* [GIT PULL] KVM/arm updates for Linux v5.1
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
Paolo, Radim,
This is the KVM/arm updates for Linux v5.1, much smaller than expected, as
most of the stuff we were hoping to queue missed the merge window. Oh well,
5.2 will be "interesting".
This time, we have a bunch of changes that prepare the landing of the some
of the nested virt code, some optimisations on the timer emulation (direct
assignment of the physical timer on VHE systems), some better type-safety
checks for the HYP code, Set/Way optimisations for silly 32bit guests, and a
bunch of cleanups here and there. Oh, and a MAINTAINERS update to officially
mention that we have a set of dedicated reviewers working with us!
Note that the first patch hits code that is outside of the KVM tree, but
that Daniel has agreed to us taking directly given that it was required for
a lot of other things in this drop.
Please pull,
M.
The following changes since commit 49a57857aeea06ca831043acbb0fa5e0f50602fd:
Linux 5.0-rc3 (2019-01-21 13:14:44 +1300)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git tags/kvmarm-for-v5.1
for you to fetch changes up to c88b093693ccbe41991ef2e9b1d251945e6e54ed:
arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2 (2019-02-22 09:59:20 +0000)
----------------------------------------------------------------
KVM/arm updates for Linux v5.1
- A number of pre-nested code rework
- Direct physical timer assignment on VHE systems
- kvm_call_hyp type safety enforcement
- Set/Way cache sanitisation for 32bit guests
- Build system cleanups
- A bunch of janitorial fixes
----------------------------------------------------------------
Andre Przywara (2):
clocksource/arm_arch_timer: Store physical timer IRQ number for KVM on VHE
KVM: arm/arm64: consolidate arch timer trap handlers
Ard Biesheuvel (2):
arm64: KVM: Expose sanitised cache type register to guest
arm64: KVM: Describe data or unified caches as having 1 set and 1 way
Christoffer Dall (6):
KVM: arm/arm64: Factor out VMID into struct kvm_vmid
KVM: arm/arm64: Simplify bg_timer programming
KVM: arm/arm64: timer: Rework data structures for multiple timers
KVM: arm/arm64: arch_timer: Assign the phys timer on VHE systems
KVM: arm/arm64: Rework the timer code to use a timer_map
KVM: arm/arm64: Move kvm_is_write_fault to header file
Colin Ian King (1):
KVM: arm/arm64: fix spelling mistake: "auxilary" -> "auxiliary"
Dave Martin (1):
arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2
Marc Zyngier (9):
arm/arm64: KVM: Introduce kvm_call_hyp_ret()
arm64: KVM: Allow for direct call of HYP functions when using VHE
arm64: KVM: Drop VHE-specific HYP call stub
ARM: KVM: Teach some form of type-safety to kvm_call_hyp
arm/arm64: KVM: Statically configure the host's view of MPIDR
KVM: arm64: Fix ICH_ELRSR_EL2 sysreg naming
KVM: arm64: Reuse sys_reg() macro when searching the trap table
KVM: arm/arm64: arch_timer: Mark physical interrupt active when a virtual interrupt is pending
KVM: arm/arm64: Update MAINTAINERS entries
Masahiro Yamada (3):
KVM: arm/arm64: Fix TRACE_INCLUDE_PATH
KVM: arm/arm64: Remove -I. header search paths
KVM: arm/arm64: Prefix header search paths with $(srctree)/
Shaokun Zhang (2):
KVM: arm/arm64: Remove unused gpa_end variable
KVM: arm/arm64: Remove unused timer variable
Zenghui Yu (1):
KVM: arm64: Fix comment for KVM_PHYS_SHIFT
MAINTAINERS | 18 +-
arch/arm/include/asm/arch_gicv3.h | 4 +-
arch/arm/include/asm/kvm_emulate.h | 8 +
arch/arm/include/asm/kvm_host.h | 53 ++-
arch/arm/include/asm/kvm_hyp.h | 4 +
arch/arm/include/asm/kvm_mmu.h | 9 +-
arch/arm/kvm/Makefile | 5 +-
arch/arm/kvm/coproc.c | 23 +-
arch/arm/kvm/hyp/cp15-sr.c | 1 -
arch/arm/kvm/hyp/hyp-entry.S | 2 +-
arch/arm/kvm/hyp/switch.c | 2 +-
arch/arm/kvm/hyp/tlb.c | 4 +-
arch/arm/kvm/interrupts.S | 4 +-
arch/arm64/include/asm/kvm_emulate.h | 12 +
arch/arm64/include/asm/kvm_host.h | 48 ++-
arch/arm64/include/asm/kvm_hyp.h | 3 +-
arch/arm64/include/asm/kvm_mmu.h | 13 +-
arch/arm64/include/asm/sysreg.h | 7 +-
arch/arm64/kvm/Makefile | 4 +-
arch/arm64/kvm/debug.c | 2 +-
arch/arm64/kvm/hyp.S | 3 -
arch/arm64/kvm/hyp/hyp-entry.S | 12 -
arch/arm64/kvm/hyp/sysreg-sr.c | 1 -
arch/arm64/kvm/sys_regs.c | 168 ++++++----
drivers/clocksource/arm_arch_timer.c | 11 +-
include/clocksource/arm_arch_timer.h | 1 +
include/kvm/arm_arch_timer.h | 68 ++--
virt/kvm/arm/arch_timer.c | 608 ++++++++++++++++++++++++-----------
virt/kvm/arm/arm.c | 64 ++--
virt/kvm/arm/hyp/vgic-v3-sr.c | 2 +-
virt/kvm/arm/mmu.c | 18 +-
virt/kvm/arm/trace.h | 107 +++++-
virt/kvm/arm/vgic/vgic-v3.c | 4 +-
33 files changed, 909 insertions(+), 384 deletions(-)
^ permalink raw reply [flat|nested] 58+ messages in thread
* [GIT PULL] KVM/arm updates for Linux v5.1
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
Paolo, Radim,
This is the KVM/arm updates for Linux v5.1, much smaller than expected, as
most of the stuff we were hoping to queue missed the merge window. Oh well,
5.2 will be "interesting".
This time, we have a bunch of changes that prepare the landing of the some
of the nested virt code, some optimisations on the timer emulation (direct
assignment of the physical timer on VHE systems), some better type-safety
checks for the HYP code, Set/Way optimisations for silly 32bit guests, and a
bunch of cleanups here and there. Oh, and a MAINTAINERS update to officially
mention that we have a set of dedicated reviewers working with us!
Note that the first patch hits code that is outside of the KVM tree, but
that Daniel has agreed to us taking directly given that it was required for
a lot of other things in this drop.
Please pull,
M.
The following changes since commit 49a57857aeea06ca831043acbb0fa5e0f50602fd:
Linux 5.0-rc3 (2019-01-21 13:14:44 +1300)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git tags/kvmarm-for-v5.1
for you to fetch changes up to c88b093693ccbe41991ef2e9b1d251945e6e54ed:
arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2 (2019-02-22 09:59:20 +0000)
----------------------------------------------------------------
KVM/arm updates for Linux v5.1
- A number of pre-nested code rework
- Direct physical timer assignment on VHE systems
- kvm_call_hyp type safety enforcement
- Set/Way cache sanitisation for 32bit guests
- Build system cleanups
- A bunch of janitorial fixes
----------------------------------------------------------------
Andre Przywara (2):
clocksource/arm_arch_timer: Store physical timer IRQ number for KVM on VHE
KVM: arm/arm64: consolidate arch timer trap handlers
Ard Biesheuvel (2):
arm64: KVM: Expose sanitised cache type register to guest
arm64: KVM: Describe data or unified caches as having 1 set and 1 way
Christoffer Dall (6):
KVM: arm/arm64: Factor out VMID into struct kvm_vmid
KVM: arm/arm64: Simplify bg_timer programming
KVM: arm/arm64: timer: Rework data structures for multiple timers
KVM: arm/arm64: arch_timer: Assign the phys timer on VHE systems
KVM: arm/arm64: Rework the timer code to use a timer_map
KVM: arm/arm64: Move kvm_is_write_fault to header file
Colin Ian King (1):
KVM: arm/arm64: fix spelling mistake: "auxilary" -> "auxiliary"
Dave Martin (1):
arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2
Marc Zyngier (9):
arm/arm64: KVM: Introduce kvm_call_hyp_ret()
arm64: KVM: Allow for direct call of HYP functions when using VHE
arm64: KVM: Drop VHE-specific HYP call stub
ARM: KVM: Teach some form of type-safety to kvm_call_hyp
arm/arm64: KVM: Statically configure the host's view of MPIDR
KVM: arm64: Fix ICH_ELRSR_EL2 sysreg naming
KVM: arm64: Reuse sys_reg() macro when searching the trap table
KVM: arm/arm64: arch_timer: Mark physical interrupt active when a virtual interrupt is pending
KVM: arm/arm64: Update MAINTAINERS entries
Masahiro Yamada (3):
KVM: arm/arm64: Fix TRACE_INCLUDE_PATH
KVM: arm/arm64: Remove -I. header search paths
KVM: arm/arm64: Prefix header search paths with $(srctree)/
Shaokun Zhang (2):
KVM: arm/arm64: Remove unused gpa_end variable
KVM: arm/arm64: Remove unused timer variable
Zenghui Yu (1):
KVM: arm64: Fix comment for KVM_PHYS_SHIFT
MAINTAINERS | 18 +-
arch/arm/include/asm/arch_gicv3.h | 4 +-
arch/arm/include/asm/kvm_emulate.h | 8 +
arch/arm/include/asm/kvm_host.h | 53 ++-
arch/arm/include/asm/kvm_hyp.h | 4 +
arch/arm/include/asm/kvm_mmu.h | 9 +-
arch/arm/kvm/Makefile | 5 +-
arch/arm/kvm/coproc.c | 23 +-
arch/arm/kvm/hyp/cp15-sr.c | 1 -
arch/arm/kvm/hyp/hyp-entry.S | 2 +-
arch/arm/kvm/hyp/switch.c | 2 +-
arch/arm/kvm/hyp/tlb.c | 4 +-
arch/arm/kvm/interrupts.S | 4 +-
arch/arm64/include/asm/kvm_emulate.h | 12 +
arch/arm64/include/asm/kvm_host.h | 48 ++-
arch/arm64/include/asm/kvm_hyp.h | 3 +-
arch/arm64/include/asm/kvm_mmu.h | 13 +-
arch/arm64/include/asm/sysreg.h | 7 +-
arch/arm64/kvm/Makefile | 4 +-
arch/arm64/kvm/debug.c | 2 +-
arch/arm64/kvm/hyp.S | 3 -
arch/arm64/kvm/hyp/hyp-entry.S | 12 -
arch/arm64/kvm/hyp/sysreg-sr.c | 1 -
arch/arm64/kvm/sys_regs.c | 168 ++++++----
drivers/clocksource/arm_arch_timer.c | 11 +-
include/clocksource/arm_arch_timer.h | 1 +
include/kvm/arm_arch_timer.h | 68 ++--
virt/kvm/arm/arch_timer.c | 608 ++++++++++++++++++++++++-----------
virt/kvm/arm/arm.c | 64 ++--
virt/kvm/arm/hyp/vgic-v3-sr.c | 2 +-
virt/kvm/arm/mmu.c | 18 +-
virt/kvm/arm/trace.h | 107 +++++-
virt/kvm/arm/vgic/vgic-v3.c | 4 +-
33 files changed, 909 insertions(+), 384 deletions(-)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 58+ messages in thread
* [PATCH 01/27] clocksource/arm_arch_timer: Store physical timer IRQ number for KVM on VHE
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Andre Przywara <andre.przywara@arm.com>
A host running in VHE mode gets the EL2 physical timer as its time
source (accessed using the EL1 sysreg accessors, which get re-directed
to the EL2 sysregs by VHE).
