 11:41, Suzuki K Poulose <suzuki.poulose@arm.com> wrote:
> --- a/Documentation/virtual/kvm/api.txt
> +++ b/Documentation/virtual/kvm/api.txt
> @@ -122,6 +122,14 @@ the default trap & emulate implementation (which changes the virtual
>  memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
>  flag KVM_VM_MIPS_VZ.
>
> +To configure the physical address space size for a VM (IPA size) on arm64,
> +check KVM_CAP_ARM_VM_PHYS_SHIFT (which returns the maximum limit for the
> +IPA shift) and use KVM_VM_TYPE_ARM_PHYS_SHIFT(PHYS_SHIFT). Bits[7-0] of the
> +machine type has been reserved for specifying the PHYS_SHIFT.
> +The supported range is [32...IPA_LIMIT], where IPA_LIMIT could be
> +identified by checking KVM_CAP_ARM_VM_PHYS_SHIFT. For backward compatibility
> +a value of 0 selects 40bits.
> +

Given this as the API documentation, I don't think I could figure out
what I as a userspace user of it need to do without looking at the
kernel code. Could I ask you to expand it so that it is a bit less
terse and a bit more detailed? (For instance, what is a PHYS_SHIFT
and why do I have to specify it rather than just telling the kernel
I want a 48 bit guest address space?)

thanks
-- PMM

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-18  1:55   ` Peter Maydell
@ 2018-09-18 15:16     ` Suzuki K Poulose
  2018-09-18 15:36       ` Peter Maydell
  0 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-18 15:16 UTC (permalink / raw)
  To: Peter Maydell
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

Hi Peter,

On 18/09/2018 02:55, Peter Maydell wrote:
> On 17 September 2018 at 11:41, Suzuki K Poulose <suzuki.poulose@arm.com> wrote:
>> --- a/Documentation/virtual/kvm/api.txt
>> +++ b/Documentation/virtual/kvm/api.txt
>> @@ -122,6 +122,14 @@ the default trap & emulate implementation (which changes the virtual
>>   memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
>>   flag KVM_VM_MIPS_VZ.
>>
>> +To configure the physical address space size for a VM (IPA size) on arm64,
>> +check KVM_CAP_ARM_VM_PHYS_SHIFT (which returns the maximum limit for the
>> +IPA shift) and use KVM_VM_TYPE_ARM_PHYS_SHIFT(PHYS_SHIFT). Bits[7-0] of the
>> +machine type has been reserved for specifying the PHYS_SHIFT.
>> +The supported range is [32...IPA_LIMIT], where IPA_LIMIT could be
>> +identified by checking KVM_CAP_ARM_VM_PHYS_SHIFT. For backward compatibility
>> +a value of 0 selects 40bits.
>> +
> 
> Given this as the API documentation, I don't think I could figure out
> what I as a userspace user of it need to do without looking at the
> kernel code. Could I ask you to expand it so that it is a bit less
> terse and a bit more detailed? (For instance, what is a PHYS_SHIFT
> and why do I have to specify it rather than just telling the kernel
> I want a 48 bit guest address space?)

Thanks for the feedback.  I acknowledge that the documentation is not
quite clear for a userspace user. How about:

"To configure the physical address space size for a VM (IPA size) on arm64,
check KVM_CAP_ARM_VM_IPA_SIZE and use KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits)
as the argument to KVM_CREATE_VM, where IPA_Bits is the maximum width of
any physical address used by the VM. The IPA_Bits is encoded in Bits[7-0]
of the machine type, and must be one of { 0, 32, ... , Host_IPA_Limit },
where :
1) IPA_Bits = 0 implies 40bits IPA (for backward compatibility)
2) Host_IPA_Limit is the maximum limit for IPA_Bits on the host, which is
    dependent on the CPU capability and the host kernel configuration.
    This can be detected by checking the extension KVM_CAP_ARM_VM_IPA_SIZE
"

note: I have renamed the KVM_CAP_ARM_VM_PHYS_SHIFT => KVM_CAP_ARM_VM_IPA_SIZE
above and can update the code, if the latter is better.

Suzuki

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-18 15:16     ` Suzuki K Poulose
@ 2018-09-18 15:36       ` Peter Maydell
  2018-09-18 16:27         ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Peter Maydell @ 2018-09-18 15:36 UTC (permalink / raw)
  To: Suzuki K Poulose
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

On 18 September 2018 at 16:16, Suzuki K Poulose <Suzuki.Poulose@arm.com> wrote:
> Hi Peter,
>
> On 18/09/2018 02:55, Peter Maydell wrote:
>>
>> On 17 September 2018 at 11:41, Suzuki K Poulose <suzuki.poulose@arm.com>
>> wrote:
>>>
>>> --- a/Documentation/virtual/kvm/api.txt
>>> +++ b/Documentation/virtual/kvm/api.txt
>>> @@ -122,6 +122,14 @@ the default trap & emulate implementation (which
>>> changes the virtual
>>>   memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
>>>   flag KVM_VM_MIPS_VZ.
>>>
>>> +To configure the physical address space size for a VM (IPA size) on
>>> arm64,
>>> +check KVM_CAP_ARM_VM_PHYS_SHIFT (which returns the maximum limit for the
>>> +IPA shift) and use KVM_VM_TYPE_ARM_PHYS_SHIFT(PHYS_SHIFT). Bits[7-0] of
>>> the
>>> +machine type has been reserved for specifying the PHYS_SHIFT.
>>> +The supported range is [32...IPA_LIMIT], where IPA_LIMIT could be
>>> +identified by checking KVM_CAP_ARM_VM_PHYS_SHIFT. For backward
>>> compatibility
>>> +a value of 0 selects 40bits.
>>> +
>>
>>
>> Given this as the API documentation, I don't think I could figure out
>> what I as a userspace user of it need to do without looking at the
>> kernel code. Could I ask you to expand it so that it is a bit less
>> terse and a bit more detailed? (For instance, what is a PHYS_SHIFT
>> and why do I have to specify it rather than just telling the kernel
>> I want a 48 bit guest address space?)
>
>
> Thanks for the feedback.  I acknowledge that the documentation is not
> quite clear for a userspace user. How about:
>
> "To configure the physical address space size for a VM (IPA size) on arm64,
> check KVM_CAP_ARM_VM_IPA_SIZE and use KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits)
> as the argument to KVM_CREATE_VM, where IPA_Bits is the maximum width of
> any physical address used by the VM. The IPA_Bits is encoded in Bits[7-0]
> of the machine type, and must be one of { 0, 32, ... , Host_IPA_Limit },
> where :
> 1) IPA_Bits = 0 implies 40bits IPA (for backward compatibility)
> 2) Host_IPA_Limit is the maximum limit for IPA_Bits on the host, which is
>    dependent on the CPU capability and the host kernel configuration.
>    This can be detected by checking the extension KVM_CAP_ARM_VM_IPA_SIZE
> "

I think this is still somewhat confusing. In particular, you're
describing both the "ask the kernel what it supports" API and
the "tell the kernel what we want" API in a single sentence
("...check KVM_CAP_ARM_VM_IPA_SIZE and use KVM_VM_TYPE_ARM_IPA_SIZE...").
There isn't a length limit on documentation, so why not describe
them both clearly in separate sentences?

Also, can I use any IPA value between 32 and Host_IPA_Limit, or
are only certain values in that range supported (if so, which)?

> note: I have renamed the KVM_CAP_ARM_VM_PHYS_SHIFT =>
> KVM_CAP_ARM_VM_IPA_SIZE
> above and can update the code, if the latter is better.

I think the new name is definitely clearer, thanks.

-- PMM

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-18 15:36       ` Peter Maydell
@ 2018-09-18 16:27         ` Suzuki K Poulose
  2018-09-18 17:15           ` Peter Maydell
  0 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-18 16:27 UTC (permalink / raw)
  To: Peter Maydell
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

Hi Peter,

On 18/09/18 16:36, Peter Maydell wrote:
> On 18 September 2018 at 16:16, Suzuki K Poulose <Suzuki.Poulose@arm.com> wrote:
>> Hi Peter,
>>
>> On 18/09/2018 02:55, Peter Maydell wrote:
>>>
>>> On 17 September 2018 at 11:41, Suzuki K Poulose <suzuki.poulose@arm.com>
>>> wrote:
>>>>
>>>> --- a/Documentation/virtual/kvm/api.txt
>>>> +++ b/Documentation/virtual/kvm/api.txt
>>>> @@ -122,6 +122,14 @@ the default trap & emulate implementation (which
>>>> changes the virtual
>>>>    memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
>>>>    flag KVM_VM_MIPS_VZ.
>>>>
>>>> +To configure the physical address space size for a VM (IPA size) on
>>>> arm64,
>>>> +check KVM_CAP_ARM_VM_PHYS_SHIFT (which returns the maximum limit for the
>>>> +IPA shift) and use KVM_VM_TYPE_ARM_PHYS_SHIFT(PHYS_SHIFT). Bits[7-0] of
>>>> the
>>>> +machine type has been reserved for specifying the PHYS_SHIFT.
>>>> +The supported range is [32...IPA_LIMIT], where IPA_LIMIT could be
>>>> +identified by checking KVM_CAP_ARM_VM_PHYS_SHIFT. For backward
>>>> compatibility
>>>> +a value of 0 selects 40bits.
>>>> +
>>>
>>>
>>> Given this as the API documentation, I don't think I could figure out
>>> what I as a userspace user of it need to do without looking at the
>>> kernel code. Could I ask you to expand it so that it is a bit less
>>> terse and a bit more detailed? (For instance, what is a PHYS_SHIFT
>>> and why do I have to specify it rather than just telling the kernel
>>> I want a 48 bit guest address space?)
>>
>>
>> Thanks for the feedback.  I acknowledge that the documentation is not
>> quite clear for a userspace user. How about:
>>
>> "To configure the physical address space size for a VM (IPA size) on arm64,
>> check KVM_CAP_ARM_VM_IPA_SIZE and use KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits)
>> as the argument to KVM_CREATE_VM, where IPA_Bits is the maximum width of
>> any physical address used by the VM. The IPA_Bits is encoded in Bits[7-0]
>> of the machine type, and must be one of { 0, 32, ... , Host_IPA_Limit },
>> where :
>> 1) IPA_Bits = 0 implies 40bits IPA (for backward compatibility)
>> 2) Host_IPA_Limit is the maximum limit for IPA_Bits on the host, which is
>>     dependent on the CPU capability and the host kernel configuration.
>>     This can be detected by checking the extension KVM_CAP_ARM_VM_IPA_SIZE
>> "
> 
> I think this is still somewhat confusing. In particular, you're
> describing both the "ask the kernel what it supports" API and
> the "tell the kernel what we want" API in a single sentence
> ("...check KVM_CAP_ARM_VM_IPA_SIZE and use KVM_VM_TYPE_ARM_IPA_SIZE...").
> There isn't a length limit on documentation, so why not describe
> them both clearly in separate sentences?

Sure, will do. I didn't want to hijack the KVM_CREATE_VM ioctl command
section to explain a kvm CAP.

> 
> Also, can I use any IPA value between 32 and Host_IPA_Limit, or
> are only certain values in that range supported (if so, which)?

Any value in the range is supported. I should probably add that
"The configured IPA size is different from what is observed by guest in
ID_AA64MMFR0_EL1[PARange] to avoid the confusion." So here it goes :

---

"On arm64, the physical address size for a VM (IPA Size limit) is limited
to 40bits by default. The limit can be configured if the host supports the
extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use
KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type
identifier, where IPA_Bits is the maximum width of any physical
address used by the VM. The IPA_Bits is encoded in bits[7-0] of the
machine type identifier.

e.g, to configure a guest to use 48bit physical address size :

	vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48));

The requested size (IPA_Bits) must be :
   0 - Implies default 40bits (for backward compatibility)

   or

   N - Implies N bits, where N is a positive integer such that 32 <= N <= Host_IPA_Limit

Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and
is dependent on the CPU capability and the kernel configuration. The limit can
be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION
ioctl() at run-time.

Please note that configuring the IPA size does not affect the capability
exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects
the guest to host physical address (stage2) translations setup by the host.
"

Suzuki

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-18 16:27         ` Suzuki K Poulose
@ 2018-09-18 17:15           ` Peter Maydell
  2018-09-19 10:03             ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Peter Maydell @ 2018-09-18 17:15 UTC (permalink / raw)
  To: Suzuki K Poulose
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

On 18 September 2018 at 17:27, Suzuki K Poulose <Suzuki.Poulose@arm.com> wrote:
> ---
>
> "On arm64, the physical address size for a VM (IPA Size limit) is limited
> to 40bits by default. The limit can be configured if the host supports the
> extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use
> KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type
> identifier, where IPA_Bits is the maximum width of any physical
> address used by the VM. The IPA_Bits is encoded in bits[7-0] of the
> machine type identifier.
>
> e.g, to configure a guest to use 48bit physical address size :
>
>         vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48));
>
> The requested size (IPA_Bits) must be :
>   0 - Implies default 40bits (for backward compatibility)
>
>   or
>
>   N - Implies N bits, where N is a positive integer such that 32 <= N <=
> Host_IPA_Limit
>
> Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and
> is dependent on the CPU capability and the kernel configuration. The limit
> can
> be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION
> ioctl() at run-time.
>
> Please note that configuring the IPA size does not affect the capability
> exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects
> the guest to host physical address (stage2) translations setup by the host.
> "

Thanks, this is much clearer. The only bit I'm not sure about is that
last paragraph -- if I ask for a VM with a 48 bit address space why
don't we tell the guest that that's what it has ?

thanks
-- PMM

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-18 17:15           ` Peter Maydell
@ 2018-09-19 10:03             ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-19 10:03 UTC (permalink / raw)
  To: Peter Maydell
  Cc: arm-mail-list, kvmarm, kvm-devel, Marc Zyngier, Christoffer Dall,
	Eric Auger, Paolo Bonzini, Radim Krčmář,
	Will Deacon, Catalin Marinas, James Morse, Dave P Martin,
	julien.grall, lkml - Kernel Mailing List

On 09/18/2018 06:15 PM, Peter Maydell wrote:
> On 18 September 2018 at 17:27, Suzuki K Poulose <Suzuki.Poulose@arm.com> wrote:
>> ---
>>
>> "On arm64, the physical address size for a VM (IPA Size limit) is limited
>> to 40bits by default. The limit can be configured if the host supports the
>> extension KVM_CAP_ARM_VM_IPA_SIZE. When supported, use
>> KVM_VM_TYPE_ARM_IPA_SIZE(IPA_Bits) to set the size in the machine type
>> identifier, where IPA_Bits is the maximum width of any physical
>> address used by the VM. The IPA_Bits is encoded in bits[7-0] of the
>> machine type identifier.
>>
>> e.g, to configure a guest to use 48bit physical address size :
>>
>>          vm_fd = ioctl(dev_fd, KVM_CREATE_VM, KVM_VM_TYPE_ARM_IPA_SIZE(48));
>>
>> The requested size (IPA_Bits) must be :
>>    0 - Implies default 40bits (for backward compatibility)
>>
>>    or
>>
>>    N - Implies N bits, where N is a positive integer such that 32 <= N <=
>> Host_IPA_Limit
>>
>> Host_IPA_Limit is the maximum possible value for IPA_Bits on the host and
>> is dependent on the CPU capability and the kernel configuration. The limit
>> can
>> be retrieved using KVM_CAP_ARM_VM_IPA_SIZE of the KVM_CHECK_EXTENSION
>> ioctl() at run-time.
>>
>> Please note that configuring the IPA size does not affect the capability
>> exposed by the guest CPUs in ID_AA64MMFR0_EL1[PARange]. It only affects
>> the guest to host physical address (stage2) translations setup by the host.
>> "
> 
> Thanks, this is much clearer. The only bit I'm not sure about is that
> last paragraph -- if I ask for a VM with a 48 bit address space why
> don't we tell the guest that that's what it has ?

The point is the IPA Size is not a limit on the CPU's PA size. e.g, if
this guest was a nested hypervisor with 48bit IPA, it could still
support a 52bit IPA nested guest within by simply looking up the CPU
PARange. The IPA size configuration here is simply a hint to the
hypervisor on where the memory banks would be kept. And this certainly
true on real platforms (e.g, my Juno has 42bit PARange on A57, while
A53 reports 40bit, with PA max at 40bits).