The EL1 physical timer remains unused by the host kernel, allowing us to
pass that on directly to a KVM guest and saves us from emulating this
timer for the guest on VHE systems.
Store the EL1 Physical Timer's IRQ number in
struct arch_timer_kvm_info on VHE systems to allow KVM to use it.
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
drivers/clocksource/arm_arch_timer.c | 11 +++++++++--
include/clocksource/arm_arch_timer.h | 1 +
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
index 9a7d4dc00b6e..b9243e2328b4 100644
--- a/drivers/clocksource/arm_arch_timer.c
+++ b/drivers/clocksource/arm_arch_timer.c
@@ -1206,6 +1206,13 @@ static enum arch_timer_ppi_nr __init arch_timer_select_ppi(void)
return ARCH_TIMER_PHYS_SECURE_PPI;
}
+static void __init arch_timer_populate_kvm_info(void)
+{
+ arch_timer_kvm_info.virtual_irq = arch_timer_ppi[ARCH_TIMER_VIRT_PPI];
+ if (is_kernel_in_hyp_mode())
+ arch_timer_kvm_info.physical_irq = arch_timer_ppi[ARCH_TIMER_PHYS_NONSECURE_PPI];
+}
+
static int __init arch_timer_of_init(struct device_node *np)
{
int i, ret;
@@ -1220,7 +1227,7 @@ static int __init arch_timer_of_init(struct device_node *np)
for (i = ARCH_TIMER_PHYS_SECURE_PPI; i < ARCH_TIMER_MAX_TIMER_PPI; i++)
arch_timer_ppi[i] = irq_of_parse_and_map(np, i);
- arch_timer_kvm_info.virtual_irq = arch_timer_ppi[ARCH_TIMER_VIRT_PPI];
+ arch_timer_populate_kvm_info();
rate = arch_timer_get_cntfrq();
arch_timer_of_configure_rate(rate, np);
@@ -1550,7 +1557,7 @@ static int __init arch_timer_acpi_init(struct acpi_table_header *table)
arch_timer_ppi[ARCH_TIMER_HYP_PPI] =
acpi_gtdt_map_ppi(ARCH_TIMER_HYP_PPI);
- arch_timer_kvm_info.virtual_irq = arch_timer_ppi[ARCH_TIMER_VIRT_PPI];
+ arch_timer_populate_kvm_info();
/*
* When probing via ACPI, we have no mechanism to override the sysreg
diff --git a/include/clocksource/arm_arch_timer.h b/include/clocksource/arm_arch_timer.h
index 349e5957c949..702967d996bb 100644
--- a/include/clocksource/arm_arch_timer.h
+++ b/include/clocksource/arm_arch_timer.h
@@ -74,6 +74,7 @@ enum arch_timer_spi_nr {
struct arch_timer_kvm_info {
struct timecounter timecounter;
int virtual_irq;
+ int physical_irq;
};
struct arch_timer_mem_frame {
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 01/27] clocksource/arm_arch_timer: Store physical timer IRQ number for KVM on VHE
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Andre Przywara <andre.przywara@arm.com>
A host running in VHE mode gets the EL2 physical timer as its time
source (accessed using the EL1 sysreg accessors, which get re-directed
to the EL2 sysregs by VHE).
The EL1 physical timer remains unused by the host kernel, allowing us to
pass that on directly to a KVM guest and saves us from emulating this
timer for the guest on VHE systems.
Store the EL1 Physical Timer's IRQ number in
struct arch_timer_kvm_info on VHE systems to allow KVM to use it.
Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
drivers/clocksource/arm_arch_timer.c | 11 +++++++++--
include/clocksource/arm_arch_timer.h | 1 +
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
index 9a7d4dc00b6e..b9243e2328b4 100644
--- a/drivers/clocksource/arm_arch_timer.c
+++ b/drivers/clocksource/arm_arch_timer.c
@@ -1206,6 +1206,13 @@ static enum arch_timer_ppi_nr __init arch_timer_select_ppi(void)
return ARCH_TIMER_PHYS_SECURE_PPI;
}
+static void __init arch_timer_populate_kvm_info(void)
+{
+ arch_timer_kvm_info.virtual_irq = arch_timer_ppi[ARCH_TIMER_VIRT_PPI];
+ if (is_kernel_in_hyp_mode())
+ arch_timer_kvm_info.physical_irq = arch_timer_ppi[ARCH_TIMER_PHYS_NONSECURE_PPI];
+}
+
static int __init arch_timer_of_init(struct device_node *np)
{
int i, ret;
@@ -1220,7 +1227,7 @@ static int __init arch_timer_of_init(struct device_node *np)
for (i = ARCH_TIMER_PHYS_SECURE_PPI; i < ARCH_TIMER_MAX_TIMER_PPI; i++)
arch_timer_ppi[i] = irq_of_parse_and_map(np, i);
- arch_timer_kvm_info.virtual_irq = arch_timer_ppi[ARCH_TIMER_VIRT_PPI];
+ arch_timer_populate_kvm_info();
rate = arch_timer_get_cntfrq();
arch_timer_of_configure_rate(rate, np);
@@ -1550,7 +1557,7 @@ static int __init arch_timer_acpi_init(struct acpi_table_header *table)
arch_timer_ppi[ARCH_TIMER_HYP_PPI] =
acpi_gtdt_map_ppi(ARCH_TIMER_HYP_PPI);
- arch_timer_kvm_info.virtual_irq = arch_timer_ppi[ARCH_TIMER_VIRT_PPI];
+ arch_timer_populate_kvm_info();
/*
* When probing via ACPI, we have no mechanism to override the sysreg
diff --git a/include/clocksource/arm_arch_timer.h b/include/clocksource/arm_arch_timer.h
index 349e5957c949..702967d996bb 100644
--- a/include/clocksource/arm_arch_timer.h
+++ b/include/clocksource/arm_arch_timer.h
@@ -74,6 +74,7 @@ enum arch_timer_spi_nr {
struct arch_timer_kvm_info {
struct timecounter timecounter;
int virtual_irq;
+ int physical_irq;
};
struct arch_timer_mem_frame {
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 02/27] arm/arm64: KVM: Introduce kvm_call_hyp_ret()
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
Until now, we haven't differentiated between HYP calls that
have a return value and those who don't. As we're about to
change this, introduce kvm_call_hyp_ret(), and change all
call sites that actually make use of a return value.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_host.h | 3 +++
arch/arm64/include/asm/kvm_host.h | 1 +
arch/arm64/kvm/debug.c | 2 +-
virt/kvm/arm/arm.c | 2 +-
virt/kvm/arm/vgic/vgic-v3.c | 4 ++--
5 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index ca56537b61bc..023c9f2b1eea 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -214,7 +214,10 @@ unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
+
unsigned long kvm_call_hyp(void *hypfn, ...);
+#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
+
void force_vm_exit(const cpumask_t *mask);
int __kvm_arm_vcpu_get_events(struct kvm_vcpu *vcpu,
struct kvm_vcpu_events *events);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 7732d0ba4e60..e54cb7c88a4e 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -371,6 +371,7 @@ void kvm_arm_resume_guest(struct kvm *kvm);
u64 __kvm_call_hyp(void *hypfn, ...);
#define kvm_call_hyp(f, ...) __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__)
+#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
void force_vm_exit(const cpumask_t *mask);
void kvm_mmu_wp_memory_region(struct kvm *kvm, int slot);
diff --git a/arch/arm64/kvm/debug.c b/arch/arm64/kvm/debug.c
index f39801e4136c..fd917d6d12af 100644
--- a/arch/arm64/kvm/debug.c
+++ b/arch/arm64/kvm/debug.c
@@ -76,7 +76,7 @@ static void restore_guest_debug_regs(struct kvm_vcpu *vcpu)
void kvm_arm_init_debug(void)
{
- __this_cpu_write(mdcr_el2, kvm_call_hyp(__kvm_get_mdcr_el2));
+ __this_cpu_write(mdcr_el2, kvm_call_hyp_ret(__kvm_get_mdcr_el2));
}
/**
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 9e350fd34504..4d55f98f97f7 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -765,7 +765,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
ret = kvm_vcpu_run_vhe(vcpu);
kvm_arm_vhe_guest_exit();
} else {
- ret = kvm_call_hyp(__kvm_vcpu_run_nvhe, vcpu);
+ ret = kvm_call_hyp_ret(__kvm_vcpu_run_nvhe, vcpu);
}
vcpu->mode = OUTSIDE_GUEST_MODE;
diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c
index 9c0dd234ebe8..67f98151c88d 100644
--- a/virt/kvm/arm/vgic/vgic-v3.c
+++ b/virt/kvm/arm/vgic/vgic-v3.c
@@ -589,7 +589,7 @@ early_param("kvm-arm.vgic_v4_enable", early_gicv4_enable);
*/
int vgic_v3_probe(const struct gic_kvm_info *info)
{
- u32 ich_vtr_el2 = kvm_call_hyp(__vgic_v3_get_ich_vtr_el2);
+ u32 ich_vtr_el2 = kvm_call_hyp_ret(__vgic_v3_get_ich_vtr_el2);
int ret;
/*
@@ -679,7 +679,7 @@ void vgic_v3_put(struct kvm_vcpu *vcpu)
struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
if (likely(cpu_if->vgic_sre))
- cpu_if->vgic_vmcr = kvm_call_hyp(__vgic_v3_read_vmcr);
+ cpu_if->vgic_vmcr = kvm_call_hyp_ret(__vgic_v3_read_vmcr);
kvm_call_hyp(__vgic_v3_save_aprs, vcpu);
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 02/27] arm/arm64: KVM: Introduce kvm_call_hyp_ret()
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
Until now, we haven't differentiated between HYP calls that
have a return value and those who don't. As we're about to
change this, introduce kvm_call_hyp_ret(), and change all
call sites that actually make use of a return value.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_host.h | 3 +++
arch/arm64/include/asm/kvm_host.h | 1 +
arch/arm64/kvm/debug.c | 2 +-
virt/kvm/arm/arm.c | 2 +-
virt/kvm/arm/vgic/vgic-v3.c | 4 ++--
5 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index ca56537b61bc..023c9f2b1eea 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -214,7 +214,10 @@ unsigned long kvm_arm_num_regs(struct kvm_vcpu *vcpu);
int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
+
unsigned long kvm_call_hyp(void *hypfn, ...);
+#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
+
void force_vm_exit(const cpumask_t *mask);
int __kvm_arm_vcpu_get_events(struct kvm_vcpu *vcpu,
struct kvm_vcpu_events *events);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 7732d0ba4e60..e54cb7c88a4e 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -371,6 +371,7 @@ void kvm_arm_resume_guest(struct kvm *kvm);
u64 __kvm_call_hyp(void *hypfn, ...);
#define kvm_call_hyp(f, ...) __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__)
+#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
void force_vm_exit(const cpumask_t *mask);
void kvm_mmu_wp_memory_region(struct kvm *kvm, int slot);
diff --git a/arch/arm64/kvm/debug.c b/arch/arm64/kvm/debug.c
index f39801e4136c..fd917d6d12af 100644
--- a/arch/arm64/kvm/debug.c
+++ b/arch/arm64/kvm/debug.c
@@ -76,7 +76,7 @@ static void restore_guest_debug_regs(struct kvm_vcpu *vcpu)
void kvm_arm_init_debug(void)
{
- __this_cpu_write(mdcr_el2, kvm_call_hyp(__kvm_get_mdcr_el2));
+ __this_cpu_write(mdcr_el2, kvm_call_hyp_ret(__kvm_get_mdcr_el2));
}
/**
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 9e350fd34504..4d55f98f97f7 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -765,7 +765,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
ret = kvm_vcpu_run_vhe(vcpu);
kvm_arm_vhe_guest_exit();
} else {
- ret = kvm_call_hyp(__kvm_vcpu_run_nvhe, vcpu);
+ ret = kvm_call_hyp_ret(__kvm_vcpu_run_nvhe, vcpu);
}
vcpu->mode = OUTSIDE_GUEST_MODE;
diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c
index 9c0dd234ebe8..67f98151c88d 100644
--- a/virt/kvm/arm/vgic/vgic-v3.c
+++ b/virt/kvm/arm/vgic/vgic-v3.c
@@ -589,7 +589,7 @@ early_param("kvm-arm.vgic_v4_enable", early_gicv4_enable);
*/
int vgic_v3_probe(const struct gic_kvm_info *info)
{
- u32 ich_vtr_el2 = kvm_call_hyp(__vgic_v3_get_ich_vtr_el2);
+ u32 ich_vtr_el2 = kvm_call_hyp_ret(__vgic_v3_get_ich_vtr_el2);
int ret;
/*
@@ -679,7 +679,7 @@ void vgic_v3_put(struct kvm_vcpu *vcpu)
struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
if (likely(cpu_if->vgic_sre))
- cpu_if->vgic_vmcr = kvm_call_hyp(__vgic_v3_read_vmcr);
+ cpu_if->vgic_vmcr = kvm_call_hyp_ret(__vgic_v3_read_vmcr);
kvm_call_hyp(__vgic_v3_save_aprs, vcpu);
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 03/27] arm64: KVM: Allow for direct call of HYP functions when using VHE
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
When running VHE, there is no need to jump via some stub to perform
a "HYP" function call, as there is a single address space.
Let's thus change kvm_call_hyp() and co to perform a direct call
in this case. Although this results in a bit of code expansion,
it allows the compiler to check for type compatibility, something
that we are missing so far.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm64/include/asm/kvm_host.h | 32 +++++++++++++++++++++++++++++--
1 file changed, 30 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index e54cb7c88a4e..8b7702bdb219 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -370,8 +370,36 @@ void kvm_arm_halt_guest(struct kvm *kvm);
void kvm_arm_resume_guest(struct kvm *kvm);
u64 __kvm_call_hyp(void *hypfn, ...);
-#define kvm_call_hyp(f, ...) __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__)
-#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
+
+/*
+ * The couple of isb() below are there to guarantee the same behaviour
+ * on VHE as on !VHE, where the eret to EL1 acts as a context
+ * synchronization event.
+ */
+#define kvm_call_hyp(f, ...) \
+ do { \
+ if (has_vhe()) { \
+ f(__VA_ARGS__); \
+ isb(); \
+ } else { \
+ __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__); \
+ } \
+ } while(0)
+
+#define kvm_call_hyp_ret(f, ...) \
+ ({ \
+ typeof(f(__VA_ARGS__)) ret; \
+ \
+ if (has_vhe()) { \
+ ret = f(__VA_ARGS__); \
+ isb(); \
+ } else { \
+ ret = __kvm_call_hyp(kvm_ksym_ref(f), \
+ ##__VA_ARGS__); \
+ } \
+ \
+ ret; \
+ })
void force_vm_exit(const cpumask_t *mask);
void kvm_mmu_wp_memory_region(struct kvm *kvm, int slot);
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 03/27] arm64: KVM: Allow for direct call of HYP functions when using VHE
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
When running VHE, there is no need to jump via some stub to perform
a "HYP" function call, as there is a single address space.
Let's thus change kvm_call_hyp() and co to perform a direct call
in this case. Although this results in a bit of code expansion,
it allows the compiler to check for type compatibility, something
that we are missing so far.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm64/include/asm/kvm_host.h | 32 +++++++++++++++++++++++++++++--
1 file changed, 30 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index e54cb7c88a4e..8b7702bdb219 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -370,8 +370,36 @@ void kvm_arm_halt_guest(struct kvm *kvm);
void kvm_arm_resume_guest(struct kvm *kvm);
u64 __kvm_call_hyp(void *hypfn, ...);
-#define kvm_call_hyp(f, ...) __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__)
-#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
+
+/*
+ * The couple of isb() below are there to guarantee the same behaviour
+ * on VHE as on !VHE, where the eret to EL1 acts as a context
+ * synchronization event.
+ */
+#define kvm_call_hyp(f, ...) \
+ do { \
+ if (has_vhe()) { \
+ f(__VA_ARGS__); \
+ isb(); \
+ } else { \
+ __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__); \
+ } \
+ } while(0)
+
+#define kvm_call_hyp_ret(f, ...) \
+ ({ \
+ typeof(f(__VA_ARGS__)) ret; \
+ \
+ if (has_vhe()) { \
+ ret = f(__VA_ARGS__); \
+ isb(); \
+ } else { \
+ ret = __kvm_call_hyp(kvm_ksym_ref(f), \
+ ##__VA_ARGS__); \
+ } \
+ \
+ ret; \
+ })
void force_vm_exit(const cpumask_t *mask);
void kvm_mmu_wp_memory_region(struct kvm *kvm, int slot);
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 04/27] arm64: KVM: Drop VHE-specific HYP call stub
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
We now call VHE code directly, without going through any central
dispatching function. Let's drop that code.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm64/kvm/hyp.S | 3 ---
arch/arm64/kvm/hyp/hyp-entry.S | 12 ------------
2 files changed, 15 deletions(-)
diff --git a/arch/arm64/kvm/hyp.S b/arch/arm64/kvm/hyp.S
index 952f6cb9cf72..2845aa680841 100644
--- a/arch/arm64/kvm/hyp.S
+++ b/arch/arm64/kvm/hyp.S
@@ -40,9 +40,6 @@
* arch/arm64/kernel/hyp_stub.S.
*/
ENTRY(__kvm_call_hyp)
-alternative_if_not ARM64_HAS_VIRT_HOST_EXTN
hvc #0
ret
-alternative_else_nop_endif
- b __vhe_hyp_call
ENDPROC(__kvm_call_hyp)
diff --git a/arch/arm64/kvm/hyp/hyp-entry.S b/arch/arm64/kvm/hyp/hyp-entry.S
index 73c1b483ec39..2b1e686772bf 100644
--- a/arch/arm64/kvm/hyp/hyp-entry.S
+++ b/arch/arm64/kvm/hyp/hyp-entry.S
@@ -43,18 +43,6 @@
ldr lr, [sp], #16
.endm
-ENTRY(__vhe_hyp_call)
- do_el2_call
- /*
- * We used to rely on having an exception return to get
- * an implicit isb. In the E2H case, we don't have it anymore.
- * rather than changing all the leaf functions, just do it here
- * before returning to the rest of the kernel.
- */
- isb
- ret
-ENDPROC(__vhe_hyp_call)
-
el1_sync: // Guest trapped into EL2
mrs x0, esr_el2
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 04/27] arm64: KVM: Drop VHE-specific HYP call stub
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
We now call VHE code directly, without going through any central
dispatching function. Let's drop that code.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm64/kvm/hyp.S | 3 ---
arch/arm64/kvm/hyp/hyp-entry.S | 12 ------------
2 files changed, 15 deletions(-)
diff --git a/arch/arm64/kvm/hyp.S b/arch/arm64/kvm/hyp.S
index 952f6cb9cf72..2845aa680841 100644
--- a/arch/arm64/kvm/hyp.S
+++ b/arch/arm64/kvm/hyp.S
@@ -40,9 +40,6 @@
* arch/arm64/kernel/hyp_stub.S.
*/
ENTRY(__kvm_call_hyp)
-alternative_if_not ARM64_HAS_VIRT_HOST_EXTN
hvc #0
ret
-alternative_else_nop_endif
- b __vhe_hyp_call
ENDPROC(__kvm_call_hyp)
diff --git a/arch/arm64/kvm/hyp/hyp-entry.S b/arch/arm64/kvm/hyp/hyp-entry.S
index 73c1b483ec39..2b1e686772bf 100644
--- a/arch/arm64/kvm/hyp/hyp-entry.S
+++ b/arch/arm64/kvm/hyp/hyp-entry.S
@@ -43,18 +43,6 @@
ldr lr, [sp], #16
.endm
-ENTRY(__vhe_hyp_call)
- do_el2_call
- /*
- * We used to rely on having an exception return to get
- * an implicit isb. In the E2H case, we don't have it anymore.
- * rather than changing all the leaf functions, just do it here
- * before returning to the rest of the kernel.
- */
- isb
- ret
-ENDPROC(__vhe_hyp_call)
-
el1_sync: // Guest trapped into EL2
mrs x0, esr_el2
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 05/27] ARM: KVM: Teach some form of type-safety to kvm_call_hyp
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
Just like on arm64, and for the same reasons, kvm_call_hyp removes
any form of type safety when calling into HYP. But we can still
try to tell the compiler what we're trying to achieve.
Here, we can add code that would do the function call if it wasn't
guarded by an always-false predicate. Hopefully, the compiler is
dumb enough to do the type checking and clever enough to not emit
the corresponding code...
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_host.h | 31 ++++++++++++++++++++++++++++---
arch/arm/kvm/hyp/hyp-entry.S | 2 +-
arch/arm/kvm/interrupts.S | 4 ++--
3 files changed, 31 insertions(+), 6 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 023c9f2b1eea..4b6193f2f0f6 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -215,8 +215,33 @@ int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
-unsigned long kvm_call_hyp(void *hypfn, ...);
-#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
+unsigned long __kvm_call_hyp(void *hypfn, ...);
+
+/*
+ * The has_vhe() part doesn't get emitted, but is used for type-checking.
+ */
+#define kvm_call_hyp(f, ...) \
+ do { \
+ if (has_vhe()) { \
+ f(__VA_ARGS__); \
+ } else { \
+ __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__); \
+ } \
+ } while(0)
+
+#define kvm_call_hyp_ret(f, ...) \
+ ({ \
+ typeof(f(__VA_ARGS__)) ret; \
+ \
+ if (has_vhe()) { \
+ ret = f(__VA_ARGS__); \
+ } else { \
+ ret = __kvm_call_hyp(kvm_ksym_ref(f), \
+ ##__VA_ARGS__); \
+ } \
+ \
+ ret; \
+ })
void force_vm_exit(const cpumask_t *mask);
int __kvm_arm_vcpu_get_events(struct kvm_vcpu *vcpu,
@@ -268,7 +293,7 @@ static inline void __cpu_init_hyp_mode(phys_addr_t pgd_ptr,
* compliant with the PCS!).