Suzuki

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 07/18] kvm: arm64: Configure VTCR_EL2 per VM
  2018-09-17 10:41 ` [PATCH v5 07/18] kvm: arm64: Configure VTCR_EL2 per VM Suzuki K Poulose
@ 2018-09-20 10:21   ` Auger Eric
  2018-09-20 10:38     ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Auger Eric @ 2018-09-20 10:21 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> Add support for setting the VTCR_EL2 per VM, rather than hard
> coding a value at boot time per CPU. This would allow us to tune
> the stage2 page table parameters per VM in the later changes.
> 
> We compute the VTCR fields based on the system wide sanitised
> feature registers, except for the hardware management of Access
> Flags (VTCR_EL2.HA). It is fine to run a system with a mix of
> CPUs that may or may not update the page table Access Flags.
> Since the bit is RES0 on CPUs that don't support it, the bit
> should be ignored on them.
> 
> Suggested-by: Marc Zyngier <marc.zyngier@arm.com>
> Acked-by: Christoffer Dall <cdall@kernel.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
>  arch/arm64/include/asm/kvm_arm.h  |  3 +-
>  arch/arm64/include/asm/kvm_asm.h  |  2 -
>  arch/arm64/include/asm/kvm_host.h | 12 ++++--
>  arch/arm64/include/asm/kvm_hyp.h  |  1 +
>  arch/arm64/kvm/hyp/Makefile       |  1 -
>  arch/arm64/kvm/hyp/s2-setup.c     | 72 -------------------------------
>  arch/arm64/kvm/reset.c            | 30 +++++++++++++
>  7 files changed, 40 insertions(+), 81 deletions(-)
>  delete mode 100644 arch/arm64/kvm/hyp/s2-setup.c
> 
> diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
> index 5f807b680a5f..14317b3a1820 100644
> --- a/arch/arm64/include/asm/kvm_arm.h
> +++ b/arch/arm64/include/asm/kvm_arm.h
> @@ -135,8 +135,7 @@
>   * 40 bits wide (T0SZ = 24).  Systems with a PARange smaller than 40 bits are
>   * not known to exist and will break with this configuration.
>   *
> - * VTCR_EL2.PS is extracted from ID_AA64MMFR0_EL1.PARange at boot time
> - * (see hyp-init.S).
> + * The VTCR_EL2 is configured per VM and is initialised in kvm_arm_config_vm().
>   *
>   * Note that when using 4K pages, we concatenate two first level page tables
>   * together. With 16K pages, we concatenate 16 first level page tables.
> diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
> index 102b5a5c47b6..0b53c72e7591 100644
> --- a/arch/arm64/include/asm/kvm_asm.h
> +++ b/arch/arm64/include/asm/kvm_asm.h
> @@ -72,8 +72,6 @@ extern void __vgic_v3_init_lrs(void);
>  
>  extern u32 __kvm_get_mdcr_el2(void);
>  
> -extern u32 __init_stage2_translation(void);
> -
>  /* Home-grown __this_cpu_{ptr,read} variants that always work at HYP */
>  #define __hyp_this_cpu_ptr(sym)						\
>  	({								\
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index b04280ae1be0..5ecd457bce7d 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -61,11 +61,13 @@ struct kvm_arch {
>  	u64    vmid_gen;
>  	u32    vmid;
>  
> -	/* 1-level 2nd stage table, protected by kvm->mmu_lock */
> +	/* stage2 entry level table */
>  	pgd_t *pgd;
>  
>  	/* VTTBR value associated with above pgd and vmid */
>  	u64    vttbr;
> +	/* VTCR_EL2 value for this VM */
> +	u64    vtcr;
>  
>  	/* The last vcpu id that ran on each physical CPU */
>  	int __percpu *last_vcpu_ran;
> @@ -442,10 +444,12 @@ int kvm_arm_vcpu_arch_has_attr(struct kvm_vcpu *vcpu,
>  
>  static inline void __cpu_init_stage2(void)
>  {
> -	u32 parange = kvm_call_hyp(__init_stage2_translation);
> +	u32 ps;
>  
> -	WARN_ONCE(parange < 40,
> -		  "PARange is %d bits, unsupported configuration!", parange);
> +	/* Sanity check for minimum IPA size support */
> +	ps = id_aa64mmfr0_parange_to_phys_shift(read_sysreg(id_aa64mmfr0_el1) & 0x7);
> +	WARN_ONCE(ps < 40,
> +		  "PARange is %d bits, unsupported configuration!", ps);
>  }
>  
>  /* Guest/host FPSIMD coordination helpers */
> diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h
> index d1bd1e0f14d7..23aca66767f9 100644
> --- a/arch/arm64/include/asm/kvm_hyp.h
> +++ b/arch/arm64/include/asm/kvm_hyp.h
> @@ -161,6 +161,7 @@ void __noreturn __hyp_do_panic(unsigned long, ...);
>   */
>  static __always_inline void __hyp_text __load_guest_stage2(struct kvm *kvm)
>  {
> +	write_sysreg(kvm->arch.vtcr, vtcr_el2);
>  	write_sysreg(kvm->arch.vttbr, vttbr_el2);
>  }
>  
> diff --git a/arch/arm64/kvm/hyp/Makefile b/arch/arm64/kvm/hyp/Makefile
> index 2fabc2dc1966..82d1904328ad 100644
> --- a/arch/arm64/kvm/hyp/Makefile
> +++ b/arch/arm64/kvm/hyp/Makefile
> @@ -19,7 +19,6 @@ obj-$(CONFIG_KVM_ARM_HOST) += switch.o
>  obj-$(CONFIG_KVM_ARM_HOST) += fpsimd.o
>  obj-$(CONFIG_KVM_ARM_HOST) += tlb.o
>  obj-$(CONFIG_KVM_ARM_HOST) += hyp-entry.o
> -obj-$(CONFIG_KVM_ARM_HOST) += s2-setup.o
>  
>  # KVM code is run at a different exception code with a different map, so
>  # compiler instrumentation that inserts callbacks or checks into the code may
> diff --git a/arch/arm64/kvm/hyp/s2-setup.c b/arch/arm64/kvm/hyp/s2-setup.c
> deleted file mode 100644
> index e1ca672e937a..000000000000
> --- a/arch/arm64/kvm/hyp/s2-setup.c
> +++ /dev/null
> @@ -1,72 +0,0 @@
> -/*
> - * Copyright (C) 2016 - ARM Ltd
> - * Author: Marc Zyngier <marc.zyngier@arm.com>
> - *
> - * This program is free software; you can redistribute it and/or modify
> - * it under the terms of the GNU General Public License version 2 as
> - * published by the Free Software Foundation.
> - *
> - * This program is distributed in the hope that it will be useful,
> - * but WITHOUT ANY WARRANTY; without even the implied warranty of
> - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> - * GNU General Public License for more details.
> - *
> - * You should have received a copy of the GNU General Public License
> - * along with this program.  If not, see <http://www.gnu.org/licenses/>.
> - */
> -
> -#include <linux/types.h>
> -#include <asm/kvm_arm.h>
> -#include <asm/kvm_asm.h>
> -#include <asm/kvm_hyp.h>
> -#include <asm/cpufeature.h>
> -
> -u32 __hyp_text __init_stage2_translation(void)
> -{
> -	u64 val = VTCR_EL2_FLAGS;
> -	u64 parange;
> -	u32 phys_shift;
> -	u64 tmp;
> -
> -	/*
> -	 * Read the PARange bits from ID_AA64MMFR0_EL1 and set the PS
> -	 * bits in VTCR_EL2. Amusingly, the PARange is 4 bits, but the
> -	 * allocated values are limited to 3bits.
> -	 */
> -	parange = read_sysreg(id_aa64mmfr0_el1) & 7;
> -	if (parange > ID_AA64MMFR0_PARANGE_MAX)
> -		parange = ID_AA64MMFR0_PARANGE_MAX;
> -	val |= parange << VTCR_EL2_PS_SHIFT;
> -
> -	/* Compute the actual PARange... */
> -	phys_shift = id_aa64mmfr0_parange_to_phys_shift(parange);
> -
> -	/*
> -	 * ... and clamp it to 40 bits, unless we have some braindead
> -	 * HW that implements less than that. In all cases, we'll
> -	 * return that value for the rest of the kernel to decide what
> -	 * to do.
> -	 */
> -	val |= VTCR_EL2_T0SZ(phys_shift > 40 ? 40 : phys_shift);
> -
> -	/*
> -	 * Check the availability of Hardware Access Flag / Dirty Bit
> -	 * Management in ID_AA64MMFR1_EL1 and enable the feature in VTCR_EL2.
> -	 */
> -	tmp = (read_sysreg(id_aa64mmfr1_el1) >> ID_AA64MMFR1_HADBS_SHIFT) & 0xf;
> -	if (tmp)
> -		val |= VTCR_EL2_HA;
> -
> -	/*
> -	 * Read the VMIDBits bits from ID_AA64MMFR1_EL1 and set the VS
> -	 * bit in VTCR_EL2.
> -	 */
> -	tmp = (read_sysreg(id_aa64mmfr1_el1) >> ID_AA64MMFR1_VMIDBITS_SHIFT) & 0xf;
> -	val |= (tmp == ID_AA64MMFR1_VMIDBITS_16) ?
> -			VTCR_EL2_VS_16BIT :
> -			VTCR_EL2_VS_8BIT;
> -
> -	write_sysreg(val, vtcr_el2);
> -
> -	return phys_shift;
> -}
> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
> index b0c07dab5cb3..e0c49377b771 100644
> --- a/arch/arm64/kvm/reset.c
> +++ b/arch/arm64/kvm/reset.c
> @@ -26,6 +26,7 @@
>  
>  #include <kvm/arm_arch_timer.h>
>  
> +#include <asm/cpufeature.h>
>  #include <asm/cputype.h>
>  #include <asm/ptrace.h>
>  #include <asm/kvm_arm.h>
> @@ -134,9 +135,38 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
>  	return kvm_timer_vcpu_reset(vcpu);
>  }
>  
> +/*
> + * Configure the VTCR_EL2 for this VM. The VTCR value is common
> + * across all the physical CPUs on the system. We use system wide
> + * sanitised values to fill in different fields, except for Hardware
> + * Management of Access Flags. HA Flag is set unconditionally on
> + * all CPUs, as it is safe to run with or without the feature and
> + * the bit is RES0 on CPUs that don't support it.
> + */
>  int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>  {
> +	u64 vtcr = VTCR_EL2_FLAGS;
#define VTCR_EL2_FLAGS			(VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)

in include/asm/kvm_arm.h

I don't see T0SZ=24 encoded there and I don't see it set either in the
code below? For bisection purpose.

Thanks

Eric
> +	u64 parange;
> +
>  	if (type)
>  		return -EINVAL;
> +
> +	parange = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1) & 7;
> +	if (parange > ID_AA64MMFR0_PARANGE_MAX)
> +		parange = ID_AA64MMFR0_PARANGE_MAX;
> +	vtcr |= parange << VTCR_EL2_PS_SHIFT;
> +
> +	/*
> +	 * Enable the Hardware Access Flag management, unconditionally
> +	 * on all CPUs. The features is RES0 on CPUs without the support
> +	 * and must be ignored by the CPUs.
> +	 */
> +	vtcr |= VTCR_EL2_HA;
> +
> +	/* Set the vmid bits */
> +	vtcr |= (kvm_get_vmid_bits() == 16) ?
> +		VTCR_EL2_VS_16BIT :
> +		VTCR_EL2_VS_8BIT;
> +	kvm->arch.vtcr = vtcr;
>  	return 0;
>  }
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 06/18] kvm: arm/arm64: Allow arch specific configurations for VM
  2018-09-17 10:41 ` [PATCH v5 06/18] kvm: arm/arm64: Allow arch specific configurations for VM Suzuki K Poulose
@ 2018-09-20 10:22   ` Auger Eric
  0 siblings, 0 replies; 50+ messages in thread
From: Auger Eric @ 2018-09-20 10:22 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> Allow the arch backends to perform VM specific initialisation.
> This will be later used to handle IPA size configuration and per-VM
> VTCR configuration on arm64.
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks

Eric

> ---
>  arch/arm/include/asm/kvm_host.h   | 7 +++++++
>  arch/arm64/include/asm/kvm_host.h | 2 ++
>  arch/arm64/kvm/reset.c            | 7 +++++++
>  virt/kvm/arm/arm.c                | 5 +++--
>  4 files changed, 19 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
> index 3ad482d2f1eb..72d46418e1ef 100644
> --- a/arch/arm/include/asm/kvm_host.h
> +++ b/arch/arm/include/asm/kvm_host.h
> @@ -354,4 +354,11 @@ static inline void kvm_vcpu_put_sysregs(struct kvm_vcpu *vcpu) {}
>  struct kvm *kvm_arch_alloc_vm(void);
>  void kvm_arch_free_vm(struct kvm *kvm);
>  
> +static inline int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
> +{
> +	if (type)
> +		return -EINVAL;
> +	return 0;
> +}
> +
>  #endif /* __ARM_KVM_HOST_H__ */
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index 3d6d7336f871..b04280ae1be0 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -513,4 +513,6 @@ void kvm_vcpu_put_sysregs(struct kvm_vcpu *vcpu);
>  struct kvm *kvm_arch_alloc_vm(void);
>  void kvm_arch_free_vm(struct kvm *kvm);
>  
> +int kvm_arm_config_vm(struct kvm *kvm, unsigned long type);
> +
>  #endif /* __ARM64_KVM_HOST_H__ */
> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
> index e37c78bbe1ca..b0c07dab5cb3 100644
> --- a/arch/arm64/kvm/reset.c
> +++ b/arch/arm64/kvm/reset.c
> @@ -133,3 +133,10 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
>  	/* Reset timer */
>  	return kvm_timer_vcpu_reset(vcpu);
>  }
> +
> +int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
> +{
> +	if (type)
> +		return -EINVAL;
> +	return 0;
> +}
> diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
> index c92053bc3f96..327d0fd28380 100644
> --- a/virt/kvm/arm/arm.c
> +++ b/virt/kvm/arm/arm.c
> @@ -120,8 +120,9 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
>  {
>  	int ret, cpu;
>  
> -	if (type)
> -		return -EINVAL;
> +	ret = kvm_arm_config_vm(kvm, type);
> +	if (ret)
> +		return ret;
>  
>  	kvm->arch.last_vcpu_ran = alloc_percpu(typeof(*kvm->arch.last_vcpu_ran));
>  	if (!kvm->arch.last_vcpu_ran)
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 07/18] kvm: arm64: Configure VTCR_EL2 per VM
  2018-09-20 10:21   ` Auger Eric
@ 2018-09-20 10:38     ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-20 10:38 UTC (permalink / raw)
  To: Auger Eric, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Eric,

On 20/09/18 11:21, Auger Eric wrote:
> Hi Suzuki,
>
> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>> Add support for setting the VTCR_EL2 per VM, rather than hard
>> coding a value at boot time per CPU. This would allow us to tune
>> the stage2 page table parameters per VM in the later changes.
>>
>> We compute the VTCR fields based on the system wide sanitised
>> feature registers, except for the hardware management of Access
>> Flags (VTCR_EL2.HA). It is fine to run a system with a mix of
>> CPUs that may or may not update the page table Access Flags.
>> Since the bit is RES0 on CPUs that don't support it, the bit
>> should be ignored on them.
>>
>> Suggested-by: Marc Zyngier <marc.zyngier@arm.com>
>> Acked-by: Christoffer Dall <cdall@kernel.org>
>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>

>> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
>> index b0c07dab5cb3..e0c49377b771 100644
>> --- a/arch/arm64/kvm/reset.c
>> +++ b/arch/arm64/kvm/reset.c
>> @@ -26,6 +26,7 @@
>>
>>   #include <kvm/arm_arch_timer.h>
>>
>> +#include <asm/cpufeature.h>
>>   #include <asm/cputype.h>
>>   #include <asm/ptrace.h>
>>   #include <asm/kvm_arm.h>
>> @@ -134,9 +135,38 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
>>      return kvm_timer_vcpu_reset(vcpu);
>>   }
>>
>> +/*
>> + * Configure the VTCR_EL2 for this VM. The VTCR value is common
>> + * across all the physical CPUs on the system. We use system wide
>> + * sanitised values to fill in different fields, except for Hardware
>> + * Management of Access Flags. HA Flag is set unconditionally on
>> + * all CPUs, as it is safe to run with or without the feature and
>> + * the bit is RES0 on CPUs that don't support it.
>> + */
>>   int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>>   {
>> +    u64 vtcr = VTCR_EL2_FLAGS;
> #define VTCR_EL2_FLAGS                        (VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
>
> in include/asm/kvm_arm.h
>
> I don't see T0SZ=24 encoded there and I don't see it set either in the
> code below? For bisection purpose.

You're right, I am missing it here. I have fixed this up for v6,
with the following hunk.


diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
index e0c49377b771..fa897eafc2c9 100644
--- a/arch/arm64/kvm/reset.c
+++ b/arch/arm64/kvm/reset.c
@@ -147,6 +147,7 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long
type)
  {
        u64 vtcr = VTCR_EL2_FLAGS;
        u64 parange;
+       u32 phys_shift;

        if (type)
                return -EINVAL;
@@ -156,6 +157,9 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long
type)
                parange = ID_AA64MMFR0_PARANGE_MAX;
        vtcr |= parange << VTCR_EL2_PS_SHIFT;

+       phys_shift = id_aa64mmfr0_parange_to_phys_shift(parange);
+
+       vtcr |= VTCR_EL2_T0SZ(phys_shift > 40 ? 40 : phys_shift);
        /*
         * Enable the Hardware Access Flag management, unconditionally
         * on all CPUs. The features is RES0 on CPUs without the support


Thanks
Suzuki
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.

^ permalink raw reply related	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask
  2018-09-17 10:41 ` [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask Suzuki K Poulose
@ 2018-09-20 14:07   ` Auger Eric
  2018-09-20 15:22     ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Auger Eric @ 2018-09-20 14:07 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,
On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> On arm64 VTTBR_EL2:BADDR holds the base address for the stage2
> translation table. The Arm ARM mandates that the bits BADDR[x-1:0]
> should be 0, where 'x' is defined for a given IPA Size and the
> number of levels for a translation granule size. It is defined
> using some magical constants. This patch is a reverse engineered
> implementation to calculate the 'x' at runtime for a given ipa and
> number of page table levels. See patch for more details.
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>

> ---
> Changes since V3:
>  - Update reference to latest ARM ARM and improve commentary
> ---
>  arch/arm64/include/asm/kvm_arm.h | 63 +++++++++++++++++++++++++++++---
>  arch/arm64/include/asm/kvm_mmu.h | 25 ++++++++++++-
>  2 files changed, 81 insertions(+), 7 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
> index 14317b3a1820..3fb1d440be6e 100644
> --- a/arch/arm64/include/asm/kvm_arm.h
> +++ b/arch/arm64/include/asm/kvm_arm.h
> @@ -123,7 +123,6 @@
>  #define VTCR_EL2_SL0_MASK	(3 << VTCR_EL2_SL0_SHIFT)
>  #define VTCR_EL2_SL0_LVL1	(1 << VTCR_EL2_SL0_SHIFT)
>  #define VTCR_EL2_T0SZ_MASK	0x3f
> -#define VTCR_EL2_T0SZ_40B	24
>  #define VTCR_EL2_VS_SHIFT	19
>  #define VTCR_EL2_VS_8BIT	(0 << VTCR_EL2_VS_SHIFT)
>  #define VTCR_EL2_VS_16BIT	(1 << VTCR_EL2_VS_SHIFT)
> @@ -140,11 +139,8 @@
>   * Note that when using 4K pages, we concatenate two first level page tables
>   * together. With 16K pages, we concatenate 16 first level page tables.
>   *
> - * The magic numbers used for VTTBR_X in this patch can be found in Tables
> - * D4-23 and D4-25 in ARM DDI 0487A.b.
>   */
>  
> -#define VTCR_EL2_T0SZ_IPA	VTCR_EL2_T0SZ_40B
>  #define VTCR_EL2_COMMON_BITS	(VTCR_EL2_SH0_INNER | VTCR_EL2_ORGN0_WBWA | \
>  				 VTCR_EL2_IRGN0_WBWA | VTCR_EL2_RES1)
>  
> @@ -175,9 +171,64 @@
>  #endif
>  
>  #define VTCR_EL2_FLAGS			(VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
> -#define VTTBR_X				(VTTBR_X_TGRAN_MAGIC - VTCR_EL2_T0SZ_IPA)
> +/*
> + * ARM VMSAv8-64 defines an algorithm for finding the translation table
> + * descriptors in section D4.2.8 in ARM DDI 0487C.a.
> + *
> + * The algorithm defines the expectations on the BaseAddress (for the page
> + * table) bits resolved at each level based on the page size, entry level
> + * and T0SZ. The variable "x" in the algorithm also affects the VTTBR:BADDR
> + * for stage2 page table.
> + *
> + * The value of "x" is calculated as :
> + *	x = Magic_N - T0SZ

What is not crystal clear to me is the "if SL0b,c = n" case where x get
a value not based on Magic_N. Please could you explain why it is not
relevant?

Thanks

Eric
> + *
> + * where Magic_N is an integer depending on the page size and the entry
> + * level of the page table as below:
> + *
> + *	--------------------------------------------
> + *	| Entry level		|  4K    16K   64K |
> + *	--------------------------------------------
> + *	| Level: 0 (4 levels)	| 28   |  -  |  -  |
> + *	--------------------------------------------
> + *	| Level: 1 (3 levels)	| 37   | 31  | 25  |
> + *	--------------------------------------------
> + *	| Level: 2 (2 levels)	| 46   | 42  | 38  |
> + *	--------------------------------------------
> + *	| Level: 3 (1 level)	| -    | 53  | 51  |
> + *	--------------------------------------------
> + *
> + * We have a magic formula for the Magic_N below:
> + *
> + *  Magic_N(PAGE_SIZE, Level) = 64 - ((PAGE_SHIFT - 3) * Number_of_levels)
> + *
> + * where Number_of_levels = (4 - Level). We are only interested in the
> + * value for Entry_Level for the stage2 page table.
> + *
> + * So, given that T0SZ = (64 - IPA_SHIFT), we can compute 'x' as follows:
> + *
> + *	x = (64 - ((PAGE_SHIFT - 3) * Number_of_levels)) - (64 - IPA_SHIFT)
> + *	  = IPA_SHIFT - ((PAGE_SHIFT - 3) * Number of levels)
> + *
> + * Here is one way to explain the Magic Formula:
> + *
> + *  x = log2(Size_of_Entry_Level_Table)
> + *
> + * Since, we can resolve (PAGE_SHIFT - 3) bits at each level, and another
> + * PAGE_SHIFT bits in the PTE, we have :
> + *
> + *  Bits_Entry_level = IPA_SHIFT - ((PAGE_SHIFT - 3) * (n - 1) + PAGE_SHIFT)
> + *		     = IPA_SHIFT - (PAGE_SHIFT - 3) * n - 3
> + *  where n = number of levels, and since each pointer is 8bytes, we have:
> + *
> + *  x = Bits_Entry_Level + 3
> + *    = IPA_SHIFT - (PAGE_SHIFT - 3) * n
> + *
> + * The only constraint here is that, we have to find the number of page table
> + * levels for a given IPA size (which we do, see stage2_pt_levels())
> + */
> +#define ARM64_VTTBR_X(ipa, levels)	((ipa) - ((levels) * (PAGE_SHIFT - 3)))
>  
> -#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)
>  
> diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
> index 7342d2c51773..ac3ca9690bad 100644
> --- a/arch/arm64/include/asm/kvm_mmu.h
> +++ b/arch/arm64/include/asm/kvm_mmu.h
> @@ -145,7 +145,6 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
>  #define kvm_phys_shift(kvm)		KVM_PHYS_SHIFT
>  #define kvm_phys_size(kvm)		(_AC(1, ULL) << kvm_phys_shift(kvm))
>  #define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - _AC(1, ULL))
> -#define kvm_vttbr_baddr_mask(kvm)	VTTBR_BADDR_MASK
>  
>  static inline bool kvm_page_empty(void *ptr)
>  {
> @@ -520,5 +519,29 @@ static inline int hyp_map_aux_data(void)
>  
>  #define kvm_phys_to_vttbr(addr)		phys_to_ttbr(addr)
>  
> +/*
> + * Get the magic number 'x' for VTTBR:BADDR of this KVM instance.
> + * With v8.2 LVA extensions, 'x' should be a minimum of 6 with
> + * 52bit IPS.
> + */
> +static inline int arm64_vttbr_x(u32 ipa_shift, u32 levels)
> +{
> +	int x = ARM64_VTTBR_X(ipa_shift, levels);
> +
> +	return (IS_ENABLED(CONFIG_ARM64_PA_BITS_52) && x < 6) ? 6 : x;
> +}
> +
> +static inline u64 vttbr_baddr_mask(u32 ipa_shift, u32 levels)
> +{
> +	unsigned int x = arm64_vttbr_x(ipa_shift, levels);
> +
> +	return GENMASK_ULL(PHYS_MASK_SHIFT - 1, x);
> +}
> +
> +static inline u64 kvm_vttbr_baddr_mask(struct kvm *kvm)
> +{
> +	return vttbr_baddr_mask(kvm_phys_shift(kvm), kvm_stage2_levels(kvm));
> +}
> +
>  #endif /* __ASSEMBLY__ */
>  #endif /* __ARM64_KVM_MMU_H__ */
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 10/18] kvm: arm64: Make stage2 page table layout dynamic
  2018-09-17 10:41 ` [PATCH v5 10/18] kvm: arm64: Make stage2 page table layout dynamic Suzuki K Poulose
@ 2018-09-20 14:07   ` Auger Eric
  0 siblings, 0 replies; 50+ messages in thread
From: Auger Eric @ 2018-09-20 14:07 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> Switch to dynamic stage2 page table layout based on the given
> VM. So far we had a common stage2 table layout determined at
> compile time. Make decision based on the VM instance depending
> on the IPA limit for the VM. Adds helpers to compute the stage2
> parameters based on the guest's IPA and uses them to make the decisions.
> 
> The IPA limit is still fixed to 40bits and the build time check
> to ensure the stage2 doesn't exceed the host kernels page table
> levels is retained. Also make sure that the host has pud/pmd helpers
> are used only when they are available at host.
needs some rewording.
> 
> Cc: Christoffer Dall <cdall@kernel.org>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
>  arch/arm64/include/asm/stage2_pgtable.h | 84 +++++++++++++++----------
>  1 file changed, 52 insertions(+), 32 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
> index 384f9e982cba..e5acda8e2e31 100644
> --- a/arch/arm64/include/asm/stage2_pgtable.h
> +++ b/arch/arm64/include/asm/stage2_pgtable.h
> @@ -22,6 +22,13 @@
>  #include <linux/hugetlb.h>
>  #include <asm/pgtable.h>
>  
> +/*
> + * PGDIR_SHIFT determines the size a top-level page table entry can map
strictly speaking it is the index of the start level, right? size of the
top-level page table entry is 1 << PGDIR_SHIFT
> + * and depends on the number of levels in the page table. Compute the
> + * PGDIR_SHIFT for a given number of levels.
> + */
> +#define pt_levels_pgdir_shift(n)	ARM64_HW_PGTABLE_LEVEL_SHIFT(4 - (n))
why not using lvls instead of n as below?