*/
- kvm_call_hyp((void*)hyp_stack_ptr, vector_ptr, pgd_ptr);
+ __kvm_call_hyp((void*)hyp_stack_ptr, vector_ptr, pgd_ptr);
}
static inline void __cpu_init_stage2(void)
diff --git a/arch/arm/kvm/hyp/hyp-entry.S b/arch/arm/kvm/hyp/hyp-entry.S
index aa3f9a9837ac..6ed3cf23fe89 100644
--- a/arch/arm/kvm/hyp/hyp-entry.S
+++ b/arch/arm/kvm/hyp/hyp-entry.S
@@ -176,7 +176,7 @@ THUMB( orr lr, lr, #PSR_T_BIT )
msr spsr_cxsf, lr
ldr lr, =panic
msr ELR_hyp, lr
- ldr lr, =kvm_call_hyp
+ ldr lr, =__kvm_call_hyp
clrex
eret
ENDPROC(__hyp_do_panic)
diff --git a/arch/arm/kvm/interrupts.S b/arch/arm/kvm/interrupts.S
index 80a1d6cd261c..a08e6419ebe9 100644
--- a/arch/arm/kvm/interrupts.S
+++ b/arch/arm/kvm/interrupts.S
@@ -42,7 +42,7 @@
* r12: caller save
* rest: callee save
*/
-ENTRY(kvm_call_hyp)
+ENTRY(__kvm_call_hyp)
hvc #0
bx lr
-ENDPROC(kvm_call_hyp)
+ENDPROC(__kvm_call_hyp)
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 05/27] ARM: KVM: Teach some form of type-safety to kvm_call_hyp
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
Just like on arm64, and for the same reasons, kvm_call_hyp removes
any form of type safety when calling into HYP. But we can still
try to tell the compiler what we're trying to achieve.
Here, we can add code that would do the function call if it wasn't
guarded by an always-false predicate. Hopefully, the compiler is
dumb enough to do the type checking and clever enough to not emit
the corresponding code...
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_host.h | 31 ++++++++++++++++++++++++++++---
arch/arm/kvm/hyp/hyp-entry.S | 2 +-
arch/arm/kvm/interrupts.S | 4 ++--
3 files changed, 31 insertions(+), 6 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 023c9f2b1eea..4b6193f2f0f6 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -215,8 +215,33 @@ int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices);
int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg);
-unsigned long kvm_call_hyp(void *hypfn, ...);
-#define kvm_call_hyp_ret(f, ...) kvm_call_hyp(f, ##__VA_ARGS__)
+unsigned long __kvm_call_hyp(void *hypfn, ...);
+
+/*
+ * The has_vhe() part doesn't get emitted, but is used for type-checking.
+ */
+#define kvm_call_hyp(f, ...) \
+ do { \
+ if (has_vhe()) { \
+ f(__VA_ARGS__); \
+ } else { \
+ __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__); \
+ } \
+ } while(0)
+
+#define kvm_call_hyp_ret(f, ...) \
+ ({ \
+ typeof(f(__VA_ARGS__)) ret; \
+ \
+ if (has_vhe()) { \
+ ret = f(__VA_ARGS__); \
+ } else { \
+ ret = __kvm_call_hyp(kvm_ksym_ref(f), \
+ ##__VA_ARGS__); \
+ } \
+ \
+ ret; \
+ })
void force_vm_exit(const cpumask_t *mask);
int __kvm_arm_vcpu_get_events(struct kvm_vcpu *vcpu,
@@ -268,7 +293,7 @@ static inline void __cpu_init_hyp_mode(phys_addr_t pgd_ptr,
* compliant with the PCS!).
*/
- kvm_call_hyp((void*)hyp_stack_ptr, vector_ptr, pgd_ptr);
+ __kvm_call_hyp((void*)hyp_stack_ptr, vector_ptr, pgd_ptr);
}
static inline void __cpu_init_stage2(void)
diff --git a/arch/arm/kvm/hyp/hyp-entry.S b/arch/arm/kvm/hyp/hyp-entry.S
index aa3f9a9837ac..6ed3cf23fe89 100644
--- a/arch/arm/kvm/hyp/hyp-entry.S
+++ b/arch/arm/kvm/hyp/hyp-entry.S
@@ -176,7 +176,7 @@ THUMB( orr lr, lr, #PSR_T_BIT )
msr spsr_cxsf, lr
ldr lr, =panic
msr ELR_hyp, lr
- ldr lr, =kvm_call_hyp
+ ldr lr, =__kvm_call_hyp
clrex
eret
ENDPROC(__hyp_do_panic)
diff --git a/arch/arm/kvm/interrupts.S b/arch/arm/kvm/interrupts.S
index 80a1d6cd261c..a08e6419ebe9 100644
--- a/arch/arm/kvm/interrupts.S
+++ b/arch/arm/kvm/interrupts.S
@@ -42,7 +42,7 @@
* r12: caller save
* rest: callee save
*/
-ENTRY(kvm_call_hyp)
+ENTRY(__kvm_call_hyp)
hvc #0
bx lr
-ENDPROC(kvm_call_hyp)
+ENDPROC(__kvm_call_hyp)
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 06/27] arm/arm64: KVM: Statically configure the host's view of MPIDR
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
We currently eagerly save/restore MPIDR. It turns out to be
slightly pointless:
- On the host, this value is known as soon as we're scheduled on a
physical CPU
- In the guest, this value cannot change, as it is set by KVM
(and this is a read-only register)
The result of the above is that we can perfectly avoid the eager
saving of MPIDR_EL1, and only keep the restore. We just have
to setup the host contexts appropriately at boot time.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_host.h | 8 ++++++++
arch/arm/kvm/hyp/cp15-sr.c | 1 -
arch/arm64/include/asm/kvm_host.h | 8 ++++++++
arch/arm64/kvm/hyp/sysreg-sr.c | 1 -
virt/kvm/arm/arm.c | 1 +
5 files changed, 17 insertions(+), 2 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 4b6193f2f0f6..43e343e00fb8 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -26,6 +26,7 @@
#include <asm/kvm_asm.h>
#include <asm/kvm_mmio.h>
#include <asm/fpstate.h>
+#include <asm/smp_plat.h>
#include <kvm/arm_arch_timer.h>
#define __KVM_HAVE_ARCH_INTC_INITIALIZED
@@ -147,6 +148,13 @@ struct kvm_cpu_context {
typedef struct kvm_cpu_context kvm_cpu_context_t;
+static inline void kvm_init_host_cpu_context(kvm_cpu_context_t *cpu_ctxt,
+ int cpu)
+{
+ /* The host's MPIDR is immutable, so let's set it up at boot time */
+ cpu_ctxt->cp15[c0_MPIDR] = cpu_logical_map(cpu);
+}
+
struct kvm_vcpu_arch {
struct kvm_cpu_context ctxt;
diff --git a/arch/arm/kvm/hyp/cp15-sr.c b/arch/arm/kvm/hyp/cp15-sr.c
index c4782812714c..8bf895ec6e04 100644
--- a/arch/arm/kvm/hyp/cp15-sr.c
+++ b/arch/arm/kvm/hyp/cp15-sr.c
@@ -27,7 +27,6 @@ static u64 *cp15_64(struct kvm_cpu_context *ctxt, int idx)
void __hyp_text __sysreg_save_state(struct kvm_cpu_context *ctxt)
{
- ctxt->cp15[c0_MPIDR] = read_sysreg(VMPIDR);
ctxt->cp15[c0_CSSELR] = read_sysreg(CSSELR);
ctxt->cp15[c1_SCTLR] = read_sysreg(SCTLR);
ctxt->cp15[c1_CPACR] = read_sysreg(CPACR);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 8b7702bdb219..f497bb31031f 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -30,6 +30,7 @@
#include <asm/kvm.h>
#include <asm/kvm_asm.h>
#include <asm/kvm_mmio.h>
+#include <asm/smp_plat.h>
#include <asm/thread_info.h>
#define __KVM_HAVE_ARCH_INTC_INITIALIZED
@@ -418,6 +419,13 @@ struct kvm_vcpu *kvm_mpidr_to_vcpu(struct kvm *kvm, unsigned long mpidr);
DECLARE_PER_CPU(kvm_cpu_context_t, kvm_host_cpu_state);
+static inline void kvm_init_host_cpu_context(kvm_cpu_context_t *cpu_ctxt,
+ int cpu)
+{
+ /* The host's MPIDR is immutable, so let's set it up at boot time */
+ cpu_ctxt->sys_regs[MPIDR_EL1] = cpu_logical_map(cpu);
+}
+
void __kvm_enable_ssbs(void);
static inline void __cpu_init_hyp_mode(phys_addr_t pgd_ptr,
diff --git a/arch/arm64/kvm/hyp/sysreg-sr.c b/arch/arm64/kvm/hyp/sysreg-sr.c
index 68d6f7c3b237..2498f86defcb 100644
--- a/arch/arm64/kvm/hyp/sysreg-sr.c
+++ b/arch/arm64/kvm/hyp/sysreg-sr.c
@@ -52,7 +52,6 @@ static void __hyp_text __sysreg_save_user_state(struct kvm_cpu_context *ctxt)
static void __hyp_text __sysreg_save_el1_state(struct kvm_cpu_context *ctxt)
{
- ctxt->sys_regs[MPIDR_EL1] = read_sysreg(vmpidr_el2);
ctxt->sys_regs[CSSELR_EL1] = read_sysreg(csselr_el1);
ctxt->sys_regs[SCTLR_EL1] = read_sysreg_el1(sctlr);
ctxt->sys_regs[ACTLR_EL1] = read_sysreg(actlr_el1);
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 4d55f98f97f7..3dd240ea9e76 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -1561,6 +1561,7 @@ static int init_hyp_mode(void)
kvm_cpu_context_t *cpu_ctxt;
cpu_ctxt = per_cpu_ptr(&kvm_host_cpu_state, cpu);
+ kvm_init_host_cpu_context(cpu_ctxt, cpu);
err = create_hyp_mappings(cpu_ctxt, cpu_ctxt + 1, PAGE_HYP);
if (err) {
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 06/27] arm/arm64: KVM: Statically configure the host's view of MPIDR
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
We currently eagerly save/restore MPIDR. It turns out to be
slightly pointless:
- On the host, this value is known as soon as we're scheduled on a
physical CPU
- In the guest, this value cannot change, as it is set by KVM
(and this is a read-only register)
The result of the above is that we can perfectly avoid the eager
saving of MPIDR_EL1, and only keep the restore. We just have
to setup the host contexts appropriately at boot time.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_host.h | 8 ++++++++
arch/arm/kvm/hyp/cp15-sr.c | 1 -
arch/arm64/include/asm/kvm_host.h | 8 ++++++++
arch/arm64/kvm/hyp/sysreg-sr.c | 1 -
virt/kvm/arm/arm.c | 1 +
5 files changed, 17 insertions(+), 2 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 4b6193f2f0f6..43e343e00fb8 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -26,6 +26,7 @@
#include <asm/kvm_asm.h>
#include <asm/kvm_mmio.h>
#include <asm/fpstate.h>
+#include <asm/smp_plat.h>
#include <kvm/arm_arch_timer.h>
#define __KVM_HAVE_ARCH_INTC_INITIALIZED
@@ -147,6 +148,13 @@ struct kvm_cpu_context {
typedef struct kvm_cpu_context kvm_cpu_context_t;
+static inline void kvm_init_host_cpu_context(kvm_cpu_context_t *cpu_ctxt,
+ int cpu)
+{
+ /* The host's MPIDR is immutable, so let's set it up at boot time */
+ cpu_ctxt->cp15[c0_MPIDR] = cpu_logical_map(cpu);
+}
+
struct kvm_vcpu_arch {
struct kvm_cpu_context ctxt;
diff --git a/arch/arm/kvm/hyp/cp15-sr.c b/arch/arm/kvm/hyp/cp15-sr.c
index c4782812714c..8bf895ec6e04 100644
--- a/arch/arm/kvm/hyp/cp15-sr.c
+++ b/arch/arm/kvm/hyp/cp15-sr.c
@@ -27,7 +27,6 @@ static u64 *cp15_64(struct kvm_cpu_context *ctxt, int idx)
void __hyp_text __sysreg_save_state(struct kvm_cpu_context *ctxt)
{
- ctxt->cp15[c0_MPIDR] = read_sysreg(VMPIDR);
ctxt->cp15[c0_CSSELR] = read_sysreg(CSSELR);
ctxt->cp15[c1_SCTLR] = read_sysreg(SCTLR);
ctxt->cp15[c1_CPACR] = read_sysreg(CPACR);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 8b7702bdb219..f497bb31031f 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -30,6 +30,7 @@
#include <asm/kvm.h>
#include <asm/kvm_asm.h>
#include <asm/kvm_mmio.h>
+#include <asm/smp_plat.h>
#include <asm/thread_info.h>
#define __KVM_HAVE_ARCH_INTC_INITIALIZED
@@ -418,6 +419,13 @@ struct kvm_vcpu *kvm_mpidr_to_vcpu(struct kvm *kvm, unsigned long mpidr);
DECLARE_PER_CPU(kvm_cpu_context_t, kvm_host_cpu_state);
+static inline void kvm_init_host_cpu_context(kvm_cpu_context_t *cpu_ctxt,
+ int cpu)
+{
+ /* The host's MPIDR is immutable, so let's set it up at boot time */
+ cpu_ctxt->sys_regs[MPIDR_EL1] = cpu_logical_map(cpu);
+}
+
void __kvm_enable_ssbs(void);
static inline void __cpu_init_hyp_mode(phys_addr_t pgd_ptr,
diff --git a/arch/arm64/kvm/hyp/sysreg-sr.c b/arch/arm64/kvm/hyp/sysreg-sr.c
index 68d6f7c3b237..2498f86defcb 100644
--- a/arch/arm64/kvm/hyp/sysreg-sr.c
+++ b/arch/arm64/kvm/hyp/sysreg-sr.c
@@ -52,7 +52,6 @@ static void __hyp_text __sysreg_save_user_state(struct kvm_cpu_context *ctxt)
static void __hyp_text __sysreg_save_el1_state(struct kvm_cpu_context *ctxt)
{
- ctxt->sys_regs[MPIDR_EL1] = read_sysreg(vmpidr_el2);
ctxt->sys_regs[CSSELR_EL1] = read_sysreg(csselr_el1);
ctxt->sys_regs[SCTLR_EL1] = read_sysreg_el1(sctlr);
ctxt->sys_regs[ACTLR_EL1] = read_sysreg(actlr_el1);
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 4d55f98f97f7..3dd240ea9e76 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -1561,6 +1561,7 @@ static int init_hyp_mode(void)
kvm_cpu_context_t *cpu_ctxt;
cpu_ctxt = per_cpu_ptr(&kvm_host_cpu_state, cpu);
+ kvm_init_host_cpu_context(cpu_ctxt, cpu);
err = create_hyp_mappings(cpu_ctxt, cpu_ctxt + 1, PAGE_HYP);
if (err) {
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 07/27] KVM: arm/arm64: Factor out VMID into struct kvm_vmid
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Christoffer Dall <christoffer.dall@arm.com>
In preparation for nested virtualization where we are going to have more
than a single VMID per VM, let's factor out the VMID data into a
separate VMID data structure and change the VMID allocator to operate on
this new structure instead of using a struct kvm.
This also means that udate_vttbr now becomes update_vmid, and that the
vttbr itself is generated on the fly based on the stage 2 page table
base address and the vmid.
We cache the physical address of the pgd when allocating the pgd to
avoid doing the calculation on every entry to the guest and to avoid
calling into potentially non-hyp-mapped code from hyp/EL2.
If we wanted to merge the VMID allocator with the arm64 ASID allocator
at some point in the future, it should actually become easier to do that
after this patch.
Note that to avoid mapping the kvm_vmid_bits variable into hyp, we
simply forego the masking of the vmid value in kvm_get_vttbr and rely on
update_vmid to always assign a valid vmid value (within the supported
range).
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
[maz: minor cleanups]
Reviewed-by: Julien Thierry <julien.thierry@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/include/asm/kvm_host.h | 13 ++++---
arch/arm/include/asm/kvm_mmu.h | 9 +++--
arch/arm/kvm/hyp/switch.c | 2 +-
arch/arm/kvm/hyp/tlb.c | 4 +--
arch/arm64/include/asm/kvm_host.h | 9 +++--
arch/arm64/include/asm/kvm_hyp.h | 3 +-
arch/arm64/include/asm/kvm_mmu.h | 10 ++++--
virt/kvm/arm/arm.c | 57 +++++++++++--------------------
virt/kvm/arm/mmu.c | 7 ++++
9 files changed, 61 insertions(+), 53 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 43e343e00fb8..8073267dc4a0 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -57,10 +57,13 @@ int __attribute_const__ kvm_target_cpu(void);
int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
void kvm_reset_coprocs(struct kvm_vcpu *vcpu);
-struct kvm_arch {
- /* VTTBR value associated with below pgd and vmid */
- u64 vttbr;
+struct kvm_vmid {
+ /* The VMID generation used for the virt. memory system */
+ u64 vmid_gen;
+ u32 vmid;
+};
+struct kvm_arch {
/* The last vcpu id that ran on each physical CPU */
int __percpu *last_vcpu_ran;
@@ -70,11 +73,11 @@ struct kvm_arch {
*/
/* The VMID generation used for the virt. memory system */
- u64 vmid_gen;
- u32 vmid;
+ struct kvm_vmid vmid;
/* Stage-2 page table */
pgd_t *pgd;
+ phys_addr_t pgd_phys;
/* Interrupt controller */
struct vgic_dist vgic;
diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
index 3a875fc1b63c..2de96a180166 100644
--- a/arch/arm/include/asm/kvm_mmu.h
+++ b/arch/arm/include/asm/kvm_mmu.h
@@ -421,9 +421,14 @@ static inline int hyp_map_aux_data(void)
static inline void kvm_set_ipa_limit(void) {}
-static inline bool kvm_cpu_has_cnp(void)
+static __always_inline u64 kvm_get_vttbr(struct kvm *kvm)
{
- return false;
+ struct kvm_vmid *vmid = &kvm->arch.vmid;
+ u64 vmid_field, baddr;
+
+ baddr = kvm->arch.pgd_phys;
+ vmid_field = (u64)vmid->vmid << VTTBR_VMID_SHIFT;
+ return kvm_phys_to_vttbr(baddr) | vmid_field;
}
#endif /* !__ASSEMBLY__ */
diff --git a/arch/arm/kvm/hyp/switch.c b/arch/arm/kvm/hyp/switch.c
index acf1c37fa49c..3b058a5d7c5f 100644
--- a/arch/arm/kvm/hyp/switch.c
+++ b/arch/arm/kvm/hyp/switch.c
@@ -77,7 +77,7 @@ static void __hyp_text __deactivate_traps(struct kvm_vcpu *vcpu)
static void __hyp_text __activate_vm(struct kvm_vcpu *vcpu)
{
struct kvm *kvm = kern_hyp_va(vcpu->kvm);
- write_sysreg(kvm->arch.vttbr, VTTBR);
+ write_sysreg(kvm_get_vttbr(kvm), VTTBR);
write_sysreg(vcpu->arch.midr, VPIDR);
}
diff --git a/arch/arm/kvm/hyp/tlb.c b/arch/arm/kvm/hyp/tlb.c
index c0edd450e104..8e4afba73635 100644
--- a/arch/arm/kvm/hyp/tlb.c
+++ b/arch/arm/kvm/hyp/tlb.c
@@ -41,7 +41,7 @@ void __hyp_text __kvm_tlb_flush_vmid(struct kvm *kvm)
/* Switch to requested VMID */
kvm = kern_hyp_va(kvm);
- write_sysreg(kvm->arch.vttbr, VTTBR);
+ write_sysreg(kvm_get_vttbr(kvm), VTTBR);
isb();
write_sysreg(0, TLBIALLIS);
@@ -61,7 +61,7 @@ void __hyp_text __kvm_tlb_flush_local_vmid(struct kvm_vcpu *vcpu)
struct kvm *kvm = kern_hyp_va(kern_hyp_va(vcpu)->kvm);
/* Switch to requested VMID */
- write_sysreg(kvm->arch.vttbr, VTTBR);
+ write_sysreg(kvm_get_vttbr(kvm), VTTBR);
isb();
write_sysreg(0, TLBIALL);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index f497bb31031f..444dd1cb1958 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -57,16 +57,19 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
int kvm_arch_vm_ioctl_check_extension(struct kvm *kvm, long ext);
void __extended_idmap_trampoline(phys_addr_t boot_pgd, phys_addr_t idmap_start);
-struct kvm_arch {
+struct kvm_vmid {
/* The VMID generation used for the virt. memory system */
u64 vmid_gen;
u32 vmid;
+};
+
+struct kvm_arch {
+ struct kvm_vmid vmid;
/* stage2 entry level table */
pgd_t *pgd;
+ phys_addr_t pgd_phys;
- /* VTTBR value associated with above pgd and vmid */
- u64 vttbr;
/* VTCR_EL2 value for this VM */
u64 vtcr;
diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h
index a80a7ef57325..4da765f2cca5 100644
--- a/arch/arm64/include/asm/kvm_hyp.h
+++ b/arch/arm64/include/asm/kvm_hyp.h
@@ -21,6 +21,7 @@
#include <linux/compiler.h>
#include <linux/kvm_host.h>
#include <asm/alternative.h>
+#include <asm/kvm_mmu.h>
#include <asm/sysreg.h>
#define __hyp_text __section(.hyp.text) notrace
@@ -163,7 +164,7 @@ 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);
+ write_sysreg(kvm_get_vttbr(kvm), vttbr_el2);
/*
* ARM erratum 1165522 requires the actual execution of the above
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index 8af4b1befa42..c423c8c4fc39 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -591,9 +591,15 @@ static inline u64 kvm_vttbr_baddr_mask(struct kvm *kvm)
return vttbr_baddr_mask(kvm_phys_shift(kvm), kvm_stage2_levels(kvm));
}
-static inline bool kvm_cpu_has_cnp(void)
+static __always_inline u64 kvm_get_vttbr(struct kvm *kvm)
{
- return system_supports_cnp();
+ struct kvm_vmid *vmid = &kvm->arch.vmid;
+ u64 vmid_field, baddr;
+ u64 cnp = system_supports_cnp() ? VTTBR_CNP_BIT : 0;
+
+ baddr = kvm->arch.pgd_phys;
+ vmid_field = (u64)vmid->vmid << VTTBR_VMID_SHIFT;
+ return kvm_phys_to_vttbr(baddr) | vmid_field | cnp;
}
#endif /* __ASSEMBLY__ */
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 3dd240ea9e76..b77db673bb03 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -65,7 +65,6 @@ static DEFINE_PER_CPU(struct kvm_vcpu *, kvm_arm_running_vcpu);
/* The VMID used in the VTTBR */
static atomic64_t kvm_vmid_gen = ATOMIC64_INIT(1);
static u32 kvm_next_vmid;
-static unsigned int kvm_vmid_bits __read_mostly;
static DEFINE_SPINLOCK(kvm_vmid_lock);
static bool vgic_present;
@@ -142,7 +141,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
kvm_vgic_early_init(kvm);
/* Mark the initial VMID generation invalid */
- kvm->arch.vmid_gen = 0;
+ kvm->arch.vmid.vmid_gen = 0;
/* The maximum number of VCPUs is limited by the host's GIC model */
kvm->arch.max_vcpus = vgic_present ?