Besides
Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks

Eric
> +
>  /*
>   * The hardware supports concatenation of up to 16 tables at stage2 entry level
>   * and we use the feature whenever possible.
> @@ -30,11 +37,13 @@
>   * On arm64, the smallest PAGE_SIZE supported is 4k, which means
>   *             (PAGE_SHIFT - 3) > 4 holds for all page sizes.
>   * This implies, the total number of page table levels at stage2 expected
> - * by the hardware is actually the number of levels required for (KVM_PHYS_SHIFT - 4)
> + * by the hardware is actually the number of levels required for (IPA_SHIFT - 4)
>   * in normal translations(e.g, stage1), since we cannot have another level in
> - * the range (KVM_PHYS_SHIFT, KVM_PHYS_SHIFT - 4).
> + * the range (IPA_SHIFT, IPA_SHIFT - 4).
>   */
> -#define STAGE2_PGTABLE_LEVELS		ARM64_HW_PGTABLE_LEVELS(KVM_PHYS_SHIFT - 4)
> +#define stage2_pgtable_levels(ipa)	ARM64_HW_PGTABLE_LEVELS((ipa) - 4)
> +#define STAGE2_PGTABLE_LEVELS		stage2_pgtable_levels(KVM_PHYS_SHIFT)
> +#define kvm_stage2_levels(kvm)		stage2_pgtable_levels(kvm_phys_shift(kvm))
>  
>  /*
>   * With all the supported VA_BITs and 40bit guest IPA, the following condition
> @@ -54,33 +63,42 @@
>  #error "Unsupported combination of guest IPA and host VA_BITS."
>  #endif
>  
> -/* S2_PGDIR_SHIFT is the size mapped by top-level stage2 entry */
> -#define S2_PGDIR_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(4 - STAGE2_PGTABLE_LEVELS)
> -#define S2_PGDIR_SIZE			(1UL << S2_PGDIR_SHIFT)
> -#define S2_PGDIR_MASK			(~(S2_PGDIR_SIZE - 1))
> +
> +/* stage2_pgdir_shift() is the size mapped by top-level stage2 entry for the VM */
> +#define stage2_pgdir_shift(kvm)		pt_levels_pgdir_shift(kvm_stage2_levels(kvm))
> +#define stage2_pgdir_size(kvm)		(1ULL << stage2_pgdir_shift(kvm))
> +#define stage2_pgdir_mask(kvm)		~(stage2_pgdir_size(kvm) - 1)
>  
>  /*
>   * The number of PTRS across all concatenated stage2 tables given by the
>   * number of bits resolved at the initial level.
>   */
> -#define PTRS_PER_S2_PGD			(1 << (KVM_PHYS_SHIFT - S2_PGDIR_SHIFT))
> +#define __s2_pgd_ptrs(ipa, lvls)	(1 << ((ipa) - pt_levels_pgdir_shift((lvls))))
> +#define __s2_pgd_size(ipa, lvls)	(__s2_pgd_ptrs((ipa), (lvls)) * sizeof(pgd_t))
> +
> +#define stage2_pgd_ptrs(kvm)		__s2_pgd_ptrs(kvm_phys_shift(kvm), kvm_stage2_levels(kvm))
> +#define stage2_pgd_size(kvm)		__s2_pgd_size(kvm_phys_shift(kvm), kvm_stage2_levels(kvm))
>  
>  /*
>   * kvm_mmmu_cache_min_pages() is the number of pages required to install
>   * a stage-2 translation. We pre-allocate the entry level page table at
>   * the VM creation.
>   */
> -#define kvm_mmu_cache_min_pages(kvm)	(STAGE2_PGTABLE_LEVELS - 1)
> +#define kvm_mmu_cache_min_pages(kvm)	(kvm_stage2_levels(kvm) - 1)
>  
>  /* Stage2 PUD definitions when the level is present */
> -#define STAGE2_PGTABLE_HAS_PUD		(STAGE2_PGTABLE_LEVELS > 3)
> +static inline bool kvm_stage2_has_pud(struct kvm *kvm)
> +{
> +	return (CONFIG_PGTABLE_LEVELS > 3) && (kvm_stage2_levels(kvm) > 3);
> +}
> +
>  #define S2_PUD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(1)
>  #define S2_PUD_SIZE			(1UL << S2_PUD_SHIFT)
>  #define S2_PUD_MASK			(~(S2_PUD_SIZE - 1))
>  
>  static inline bool stage2_pgd_none(struct kvm *kvm, pgd_t pgd)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD)
> +	if (kvm_stage2_has_pud(kvm))
>  		return pgd_none(pgd);
>  	else
>  		return 0;
> @@ -88,13 +106,13 @@ static inline bool stage2_pgd_none(struct kvm *kvm, pgd_t pgd)
>  
>  static inline void stage2_pgd_clear(struct kvm *kvm, pgd_t *pgdp)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD)
> +	if (kvm_stage2_has_pud(kvm))
>  		pgd_clear(pgdp);
>  }
>  
>  static inline bool stage2_pgd_present(struct kvm *kvm, pgd_t pgd)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD)
> +	if (kvm_stage2_has_pud(kvm))
>  		return pgd_present(pgd);
>  	else
>  		return 1;
> @@ -102,14 +120,14 @@ static inline bool stage2_pgd_present(struct kvm *kvm, pgd_t pgd)
>  
>  static inline void stage2_pgd_populate(struct kvm *kvm, pgd_t *pgd, pud_t *pud)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD)
> +	if (kvm_stage2_has_pud(kvm))
>  		pgd_populate(NULL, pgd, pud);
>  }
>  
>  static inline pud_t *stage2_pud_offset(struct kvm *kvm,
>  				       pgd_t *pgd, unsigned long address)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD)
> +	if (kvm_stage2_has_pud(kvm))
>  		return pud_offset(pgd, address);
>  	else
>  		return (pud_t *)pgd;
> @@ -117,13 +135,13 @@ static inline pud_t *stage2_pud_offset(struct kvm *kvm,
>  
>  static inline void stage2_pud_free(struct kvm *kvm, pud_t *pud)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD)
> +	if (kvm_stage2_has_pud(kvm))
>  		pud_free(NULL, pud);
>  }
>  
>  static inline bool stage2_pud_table_empty(struct kvm *kvm, pud_t *pudp)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD)
> +	if (kvm_stage2_has_pud(kvm))
>  		return kvm_page_empty(pudp);
>  	else
>  		return false;
> @@ -132,7 +150,7 @@ static inline bool stage2_pud_table_empty(struct kvm *kvm, pud_t *pudp)
>  static inline phys_addr_t
>  stage2_pud_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
> -	if (STAGE2_PGTABLE_HAS_PUD) {
> +	if (kvm_stage2_has_pud(kvm)) {
>  		phys_addr_t boundary = (addr + S2_PUD_SIZE) & S2_PUD_MASK;
>  
>  		return (boundary - 1 < end - 1) ? boundary : end;
> @@ -142,14 +160,18 @@ stage2_pud_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  }
>  
>  /* Stage2 PMD definitions when the level is present */
> -#define STAGE2_PGTABLE_HAS_PMD		(STAGE2_PGTABLE_LEVELS > 2)
> +static inline bool kvm_stage2_has_pmd(struct kvm *kvm)
> +{
> +	return (CONFIG_PGTABLE_LEVELS > 2) && (kvm_stage2_levels(kvm) > 2);
> +}
> +
>  #define S2_PMD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(2)
>  #define S2_PMD_SIZE			(1UL << S2_PMD_SHIFT)
>  #define S2_PMD_MASK			(~(S2_PMD_SIZE - 1))
>  
>  static inline bool stage2_pud_none(struct kvm *kvm, pud_t pud)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		return pud_none(pud);
>  	else
>  		return 0;
> @@ -157,13 +179,13 @@ static inline bool stage2_pud_none(struct kvm *kvm, pud_t pud)
>  
>  static inline void stage2_pud_clear(struct kvm *kvm, pud_t *pud)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		pud_clear(pud);
>  }
>  
>  static inline bool stage2_pud_present(struct kvm *kvm, pud_t pud)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		return pud_present(pud);
>  	else
>  		return 1;
> @@ -171,14 +193,14 @@ static inline bool stage2_pud_present(struct kvm *kvm, pud_t pud)
>  
>  static inline void stage2_pud_populate(struct kvm *kvm, pud_t *pud, pmd_t *pmd)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		pud_populate(NULL, pud, pmd);
>  }
>  
>  static inline pmd_t *stage2_pmd_offset(struct kvm *kvm,
>  				       pud_t *pud, unsigned long address)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		return pmd_offset(pud, address);
>  	else
>  		return (pmd_t *)pud;
> @@ -186,13 +208,13 @@ static inline pmd_t *stage2_pmd_offset(struct kvm *kvm,
>  
>  static inline void stage2_pmd_free(struct kvm *kvm, pmd_t *pmd)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		pmd_free(NULL, pmd);
>  }
>  
>  static inline bool stage2_pud_huge(struct kvm *kvm, pud_t pud)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		return pud_huge(pud);
>  	else
>  		return 0;
> @@ -200,7 +222,7 @@ static inline bool stage2_pud_huge(struct kvm *kvm, pud_t pud)
>  
>  static inline bool stage2_pmd_table_empty(struct kvm *kvm, pmd_t *pmdp)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD)
> +	if (kvm_stage2_has_pmd(kvm))
>  		return kvm_page_empty(pmdp);
>  	else
>  		return 0;
> @@ -209,7 +231,7 @@ static inline bool stage2_pmd_table_empty(struct kvm *kvm, pmd_t *pmdp)
>  static inline phys_addr_t
>  stage2_pmd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
> -	if (STAGE2_PGTABLE_HAS_PMD) {
> +	if (kvm_stage2_has_pmd(kvm)) {
>  		phys_addr_t boundary = (addr + S2_PMD_SIZE) & S2_PMD_MASK;
>  
>  		return (boundary - 1 < end - 1) ? boundary : end;
> @@ -223,17 +245,15 @@ static inline bool stage2_pte_table_empty(struct kvm *kvm, pte_t *ptep)
>  	return kvm_page_empty(ptep);
>  }
>  
> -#define stage2_pgd_size(kvm)	(PTRS_PER_S2_PGD * sizeof(pgd_t))
> -
>  static inline unsigned long stage2_pgd_index(struct kvm *kvm, phys_addr_t addr)
>  {
> -	return (((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1));
> +	return (((addr) >> stage2_pgdir_shift(kvm)) & (stage2_pgd_ptrs(kvm) - 1));
>  }
>  
>  static inline phys_addr_t
>  stage2_pgd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
> -	phys_addr_t boundary = (addr + S2_PGDIR_SIZE) & S2_PGDIR_MASK;
> +	phys_addr_t boundary = (addr + stage2_pgdir_size(kvm)) & stage2_pgdir_mask(kvm);
>  
>  	return (boundary - 1 < end - 1) ? boundary : end;
>  }
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 08/18] kvm: arm/arm64: Prepare for VM specific stage2 translations
  2018-09-17 10:41 ` [PATCH v5 08/18] kvm: arm/arm64: Prepare for VM specific stage2 translations Suzuki K Poulose
@ 2018-09-20 14:07   ` Auger Eric
  0 siblings, 0 replies; 50+ messages in thread
From: Auger Eric @ 2018-09-20 14:07 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> Right now the stage2 page table for a VM is hard coded, assuming
> an IPA of 40bits. As we are about to add support for per VM IPA,
> prepare the stage2 page table helpers to accept the kvm instance
> to make the right decision for the VM. No functional changes.
> Adds stage2_pgd_size(kvm) to replace S2_PGD_SIZE. Also, moves
> some of the definitions in arm32 to align with the arm64.
> Also drop the _AC() specifier constants wherever possible.
> 
> Cc: Christoffer Dall <cdall@kernel.org>
> Acked-by: Marc Zyngier <marc.zyngier@arm.com>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks

Eric
> ---
> Changes since V3:
>  - Improve the comment about kvm_mmu_cache_min_pages()
>  - Drop _AC() in arm64 definitions
>  - Move kvm_mmu_cache_min_pages() in arm to stage2_pgtable.h in
>    line with arm64.
> ---
>  arch/arm/include/asm/kvm_arm.h                |   3 +-
>  arch/arm/include/asm/kvm_mmu.h                |  13 +-
>  arch/arm/include/asm/stage2_pgtable.h         |  50 +++++---
>  arch/arm64/include/asm/kvm_mmu.h              |   7 +-
>  arch/arm64/include/asm/stage2_pgtable-nopmd.h |  18 +--
>  arch/arm64/include/asm/stage2_pgtable-nopud.h |  16 +--
>  arch/arm64/include/asm/stage2_pgtable.h       |  58 +++++----
>  virt/kvm/arm/arm.c                            |   2 +-
>  virt/kvm/arm/mmu.c                            | 119 +++++++++---------
>  virt/kvm/arm/vgic/vgic-kvm-device.c           |   2 +-
>  10 files changed, 156 insertions(+), 132 deletions(-)
> 
> diff --git a/arch/arm/include/asm/kvm_arm.h b/arch/arm/include/asm/kvm_arm.h
> index 3ab8b3781bfe..c3f1f9b304b7 100644
> --- a/arch/arm/include/asm/kvm_arm.h
> +++ b/arch/arm/include/asm/kvm_arm.h
> @@ -133,8 +133,7 @@
>   * space.
>   */
>  #define KVM_PHYS_SHIFT	(40)
> -#define KVM_PHYS_SIZE	(_AC(1, ULL) << KVM_PHYS_SHIFT)
> -#define KVM_PHYS_MASK	(KVM_PHYS_SIZE - _AC(1, ULL))
> +
>  #define PTRS_PER_S2_PGD	(_AC(1, ULL) << (KVM_PHYS_SHIFT - 30))
>  
>  /* Virtualization Translation Control Register (VTCR) bits */
> diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
> index 265ea9cf7df7..12ae5fbbcf01 100644
> --- a/arch/arm/include/asm/kvm_mmu.h
> +++ b/arch/arm/include/asm/kvm_mmu.h
> @@ -35,16 +35,12 @@
>  		addr;							\
>  	})
>  
> -/*
> - * KVM_MMU_CACHE_MIN_PAGES is the number of stage2 page table translation levels.
> - */
> -#define KVM_MMU_CACHE_MIN_PAGES	2
> -
>  #ifndef __ASSEMBLY__
>  
>  #include <linux/highmem.h>
>  #include <asm/cacheflush.h>
>  #include <asm/cputype.h>
> +#include <asm/kvm_arm.h>
>  #include <asm/kvm_hyp.h>
>  #include <asm/pgalloc.h>
>  #include <asm/stage2_pgtable.h>
> @@ -52,6 +48,13 @@
>  /* Ensure compatibility with arm64 */
>  #define VA_BITS			32
>  
> +#define kvm_phys_shift(kvm)		KVM_PHYS_SHIFT
> +#define kvm_phys_size(kvm)		(1ULL << kvm_phys_shift(kvm))
> +#define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - 1ULL)
> +#define kvm_vttbr_baddr_mask(kvm)	VTTBR_BADDR_MASK
> +
> +#define stage2_pgd_size(kvm)		(PTRS_PER_S2_PGD * sizeof(pgd_t))
> +
>  int create_hyp_mappings(void *from, void *to, pgprot_t prot);
>  int create_hyp_io_mappings(phys_addr_t phys_addr, size_t size,
>  			   void __iomem **kaddr,
> diff --git a/arch/arm/include/asm/stage2_pgtable.h b/arch/arm/include/asm/stage2_pgtable.h
> index 460d616bb2d6..f6a7ea805232 100644
> --- a/arch/arm/include/asm/stage2_pgtable.h
> +++ b/arch/arm/include/asm/stage2_pgtable.h
> @@ -19,43 +19,53 @@
>  #ifndef __ARM_S2_PGTABLE_H_
>  #define __ARM_S2_PGTABLE_H_
>  
> -#define stage2_pgd_none(pgd)			pgd_none(pgd)
> -#define stage2_pgd_clear(pgd)			pgd_clear(pgd)
> -#define stage2_pgd_present(pgd)			pgd_present(pgd)
> -#define stage2_pgd_populate(pgd, pud)		pgd_populate(NULL, pgd, pud)
> -#define stage2_pud_offset(pgd, address)		pud_offset(pgd, address)
> -#define stage2_pud_free(pud)			pud_free(NULL, pud)
> +/*
> + * kvm_mmu_cache_min_pages() is the number of pages required
> + * to install a stage-2 translation. We pre-allocate the entry
> + * level table at VM creation. Since we have a 3 level page-table,
> + * we need only two pages to add a new mapping.
> + */
> +#define kvm_mmu_cache_min_pages(kvm)	2
>  
> -#define stage2_pud_none(pud)			pud_none(pud)
> -#define stage2_pud_clear(pud)			pud_clear(pud)
> -#define stage2_pud_present(pud)			pud_present(pud)
> -#define stage2_pud_populate(pud, pmd)		pud_populate(NULL, pud, pmd)
> -#define stage2_pmd_offset(pud, address)		pmd_offset(pud, address)
> -#define stage2_pmd_free(pmd)			pmd_free(NULL, pmd)
> +#define stage2_pgd_none(kvm, pgd)		pgd_none(pgd)
> +#define stage2_pgd_clear(kvm, pgd)		pgd_clear(pgd)
> +#define stage2_pgd_present(kvm, pgd)		pgd_present(pgd)
> +#define stage2_pgd_populate(kvm, pgd, pud)	pgd_populate(NULL, pgd, pud)
> +#define stage2_pud_offset(kvm, pgd, address)	pud_offset(pgd, address)
> +#define stage2_pud_free(kvm, pud)		pud_free(NULL, pud)
>  
> -#define stage2_pud_huge(pud)			pud_huge(pud)
> +#define stage2_pud_none(kvm, pud)		pud_none(pud)
> +#define stage2_pud_clear(kvm, pud)		pud_clear(pud)
> +#define stage2_pud_present(kvm, pud)		pud_present(pud)
> +#define stage2_pud_populate(kvm, pud, pmd)	pud_populate(NULL, pud, pmd)
> +#define stage2_pmd_offset(kvm, pud, address)	pmd_offset(pud, address)
> +#define stage2_pmd_free(kvm, pmd)		pmd_free(NULL, pmd)
> +
> +#define stage2_pud_huge(kvm, pud)		pud_huge(pud)
>  
>  /* Open coded p*d_addr_end that can deal with 64bit addresses */
> -static inline phys_addr_t stage2_pgd_addr_end(phys_addr_t addr, phys_addr_t end)
> +static inline phys_addr_t
> +stage2_pgd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
>  	phys_addr_t boundary = (addr + PGDIR_SIZE) & PGDIR_MASK;
>  
>  	return (boundary - 1 < end - 1) ? boundary : end;
>  }
>  
> -#define stage2_pud_addr_end(addr, end)		(end)
> +#define stage2_pud_addr_end(kvm, addr, end)	(end)
>  
> -static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
> +static inline phys_addr_t
> +stage2_pmd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
>  	phys_addr_t boundary = (addr + PMD_SIZE) & PMD_MASK;
>  
>  	return (boundary - 1 < end - 1) ? boundary : end;
>  }
>  
> -#define stage2_pgd_index(addr)				pgd_index(addr)
> +#define stage2_pgd_index(kvm, addr)		pgd_index(addr)
>  
> -#define stage2_pte_table_empty(ptep)			kvm_page_empty(ptep)
> -#define stage2_pmd_table_empty(pmdp)			kvm_page_empty(pmdp)
> -#define stage2_pud_table_empty(pudp)			false
> +#define stage2_pte_table_empty(kvm, ptep)	kvm_page_empty(ptep)
> +#define stage2_pmd_table_empty(kvm, pmdp)	kvm_page_empty(pmdp)
> +#define stage2_pud_table_empty(kvm, pudp)	false
>  
>  #endif	/* __ARM_S2_PGTABLE_H_ */
> diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
> index d6fff7de5539..3a032066e52c 100644
> --- a/arch/arm64/include/asm/kvm_mmu.h
> +++ b/arch/arm64/include/asm/kvm_mmu.h
> @@ -141,8 +141,11 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
>   * We currently only support a 40bit IPA.
>   */
>  #define KVM_PHYS_SHIFT	(40)
> -#define KVM_PHYS_SIZE	(1UL << KVM_PHYS_SHIFT)
> -#define KVM_PHYS_MASK	(KVM_PHYS_SIZE - 1UL)
> +
> +#define kvm_phys_shift(kvm)		KVM_PHYS_SHIFT
> +#define kvm_phys_size(kvm)		(_AC(1, ULL) << kvm_phys_shift(kvm))
> +#define kvm_phys_mask(kvm)		(kvm_phys_size(kvm) - _AC(1, ULL))
> +#define kvm_vttbr_baddr_mask(kvm)	VTTBR_BADDR_MASK
>  
>  #include <asm/stage2_pgtable.h>
>  
> diff --git a/arch/arm64/include/asm/stage2_pgtable-nopmd.h b/arch/arm64/include/asm/stage2_pgtable-nopmd.h
> index 2656a0fd05a6..0280dedbf75f 100644
> --- a/arch/arm64/include/asm/stage2_pgtable-nopmd.h
> +++ b/arch/arm64/include/asm/stage2_pgtable-nopmd.h
> @@ -26,17 +26,17 @@
>  #define S2_PMD_SIZE		(1UL << S2_PMD_SHIFT)
>  #define S2_PMD_MASK		(~(S2_PMD_SIZE-1))
>  
> -#define stage2_pud_none(pud)			(0)
> -#define stage2_pud_present(pud)			(1)
> -#define stage2_pud_clear(pud)			do { } while (0)
> -#define stage2_pud_populate(pud, pmd)		do { } while (0)
> -#define stage2_pmd_offset(pud, address)		((pmd_t *)(pud))
> +#define stage2_pud_none(kvm, pud)		(0)
> +#define stage2_pud_present(kvm, pud)		(1)
> +#define stage2_pud_clear(kvm, pud)		do { } while (0)
> +#define stage2_pud_populate(kvm, pud, pmd)	do { } while (0)
> +#define stage2_pmd_offset(kvm, pud, address)	((pmd_t *)(pud))
>  
> -#define stage2_pmd_free(pmd)			do { } while (0)
> +#define stage2_pmd_free(kvm, pmd)		do { } while (0)
>  
> -#define stage2_pmd_addr_end(addr, end)		(end)
> +#define stage2_pmd_addr_end(kvm, addr, end)	(end)
>  
> -#define stage2_pud_huge(pud)			(0)
> -#define stage2_pmd_table_empty(pmdp)		(0)
> +#define stage2_pud_huge(kvm, pud)		(0)
> +#define stage2_pmd_table_empty(kvm, pmdp)	(0)
>  
>  #endif
> diff --git a/arch/arm64/include/asm/stage2_pgtable-nopud.h b/arch/arm64/include/asm/stage2_pgtable-nopud.h
> index 5ee87b54ebf3..cd6304e203be 100644
> --- a/arch/arm64/include/asm/stage2_pgtable-nopud.h
> +++ b/arch/arm64/include/asm/stage2_pgtable-nopud.h
> @@ -24,16 +24,16 @@
>  #define S2_PUD_SIZE		(_AC(1, UL) << S2_PUD_SHIFT)
>  #define S2_PUD_MASK		(~(S2_PUD_SIZE-1))
>  
> -#define stage2_pgd_none(pgd)			(0)
> -#define stage2_pgd_present(pgd)			(1)
> -#define stage2_pgd_clear(pgd)			do { } while (0)
> -#define stage2_pgd_populate(pgd, pud)	do { } while (0)
> +#define stage2_pgd_none(kvm, pgd)		(0)
> +#define stage2_pgd_present(kvm, pgd)		(1)
> +#define stage2_pgd_clear(kvm, pgd)		do { } while (0)
> +#define stage2_pgd_populate(kvm, pgd, pud)	do { } while (0)
>  
> -#define stage2_pud_offset(pgd, address)		((pud_t *)(pgd))
> +#define stage2_pud_offset(kvm, pgd, address)	((pud_t *)(pgd))
>  
> -#define stage2_pud_free(x)			do { } while (0)
> +#define stage2_pud_free(kvm, x)			do { } while (0)
>  
> -#define stage2_pud_addr_end(addr, end)		(end)
> -#define stage2_pud_table_empty(pmdp)		(0)
> +#define stage2_pud_addr_end(kvm, addr, end)	(end)
> +#define stage2_pud_table_empty(kvm, pmdp)	(0)
>  
>  #endif
> diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
> index 8b68099348e5..11891612be14 100644
> --- a/arch/arm64/include/asm/stage2_pgtable.h
> +++ b/arch/arm64/include/asm/stage2_pgtable.h
> @@ -55,7 +55,7 @@
>  
>  /* S2_PGDIR_SHIFT is the size mapped by top-level stage2 entry */
>  #define S2_PGDIR_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(4 - STAGE2_PGTABLE_LEVELS)
> -#define S2_PGDIR_SIZE			(_AC(1, UL) << S2_PGDIR_SHIFT)
> +#define S2_PGDIR_SIZE			(1UL << S2_PGDIR_SHIFT)
>  #define S2_PGDIR_MASK			(~(S2_PGDIR_SIZE - 1))
>  
>  /*
> @@ -65,28 +65,30 @@
>  #define PTRS_PER_S2_PGD			(1 << (KVM_PHYS_SHIFT - S2_PGDIR_SHIFT))
>  
>  /*
> - * KVM_MMU_CACHE_MIN_PAGES is the number of stage2 page table translation
> - * levels in addition to the PGD.
> + * kvm_mmmu_cache_min_pages() is the number of pages required to install
> + * a stage-2 translation. We pre-allocate the entry level page table at
> + * the VM creation.
>   */
> -#define KVM_MMU_CACHE_MIN_PAGES		(STAGE2_PGTABLE_LEVELS - 1)
> +#define kvm_mmu_cache_min_pages(kvm)	(STAGE2_PGTABLE_LEVELS - 1)
>  
>  
>  #if STAGE2_PGTABLE_LEVELS > 3
>  
>  #define S2_PUD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(1)
> -#define S2_PUD_SIZE			(_AC(1, UL) << S2_PUD_SHIFT)
> +#define S2_PUD_SIZE			(1UL << S2_PUD_SHIFT)
>  #define S2_PUD_MASK			(~(S2_PUD_SIZE - 1))
>  
> -#define stage2_pgd_none(pgd)				pgd_none(pgd)
> -#define stage2_pgd_clear(pgd)				pgd_clear(pgd)
> -#define stage2_pgd_present(pgd)				pgd_present(pgd)
> -#define stage2_pgd_populate(pgd, pud)			pgd_populate(NULL, pgd, pud)
> -#define stage2_pud_offset(pgd, address)			pud_offset(pgd, address)
> -#define stage2_pud_free(pud)				pud_free(NULL, pud)
> +#define stage2_pgd_none(kvm, pgd)		pgd_none(pgd)
> +#define stage2_pgd_clear(kvm, pgd)		pgd_clear(pgd)
> +#define stage2_pgd_present(kvm, pgd)		pgd_present(pgd)
> +#define stage2_pgd_populate(kvm, pgd, pud)	pgd_populate(NULL, pgd, pud)
> +#define stage2_pud_offset(kvm, pgd, address)	pud_offset(pgd, address)
> +#define stage2_pud_free(kvm, pud)		pud_free(NULL, pud)
>  
> -#define stage2_pud_table_empty(pudp)			kvm_page_empty(pudp)
> +#define stage2_pud_table_empty(kvm, pudp)	kvm_page_empty(pudp)
>  
> -static inline phys_addr_t stage2_pud_addr_end(phys_addr_t addr, phys_addr_t end)
> +static inline phys_addr_t
> +stage2_pud_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
>  	phys_addr_t boundary = (addr + S2_PUD_SIZE) & S2_PUD_MASK;
>  
> @@ -99,20 +101,21 @@ static inline phys_addr_t stage2_pud_addr_end(phys_addr_t addr, phys_addr_t end)
>  #if STAGE2_PGTABLE_LEVELS > 2
>  
>  #define S2_PMD_SHIFT			ARM64_HW_PGTABLE_LEVEL_SHIFT(2)
> -#define S2_PMD_SIZE			(_AC(1, UL) << S2_PMD_SHIFT)
> +#define S2_PMD_SIZE			(1UL << S2_PMD_SHIFT)
>  #define S2_PMD_MASK			(~(S2_PMD_SIZE - 1))
>  
> -#define stage2_pud_none(pud)				pud_none(pud)
> -#define stage2_pud_clear(pud)				pud_clear(pud)
> -#define stage2_pud_present(pud)				pud_present(pud)
> -#define stage2_pud_populate(pud, pmd)			pud_populate(NULL, pud, pmd)
> -#define stage2_pmd_offset(pud, address)			pmd_offset(pud, address)
> -#define stage2_pmd_free(pmd)				pmd_free(NULL, pmd)
> +#define stage2_pud_none(kvm, pud)		pud_none(pud)
> +#define stage2_pud_clear(kvm, pud)		pud_clear(pud)
> +#define stage2_pud_present(kvm, pud)		pud_present(pud)
> +#define stage2_pud_populate(kvm, pud, pmd)	pud_populate(NULL, pud, pmd)
> +#define stage2_pmd_offset(kvm, pud, address)	pmd_offset(pud, address)
> +#define stage2_pmd_free(kvm, pmd)		pmd_free(NULL, pmd)
>  
> -#define stage2_pud_huge(pud)				pud_huge(pud)
> -#define stage2_pmd_table_empty(pmdp)			kvm_page_empty(pmdp)
> +#define stage2_pud_huge(kvm, pud)		pud_huge(pud)
> +#define stage2_pmd_table_empty(kvm, pmdp)	kvm_page_empty(pmdp)
>  
> -static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
> +static inline phys_addr_t
> +stage2_pmd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
>  	phys_addr_t boundary = (addr + S2_PMD_SIZE) & S2_PMD_MASK;
>  
> @@ -121,7 +124,7 @@ static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
>  
>  #endif		/* STAGE2_PGTABLE_LEVELS > 2 */
>  
> -#define stage2_pte_table_empty(ptep)			kvm_page_empty(ptep)
> +#define stage2_pte_table_empty(kvm, ptep)	kvm_page_empty(ptep)
>  
>  #if STAGE2_PGTABLE_LEVELS == 2
>  #include <asm/stage2_pgtable-nopmd.h>
> @@ -129,10 +132,13 @@ static inline phys_addr_t stage2_pmd_addr_end(phys_addr_t addr, phys_addr_t end)
>  #include <asm/stage2_pgtable-nopud.h>
>  #endif
>  
> +#define stage2_pgd_size(kvm)	(PTRS_PER_S2_PGD * sizeof(pgd_t))
>  
> -#define stage2_pgd_index(addr)				(((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1))
> +#define stage2_pgd_index(kvm, addr) \
> +	(((addr) >> S2_PGDIR_SHIFT) & (PTRS_PER_S2_PGD - 1))
>  
> -static inline phys_addr_t stage2_pgd_addr_end(phys_addr_t addr, phys_addr_t end)
> +static inline phys_addr_t
> +stage2_pgd_addr_end(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  {
>  	phys_addr_t boundary = (addr + S2_PGDIR_SIZE) & S2_PGDIR_MASK;
>  
> diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
> index 327d0fd28380..43e716bc3f08 100644
> --- a/virt/kvm/arm/arm.c
> +++ b/virt/kvm/arm/arm.c
> @@ -545,7 +545,7 @@ static void update_vttbr(struct kvm *kvm)
>  
>  	/* update vttbr to be used with the new vmid */
>  	pgd_phys = virt_to_phys(kvm->arch.pgd);
> -	BUG_ON(pgd_phys & ~VTTBR_BADDR_MASK);
> +	BUG_ON(pgd_phys & ~kvm_vttbr_baddr_mask(kvm));
>  	vmid = ((u64)(kvm->arch.vmid) << VTTBR_VMID_SHIFT) & VTTBR_VMID_MASK(kvm_vmid_bits);
>  	kvm->arch.vttbr = kvm_phys_to_vttbr(pgd_phys) | vmid;
>  
> diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
> index 4a285d760ce0..7e477b3cae5b 100644
> --- a/virt/kvm/arm/mmu.c
> +++ b/virt/kvm/arm/mmu.c
> @@ -45,7 +45,6 @@ static phys_addr_t hyp_idmap_vector;
>  
>  static unsigned long io_map_base;
>  
> -#define S2_PGD_SIZE	(PTRS_PER_S2_PGD * sizeof(pgd_t))
>  #define hyp_pgd_order get_order(PTRS_PER_PGD * sizeof(pgd_t))
>  
>  #define KVM_S2PTE_FLAG_IS_IOMAP		(1UL << 0)
> @@ -150,20 +149,20 @@ static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc)
>  
>  static void clear_stage2_pgd_entry(struct kvm *kvm, pgd_t *pgd, phys_addr_t addr)
>  {
> -	pud_t *pud_table __maybe_unused = stage2_pud_offset(pgd, 0UL);
> -	stage2_pgd_clear(pgd);
> +	pud_t *pud_table __maybe_unused = stage2_pud_offset(kvm, pgd, 0UL);
> +	stage2_pgd_clear(kvm, pgd);
>  	kvm_tlb_flush_vmid_ipa(kvm, addr);
> -	stage2_pud_free(pud_table);
> +	stage2_pud_free(kvm, pud_table);
>  	put_page(virt_to_page(pgd));
>  }
>  
>  static void clear_stage2_pud_entry(struct kvm *kvm, pud_t *pud, phys_addr_t addr)
>  {
> -	pmd_t *pmd_table __maybe_unused = stage2_pmd_offset(pud, 0);
> -	VM_BUG_ON(stage2_pud_huge(*pud));
> -	stage2_pud_clear(pud);
> +	pmd_t *pmd_table __maybe_unused = stage2_pmd_offset(kvm, pud, 0);
> +	VM_BUG_ON(stage2_pud_huge(kvm, *pud));
> +	stage2_pud_clear(kvm, pud);
>  	kvm_tlb_flush_vmid_ipa(kvm, addr);
> -	stage2_pmd_free(pmd_table);
> +	stage2_pmd_free(kvm, pmd_table);
>  	put_page(virt_to_page(pud));
>  }
>  
> @@ -252,7 +251,7 @@ static void unmap_stage2_ptes(struct kvm *kvm, pmd_t *pmd,
>  		}
>  	} while (pte++, addr += PAGE_SIZE, addr != end);
>  
> -	if (stage2_pte_table_empty(start_pte))
> +	if (stage2_pte_table_empty(kvm, start_pte))
>  		clear_stage2_pmd_entry(kvm, pmd, start_addr);
>  }
>  
> @@ -262,9 +261,9 @@ static void unmap_stage2_pmds(struct kvm *kvm, pud_t *pud,
>  	phys_addr_t next, start_addr = addr;
>  	pmd_t *pmd, *start_pmd;
>  
> -	start_pmd = pmd = stage2_pmd_offset(pud, addr);
> +	start_pmd = pmd = stage2_pmd_offset(kvm, pud, addr);
>  	do {
> -		next = stage2_pmd_addr_end(addr, end);
> +		next = stage2_pmd_addr_end(kvm, addr, end);
>  		if (!pmd_none(*pmd)) {
>  			if (pmd_thp_or_huge(*pmd)) {
>  				pmd_t old_pmd = *pmd;
> @@ -281,7 +280,7 @@ static void unmap_stage2_pmds(struct kvm *kvm, pud_t *pud,
>  		}
>  	} while (pmd++, addr = next, addr != end);
>  
> -	if (stage2_pmd_table_empty(start_pmd))
> +	if (stage2_pmd_table_empty(kvm, start_pmd))
>  		clear_stage2_pud_entry(kvm, pud, start_addr);
>  }
>  
> @@ -291,14 +290,14 @@ static void unmap_stage2_puds(struct kvm *kvm, pgd_t *pgd,
>  	phys_addr_t next, start_addr = addr;
>  	pud_t *pud, *start_pud;
>  
> -	start_pud = pud = stage2_pud_offset(pgd, addr);
> +	start_pud = pud = stage2_pud_offset(kvm, pgd, addr);
>  	do {
> -		next = stage2_pud_addr_end(addr, end);
> -		if (!stage2_pud_none(*pud)) {
> -			if (stage2_pud_huge(*pud)) {
> +		next = stage2_pud_addr_end(kvm, addr, end);
> +		if (!stage2_pud_none(kvm, *pud)) {
> +			if (stage2_pud_huge(kvm, *pud)) {
>  				pud_t old_pud = *pud;
>  
> -				stage2_pud_clear(pud);
> +				stage2_pud_clear(kvm, pud);
>  				kvm_tlb_flush_vmid_ipa(kvm, addr);
>  				kvm_flush_dcache_pud(old_pud);
>  				put_page(virt_to_page(pud));
> @@ -308,7 +307,7 @@ static void unmap_stage2_puds(struct kvm *kvm, pgd_t *pgd,
>  		}
>  	} while (pud++, addr = next, addr != end);
>  
> -	if (stage2_pud_table_empty(start_pud))
> +	if (stage2_pud_table_empty(kvm, start_pud))
>  		clear_stage2_pgd_entry(kvm, pgd, start_addr);
>  }
>  
> @@ -332,7 +331,7 @@ static void unmap_stage2_range(struct kvm *kvm, phys_addr_t start, u64 size)
>  	assert_spin_locked(&kvm->mmu_lock);
>  	WARN_ON(size & ~PAGE_MASK);
>  
> -	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
> +	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
>  	do {
>  		/*
>  		 * Make sure the page table is still active, as another thread
> @@ -341,8 +340,8 @@ static void unmap_stage2_range(struct kvm *kvm, phys_addr_t start, u64 size)
>  		 */
>  		if (!READ_ONCE(kvm->arch.pgd))
>  			break;
> -		next = stage2_pgd_addr_end(addr, end);
> -		if (!stage2_pgd_none(*pgd))
> +		next = stage2_pgd_addr_end(kvm, addr, end);
> +		if (!stage2_pgd_none(kvm, *pgd))
>  			unmap_stage2_puds(kvm, pgd, addr, next);
>  		/*
>  		 * If the range is too large, release the kvm->mmu_lock
> @@ -371,9 +370,9 @@ static void stage2_flush_pmds(struct kvm *kvm, pud_t *pud,
>  	pmd_t *pmd;
>  	phys_addr_t next;
>  
> -	pmd = stage2_pmd_offset(pud, addr);
> +	pmd = stage2_pmd_offset(kvm, pud, addr);
>  	do {
> -		next = stage2_pmd_addr_end(addr, end);
> +		next = stage2_pmd_addr_end(kvm, addr, end);
>  		if (!pmd_none(*pmd)) {
>  			if (pmd_thp_or_huge(*pmd))
>  				kvm_flush_dcache_pmd(*pmd);
> @@ -389,11 +388,11 @@ static void stage2_flush_puds(struct kvm *kvm, pgd_t *pgd,
>  	pud_t *pud;
>  	phys_addr_t next;
>  
> -	pud = stage2_pud_offset(pgd, addr);
> +	pud = stage2_pud_offset(kvm, pgd, addr);
>  	do {
> -		next = stage2_pud_addr_end(addr, end);
> -		if (!stage2_pud_none(*pud)) {
> -			if (stage2_pud_huge(*pud))
> +		next = stage2_pud_addr_end(kvm, addr, end);
> +		if (!stage2_pud_none(kvm, *pud)) {
> +			if (stage2_pud_huge(kvm, *pud))
>  				kvm_flush_dcache_pud(*pud);
>  			else
>  				stage2_flush_pmds(kvm, pud, addr, next);
> @@ -409,10 +408,10 @@ static void stage2_flush_memslot(struct kvm *kvm,
>  	phys_addr_t next;
>  	pgd_t *pgd;
>  
> -	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
> +	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
>  	do {
> -		next = stage2_pgd_addr_end(addr, end);
> -		if (!stage2_pgd_none(*pgd))
> +		next = stage2_pgd_addr_end(kvm, addr, end);
> +		if (!stage2_pgd_none(kvm, *pgd))
>  			stage2_flush_puds(kvm, pgd, addr, next);
>  	} while (pgd++, addr = next, addr != end);
>  }
> @@ -898,7 +897,7 @@ int kvm_alloc_stage2_pgd(struct kvm *kvm)
>  	}
>  
>  	/* Allocate the HW PGD, making sure that each page gets its own refcount */
> -	pgd = alloc_pages_exact(S2_PGD_SIZE, GFP_KERNEL | __GFP_ZERO);
> +	pgd = alloc_pages_exact(stage2_pgd_size(kvm), GFP_KERNEL | __GFP_ZERO);
>  	if (!pgd)
>  		return -ENOMEM;
>  
> @@ -987,7 +986,7 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
>  
>  	spin_lock(&kvm->mmu_lock);
>  	if (kvm->arch.pgd) {
> -		unmap_stage2_range(kvm, 0, KVM_PHYS_SIZE);
> +		unmap_stage2_range(kvm, 0, kvm_phys_size(kvm));
>  		pgd = READ_ONCE(kvm->arch.pgd);
>  		kvm->arch.pgd = NULL;
>  	}
> @@ -995,7 +994,7 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
>  
>  	/* Free the HW pgd, one page at a time */
>  	if (pgd)
> -		free_pages_exact(pgd, S2_PGD_SIZE);
> +		free_pages_exact(pgd, stage2_pgd_size(kvm));
>  }
>  
>  static pud_t *stage2_get_pud(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
> @@ -1004,16 +1003,16 @@ static pud_t *stage2_get_pud(struct kvm *kvm, struct kvm_mmu_memory_cache *cache
>  	pgd_t *pgd;
>  	pud_t *pud;
>  
> -	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
> -	if (stage2_pgd_none(*pgd)) {
> +	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
> +	if (stage2_pgd_none(kvm, *pgd)) {
>  		if (!cache)
>  			return NULL;
>  		pud = mmu_memory_cache_alloc(cache);
> -		stage2_pgd_populate(pgd, pud);
> +		stage2_pgd_populate(kvm, pgd, pud);
>  		get_page(virt_to_page(pgd));
>  	}
>  
> -	return stage2_pud_offset(pgd, addr);
> +	return stage2_pud_offset(kvm, pgd, addr);
>  }
>  
>  static pmd_t *stage2_get_pmd(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
> @@ -1026,15 +1025,15 @@ static pmd_t *stage2_get_pmd(struct kvm *kvm, struct kvm_mmu_memory_cache *cache
>  	if (!pud)
>  		return NULL;
>  
> -	if (stage2_pud_none(*pud)) {
> +	if (stage2_pud_none(kvm, *pud)) {
>  		if (!cache)
>  			return NULL;
>  		pmd = mmu_memory_cache_alloc(cache);
> -		stage2_pud_populate(pud, pmd);
> +		stage2_pud_populate(kvm, pud, pmd);
>  		get_page(virt_to_page(pud));
>  	}
>  
> -	return stage2_pmd_offset(pud, addr);
> +	return stage2_pmd_offset(kvm, pud, addr);
>  }
>  
>  static int stage2_set_pmd_huge(struct kvm *kvm, struct kvm_mmu_memory_cache
> @@ -1208,8 +1207,9 @@ int kvm_phys_addr_ioremap(struct kvm *kvm, phys_addr_t guest_ipa,
>  		if (writable)
>  			pte = kvm_s2pte_mkwrite(pte);
>  
> -		ret = mmu_topup_memory_cache(&cache, KVM_MMU_CACHE_MIN_PAGES,
> -						KVM_NR_MEM_OBJS);
> +		ret = mmu_topup_memory_cache(&cache,
> +					     kvm_mmu_cache_min_pages(kvm),
> +					     KVM_NR_MEM_OBJS);
>  		if (ret)
>  			goto out;
>  		spin_lock(&kvm->mmu_lock);
> @@ -1297,19 +1297,21 @@ static void stage2_wp_ptes(pmd_t *pmd, phys_addr_t addr, phys_addr_t end)
>  
>  /**
>   * stage2_wp_pmds - write protect PUD range
> + * kvm:		kvm instance for the VM
>   * @pud:	pointer to pud entry
>   * @addr:	range start address
>   * @end:	range end address
>   */
> -static void stage2_wp_pmds(pud_t *pud, phys_addr_t addr, phys_addr_t end)
> +static void stage2_wp_pmds(struct kvm *kvm, pud_t *pud,
> +			   phys_addr_t addr, phys_addr_t end)
>  {
>  	pmd_t *pmd;
>  	phys_addr_t next;
>  
> -	pmd = stage2_pmd_offset(pud, addr);
> +	pmd = stage2_pmd_offset(kvm, pud, addr);
>  
>  	do {
> -		next = stage2_pmd_addr_end(addr, end);
> +		next = stage2_pmd_addr_end(kvm, addr, end);
>  		if (!pmd_none(*pmd)) {
>  			if (pmd_thp_or_huge(*pmd)) {
>  				if (!kvm_s2pmd_readonly(pmd))
> @@ -1329,18 +1331,19 @@ static void stage2_wp_pmds(pud_t *pud, phys_addr_t addr, phys_addr_t end)
>    *
>    * Process PUD entries, for a huge PUD we cause a panic.
>    */
> -static void  stage2_wp_puds(pgd_t *pgd, phys_addr_t addr, phys_addr_t end)
> +static void  stage2_wp_puds(struct kvm *kvm, pgd_t *pgd,
> +			    phys_addr_t addr, phys_addr_t end)
>  {
>  	pud_t *pud;
>  	phys_addr_t next;
>  
> -	pud = stage2_pud_offset(pgd, addr);
> +	pud = stage2_pud_offset(kvm, pgd, addr);
>  	do {
> -		next = stage2_pud_addr_end(addr, end);
> -		if (!stage2_pud_none(*pud)) {
> +		next = stage2_pud_addr_end(kvm, addr, end);
> +		if (!stage2_pud_none(kvm, *pud)) {
>  			/* TODO:PUD not supported, revisit later if supported */
> -			BUG_ON(stage2_pud_huge(*pud));
> -			stage2_wp_pmds(pud, addr, next);
> +			BUG_ON(stage2_pud_huge(kvm, *pud));
> +			stage2_wp_pmds(kvm, pud, addr, next);
>  		}
>  	} while (pud++, addr = next, addr != end);
>  }
> @@ -1356,7 +1359,7 @@ static void stage2_wp_range(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  	pgd_t *pgd;
>  	phys_addr_t next;
>  
> -	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
> +	pgd = kvm->arch.pgd + stage2_pgd_index(kvm, addr);
>  	do {
>  		/*
>  		 * Release kvm_mmu_lock periodically if the memory region is
> @@ -1370,9 +1373,9 @@ static void stage2_wp_range(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
>  		cond_resched_lock(&kvm->mmu_lock);
>  		if (!READ_ONCE(kvm->arch.pgd))
>  			break;
> -		next = stage2_pgd_addr_end(addr, end);
> -		if (stage2_pgd_present(*pgd))
> -			stage2_wp_puds(pgd, addr, next);
> +		next = stage2_pgd_addr_end(kvm, addr, end);
> +		if (stage2_pgd_present(kvm, *pgd))
> +			stage2_wp_puds(kvm, pgd, addr, next);
>  	} while (pgd++, addr = next, addr != end);
>  }
>  
> @@ -1521,7 +1524,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
>  	up_read(&current->mm->mmap_sem);
>  
>  	/* We need minimum second+third level pages */
> -	ret = mmu_topup_memory_cache(memcache, KVM_MMU_CACHE_MIN_PAGES,
> +	ret = mmu_topup_memory_cache(memcache, kvm_mmu_cache_min_pages(kvm),
>  				     KVM_NR_MEM_OBJS);
>  	if (ret)
>  		return ret;
> @@ -1764,7 +1767,7 @@ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
>  	}
>  
>  	/* Userspace should not be able to register out-of-bounds IPAs */
> -	VM_BUG_ON(fault_ipa >= KVM_PHYS_SIZE);
> +	VM_BUG_ON(fault_ipa >= kvm_phys_size(vcpu->kvm));
>  
>  	if (fault_status == FSC_ACCESS) {
>  		handle_access_fault(vcpu, fault_ipa);
> @@ -2063,7 +2066,7 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
>  	 * space addressable by the KVM guest IPA space.
>  	 */
>  	if (memslot->base_gfn + memslot->npages >=
> -	    (KVM_PHYS_SIZE >> PAGE_SHIFT))
> +	    (kvm_phys_size(kvm) >> PAGE_SHIFT))
>  		return -EFAULT;
>  
>  	down_read(&current->mm->mmap_sem);
> diff --git a/virt/kvm/arm/vgic/vgic-kvm-device.c b/virt/kvm/arm/vgic/vgic-kvm-device.c
> index 6ada2432e37c..114dce9f4bf5 100644
> --- a/virt/kvm/arm/vgic/vgic-kvm-device.c
> +++ b/virt/kvm/arm/vgic/vgic-kvm-device.c
> @@ -25,7 +25,7 @@
>  int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
>  		      phys_addr_t addr, phys_addr_t alignment)
>  {
> -	if (addr & ~KVM_PHYS_MASK)
> +	if (addr & ~kvm_phys_mask(kvm))
>  		return -E2BIG;
>  
>  	if (!IS_ALIGNED(addr, alignment))
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 12/18] kvm: arm64: Configure VTCR_EL2.SL0 per VM
  2018-09-17 10:41 ` [PATCH v5 12/18] kvm: arm64: Configure VTCR_EL2.SL0 per VM Suzuki K Poulose
@ 2018-09-20 14:25   ` Auger Eric
  2018-09-20 15:25     ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Auger Eric @ 2018-09-20 14:25 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,
On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> VTCR_EL2 holds the following key stage2 translation table
> parameters:
>   SL0  - Entry level in the page table lookup.
>   T0SZ - Denotes the size of the memory addressed by the table.
> 
> We have been using fixed values for the SL0 depending on the
> page size as we have a fixed IPA size. But since we are about
> to make it dynamic, we need to calculate the SL0 at runtime
> per VM. This patch adds a helper to compute the value of SL0
> for a VM based on the IPA size.
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Cc: Eric Auger <eric.auger@redhat.com>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
> Changes since v3:
>  - Update reference to latest ARM ARM.
>  - Update per-vm VTCR value of SL0.
>  - Add helpers to decode levels from SL0.
>  - Didn't pick up Reviewed-by tag from Eric, as there
>    are some new changes in this version
> ---
>  arch/arm64/include/asm/kvm_arm.h | 51 +++++++++++++++++++++++++-------
>  arch/arm64/kvm/reset.c           |  1 +
>  2 files changed, 41 insertions(+), 11 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
> index 3fb1d440be6e..5c1487dc5dca 100644
> --- a/arch/arm64/include/asm/kvm_arm.h
> +++ b/arch/arm64/include/asm/kvm_arm.h
> @@ -121,7 +121,6 @@
>  #define VTCR_EL2_IRGN0_WBWA	TCR_IRGN0_WBWA
>  #define VTCR_EL2_SL0_SHIFT	6
>  #define VTCR_EL2_SL0_MASK	(3 << VTCR_EL2_SL0_SHIFT)
> -#define VTCR_EL2_SL0_LVL1	(1 << VTCR_EL2_SL0_SHIFT)
>  #define VTCR_EL2_T0SZ_MASK	0x3f
>  #define VTCR_EL2_VS_SHIFT	19
>  #define VTCR_EL2_VS_8BIT	(0 << VTCR_EL2_VS_SHIFT)
> @@ -148,29 +147,59 @@
>  /*
>   * Stage2 translation configuration:
>   * 64kB pages (TG0 = 1)
> - * 2 level page tables (SL = 1)
>   */
> -#define VTCR_EL2_TGRAN_FLAGS		(VTCR_EL2_TG0_64K | VTCR_EL2_SL0_LVL1)
> -#define VTTBR_X_TGRAN_MAGIC		38
> +#define VTCR_EL2_TGRAN			VTCR_EL2_TG0_64K
> +#define VTCR_EL2_TGRAN_SL0_BASE		3UL
the name if not obvious. I understand this is yet another magic number
used in the formulae below:
SL0(PAGE_SIZE, Entry_level) = SL0_BASE(PAGE_SIZE) - Entry_Level
I first tried to map this onto some spec fields. May be worth a comment?