@@ -472,37 +471,31 @@ void force_vm_exit(const cpumask_t *mask)
/**
* need_new_vmid_gen - check that the VMID is still valid
- * @kvm: The VM's VMID to check
+ * @vmid: The VMID to check
*
* return true if there is a new generation of VMIDs being used
*
- * The hardware supports only 256 values with the value zero reserved for the
- * host, so we check if an assigned value belongs to a previous generation,
- * which which requires us to assign a new value. If we're the first to use a
- * VMID for the new generation, we must flush necessary caches and TLBs on all
- * CPUs.
+ * The hardware supports a limited set of values with the value zero reserved
+ * for the host, so we check if an assigned value belongs to a previous
+ * generation, which which requires us to assign a new value. If we're the
+ * first to use a VMID for the new generation, we must flush necessary caches
+ * and TLBs on all CPUs.
*/
-static bool need_new_vmid_gen(struct kvm *kvm)
+static bool need_new_vmid_gen(struct kvm_vmid *vmid)
{
u64 current_vmid_gen = atomic64_read(&kvm_vmid_gen);
smp_rmb(); /* Orders read of kvm_vmid_gen and kvm->arch.vmid */
- return unlikely(READ_ONCE(kvm->arch.vmid_gen) != current_vmid_gen);
+ return unlikely(READ_ONCE(vmid->vmid_gen) != current_vmid_gen);
}
/**
- * update_vttbr - Update the VTTBR with a valid VMID before the guest runs
- * @kvm The guest that we are about to run
- *
- * Called from kvm_arch_vcpu_ioctl_run before entering the guest to ensure the
- * VM has a valid VMID, otherwise assigns a new one and flushes corresponding
- * caches and TLBs.
+ * update_vmid - Update the vmid with a valid VMID for the current generation
+ * @kvm: The guest that struct vmid belongs to
+ * @vmid: The stage-2 VMID information struct
*/
-static void update_vttbr(struct kvm *kvm)
+static void update_vmid(struct kvm_vmid *vmid)
{
- phys_addr_t pgd_phys;
- u64 vmid, cnp = kvm_cpu_has_cnp() ? VTTBR_CNP_BIT : 0;
-
- if (!need_new_vmid_gen(kvm))
+ if (!need_new_vmid_gen(vmid))
return;
spin_lock(&kvm_vmid_lock);
@@ -512,7 +505,7 @@ static void update_vttbr(struct kvm *kvm)
* already allocated a valid vmid for this vm, then this vcpu should
* use the same vmid.
*/
- if (!need_new_vmid_gen(kvm)) {
+ if (!need_new_vmid_gen(vmid)) {
spin_unlock(&kvm_vmid_lock);
return;
}
@@ -536,18 +529,12 @@ static void update_vttbr(struct kvm *kvm)
kvm_call_hyp(__kvm_flush_vm_context);
}
- kvm->arch.vmid = kvm_next_vmid;
+ vmid->vmid = kvm_next_vmid;
kvm_next_vmid++;
- kvm_next_vmid &= (1 << kvm_vmid_bits) - 1;
-
- /* update vttbr to be used with the new vmid */
- pgd_phys = virt_to_phys(kvm->arch.pgd);
- 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 | cnp;
+ kvm_next_vmid &= (1 << kvm_get_vmid_bits()) - 1;
smp_wmb();
- WRITE_ONCE(kvm->arch.vmid_gen, atomic64_read(&kvm_vmid_gen));
+ WRITE_ONCE(vmid->vmid_gen, atomic64_read(&kvm_vmid_gen));
spin_unlock(&kvm_vmid_lock);
}
@@ -690,7 +677,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
*/
cond_resched();
- update_vttbr(vcpu->kvm);
+ update_vmid(&vcpu->kvm->arch.vmid);
check_vcpu_requests(vcpu);
@@ -739,7 +726,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
*/
smp_store_mb(vcpu->mode, IN_GUEST_MODE);
- if (ret <= 0 || need_new_vmid_gen(vcpu->kvm) ||
+ if (ret <= 0 || need_new_vmid_gen(&vcpu->kvm->arch.vmid) ||
kvm_request_pending(vcpu)) {
vcpu->mode = OUTSIDE_GUEST_MODE;
isb(); /* Ensure work in x_flush_hwstate is committed */
@@ -1417,10 +1404,6 @@ static inline void hyp_cpu_pm_exit(void)
static int init_common_resources(void)
{
- /* set size of VMID supported by CPU */
- kvm_vmid_bits = kvm_get_vmid_bits();
- kvm_info("%d-bit VMID\n", kvm_vmid_bits);
-
kvm_set_ipa_limit();
return 0;
diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index fbdf3ac2f001..f8dda452ea24 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -908,6 +908,7 @@ int create_hyp_exec_mappings(phys_addr_t phys_addr, size_t size,
*/
int kvm_alloc_stage2_pgd(struct kvm *kvm)
{
+ phys_addr_t pgd_phys;
pgd_t *pgd;
if (kvm->arch.pgd != NULL) {
@@ -920,7 +921,12 @@ int kvm_alloc_stage2_pgd(struct kvm *kvm)
if (!pgd)
return -ENOMEM;
+ pgd_phys = virt_to_phys(pgd);
+ if (WARN_ON(pgd_phys & ~kvm_vttbr_baddr_mask(kvm)))
+ return -EINVAL;
+
kvm->arch.pgd = pgd;
+ kvm->arch.pgd_phys = pgd_phys;
return 0;
}
@@ -1008,6 +1014,7 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
unmap_stage2_range(kvm, 0, kvm_phys_size(kvm));
pgd = READ_ONCE(kvm->arch.pgd);
kvm->arch.pgd = NULL;
+ kvm->arch.pgd_phys = 0;
}
spin_unlock(&kvm->mmu_lock);
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 07/27] KVM: arm/arm64: Factor out VMID into struct kvm_vmid
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Christoffer Dall <christoffer.dall@arm.com>
In preparation for nested virtualization where we are going to have more
than a single VMID per VM, let's factor out the VMID data into a
separate VMID data structure and change the VMID allocator to operate on
this new structure instead of using a struct kvm.
This also means that udate_vttbr now becomes update_vmid, and that the
vttbr itself is generated on the fly based on the stage 2 page table
base address and the vmid.
We cache the physical address of the pgd when allocating the pgd to
avoid doing the calculation on every entry to the guest and to avoid
calling into potentially non-hyp-mapped code from hyp/EL2.
If we wanted to merge the VMID allocator with the arm64 ASID allocator
at some point in the future, it should actually become easier to do that
after this patch.
Note that to avoid mapping the kvm_vmid_bits variable into hyp, we
simply forego the masking of the vmid value in kvm_get_vttbr and rely on
update_vmid to always assign a valid vmid value (within the supported
range).