Besides
Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks

Eric
> +
>  #elif defined(CONFIG_ARM64_16K_PAGES)
>  /*
>   * Stage2 translation configuration:
>   * 16kB pages (TG0 = 2)
> - * 2 level page tables (SL = 1)
>   */
> -#define VTCR_EL2_TGRAN_FLAGS		(VTCR_EL2_TG0_16K | VTCR_EL2_SL0_LVL1)
> -#define VTTBR_X_TGRAN_MAGIC		42
> +#define VTCR_EL2_TGRAN			VTCR_EL2_TG0_16K
> +#define VTCR_EL2_TGRAN_SL0_BASE		3UL
>  #else	/* 4K */
>  /*
>   * Stage2 translation configuration:
>   * 4kB pages (TG0 = 0)
> - * 3 level page tables (SL = 1)
>   */
> -#define VTCR_EL2_TGRAN_FLAGS		(VTCR_EL2_TG0_4K | VTCR_EL2_SL0_LVL1)
> -#define VTTBR_X_TGRAN_MAGIC		37
> +#define VTCR_EL2_TGRAN			VTCR_EL2_TG0_4K
> +#define VTCR_EL2_TGRAN_SL0_BASE		2UL
>  #endif
>  
> -#define VTCR_EL2_FLAGS			(VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
> +#define VTCR_EL2_FLAGS			(VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN)
> +/*
> + * VTCR_EL2:SL0 indicates the entry level for Stage2 translation.
> + * Interestingly, it depends on the page size.
> + * See D.10.2.121, VTCR_EL2, in ARM DDI 0487C.a
> + *
> + *	-----------------------------------------
> + *	| Entry level		|  4K  | 16K/64K |
> + *	------------------------------------------
> + *	| Level: 0		|  2   |   -     |
> + *	------------------------------------------
> + *	| Level: 1		|  1   |   2     |
> + *	------------------------------------------
> + *	| Level: 2		|  0   |   1     |
> + *	------------------------------------------
> + *	| Level: 3		|  -   |   0     |
> + *	------------------------------------------
> + *
> + * That table roughly translates to :
> + *
> + *	SL0(PAGE_SIZE, Entry_level) = SL0_BASE(PAGE_SIZE) - Entry_Level
> + *
> + * Where SL0_BASE(4K) = 2 and SL0_BASE(16K) = 3, SL0_BASE(64K) = 3, provided
> + * we take care of ruling out the unsupported cases and
> + * Entry_Level = 4 - Number_of_levels.
> + *
> + */
> +#define VTCR_EL2_LVLS_TO_SL0(levels)	\
> +	((VTCR_EL2_TGRAN_SL0_BASE - (4 - (levels))) << VTCR_EL2_SL0_SHIFT)
> +#define VTCR_EL2_SL0_TO_LVLS(sl0)	\
> +	((sl0) + 4 - VTCR_EL2_TGRAN_SL0_BASE)
> +#define VTCR_EL2_LVLS(vtcr)		\
> +	VTCR_EL2_SL0_TO_LVLS(((vtcr) & VTCR_EL2_SL0_MASK) >> VTCR_EL2_SL0_SHIFT)
>  /*
>   * ARM VMSAv8-64 defines an algorithm for finding the translation table
>   * descriptors in section D4.2.8 in ARM DDI 0487C.a.
> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
> index e0c49377b771..d9b7a00993b6 100644
> --- a/arch/arm64/kvm/reset.c
> +++ b/arch/arm64/kvm/reset.c
> @@ -167,6 +167,7 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>  	vtcr |= (kvm_get_vmid_bits() == 16) ?
>  		VTCR_EL2_VS_16BIT :
>  		VTCR_EL2_VS_8BIT;
> +	vtcr |= VTCR_EL2_LVLS_TO_SL0(kvm_stage2_levels(kvm));
>  	kvm->arch.vtcr = vtcr;
>  	return 0;
>  }
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask
  2018-09-20 14:07   ` Auger Eric
@ 2018-09-20 15:22     ` Suzuki K Poulose
  2018-09-25 11:56       ` Auger Eric
  0 siblings, 1 reply; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-20 15:22 UTC (permalink / raw)
  To: Auger Eric, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel



On 20/09/18 15:07, Auger Eric wrote:
> Hi Suzuki,
> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>> On arm64 VTTBR_EL2:BADDR holds the base address for the stage2
>> translation table. The Arm ARM mandates that the bits BADDR[x-1:0]
>> should be 0, where 'x' is defined for a given IPA Size and the
>> number of levels for a translation granule size. It is defined
>> using some magical constants. This patch is a reverse engineered
>> implementation to calculate the 'x' at runtime for a given ipa and
>> number of page table levels. See patch for more details.
>>
>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>> Cc: Christoffer Dall <cdall@kernel.org>
>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
>
>> ---
>> Changes since V3:
>>   - Update reference to latest ARM ARM and improve commentary
>> ---
>>   arch/arm64/include/asm/kvm_arm.h | 63 +++++++++++++++++++++++++++++---
>>   arch/arm64/include/asm/kvm_mmu.h | 25 ++++++++++++-
>>   2 files changed, 81 insertions(+), 7 deletions(-)
>>
>> diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
>> index 14317b3a1820..3fb1d440be6e 100644
>> --- a/arch/arm64/include/asm/kvm_arm.h
>> +++ b/arch/arm64/include/asm/kvm_arm.h
>> @@ -123,7 +123,6 @@
>>   #define VTCR_EL2_SL0_MASK  (3 << VTCR_EL2_SL0_SHIFT)
>>   #define VTCR_EL2_SL0_LVL1  (1 << VTCR_EL2_SL0_SHIFT)
>>   #define VTCR_EL2_T0SZ_MASK 0x3f
>> -#define VTCR_EL2_T0SZ_40B   24
>>   #define VTCR_EL2_VS_SHIFT  19
>>   #define VTCR_EL2_VS_8BIT   (0 << VTCR_EL2_VS_SHIFT)
>>   #define VTCR_EL2_VS_16BIT  (1 << VTCR_EL2_VS_SHIFT)
>> @@ -140,11 +139,8 @@
>>    * Note that when using 4K pages, we concatenate two first level page tables
>>    * together. With 16K pages, we concatenate 16 first level page tables.
>>    *
>> - * The magic numbers used for VTTBR_X in this patch can be found in Tables
>> - * D4-23 and D4-25 in ARM DDI 0487A.b.
>>    */
>>
>> -#define VTCR_EL2_T0SZ_IPA   VTCR_EL2_T0SZ_40B
>>   #define VTCR_EL2_COMMON_BITS       (VTCR_EL2_SH0_INNER | VTCR_EL2_ORGN0_WBWA | \
>>                               VTCR_EL2_IRGN0_WBWA | VTCR_EL2_RES1)
>>
>> @@ -175,9 +171,64 @@
>>   #endif
>>
>>   #define VTCR_EL2_FLAGS                     (VTCR_EL2_COMMON_BITS | VTCR_EL2_TGRAN_FLAGS)
>> -#define VTTBR_X                             (VTTBR_X_TGRAN_MAGIC - VTCR_EL2_T0SZ_IPA)
>> +/*
>> + * ARM VMSAv8-64 defines an algorithm for finding the translation table
>> + * descriptors in section D4.2.8 in ARM DDI 0487C.a.
>> + *
>> + * The algorithm defines the expectations on the BaseAddress (for the page
>> + * table) bits resolved at each level based on the page size, entry level
>> + * and T0SZ. The variable "x" in the algorithm also affects the VTTBR:BADDR
>> + * for stage2 page table.
>> + *
>> + * The value of "x" is calculated as :
>> + *  x = Magic_N - T0SZ
>
> What is not crystal clear to me is the "if SL0b,c = n" case where x get
> a value not based on Magic_N. Please could you explain why it is not
> relevant?

We only care about the "x" for the "entry" level of the table look up
to make sure that the VTTBR is physical address meets the required
alignment. In both cases, if SL0 b,c == n, x is (PAGE_SHIFT) iff the
level you are looking at is not the "entry level". So this should always
be page aligned, like any intermediate level table.

The Magic value is needed only needed for the "entry" level due to the
fact that we may have lesser bits to resolve (i.e, depending on your
PAMax or in other words T0SZ) than the intermediate levels (where we
always resolve {PAGE_SHIFT - 3} bits. This is further complicated by the
fact that Stage2 could use different number of levels for a given T0SZ
than the stage1.
I acknowledge that the algorithm is a bit too cryptic and I spent quite
sometime decode it to the formula we use below ;-).

I could update the comment to :

/*
  * ARM VMSAv8-64 defines an algorithm for finding the translation table
  * descriptors in section D4.2.8 in ARM DDI 0487C.a.
  *
  * The algorithm defines the expectations on the translation table
  * addresses for each level, based on PAGE_SIZE, entry level
  * and the translation table size (T0SZ). The variable "x" in the
  * algorithm determines the alignment of a table base address at a given
  * level and thus determines the alignment of VTTBR:BADDR for stage2
  * page table entry level.
  * Since the number of bits resolved at the entry level could vary
  * depending on the T0SZ, the value of "x" is defined based on a
  * Magic constant for a given PAGE_SIZE and Entry Level. The
  * intermediate levels must be always aligned to the PAGE_SIZE (i.e,
  * x = PAGE_SHIFT).
  *
  * The value of "x" for entry level is calculated as :
  *     x = Magic_N - T0SZ
  *

...

Suzuki
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 12/18] kvm: arm64: Configure VTCR_EL2.SL0 per VM
  2018-09-20 14:25   ` Auger Eric
@ 2018-09-20 15:25     ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-20 15:25 UTC (permalink / raw)
  To: Auger Eric, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Eric,

On 20/09/18 15:25, Auger Eric wrote:
> Hi Suzuki,
> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>> VTCR_EL2 holds the following key stage2 translation table
>> parameters:
>>    SL0  - Entry level in the page table lookup.
>>    T0SZ - Denotes the size of the memory addressed by the table.
>>
>> We have been using fixed values for the SL0 depending on the
>> page size as we have a fixed IPA size. But since we are about
>> to make it dynamic, we need to calculate the SL0 at runtime
>> per VM. This patch adds a helper to compute the value of SL0
>> for a VM based on the IPA size.
>>
>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>> Cc: Christoffer Dall <cdall@kernel.org>
>> Cc: Eric Auger <eric.auger@redhat.com>
>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
>> ---
>> Changes since v3:
>>   - Update reference to latest ARM ARM.
>>   - Update per-vm VTCR value of SL0.
>>   - Add helpers to decode levels from SL0.
>>   - Didn't pick up Reviewed-by tag from Eric, as there
>>     are some new changes in this version


(-)
>>
>> diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
>> index 3fb1d440be6e..5c1487dc5dca 100644
>> --- a/arch/arm64/include/asm/kvm_arm.h
>> +++ b/arch/arm64/include/asm/kvm_arm.h
>> @@ -121,7 +121,6 @@
>>   #define VTCR_EL2_IRGN0_WBWA        TCR_IRGN0_WBWA
>>   #define VTCR_EL2_SL0_SHIFT 6
>>   #define VTCR_EL2_SL0_MASK  (3 << VTCR_EL2_SL0_SHIFT)
>> -#define VTCR_EL2_SL0_LVL1   (1 << VTCR_EL2_SL0_SHIFT)
>>   #define VTCR_EL2_T0SZ_MASK 0x3f
>>   #define VTCR_EL2_VS_SHIFT  19
>>   #define VTCR_EL2_VS_8BIT   (0 << VTCR_EL2_VS_SHIFT)
>> @@ -148,29 +147,59 @@
>>   /*
>>    * Stage2 translation configuration:
>>    * 64kB pages (TG0 = 1)
>> - * 2 level page tables (SL = 1)
>>    */
>> -#define VTCR_EL2_TGRAN_FLAGS                (VTCR_EL2_TG0_64K | VTCR_EL2_SL0_LVL1)
>> -#define VTTBR_X_TGRAN_MAGIC         38
>> +#define VTCR_EL2_TGRAN                      VTCR_EL2_TG0_64K
>> +#define VTCR_EL2_TGRAN_SL0_BASE             3UL
> the name if not obvious. I understand this is yet another magic number
> used in the formulae below:
> SL0(PAGE_SIZE, Entry_level) = SL0_BASE(PAGE_SIZE) - Entry_Level
> I first tried to map this onto some spec fields. May be worth a comment?

Sure, I will make that clear.


>
> Besides
> Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks
Suzuki
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 14/18] vgic: Add support for 52bit guest physical address
  2018-09-17 10:41 ` [PATCH v5 14/18] vgic: Add support for 52bit guest physical address Suzuki K Poulose
@ 2018-09-21 14:57   ` Auger Eric
  2018-09-25 10:49     ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Auger Eric @ 2018-09-21 14:57 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel, Kristina Martsenko

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> From: Kristina Martsenko <kristina.martsenko@arm.com>
> 
> Add support for handling 52bit guest physical address to the
> VGIC layer. So far we have limited the guest physical address
> to 48bits, by explicitly masking the upper bits. This patch
> removes the restriction. We do not have to check if the host
> supports 52bit as the gpa is always validated during an access.
> (e.g, kvm_{read/write}_guest, kvm_is_visible_gfn()).
> Also, the ITS table save-restore is also not affected with
> the enhancement. The DTE entries already store the bits[51:8]
> of the ITT_addr (with a 256byte alignment).
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Signed-off-by: Kristina Martsenko <kristina.martsenko@arm.com>
> [ Macro clean ups, fix PROPBASER and PENDBASER accesses ]
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
>  include/linux/irqchip/arm-gic-v3.h |  5 +++++
>  virt/kvm/arm/vgic/vgic-its.c       | 36 +++++++++---------------------
>  virt/kvm/arm/vgic/vgic-mmio-v3.c   |  2 --
>  3 files changed, 15 insertions(+), 28 deletions(-)
> 
> diff --git a/include/linux/irqchip/arm-gic-v3.h b/include/linux/irqchip/arm-gic-v3.h
> index 8bdbb5f29494..e961f40992d7 100644
> --- a/include/linux/irqchip/arm-gic-v3.h
> +++ b/include/linux/irqchip/arm-gic-v3.h
> @@ -357,6 +357,8 @@
>  #define GITS_CBASER_RaWaWt	GIC_BASER_CACHEABILITY(GITS_CBASER, INNER, RaWaWt)
>  #define GITS_CBASER_RaWaWb	GIC_BASER_CACHEABILITY(GITS_CBASER, INNER, RaWaWb)
>  
> +#define GITS_CBASER_ADDRESS(cbaser)	((cbaser) & GENMASK_ULL(52, 12))
nit GENMASK_ULL(51, 12), bit 52 is RES0
> +
>  #define GITS_BASER_NR_REGS		8
>  
>  #define GITS_BASER_VALID			(1ULL << 63)
> @@ -388,6 +390,9 @@
>  #define GITS_BASER_ENTRY_SIZE_MASK	GENMASK_ULL(52, 48)
>  #define GITS_BASER_PHYS_52_to_48(phys)					\
>  	(((phys) & GENMASK_ULL(47, 16)) | (((phys) >> 48) & 0xf) << 12)
> +#define GITS_BASER_ADDR_48_to_52(baser)					\
> +	(((baser) & GENMASK_ULL(47, 16)) | (((baser) >> 12) & 0xf) << 48)
> +
>  #define GITS_BASER_SHAREABILITY_SHIFT	(10)
>  #define GITS_BASER_InnerShareable					\
>  	GIC_BASER_SHAREABILITY(GITS_BASER, InnerShareable)
> diff --git a/virt/kvm/arm/vgic/vgic-its.c b/virt/kvm/arm/vgic/vgic-its.c
> index 12502251727e..eb2a390a6c86 100644
> --- a/virt/kvm/arm/vgic/vgic-its.c
> +++ b/virt/kvm/arm/vgic/vgic-its.c
> @@ -241,13 +241,6 @@ static struct its_ite *find_ite(struct vgic_its *its, u32 device_id,
>  	list_for_each_entry(dev, &(its)->device_list, dev_list) \
>  		list_for_each_entry(ite, &(dev)->itt_head, ite_list)
>  
> -/*
> - * We only implement 48 bits of PA at the moment, although the ITS
> - * supports more. Let's be restrictive here.
> - */
> -#define BASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 16))
> -#define CBASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 12))
> -
>  #define GIC_LPI_OFFSET 8192
>  
>  #define VITS_TYPER_IDBITS 16
> @@ -759,6 +752,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
>  {
>  	int l1_tbl_size = GITS_BASER_NR_PAGES(baser) * SZ_64K;
>  	u64 indirect_ptr, type = GITS_BASER_TYPE(baser);
> +	phys_addr_t base = GITS_BASER_ADDR_48_to_52(baser);
>  	int esz = GITS_BASER_ENTRY_SIZE(baser);
>  	int index;
>  	gfn_t gfn;
> @@ -783,7 +777,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
>  		if (id >= (l1_tbl_size / esz))
>  			return false;
>  
> -		addr = BASER_ADDRESS(baser) + id * esz;
> +		addr = base + id * esz;
>  		gfn = addr >> PAGE_SHIFT;
>  
>  		if (eaddr)
> @@ -798,7 +792,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
>  
>  	/* Each 1st level entry is represented by a 64-bit value. */
>  	if (kvm_read_guest_lock(its->dev->kvm,
> -			   BASER_ADDRESS(baser) + index * sizeof(indirect_ptr),
> +			   base + index * sizeof(indirect_ptr),
>  			   &indirect_ptr, sizeof(indirect_ptr)))
>  		return false;
>  
> @@ -808,11 +802,7 @@ static bool vgic_its_check_id(struct vgic_its *its, u64 baser, u32 id,
>  	if (!(indirect_ptr & BIT_ULL(63)))
>  		return false;
>  
> -	/*
> -	 * Mask the guest physical address and calculate the frame number.
> -	 * Any address beyond our supported 48 bits of PA will be caught
> -	 * by the actual check in the final step.
> -	 */
> +	/* Mask the guest physical address and calculate the frame number. */
>  	indirect_ptr &= GENMASK_ULL(51, 16);
>  
>  	/* Find the address of the actual entry */
> @@ -1304,9 +1294,6 @@ static u64 vgic_sanitise_its_baser(u64 reg)
>  				  GITS_BASER_OUTER_CACHEABILITY_SHIFT,
>  				  vgic_sanitise_outer_cacheability);
>  
> -	/* Bits 15:12 contain bits 51:48 of the PA, which we don't support. */
> -	reg &= ~GENMASK_ULL(15, 12);
> -
>  	/* We support only one (ITS) page size: 64K */
>  	reg = (reg & ~GITS_BASER_PAGE_SIZE_MASK) | GITS_BASER_PAGE_SIZE_64K;
>  
> @@ -1325,11 +1312,8 @@ static u64 vgic_sanitise_its_cbaser(u64 reg)
>  				  GITS_CBASER_OUTER_CACHEABILITY_SHIFT,
>  				  vgic_sanitise_outer_cacheability);
>  
> -	/*
> -	 * Sanitise the physical address to be 64k aligned.
> -	 * Also limit the physical addresses to 48 bits.
> -	 */
> -	reg &= ~(GENMASK_ULL(51, 48) | GENMASK_ULL(15, 12));
> +	/* Sanitise the physical address to be 64k aligned. */
> +	reg &= ~GENMASK_ULL(15, 12);
>  
>  	return reg;
>  }
> @@ -1375,7 +1359,7 @@ static void vgic_its_process_commands(struct kvm *kvm, struct vgic_its *its)
>  	if (!its->enabled)
>  		return;
>  
> -	cbaser = CBASER_ADDRESS(its->cbaser);
> +	cbaser = GITS_CBASER_ADDRESS(its->cbaser);
>  
>  	while (its->cwriter != its->creadr) {
>  		int ret = kvm_read_guest_lock(kvm, cbaser + its->creadr,
> @@ -2233,7 +2217,7 @@ static int vgic_its_restore_device_tables(struct vgic_its *its)
>  	if (!(baser & GITS_BASER_VALID))
>  		return 0;
>  
> -	l1_gpa = BASER_ADDRESS(baser);
> +	l1_gpa = GITS_BASER_ADDR_48_to_52(baser);
>  
>  	if (baser & GITS_BASER_INDIRECT) {
>  		l1_esz = GITS_LVL1_ENTRY_SIZE;
> @@ -2305,7 +2289,7 @@ static int vgic_its_save_collection_table(struct vgic_its *its)
>  {
>  	const struct vgic_its_abi *abi = vgic_its_get_abi(its);
>  	u64 baser = its->baser_coll_table;
> -	gpa_t gpa = BASER_ADDRESS(baser);
> +	gpa_t gpa = GITS_BASER_ADDR_48_to_52(baser);
>  	struct its_collection *collection;
>  	u64 val;
>  	size_t max_size, filled = 0;
> @@ -2354,7 +2338,7 @@ static int vgic_its_restore_collection_table(struct vgic_its *its)
>  	if (!(baser & GITS_BASER_VALID))
>  		return 0;
>  
> -	gpa = BASER_ADDRESS(baser);
> +	gpa = GITS_BASER_ADDR_48_to_52(baser);
>  
>  	max_size = GITS_BASER_NR_PAGES(baser) * SZ_64K;
>  
> diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c b/virt/kvm/arm/vgic/vgic-mmio-v3.c
> index a2a175b08b17..b3d1f0985117 100644
> --- a/virt/kvm/arm/vgic/vgic-mmio-v3.c
> +++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c
> @@ -364,7 +364,6 @@ static u64 vgic_sanitise_pendbaser(u64 reg)
>  				  vgic_sanitise_outer_cacheability);
>  
>  	reg &= ~PENDBASER_RES0_MASK;
> -	reg &= ~GENMASK_ULL(51, 48);
>  
>  	return reg;
>  }
> @@ -382,7 +381,6 @@ static u64 vgic_sanitise_propbaser(u64 reg)
>  				  vgic_sanitise_outer_cacheability);
>  
>  	reg &= ~PROPBASER_RES0_MASK;
> -	reg &= ~GENMASK_ULL(51, 48);
>  	return reg;
>  }
>  
> 
Besides looks good to me.
Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks

Eric


^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 16/18] kvm: arm64: Set a limit on the IPA size
  2018-09-17 10:41 ` [PATCH v5 16/18] kvm: arm64: Set a limit on the IPA size Suzuki K Poulose
@ 2018-09-25  9:59   ` Auger Eric
  2018-09-25 11:10     ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Auger Eric @ 2018-09-25  9:59 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> So far we have restricted the IPA size of the VM to the default
> value (40bits). Now that we can manage the IPA size per VM and
> support dynamic stage2 page tables, we can allow VMs to have
> larger IPA. This patch introduces a the maximum IPA size
> supported on the host.
to be reworded
 This is decided by the following factors :
> 
>  1) Maximum PARange supported by the CPUs - This can be inferred
>     from the system wide safe value.
>  2) Maximum PA size supported by the host kernel (48 vs 52)
>  3) Number of levels in the host page table (as we base our
>     stage2 tables on the host table helpers).
> 
> Since the stage2 page table code is dependent on the stage1
> page table, we always ensure that :
> 
>   Number of Levels at Stage1 >= Number of Levels at Stage2
> 
> So we limit the IPA to make sure that the above condition
> is satisfied. This will affect the following combinations
> of VA_BITS and IPA for different page sizes.
> 
>   Host configuration | Unsupported IPA ranges
>   39bit VA, 4K       | [44, 48]
>   36bit VA, 16K      | [41, 48]
>   42bit VA, 64K      | [47, 52]
> 
> Supporting the above combinations need independent stage2
> page table manipulation code, which would need substantial
> changes. We could purse the solution independently and
> switch the page table code once we have it ready.
> 
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
> Changes since V2:
>  - Restrict the IPA size to limit the number of page table
>    levels in stage2 to that of stage1 or less.
> ---
>  arch/arm/include/asm/kvm_mmu.h    |  2 ++
>  arch/arm64/include/asm/kvm_host.h |  2 ++
>  arch/arm64/kvm/reset.c            | 43 +++++++++++++++++++++++++++++++
>  virt/kvm/arm/arm.c                |  2 ++
>  4 files changed, 49 insertions(+)
> 
> diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
> index 12ae5fbbcf01..5ad1a54f98dc 100644
> --- a/arch/arm/include/asm/kvm_mmu.h
> +++ b/arch/arm/include/asm/kvm_mmu.h
> @@ -358,6 +358,8 @@ static inline int hyp_map_aux_data(void)
>  
>  #define kvm_phys_to_vttbr(addr)		(addr)
>  
> +static inline void kvm_set_ipa_limit(void) {}
> +
>  #endif	/* !__ASSEMBLY__ */
>  
>  #endif /* __ARM_KVM_MMU_H__ */
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index 5ecd457bce7d..f0474061851d 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -513,6 +513,8 @@ static inline int kvm_arm_have_ssbd(void)
>  void kvm_vcpu_load_sysregs(struct kvm_vcpu *vcpu);
>  void kvm_vcpu_put_sysregs(struct kvm_vcpu *vcpu);
>  
> +void kvm_set_ipa_limit(void);
> +
>  #define __KVM_HAVE_ARCH_VM_ALLOC
>  struct kvm *kvm_arch_alloc_vm(void);
>  void kvm_arch_free_vm(struct kvm *kvm);
> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
> index 51ecf0f7c912..76972b19bdd7 100644
> --- a/arch/arm64/kvm/reset.c
> +++ b/arch/arm64/kvm/reset.c
> @@ -34,6 +34,9 @@
>  #include <asm/kvm_coproc.h>
>  #include <asm/kvm_mmu.h>
>  
> +/* Maximum phys_shift supported for any VM on this host */
> +static u32 kvm_ipa_limit;
> +
>  /*
>   * ARMv8 Reset Values
>   */
> @@ -135,6 +138,46 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
>  	return kvm_timer_vcpu_reset(vcpu);
>  }
>  
> +void kvm_set_ipa_limit(void)
> +{
> +	unsigned int ipa_max, va_max, parange;
> +
> +	parange = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1) & 0x7;
> +	ipa_max = id_aa64mmfr0_parange_to_phys_shift(parange);
> +
> +	/* Raise the limit to the default size for backward compatibility */
> +	if (ipa_max < KVM_PHYS_SHIFT) {
> +		WARN_ONCE(1,
> +			  "PARange is %d bits, unsupported configuration!",
> +			  ipa_max);
> +		ipa_max = KVM_PHYS_SHIFT;
I don't really get what does happen in this case. The CPU cannot handle
PA up to ipa_max so can the VM run properly? In case it is a
showstopper, kvm_set_ipa_limit should return an error, cascaded by
init_common_resources. Otherwise the warning message may be reworded.
> +	}
> +
> +	/* Clamp it to the PA size supported by the kernel */
> +	ipa_max = (ipa_max > PHYS_MASK_SHIFT) ? PHYS_MASK_SHIFT : ipa_max;
> +	/*
> +	 * Since our stage2 table is dependent on the stage1 page table code,
> +	 * we must always honor the following condition:
> +	 *
> +	 *  Number of levels in Stage1 >= Number of levels in Stage2.
> +	 *
> +	 * So clamp the ipa limit further down to limit the number of levels.
> +	 * Since we can concatenate upto 16 tables at entry level, we could
> +	 * go upto 4bits above the maximum VA addressible with the current
addressable?
> +	 * number of levels.
> +	 */
> +	va_max = PGDIR_SHIFT + PAGE_SHIFT - 3;
> +	va_max += 4;
> +
> +	if (va_max < ipa_max) {
> +		kvm_info("Limiting IPA limit to %dbytes due to host VA bits limitation\n",> +			 va_max);
> +		ipa_max = va_max;
you have a trace for this limitation but none for the comparison against
PHYS_MASK_SHIFT.
> +	}
> +
> +	kvm_ipa_limit = ipa_max;
> +}
> +
>  /*
>   * Configure the VTCR_EL2 for this VM. The VTCR value is common
>   * across all the physical CPUs on the system. We use system wide
> diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
> index 43e716bc3f08..631f9a3ad99a 100644
> --- a/virt/kvm/arm/arm.c
> +++ b/virt/kvm/arm/arm.c
> @@ -1413,6 +1413,8 @@ static int init_common_resources(void)
>  	kvm_vmid_bits = kvm_get_vmid_bits();
>  	kvm_info("%d-bit VMID\n", kvm_vmid_bits);
>  
> +	kvm_set_ipa_limit();
As we have a kvm_info for the supported vmid_bits, may be good to output
the max IPA size supported by the host whatever the applied clamps?