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
[maz: minor cleanups]
Reviewed-by: Julien Thierry <julien.thierry@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/include/asm/kvm_host.h | 13 ++++---
arch/arm/include/asm/kvm_mmu.h | 9 +++--
arch/arm/kvm/hyp/switch.c | 2 +-
arch/arm/kvm/hyp/tlb.c | 4 +--
arch/arm64/include/asm/kvm_host.h | 9 +++--
arch/arm64/include/asm/kvm_hyp.h | 3 +-
arch/arm64/include/asm/kvm_mmu.h | 10 ++++--
virt/kvm/arm/arm.c | 57 +++++++++++--------------------
virt/kvm/arm/mmu.c | 7 ++++
9 files changed, 61 insertions(+), 53 deletions(-)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index 43e343e00fb8..8073267dc4a0 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -57,10 +57,13 @@ int __attribute_const__ kvm_target_cpu(void);
int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
void kvm_reset_coprocs(struct kvm_vcpu *vcpu);
-struct kvm_arch {
- /* VTTBR value associated with below pgd and vmid */
- u64 vttbr;
+struct kvm_vmid {
+ /* The VMID generation used for the virt. memory system */
+ u64 vmid_gen;
+ u32 vmid;
+};
+struct kvm_arch {
/* The last vcpu id that ran on each physical CPU */
int __percpu *last_vcpu_ran;
@@ -70,11 +73,11 @@ struct kvm_arch {
*/
/* The VMID generation used for the virt. memory system */
- u64 vmid_gen;
- u32 vmid;
+ struct kvm_vmid vmid;
/* Stage-2 page table */
pgd_t *pgd;
+ phys_addr_t pgd_phys;
/* Interrupt controller */
struct vgic_dist vgic;
diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
index 3a875fc1b63c..2de96a180166 100644
--- a/arch/arm/include/asm/kvm_mmu.h
+++ b/arch/arm/include/asm/kvm_mmu.h
@@ -421,9 +421,14 @@ static inline int hyp_map_aux_data(void)
static inline void kvm_set_ipa_limit(void) {}
-static inline bool kvm_cpu_has_cnp(void)
+static __always_inline u64 kvm_get_vttbr(struct kvm *kvm)
{
- return false;
+ struct kvm_vmid *vmid = &kvm->arch.vmid;
+ u64 vmid_field, baddr;
+
+ baddr = kvm->arch.pgd_phys;
+ vmid_field = (u64)vmid->vmid << VTTBR_VMID_SHIFT;
+ return kvm_phys_to_vttbr(baddr) | vmid_field;
}
#endif /* !__ASSEMBLY__ */
diff --git a/arch/arm/kvm/hyp/switch.c b/arch/arm/kvm/hyp/switch.c
index acf1c37fa49c..3b058a5d7c5f 100644
--- a/arch/arm/kvm/hyp/switch.c
+++ b/arch/arm/kvm/hyp/switch.c
@@ -77,7 +77,7 @@ static void __hyp_text __deactivate_traps(struct kvm_vcpu *vcpu)
static void __hyp_text __activate_vm(struct kvm_vcpu *vcpu)
{
struct kvm *kvm = kern_hyp_va(vcpu->kvm);
- write_sysreg(kvm->arch.vttbr, VTTBR);
+ write_sysreg(kvm_get_vttbr(kvm), VTTBR);
write_sysreg(vcpu->arch.midr, VPIDR);
}
diff --git a/arch/arm/kvm/hyp/tlb.c b/arch/arm/kvm/hyp/tlb.c
index c0edd450e104..8e4afba73635 100644
--- a/arch/arm/kvm/hyp/tlb.c
+++ b/arch/arm/kvm/hyp/tlb.c
@@ -41,7 +41,7 @@ void __hyp_text __kvm_tlb_flush_vmid(struct kvm *kvm)
/* Switch to requested VMID */
kvm = kern_hyp_va(kvm);
- write_sysreg(kvm->arch.vttbr, VTTBR);
+ write_sysreg(kvm_get_vttbr(kvm), VTTBR);
isb();
write_sysreg(0, TLBIALLIS);
@@ -61,7 +61,7 @@ void __hyp_text __kvm_tlb_flush_local_vmid(struct kvm_vcpu *vcpu)
struct kvm *kvm = kern_hyp_va(kern_hyp_va(vcpu)->kvm);
/* Switch to requested VMID */
- write_sysreg(kvm->arch.vttbr, VTTBR);
+ write_sysreg(kvm_get_vttbr(kvm), VTTBR);
isb();
write_sysreg(0, TLBIALL);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index f497bb31031f..444dd1cb1958 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -57,16 +57,19 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
int kvm_arch_vm_ioctl_check_extension(struct kvm *kvm, long ext);
void __extended_idmap_trampoline(phys_addr_t boot_pgd, phys_addr_t idmap_start);
-struct kvm_arch {
+struct kvm_vmid {
/* The VMID generation used for the virt. memory system */
u64 vmid_gen;
u32 vmid;
+};
+
+struct kvm_arch {
+ struct kvm_vmid vmid;
/* stage2 entry level table */
pgd_t *pgd;
+ phys_addr_t pgd_phys;
- /* VTTBR value associated with above pgd and vmid */
- u64 vttbr;
/* VTCR_EL2 value for this VM */
u64 vtcr;
diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h
index a80a7ef57325..4da765f2cca5 100644
--- a/arch/arm64/include/asm/kvm_hyp.h
+++ b/arch/arm64/include/asm/kvm_hyp.h
@@ -21,6 +21,7 @@
#include <linux/compiler.h>
#include <linux/kvm_host.h>
#include <asm/alternative.h>
+#include <asm/kvm_mmu.h>
#include <asm/sysreg.h>
#define __hyp_text __section(.hyp.text) notrace
@@ -163,7 +164,7 @@ 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);
+ write_sysreg(kvm_get_vttbr(kvm), vttbr_el2);
/*
* ARM erratum 1165522 requires the actual execution of the above
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index 8af4b1befa42..c423c8c4fc39 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -591,9 +591,15 @@ static inline u64 kvm_vttbr_baddr_mask(struct kvm *kvm)
return vttbr_baddr_mask(kvm_phys_shift(kvm), kvm_stage2_levels(kvm));
}
-static inline bool kvm_cpu_has_cnp(void)
+static __always_inline u64 kvm_get_vttbr(struct kvm *kvm)
{
- return system_supports_cnp();
+ struct kvm_vmid *vmid = &kvm->arch.vmid;
+ u64 vmid_field, baddr;
+ u64 cnp = system_supports_cnp() ? VTTBR_CNP_BIT : 0;
+
+ baddr = kvm->arch.pgd_phys;
+ vmid_field = (u64)vmid->vmid << VTTBR_VMID_SHIFT;
+ return kvm_phys_to_vttbr(baddr) | vmid_field | cnp;
}
#endif /* __ASSEMBLY__ */
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 3dd240ea9e76..b77db673bb03 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -65,7 +65,6 @@ static DEFINE_PER_CPU(struct kvm_vcpu *, kvm_arm_running_vcpu);
/* The VMID used in the VTTBR */
static atomic64_t kvm_vmid_gen = ATOMIC64_INIT(1);
static u32 kvm_next_vmid;
-static unsigned int kvm_vmid_bits __read_mostly;
static DEFINE_SPINLOCK(kvm_vmid_lock);
static bool vgic_present;
@@ -142,7 +141,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
kvm_vgic_early_init(kvm);
/* Mark the initial VMID generation invalid */
- kvm->arch.vmid_gen = 0;
+ kvm->arch.vmid.vmid_gen = 0;
/* The maximum number of VCPUs is limited by the host's GIC model */
kvm->arch.max_vcpus = vgic_present ?
@@ -472,37 +471,31 @@ void force_vm_exit(const cpumask_t *mask)
/**
* need_new_vmid_gen - check that the VMID is still valid
- * @kvm: The VM's VMID to check
+ * @vmid: The VMID to check
*
* return true if there is a new generation of VMIDs being used
*
- * The hardware supports only 256 values with the value zero reserved for the
- * host, so we check if an assigned value belongs to a previous generation,
- * which which requires us to assign a new value. If we're the first to use a
- * VMID for the new generation, we must flush necessary caches and TLBs on all
- * CPUs.
+ * The hardware supports a limited set of values with the value zero reserved
+ * for the host, so we check if an assigned value belongs to a previous
+ * generation, which which requires us to assign a new value. If we're the
+ * first to use a VMID for the new generation, we must flush necessary caches
+ * and TLBs on all CPUs.
*/
-static bool need_new_vmid_gen(struct kvm *kvm)
+static bool need_new_vmid_gen(struct kvm_vmid *vmid)
{
u64 current_vmid_gen = atomic64_read(&kvm_vmid_gen);
smp_rmb(); /* Orders read of kvm_vmid_gen and kvm->arch.vmid */
- return unlikely(READ_ONCE(kvm->arch.vmid_gen) != current_vmid_gen);
+ return unlikely(READ_ONCE(vmid->vmid_gen) != current_vmid_gen);
}
/**
- * update_vttbr - Update the VTTBR with a valid VMID before the guest runs
- * @kvm The guest that we are about to run
- *
- * Called from kvm_arch_vcpu_ioctl_run before entering the guest to ensure the
- * VM has a valid VMID, otherwise assigns a new one and flushes corresponding
- * caches and TLBs.
+ * update_vmid - Update the vmid with a valid VMID for the current generation
+ * @kvm: The guest that struct vmid belongs to
+ * @vmid: The stage-2 VMID information struct
*/
-static void update_vttbr(struct kvm *kvm)
+static void update_vmid(struct kvm_vmid *vmid)
{
- phys_addr_t pgd_phys;
- u64 vmid, cnp = kvm_cpu_has_cnp() ? VTTBR_CNP_BIT : 0;
-
- if (!need_new_vmid_gen(kvm))
+ if (!need_new_vmid_gen(vmid))
return;
spin_lock(&kvm_vmid_lock);
@@ -512,7 +505,7 @@ static void update_vttbr(struct kvm *kvm)
* already allocated a valid vmid for this vm, then this vcpu should
* use the same vmid.
*/
- if (!need_new_vmid_gen(kvm)) {
+ if (!need_new_vmid_gen(vmid)) {
spin_unlock(&kvm_vmid_lock);
return;
}
@@ -536,18 +529,12 @@ static void update_vttbr(struct kvm *kvm)
kvm_call_hyp(__kvm_flush_vm_context);
}
- kvm->arch.vmid = kvm_next_vmid;
+ vmid->vmid = kvm_next_vmid;
kvm_next_vmid++;
- kvm_next_vmid &= (1 << kvm_vmid_bits) - 1;
-
- /* update vttbr to be used with the new vmid */
- pgd_phys = virt_to_phys(kvm->arch.pgd);
- 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 | cnp;
+ kvm_next_vmid &= (1 << kvm_get_vmid_bits()) - 1;
smp_wmb();
- WRITE_ONCE(kvm->arch.vmid_gen, atomic64_read(&kvm_vmid_gen));
+ WRITE_ONCE(vmid->vmid_gen, atomic64_read(&kvm_vmid_gen));
spin_unlock(&kvm_vmid_lock);
}
@@ -690,7 +677,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
*/
cond_resched();
- update_vttbr(vcpu->kvm);
+ update_vmid(&vcpu->kvm->arch.vmid);
check_vcpu_requests(vcpu);
@@ -739,7 +726,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
*/
smp_store_mb(vcpu->mode, IN_GUEST_MODE);
- if (ret <= 0 || need_new_vmid_gen(vcpu->kvm) ||
+ if (ret <= 0 || need_new_vmid_gen(&vcpu->kvm->arch.vmid) ||
kvm_request_pending(vcpu)) {
vcpu->mode = OUTSIDE_GUEST_MODE;
isb(); /* Ensure work in x_flush_hwstate is committed */
@@ -1417,10 +1404,6 @@ static inline void hyp_cpu_pm_exit(void)
static int init_common_resources(void)
{
- /* set size of VMID supported by CPU */
- kvm_vmid_bits = kvm_get_vmid_bits();
- kvm_info("%d-bit VMID\n", kvm_vmid_bits);
-
kvm_set_ipa_limit();
return 0;
diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index fbdf3ac2f001..f8dda452ea24 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -908,6 +908,7 @@ int create_hyp_exec_mappings(phys_addr_t phys_addr, size_t size,
*/
int kvm_alloc_stage2_pgd(struct kvm *kvm)
{
+ phys_addr_t pgd_phys;
pgd_t *pgd;
if (kvm->arch.pgd != NULL) {
@@ -920,7 +921,12 @@ int kvm_alloc_stage2_pgd(struct kvm *kvm)
if (!pgd)
return -ENOMEM;
+ pgd_phys = virt_to_phys(pgd);
+ if (WARN_ON(pgd_phys & ~kvm_vttbr_baddr_mask(kvm)))
+ return -EINVAL;
+
kvm->arch.pgd = pgd;
+ kvm->arch.pgd_phys = pgd_phys;
return 0;
}
@@ -1008,6 +1014,7 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
unmap_stage2_range(kvm, 0, kvm_phys_size(kvm));
pgd = READ_ONCE(kvm->arch.pgd);
kvm->arch.pgd = NULL;
+ kvm->arch.pgd_phys = 0;
}
spin_unlock(&kvm->mmu_lock);
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 08/27] KVM: arm/arm64: Simplify bg_timer programming
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Christoffer Dall <christoffer.dall@arm.com>
Instead of calling into kvm_timer_[un]schedule from the main kvm
blocking path, test if the VCPU is on the wait queue from the load/put
path and perform the background timer setup/cancel in this path.
This has the distinct advantage that we no longer race between load/put
and schedule/unschedule and programming and canceling of the bg_timer
always happens when the timer state is not loaded.