Thanks

Eric
> +
>  	return 0;
>  }
>  
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 15/18] kvm: arm64: Add 52bit support for PAR to HPFAR conversoin
  2018-09-17 10:41 ` [PATCH v5 15/18] kvm: arm64: Add 52bit support for PAR to HPFAR conversoin Suzuki K Poulose
@ 2018-09-25  9:59   ` Auger Eric
  0 siblings, 0 replies; 50+ messages in thread
From: Auger Eric @ 2018-09-25  9:59 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> Add support for handling 52bit addresses in PAR to HPFAR
> conversion. Instead of hardcoding the address limits, we
> now use PHYS_MASK_SHIFT.
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks

Eric

> ---
>  arch/arm64/include/asm/kvm_arm.h | 7 +++++++
>  arch/arm64/kvm/hyp/switch.c      | 2 +-
>  2 files changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
> index 0a37c0513ede..241d1622fa19 100644
> --- a/arch/arm64/include/asm/kvm_arm.h
> +++ b/arch/arm64/include/asm/kvm_arm.h
> @@ -308,6 +308,13 @@
>  
>  /* Hyp Prefetch Fault Address Register (HPFAR/HDFAR) */
>  #define HPFAR_MASK	(~UL(0xf))
> +/*
> + * We have
> + *	PAR	[PA_Shift - 1	: 12] = PA	[PA_Shift - 1 : 12]
> + *	HPFAR	[PA_Shift - 9	: 4]  = FIPA	[PA_Shift - 1 : 12]
> + */
> +#define PAR_TO_HPFAR(par)		\
> +	(((par) & GENMASK_ULL(PHYS_MASK_SHIFT - 1, 12)) >> 8)
>  
>  #define kvm_arm_exception_type	\
>  	{0, "IRQ" }, 		\
> diff --git a/arch/arm64/kvm/hyp/switch.c b/arch/arm64/kvm/hyp/switch.c
> index 9d5ce1a3039a..7cc175c88a37 100644
> --- a/arch/arm64/kvm/hyp/switch.c
> +++ b/arch/arm64/kvm/hyp/switch.c
> @@ -263,7 +263,7 @@ static bool __hyp_text __translate_far_to_hpfar(u64 far, u64 *hpfar)
>  		return false; /* Translation failed, back to guest */
>  
>  	/* Convert PAR to HPFAR format */
> -	*hpfar = ((tmp >> 12) & ((1UL << 36) - 1)) << 4;
> +	*hpfar = PAR_TO_HPFAR(tmp);
>  	return true;
>  }
>  
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-17 10:41 ` [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM Suzuki K Poulose
  2018-09-17 14:20   ` Peter Maydell
  2018-09-18  1:55   ` Peter Maydell
@ 2018-09-25 10:00   ` Auger Eric
  2018-09-25 10:24     ` Suzuki K Poulose
  2 siblings, 1 reply; 50+ messages in thread
From: Auger Eric @ 2018-09-25 10:00 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel, Peter Maydel

Hi Suzuki,
On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> Allow specifying the physical address size limit for a new
> VM via the kvm_type argument for the KVM_CREATE_VM ioctl. This
> allows us to finalise the stage2 page table as early as possible
> and hence perform the right checks on the memory slots
> without complication. The size is ecnoded as Log2(PA_Size) in
encoded
> bits[7:0] of the type field. For backward compatibility the
> value 0 is reserved and implies 40bits. Also, lift the limit
> of the IPA to host limit and allow lower IPA sizes (e.g, 32).
> 
> The userspace could check the extension KVM_CAP_ARM_VM_PHYS_SHIFT
> for the availability of this feature. The cap check returns the
> maximum limit for the physical address shift supported by the host.
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Cc: Peter Maydel <peter.maydell@linaro.org>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
> Changes since v4:
>  - Fold the introduction of the KVM_CAP_ARM_VM_PHYS_SHIFT to this
>    patch to allow detection of the availability of the feature for
>    userspace.
>  - Document the API
>  - Restrict the feature only to arm64.
> Changes since V3:
>  - Switch to a CAP, that can be checkd via EXTENSIONS on KVM device
>    fd, rather than a dedicated ioctl.
> ---
>  Documentation/virtual/kvm/api.txt       |  8 ++++++++
>  arch/arm64/include/asm/stage2_pgtable.h | 20 --------------------
>  arch/arm64/kvm/reset.c                  | 20 ++++++++++++++++----
>  include/uapi/linux/kvm.h                | 10 ++++++++++
>  4 files changed, 34 insertions(+), 24 deletions(-)
> 
> diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
> index c664064f76fb..f860251ff27c 100644
> --- a/Documentation/virtual/kvm/api.txt
> +++ b/Documentation/virtual/kvm/api.txt
> @@ -122,6 +122,14 @@ the default trap & emulate implementation (which changes the virtual
>  memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
>  flag KVM_VM_MIPS_VZ.
>  
> +To configure the physical address space size for a VM (IPA size) on arm64,
> +check KVM_CAP_ARM_VM_PHYS_SHIFT (which returns the maximum limit for the
> +IPA shift) and use KVM_VM_TYPE_ARM_PHYS_SHIFT(PHYS_SHIFT). Bits[7-0] of the
> +machine type has been reserved for specifying the PHYS_SHIFT.
are reserved to pass the PHYS_SHIFT?
> +The supported range is [32...IPA_LIMIT], where IPA_LIMIT could be
s/could be/is
> +identified by checking KVM_CAP_ARM_VM_PHYS_SHIFT. For backward compatibility
> +a value of 0 selects 40bits.
> +
>  
>  4.3 KVM_GET_MSR_INDEX_LIST, KVM_GET_MSR_FEATURE_INDEX_LIST
>  
> diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
> index 6a56fdff0823..0b339f5a4a7c 100644
> --- a/arch/arm64/include/asm/stage2_pgtable.h
> +++ b/arch/arm64/include/asm/stage2_pgtable.h
> @@ -42,28 +42,8 @@
>   * the range (IPA_SHIFT, IPA_SHIFT - 4).
>   */
>  #define stage2_pgtable_levels(ipa)	ARM64_HW_PGTABLE_LEVELS((ipa) - 4)
> -#define STAGE2_PGTABLE_LEVELS		stage2_pgtable_levels(KVM_PHYS_SHIFT)
>  #define kvm_stage2_levels(kvm)		VTCR_EL2_LVLS(kvm->arch.vtcr)
>  
> -/*
> - * With all the supported VA_BITs and 40bit guest IPA, the following condition
> - * is always true:
> - *
> - *       STAGE2_PGTABLE_LEVELS <= CONFIG_PGTABLE_LEVELS
> - *
> - * We base our stage-2 page table walker helpers on this assumption and
> - * fall back to using the host version of the helper wherever possible.
> - * i.e, if a particular level is not folded (e.g, PUD) at stage2, we fall back
> - * to using the host version, since it is guaranteed it is not folded at host.
> - *
> - * If the condition breaks in the future, we can rearrange the host level
> - * definitions and reuse them for stage2. Till then...
> - */
> -#if STAGE2_PGTABLE_LEVELS > CONFIG_PGTABLE_LEVELS
> -#error "Unsupported combination of guest IPA and host VA_BITS."
> -#endif
> -
> -
>  /* stage2_pgdir_shift() is the size mapped by top-level stage2 entry for the VM */
>  #define stage2_pgdir_shift(kvm)		pt_levels_pgdir_shift(kvm_stage2_levels(kvm))
>  #define stage2_pgdir_size(kvm)		(1ULL << stage2_pgdir_shift(kvm))
> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
> index 0393bb974b23..c9640159e11f 100644
> --- a/arch/arm64/kvm/reset.c
> +++ b/arch/arm64/kvm/reset.c
> @@ -89,6 +89,9 @@ int kvm_arch_dev_ioctl_check_extension(struct kvm *kvm, long ext)
>  	case KVM_CAP_VCPU_EVENTS:
>  		r = 1;
>  		break;
> +	case KVM_CAP_ARM_VM_PHYS_SHIFT:
> +		r = kvm_ipa_limit;
> +		break;
>  	default:
>  		r = 0;
>  	}
> @@ -190,16 +193,25 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>  {
>  	u64 vtcr = VTCR_EL2_FLAGS;
>  	u64 parange;
> -	u8 lvls;
> +	u8 lvls, ipa_shift;
>  
> -	if (type)
> +	if (type & ~KVM_VM_TYPE_ARM_PHYS_SHIFT_MASK)
>  		return -EINVAL;
>  
> +	ipa_shift = KVM_VM_TYPE_ARM_PHYS_SHIFT(type);
> +	if (ipa_shift) {
> +		if (ipa_shift > kvm_ipa_limit ||
> +		    ipa_shift < 32)
> +			return -EINVAL;
> +	} else {
> +		ipa_shift = KVM_PHYS_SHIFT;
> +	}
> +
>  	/*
>  	 * Use a minimum 2 level page table to prevent splitting
>  	 * host PMD huge pages at stage2.
>  	 */
> -	lvls = stage2_pgtable_levels(KVM_PHYS_SHIFT);
> +	lvls = stage2_pgtable_levels(ipa_shift);
>  	if (lvls < 2)
>  		lvls = 2;
>  
> @@ -221,7 +233,7 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>  		VTCR_EL2_VS_8BIT;
>  
>  	vtcr |= VTCR_EL2_LVLS_TO_SL0(lvls);
> -	vtcr |= VTCR_EL2_T0SZ(KVM_PHYS_SHIFT);
> +	vtcr |= VTCR_EL2_T0SZ(ipa_shift);
>  
>  	kvm->arch.vtcr = vtcr;
>  	return 0;
> diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
> index 07548de5c988..2a6b29c446db 100644
> --- a/include/uapi/linux/kvm.h
> +++ b/include/uapi/linux/kvm.h
> @@ -750,6 +750,15 @@ struct kvm_ppc_resize_hpt {
>  
>  #define KVM_S390_SIE_PAGE_OFFSET 1
>  
> +/*
> + * On arm64, machine type can be used to request the physical
> + * address size for the VM. Bits[7-0] has been reserved for the PA
s/has been reserved/are?

Thanks

Eric
> + * size shift (i.e, log2(PA_Size)). For backward compatibility,
> + * value 0 implies the default IPA size, 40bits.
> + */
> +#define KVM_VM_TYPE_ARM_PHYS_SHIFT_MASK	0xffULL
> +#define KVM_VM_TYPE_ARM_PHYS_SHIFT(x)		\
> +	((x) & KVM_VM_TYPE_ARM_PHYS_SHIFT_MASK)
>  /*
>   * ioctls for /dev/kvm fds:
>   */
> @@ -952,6 +961,7 @@ struct kvm_ppc_resize_hpt {
>  #define KVM_CAP_S390_HPAGE_1M 156
>  #define KVM_CAP_NESTED_STATE 157
>  #define KVM_CAP_ARM_INJECT_SERROR_ESR 158
> +#define KVM_CAP_ARM_VM_PHYS_SHIFT 159 /* returns maximum PA shift for a VM */
>  
>  #ifdef KVM_CAP_IRQ_ROUTING
>  
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 17/18] kvm: arm64: Limit the minimum number of page table levels
  2018-09-17 10:41 ` [PATCH v5 17/18] kvm: arm64: Limit the minimum number of page table levels Suzuki K Poulose
@ 2018-09-25 10:00   ` Auger Eric
  2018-09-25 10:25     ` Suzuki K Poulose
  0 siblings, 1 reply; 50+ messages in thread
From: Auger Eric @ 2018-09-25 10:00 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
> Since we are about to remove the lower limit on the IPA size,
> make sure that we do not go to 1 level page table (e.g, with
> 32bit IPA on 64K host with concatenation) to avoid splitting
> the host PMD huge pages at stage2.
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Christoffer Dall <cdall@kernel.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> ---
>  arch/arm64/include/asm/stage2_pgtable.h |  8 +++++++-
>  arch/arm64/kvm/reset.c                  | 12 +++++++++++-
>  2 files changed, 18 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
> index 352ec4158fdf..6a56fdff0823 100644
> --- a/arch/arm64/include/asm/stage2_pgtable.h
> +++ b/arch/arm64/include/asm/stage2_pgtable.h
> @@ -72,8 +72,14 @@
>  /*
>   * The number of PTRS across all concatenated stage2 tables given by the
>   * number of bits resolved at the initial level.
> + * If we force more number of levels than necessary, we may have
more levels?
> + * stage2_pgdir_shift > IPA, in which case, stage2_pgd_ptrs will have
> + * one entry.
>   */
> -#define __s2_pgd_ptrs(ipa, lvls)	(1 << ((ipa) - pt_levels_pgdir_shift((lvls))))
> +#define pgd_ptrs_shift(ipa, pgdir_shift)	\
> +	((ipa) > (pgdir_shift) ? ((ipa) - (pgdir_shift)) : 0)
> +#define __s2_pgd_ptrs(ipa, lvls)		\
> +	(1 << (pgd_ptrs_shift((ipa), pt_levels_pgdir_shift(lvls))))
>  #define __s2_pgd_size(ipa, lvls)	(__s2_pgd_ptrs((ipa), (lvls)) * sizeof(pgd_t))
>  
>  #define stage2_pgd_ptrs(kvm)		__s2_pgd_ptrs(kvm_phys_shift(kvm), kvm_stage2_levels(kvm))
> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
> index 76972b19bdd7..0393bb974b23 100644
> --- a/arch/arm64/kvm/reset.c
> +++ b/arch/arm64/kvm/reset.c
> @@ -190,10 +190,19 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>  {
>  	u64 vtcr = VTCR_EL2_FLAGS;
>  	u64 parange;
> +	u8 lvls;
>  
>  	if (type)
>  		return -EINVAL;
>  
> +	/*
> +	 * Use a minimum 2 level page table to prevent splitting
> +	 * host PMD huge pages at stage2.
> +	 */
> +	lvls = stage2_pgtable_levels(KVM_PHYS_SHIFT);
> +	if (lvls < 2)
> +		lvls = 2;
> +
>  	parange = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1) & 7;
>  	if (parange > ID_AA64MMFR0_PARANGE_MAX)
>  		parange = ID_AA64MMFR0_PARANGE_MAX;
> @@ -210,7 +219,8 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>  	vtcr |= (kvm_get_vmid_bits() == 16) ?
>  		VTCR_EL2_VS_16BIT :
>  		VTCR_EL2_VS_8BIT;
> -	vtcr |= VTCR_EL2_LVLS_TO_SL0(stage2_pgtable_levels(KVM_PHYS_SHIFT));
> +
nit: new line not requested

Thanks

Eric
> +	vtcr |= VTCR_EL2_LVLS_TO_SL0(lvls);
>  	vtcr |= VTCR_EL2_T0SZ(KVM_PHYS_SHIFT);
>  
>  	kvm->arch.vtcr = vtcr;
> 

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM
  2018-09-25 10:00   ` Auger Eric
@ 2018-09-25 10:24     ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-25 10:24 UTC (permalink / raw)
  To: Auger Eric, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel, Peter Maydel

On 09/25/2018 11:00 AM, Auger Eric wrote:
> Hi Suzuki,
> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>> Allow specifying the physical address size limit for a new
>> VM via the kvm_type argument for the KVM_CREATE_VM ioctl. This
>> allows us to finalise the stage2 page table as early as possible
>> and hence perform the right checks on the memory slots
>> without complication. The size is ecnoded as Log2(PA_Size) in
> encoded

...

>>
>> diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
>> index c664064f76fb..f860251ff27c 100644
>> --- a/Documentation/virtual/kvm/api.txt
>> +++ b/Documentation/virtual/kvm/api.txt
>> @@ -122,6 +122,14 @@ the default trap & emulate implementation (which changes the virtual
>>   memory layout to fit in user mode), check KVM_CAP_MIPS_VZ and use the
>>   flag KVM_VM_MIPS_VZ.
>>   
>> +To configure the physical address space size for a VM (IPA size) on arm64,
>> +check KVM_CAP_ARM_VM_PHYS_SHIFT (which returns the maximum limit for the
>> +IPA shift) and use KVM_VM_TYPE_ARM_PHYS_SHIFT(PHYS_SHIFT). Bits[7-0] of the
>> +machine type has been reserved for specifying the PHYS_SHIFT.
> are reserved to pass the PHYS_SHIFT?
>> +The supported range is [32...IPA_LIMIT], where IPA_LIMIT could be
> s/could be/is



>> diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
>> index 07548de5c988..2a6b29c446db 100644
>> --- a/include/uapi/linux/kvm.h
>> +++ b/include/uapi/linux/kvm.h
>> @@ -750,6 +750,15 @@ struct kvm_ppc_resize_hpt {
>>   
>>   #define KVM_S390_SIE_PAGE_OFFSET 1
>>   
>> +/*
>> + * On arm64, machine type can be used to request the physical
>> + * address size for the VM. Bits[7-0] has been reserved for the PA
> s/has been reserved/are?

Thanks for spotting, fixed all the above.

Suzuki

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 17/18] kvm: arm64: Limit the minimum number of page table levels
  2018-09-25 10:00   ` Auger Eric
@ 2018-09-25 10:25     ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-25 10:25 UTC (permalink / raw)
  To: Auger Eric, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

On 09/25/2018 11:00 AM, Auger Eric wrote:
> Hi Suzuki,
> 
> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>> Since we are about to remove the lower limit on the IPA size,
>> make sure that we do not go to 1 level page table (e.g, with
>> 32bit IPA on 64K host with concatenation) to avoid splitting
>> the host PMD huge pages at stage2.
>>
>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>> Cc: Christoffer Dall <cdall@kernel.org>
>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
>> ---
>>   arch/arm64/include/asm/stage2_pgtable.h |  8 +++++++-
>>   arch/arm64/kvm/reset.c                  | 12 +++++++++++-
>>   2 files changed, 18 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/arm64/include/asm/stage2_pgtable.h b/arch/arm64/include/asm/stage2_pgtable.h
>> index 352ec4158fdf..6a56fdff0823 100644
>> --- a/arch/arm64/include/asm/stage2_pgtable.h
>> +++ b/arch/arm64/include/asm/stage2_pgtable.h
>> @@ -72,8 +72,14 @@
>>   /*
>>    * The number of PTRS across all concatenated stage2 tables given by the
>>    * number of bits resolved at the initial level.
>> + * If we force more number of levels than necessary, we may have
> more levels?


>> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
>> index 76972b19bdd7..0393bb974b23 100644
>> --- a/arch/arm64/kvm/reset.c
>> +++ b/arch/arm64/kvm/reset.c

>> @@ -210,7 +219,8 @@ int kvm_arm_config_vm(struct kvm *kvm, unsigned long type)
>>   	vtcr |= (kvm_get_vmid_bits() == 16) ?
>>   		VTCR_EL2_VS_16BIT :
>>   		VTCR_EL2_VS_8BIT;
>> -	vtcr |= VTCR_EL2_LVLS_TO_SL0(stage2_pgtable_levels(KVM_PHYS_SHIFT));
>> +
> nit: new line not requested
> 

Fixed all the above

Suzuki

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 14/18] vgic: Add support for 52bit guest physical address
  2018-09-21 14:57   ` Auger Eric
@ 2018-09-25 10:49     ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-25 10:49 UTC (permalink / raw)
  To: Auger Eric, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel, Kristina Martsenko

Hi Eric

On 09/21/2018 03:57 PM, Auger Eric wrote:
> Hi Suzuki,
> 
> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>> From: Kristina Martsenko <kristina.martsenko@arm.com>
>>
>> Add support for handling 52bit guest physical address to the
>> VGIC layer. So far we have limited the guest physical address
>> to 48bits, by explicitly masking the upper bits. This patch
>> removes the restriction. We do not have to check if the host
>> supports 52bit as the gpa is always validated during an access.
>> (e.g, kvm_{read/write}_guest, kvm_is_visible_gfn()).
>> Also, the ITS table save-restore is also not affected with
>> the enhancement. The DTE entries already store the bits[51:8]
>> of the ITT_addr (with a 256byte alignment).
>>
>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>> Cc: Christoffer Dall <cdall@kernel.org>
>> Signed-off-by: Kristina Martsenko <kristina.martsenko@arm.com>
>> [ Macro clean ups, fix PROPBASER and PENDBASER accesses ]
>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
>> ---
>>   include/linux/irqchip/arm-gic-v3.h |  5 +++++
>>   virt/kvm/arm/vgic/vgic-its.c       | 36 +++++++++---------------------
>>   virt/kvm/arm/vgic/vgic-mmio-v3.c   |  2 --
>>   3 files changed, 15 insertions(+), 28 deletions(-)
>>
>> diff --git a/include/linux/irqchip/arm-gic-v3.h b/include/linux/irqchip/arm-gic-v3.h
>> index 8bdbb5f29494..e961f40992d7 100644
>> --- a/include/linux/irqchip/arm-gic-v3.h
>> +++ b/include/linux/irqchip/arm-gic-v3.h
>> @@ -357,6 +357,8 @@
>>   #define GITS_CBASER_RaWaWt	GIC_BASER_CACHEABILITY(GITS_CBASER, INNER, RaWaWt)
>>   #define GITS_CBASER_RaWaWb	GIC_BASER_CACHEABILITY(GITS_CBASER, INNER, RaWaWb)
>>   
>> +#define GITS_CBASER_ADDRESS(cbaser)	((cbaser) & GENMASK_ULL(52, 12))
> nit GENMASK_ULL(51, 12), bit 52 is RES0

I will fix this.

>> diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c b/virt/kvm/arm/vgic/vgic-mmio-v3.c
>> index a2a175b08b17..b3d1f0985117 100644
>> --- a/virt/kvm/arm/vgic/vgic-mmio-v3.c
>> +++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c
>> @@ -364,7 +364,6 @@ static u64 vgic_sanitise_pendbaser(u64 reg)
>>   				  vgic_sanitise_outer_an);
>>   
>>   	reg &= ~PENDBASER_RES0_MASK;
>> -	reg &= ~GENMASK_ULL(51, 48);
>>   
>>   	return reg;
>>   }
>> @@ -382,7 +381,6 @@ static u64 vgic_sanitise_propbaser(u64 reg)
>>   				  vgic_sanitise_outer_cacheability);
>>   
>>   	reg &= ~PROPBASER_RES0_MASK;
>> -	reg &= ~GENMASK_ULL(51, 48);
>>   	return reg;
>>   }
>>   
>>
> Besides looks good to me.
> Reviewed-by: Eric Auger <eric.auger@redhat.com>

Thanks
Suzuki

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 16/18] kvm: arm64: Set a limit on the IPA size
  2018-09-25  9:59   ` Auger Eric
@ 2018-09-25 11:10     ` Suzuki K Poulose
  0 siblings, 0 replies; 50+ messages in thread
From: Suzuki K Poulose @ 2018-09-25 11:10 UTC (permalink / raw)
  To: Auger Eric, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

On 09/25/2018 10:59 AM, Auger Eric wrote:
> Hi Suzuki,
> 
> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>> So far we have restricted the IPA size of the VM to the default
>> value (40bits). Now that we can manage the IPA size per VM and
>> support dynamic stage2 page tables, we can allow VMs to have
>> larger IPA. This patch introduces a the maximum IPA size
>> supported on the host.
> to be reworded
>   This is decided by the following factors :

Sure


>> diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c
>> index 51ecf0f7c912..76972b19bdd7 100644
>> --- a/arch/arm64/kvm/reset.c
>> +++ b/arch/arm64/kvm/reset.c
>> @@ -34,6 +34,9 @@
>>   #include <asm/kvm_coproc.h>
>>   #include <asm/kvm_mmu.h>
>>   
>> +/* Maximum phys_shift supported for any VM on this host */
>> +static u32 kvm_ipa_limit;
>> +
>>   /*
>>    * ARMv8 Reset Values
>>    */
>> @@ -135,6 +138,46 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu)
>>   	return kvm_timer_vcpu_reset(vcpu);
>>   }
>>   
>> +void kvm_set_ipa_limit(void)
>> +{
>> +	unsigned int ipa_max, va_max, parange;
>> +
>> +	parange = read_sanitised_ftr_reg(SYS_ID_AA64MMFR0_EL1) & 0x7;
>> +	ipa_max = id_aa64mmfr0_parange_to_phys_shift(parange);
>> +
>> +	/* Raise the limit to the default size for backward compatibility */
>> +	if (ipa_max < KVM_PHYS_SHIFT) {
>> +		WARN_ONCE(1,
>> +			  "PARange is %d bits, unsupported configuration!",
>> +			  ipa_max);
>> +		ipa_max = KVM_PHYS_SHIFT;
> I don't really get what does happen in this case. The CPU cannot handle
> PA up to ipa_max so can the VM run properly? In case it is a
> showstopper, kvm_set_ipa_limit should return an error, cascaded by
> init_common_resources. Otherwise the warning message may be reworded.

I think this was a warning added to warn against the older
Foundation model which had a 36bit PA size. So the VTCR was progammed
with a 36bit limit, while the KVM guest was allowed to create 40bit
IPA space, though it wouldn't fly well if someone tried to.

With this series, I think we may expose the real IPA_MAX (which could
be < 40bit) and warn the user if someone tried to create a VM with
40bit IPA (vm_type == 0) and let the call succeed (for the sake of ABI).

Marc, Christoffer, Eric

Thoughts ?

>> +	}
>> +
>> +	/* Clamp it to the PA size supported by the kernel */
>> +	ipa_max = (ipa_max > PHYS_MASK_SHIFT) ? PHYS_MASK_SHIFT : ipa_max;
>> +	/*
>> +	 * Since our stage2 table is dependent on the stage1 page table code,
>> +	 * we must always honor the following condition:
>> +	 *
>> +	 *  Number of levels in Stage1 >= Number of levels in Stage2.
>> +	 *
>> +	 * So clamp the ipa limit further down to limit the number of levels.
>> +	 * Since we can concatenate upto 16 tables at entry level, we could
>> +	 * go upto 4bits above the maximum VA addressible with the current
> addressable?

Sure

>> +	 * number of levels.
>> +	 */
>> +	va_max = PGDIR_SHIFT + PAGE_SHIFT - 3;
>> +	va_max += 4;
>> +
>> +	if (va_max < ipa_max) {
>> +		kvm_info("Limiting IPA limit to %dbytes due to host VA bits limitation\n",
>> +			 va_max);
>> +		ipa_max = va_max;
> you have a trace for this limitation but none for the comparison against
> PHYS_MASK_SHIFT.

May be I could add a message which only mentions what is the limiting
factor kernel VA vs kernel PA support

>> +	}
>> +
>> +	kvm_ipa_limit = ipa_max;
>> +}
>> +
>>   /*
>>    * Configure the VTCR_EL2 for this VM. The VTCR value is common
>>    * across all the physical CPUs on the system. We use system wide
>> diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
>> index 43e716bc3f08..631f9a3ad99a 100644
>> --- a/virt/kvm/arm/arm.c
>> +++ b/virt/kvm/arm/arm.c
>> @@ -1413,6 +1413,8 @@ static int init_common_resources(void)
>>   	kvm_vmid_bits = kvm_get_vmid_bits();
>>   	kvm_info("%d-bit VMID\n", kvm_vmid_bits);
>>   
>> +	kvm_set_ipa_limit();
> As we have a kvm_info for the supported vmid_bits, may be good to output
> the max IPA size supported by the host whatever the applied clamps?

Sure, will do that.

Thanks
Suzuki

^ permalink raw reply	[flat|nested] 50+ messages in thread

* Re: [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask
  2018-09-20 15:22     ` Suzuki K Poulose
@ 2018-09-25 11:56       ` Auger Eric
  0 siblings, 0 replies; 50+ messages in thread
From: Auger Eric @ 2018-09-25 11:56 UTC (permalink / raw)
  To: Suzuki K Poulose, linux-arm-kernel
  Cc: kvmarm, kvm, marc.zyngier, cdall, pbonzini, rkrcmar, will.deacon,
	catalin.marinas, james.morse, dave.martin, julien.grall,
	linux-kernel

Hi Suzuki,

On 9/20/18 5:22 PM, Suzuki K Poulose wrote:
> 
> 
> On 20/09/18 15:07, Auger Eric wrote:
>> Hi Suzuki,
>> On 9/17/18 12:41 PM, Suzuki K Poulose wrote:
>>> On arm64 VTTBR_EL2:BADDR holds the base address for the stage2
>>> translation table. The Arm ARM mandates that the bits BADDR[x-1:0]
>>> should be 0, where 'x' is defined for a given IPA Size and the
>>> number of levels for a translation granule size. It is defined
>>> using some magical constants. This patch is a reverse engineered
>>> implementation to calculate the 'x' at runtime for a given ipa and
>>> number of page table levels. See patch for more details.
>>>
>>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>>> Cc: Christoffer Dall <cdall@kernel.org>
>>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
>>
>>> ---
>>> Changes since V3:
>>>   - Update reference to latest ARM ARM and improve commentary
>>> ---
>>>   arch/arm64/include/asm/kvm_arm.h | 63 +++++++++++++++++++++++++++++---
>>>   arch/arm64/include/asm/kvm_mmu.h | 25 ++++++++++++-
>>>   2 files changed, 81 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/arch/arm64/include/asm/kvm_arm.h
>>> b/arch/arm64/include/asm/kvm_arm.h
>>> index 14317b3a1820..3fb1d440be6e 100644
>>> --- a/arch/arm64/include/asm/kvm_arm.h
>>> +++ b/arch/arm64/include/asm/kvm_arm.h
>>> @@ -123,7 +123,6 @@
>>>   #define VTCR_EL2_SL0_MASK  (3 << VTCR_EL2_SL0_SHIFT)
>>>   #define VTCR_EL2_SL0_LVL1  (1 << VTCR_EL2_SL0_SHIFT)
>>>   #define VTCR_EL2_T0SZ_MASK 0x3f
>>> -#define VTCR_EL2_T0SZ_40B   24
>>>   #define VTCR_EL2_VS_SHIFT  19
>>>   #define VTCR_EL2_VS_8BIT   (0 << VTCR_EL2_VS_SHIFT)
>>>   #define VTCR_EL2_VS_16BIT  (1 << VTCR_EL2_VS_SHIFT)
>>> @@ -140,11 +139,8 @@
>>>    * Note that when using 4K pages, we concatenate two first level
>>> page tables
>>>    * together. With 16K pages, we concatenate 16 first level page
>>> tables.
>>>    *
>>> - * The magic numbers used for VTTBR_X in this patch can be found in
>>> Tables
>>> - * D4-23 and D4-25 in ARM DDI 0487A.b.
>>>    */
>>>
>>> -#define VTCR_EL2_T0SZ_IPA   VTCR_EL2_T0SZ_40B
>>>   #define VTCR_EL2_COMMON_BITS       (VTCR_EL2_SH0_INNER |
>>> VTCR_EL2_ORGN0_WBWA | \
>>>                               VTCR_EL2_IRGN0_WBWA | VTCR_EL2_RES1)
>>>
>>> @@ -175,9 +171,64 @@
>>>   #endif
>>>
>>>   #define VTCR_EL2_FLAGS                     (VTCR_EL2_COMMON_BITS |
>>> VTCR_EL2_TGRAN_FLAGS)
>>> -#define VTTBR_X                             (VTTBR_X_TGRAN_MAGIC -
>>> VTCR_EL2_T0SZ_IPA)
>>> +/*
>>> + * ARM VMSAv8-64 defines an algorithm for finding the translation table
>>> + * descriptors in section D4.2.8 in ARM DDI 0487C.a.
>>> + *
>>> + * The algorithm defines the expectations on the BaseAddress (for
>>> the page
>>> + * table) bits resolved at each level based on the page size, entry
>>> level
>>> + * and T0SZ. The variable "x" in the algorithm also affects the
>>> VTTBR:BADDR
>>> + * for stage2 page table.
>>> + *
>>> + * The value of "x" is calculated as :
>>> + *  x = Magic_N - T0SZ
>>
>> What is not crystal clear to me is the "if SL0b,c = n" case where x get
>> a value not based on Magic_N. Please could you explain why it is not
>> relevant?
> 
> We only care about the "x" for the "entry" level of the table look up
> to make sure that the VTTBR is physical address meets the required
> alignment. In both cases, if SL0 b,c == n, x is (PAGE_SHIFT) iff the
> level you are looking at is not the "entry level". So this should always
> be page aligned, like any intermediate level table.

Oh OK I get it now.
> 
> The Magic value is needed only needed for the "entry" level due to the
> fact that we may have lesser bits to resolve (i.e, depending on your
> PAMax or in other words T0SZ) than the intermediate levels (where we
> always resolve {PAGE_SHIFT - 3} bits. This is further complicated by the
> fact that Stage2 could use different number of levels for a given T0SZ
> than the stage1.
> I acknowledge that the algorithm is a bit too cryptic and I spent quite
> sometime decode it to the formula we use below ;-).
> 
> I could update the comment to :
> 
> /*
>  * ARM VMSAv8-64 defines an algorithm for finding the translation table
>  * descriptors in section D4.2.8 in ARM DDI 0487C.a.
>  *
>  * The algorithm defines the expectations on the translation table
>  * addresses for each level, based on PAGE_SIZE, entry level
>  * and the translation table size (T0SZ). The variable "x" in the
>  * algorithm determines the alignment of a table base address at a given
>  * level and thus determines the alignment of VTTBR:BADDR for stage2
>  * page table entry level.
>  * Since the number of bits resolved at the entry level could vary
>  * depending on the T0SZ, the value of "x" is defined based on a
>  * Magic constant for a given PAGE_SIZE and Entry Level. The
>  * intermediate levels must be always aligned to the PAGE_SIZE (i.e,
>  * x = PAGE_SHIFT).
>  *
>  * The value of "x" for entry level is calculated as :
>  *     x = Magic_N - T0SZ
>  *
Looks OK.

Thank you for the explanation.

Eric
> 
> ...
> 
> Suzuki
> IMPORTANT NOTICE: The contents of this email and any attachments are
> confidential and may also be privileged. If you are not the intended
> recipient, please notify the sender immediately and do not disclose the
> contents to any other person, use it for any purpose, or store or copy
> the information in any medium. Thank you.

^ permalink raw reply	[flat|nested] 50+ messages in thread

end of thread, other threads:[~2018-09-25 11:56 UTC | newest]

Thread overview: 50+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-09-17 10:41 [PATCH v5 00/18] kvm: arm64: Dynamic IPA and 52bit IPA Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 01/18] kvm: arm/arm64: Fix stage2_flush_memslot for 4 level page table Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 02/18] kvm: arm/arm64: Remove spurious WARN_ON Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 03/18] kvm: arm64: Add helper for loading the stage2 setting for a VM Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 04/18] arm64: Add a helper for PARange to physical shift conversion Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 05/18] kvm: arm64: Clean up VTCR_EL2 initialisation Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 06/18] kvm: arm/arm64: Allow arch specific configurations for VM Suzuki K Poulose
2018-09-20 10:22   ` Auger Eric
2018-09-17 10:41 ` [PATCH v5 07/18] kvm: arm64: Configure VTCR_EL2 per VM Suzuki K Poulose
2018-09-20 10:21   ` Auger Eric
2018-09-20 10:38     ` Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 08/18] kvm: arm/arm64: Prepare for VM specific stage2 translations Suzuki K Poulose
2018-09-20 14:07   ` Auger Eric
2018-09-17 10:41 ` [PATCH v5 09/18] kvm: arm64: Prepare for dynamic stage2 page table layout Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 10/18] kvm: arm64: Make stage2 page table layout dynamic Suzuki K Poulose
2018-09-20 14:07   ` Auger Eric
2018-09-17 10:41 ` [PATCH v5 11/18] kvm: arm64: Dynamic configuration of VTTBR mask Suzuki K Poulose
2018-09-20 14:07   ` Auger Eric
2018-09-20 15:22     ` Suzuki K Poulose
2018-09-25 11:56       ` Auger Eric
2018-09-17 10:41 ` [PATCH v5 12/18] kvm: arm64: Configure VTCR_EL2.SL0 per VM Suzuki K Poulose
2018-09-20 14:25   ` Auger Eric
2018-09-20 15:25     ` Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 13/18] kvm: arm64: Switch to per VM IPA limit Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 14/18] vgic: Add support for 52bit guest physical address Suzuki K Poulose
2018-09-21 14:57   ` Auger Eric
2018-09-25 10:49     ` Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 15/18] kvm: arm64: Add 52bit support for PAR to HPFAR conversoin Suzuki K Poulose
2018-09-25  9:59   ` Auger Eric
2018-09-17 10:41 ` [PATCH v5 16/18] kvm: arm64: Set a limit on the IPA size Suzuki K Poulose
2018-09-25  9:59   ` Auger Eric
2018-09-25 11:10     ` Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 17/18] kvm: arm64: Limit the minimum number of page table levels Suzuki K Poulose
2018-09-25 10:00   ` Auger Eric
2018-09-25 10:25     ` Suzuki K Poulose
2018-09-17 10:41 ` [PATCH v5 18/18] kvm: arm64: Allow tuning the physical address size for VM Suzuki K Poulose
2018-09-17 14:20   ` Peter Maydell
2018-09-17 14:43     ` Suzuki K Poulose
2018-09-18  1:55   ` Peter Maydell
2018-09-18 15:16     ` Suzuki K Poulose
2018-09-18 15:36       ` Peter Maydell
2018-09-18 16:27         ` Suzuki K Poulose
2018-09-18 17:15           ` Peter Maydell
2018-09-19 10:03             ` Suzuki K Poulose
2018-09-25 10:00   ` Auger Eric
2018-09-25 10:24     ` Suzuki K Poulose
2018-09-17 10:41 ` [kvmtool PATCH v5 19/18] kvmtool: Allow backends to run checks on the KVM device fd Suzuki K Poulose
2018-09-17 10:41 ` [kvmtool PATCH v5 20/18] kvmtool: arm64: Add support for guest physical address size Suzuki K Poulose
2018-09-17 10:41 ` [kvmtool PATCH v5 21/18] kvmtool: arm64: Switch memory layout Suzuki K Poulose
2018-09-17 10:41 ` [kvmtool PATCH v5 22/18] kvmtool: arm: Add support for creating VM with PA size Suzuki K Poulose

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).