Note that we must now remove the checks in kvm_timer_blocking that do
not schedule a background timer if one of the timers can fire, because
we no longer have a guarantee that kvm_vcpu_check_block() will be called
before kvm_timer_blocking.
Reported-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
include/kvm/arm_arch_timer.h | 3 ---
virt/kvm/arm/arch_timer.c | 35 ++++++++++++++---------------------
virt/kvm/arm/arm.c | 2 --
3 files changed, 14 insertions(+), 26 deletions(-)
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index 33771352dcd6..d6e6a45d1d24 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -76,9 +76,6 @@ int kvm_arm_timer_has_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr);
bool kvm_timer_is_pending(struct kvm_vcpu *vcpu);
-void kvm_timer_schedule(struct kvm_vcpu *vcpu);
-void kvm_timer_unschedule(struct kvm_vcpu *vcpu);
-
u64 kvm_phys_timer_read(void);
void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu);
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index b07ac4614e1c..4986028d9829 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -349,22 +349,12 @@ static void vtimer_save_state(struct kvm_vcpu *vcpu)
* thread is removed from its waitqueue and made runnable when there's a timer
* interrupt to handle.
*/
-void kvm_timer_schedule(struct kvm_vcpu *vcpu)
+static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
- vtimer_save_state(vcpu);
-
- /*
- * No need to schedule a background timer if any guest timer has
- * already expired, because kvm_vcpu_block will return before putting
- * the thread to sleep.
- */
- if (kvm_timer_should_fire(vtimer) || kvm_timer_should_fire(ptimer))
- return;
-
/*
* If both timers are not capable of raising interrupts (disabled or
* masked), then there's no more work for us to do.
@@ -373,12 +363,19 @@ void kvm_timer_schedule(struct kvm_vcpu *vcpu)
return;
/*
- * The guest timers have not yet expired, schedule a background timer.
+ * At least one guest time will expire. Schedule a background timer.
* Set the earliest expiration time among the guest timers.
*/
soft_timer_start(&timer->bg_timer, kvm_timer_earliest_exp(vcpu));
}
+static void kvm_timer_unblocking(struct kvm_vcpu *vcpu)
+{
+ struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+
+ soft_timer_cancel(&timer->bg_timer);
+}
+
static void vtimer_restore_state(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
@@ -401,15 +398,6 @@ static void vtimer_restore_state(struct kvm_vcpu *vcpu)
local_irq_restore(flags);
}
-void kvm_timer_unschedule(struct kvm_vcpu *vcpu)
-{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
-
- vtimer_restore_state(vcpu);
-
- soft_timer_cancel(&timer->bg_timer);
-}
-
static void set_cntvoff(u64 cntvoff)
{
u32 low = lower_32_bits(cntvoff);
@@ -485,6 +473,8 @@ void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
/* Set the background timer for the physical timer emulation. */
phys_timer_emulate(vcpu);
+ kvm_timer_unblocking(vcpu);
+
/* If the timer fired while we weren't running, inject it now */
if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
@@ -527,6 +517,9 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
*/
soft_timer_cancel(&timer->phys_timer);
+ if (swait_active(kvm_arch_vcpu_wq(vcpu)))
+ kvm_timer_blocking(vcpu);
+
/*
* The kernel may decide to run userspace after calling vcpu_put, so
* we reset cntvoff to 0 to ensure a consistent read between user
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index b77db673bb03..9fbdb9e1c51f 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -335,13 +335,11 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu)
{
- kvm_timer_schedule(vcpu);
kvm_vgic_v4_enable_doorbell(vcpu);
}
void kvm_arch_vcpu_unblocking(struct kvm_vcpu *vcpu)
{
- kvm_timer_unschedule(vcpu);
kvm_vgic_v4_disable_doorbell(vcpu);
}
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 08/27] KVM: arm/arm64: Simplify bg_timer programming
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Christoffer Dall <christoffer.dall@arm.com>
Instead of calling into kvm_timer_[un]schedule from the main kvm
blocking path, test if the VCPU is on the wait queue from the load/put
path and perform the background timer setup/cancel in this path.
This has the distinct advantage that we no longer race between load/put
and schedule/unschedule and programming and canceling of the bg_timer
always happens when the timer state is not loaded.
Note that we must now remove the checks in kvm_timer_blocking that do
not schedule a background timer if one of the timers can fire, because
we no longer have a guarantee that kvm_vcpu_check_block() will be called
before kvm_timer_blocking.
Reported-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
include/kvm/arm_arch_timer.h | 3 ---
virt/kvm/arm/arch_timer.c | 35 ++++++++++++++---------------------
virt/kvm/arm/arm.c | 2 --
3 files changed, 14 insertions(+), 26 deletions(-)
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index 33771352dcd6..d6e6a45d1d24 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -76,9 +76,6 @@ int kvm_arm_timer_has_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr);
bool kvm_timer_is_pending(struct kvm_vcpu *vcpu);
-void kvm_timer_schedule(struct kvm_vcpu *vcpu);
-void kvm_timer_unschedule(struct kvm_vcpu *vcpu);
-
u64 kvm_phys_timer_read(void);
void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu);
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index b07ac4614e1c..4986028d9829 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -349,22 +349,12 @@ static void vtimer_save_state(struct kvm_vcpu *vcpu)
* thread is removed from its waitqueue and made runnable when there's a timer
* interrupt to handle.
*/
-void kvm_timer_schedule(struct kvm_vcpu *vcpu)
+static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
- vtimer_save_state(vcpu);
-
- /*
- * No need to schedule a background timer if any guest timer has
- * already expired, because kvm_vcpu_block will return before putting
- * the thread to sleep.
- */
- if (kvm_timer_should_fire(vtimer) || kvm_timer_should_fire(ptimer))
- return;
-
/*
* If both timers are not capable of raising interrupts (disabled or
* masked), then there's no more work for us to do.
@@ -373,12 +363,19 @@ void kvm_timer_schedule(struct kvm_vcpu *vcpu)
return;
/*
- * The guest timers have not yet expired, schedule a background timer.
+ * At least one guest time will expire. Schedule a background timer.
* Set the earliest expiration time among the guest timers.
*/
soft_timer_start(&timer->bg_timer, kvm_timer_earliest_exp(vcpu));
}
+static void kvm_timer_unblocking(struct kvm_vcpu *vcpu)
+{
+ struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+
+ soft_timer_cancel(&timer->bg_timer);
+}
+
static void vtimer_restore_state(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
@@ -401,15 +398,6 @@ static void vtimer_restore_state(struct kvm_vcpu *vcpu)
local_irq_restore(flags);
}
-void kvm_timer_unschedule(struct kvm_vcpu *vcpu)
-{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
-
- vtimer_restore_state(vcpu);
-
- soft_timer_cancel(&timer->bg_timer);
-}
-
static void set_cntvoff(u64 cntvoff)
{
u32 low = lower_32_bits(cntvoff);
@@ -485,6 +473,8 @@ void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
/* Set the background timer for the physical timer emulation. */
phys_timer_emulate(vcpu);
+ kvm_timer_unblocking(vcpu);
+
/* If the timer fired while we weren't running, inject it now */
if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
@@ -527,6 +517,9 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
*/
soft_timer_cancel(&timer->phys_timer);
+ if (swait_active(kvm_arch_vcpu_wq(vcpu)))
+ kvm_timer_blocking(vcpu);
+
/*
* The kernel may decide to run userspace after calling vcpu_put, so
* we reset cntvoff to 0 to ensure a consistent read between user
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index b77db673bb03..9fbdb9e1c51f 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -335,13 +335,11 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu)
{
- kvm_timer_schedule(vcpu);
kvm_vgic_v4_enable_doorbell(vcpu);
}
void kvm_arch_vcpu_unblocking(struct kvm_vcpu *vcpu)
{
- kvm_timer_unschedule(vcpu);
kvm_vgic_v4_disable_doorbell(vcpu);
}
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 09/27] KVM: arm64: Fix ICH_ELRSR_EL2 sysreg naming
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
We previously incorrectly named the define for this system register.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/arch_gicv3.h | 4 ++--
arch/arm64/include/asm/sysreg.h | 2 +-
virt/kvm/arm/hyp/vgic-v3-sr.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/arm/include/asm/arch_gicv3.h b/arch/arm/include/asm/arch_gicv3.h
index 0bd530702118..bdc87700def2 100644
--- a/arch/arm/include/asm/arch_gicv3.h
+++ b/arch/arm/include/asm/arch_gicv3.h
@@ -54,7 +54,7 @@
#define ICH_VTR __ACCESS_CP15(c12, 4, c11, 1)
#define ICH_MISR __ACCESS_CP15(c12, 4, c11, 2)
#define ICH_EISR __ACCESS_CP15(c12, 4, c11, 3)
-#define ICH_ELSR __ACCESS_CP15(c12, 4, c11, 5)
+#define ICH_ELRSR __ACCESS_CP15(c12, 4, c11, 5)
#define ICH_VMCR __ACCESS_CP15(c12, 4, c11, 7)
#define __LR0(x) __ACCESS_CP15(c12, 4, c12, x)
@@ -151,7 +151,7 @@ CPUIF_MAP(ICH_HCR, ICH_HCR_EL2)
CPUIF_MAP(ICH_VTR, ICH_VTR_EL2)
CPUIF_MAP(ICH_MISR, ICH_MISR_EL2)
CPUIF_MAP(ICH_EISR, ICH_EISR_EL2)
-CPUIF_MAP(ICH_ELSR, ICH_ELSR_EL2)
+CPUIF_MAP(ICH_ELRSR, ICH_ELRSR_EL2)
CPUIF_MAP(ICH_VMCR, ICH_VMCR_EL2)
CPUIF_MAP(ICH_AP0R3, ICH_AP0R3_EL2)
CPUIF_MAP(ICH_AP0R2, ICH_AP0R2_EL2)
diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h
index 72dc4c011014..3e5650903d6d 100644
--- a/arch/arm64/include/asm/sysreg.h
+++ b/arch/arm64/include/asm/sysreg.h
@@ -426,7 +426,7 @@
#define SYS_ICH_VTR_EL2 sys_reg(3, 4, 12, 11, 1)
#define SYS_ICH_MISR_EL2 sys_reg(3, 4, 12, 11, 2)
#define SYS_ICH_EISR_EL2 sys_reg(3, 4, 12, 11, 3)
-#define SYS_ICH_ELSR_EL2 sys_reg(3, 4, 12, 11, 5)
+#define SYS_ICH_ELRSR_EL2 sys_reg(3, 4, 12, 11, 5)
#define SYS_ICH_VMCR_EL2 sys_reg(3, 4, 12, 11, 7)
#define __SYS__LR0_EL2(x) sys_reg(3, 4, 12, 12, x)
diff --git a/virt/kvm/arm/hyp/vgic-v3-sr.c b/virt/kvm/arm/hyp/vgic-v3-sr.c
index 9652c453480f..264d92da3240 100644
--- a/virt/kvm/arm/hyp/vgic-v3-sr.c
+++ b/virt/kvm/arm/hyp/vgic-v3-sr.c
@@ -226,7 +226,7 @@ void __hyp_text __vgic_v3_save_state(struct kvm_vcpu *vcpu)
int i;
u32 elrsr;
- elrsr = read_gicreg(ICH_ELSR_EL2);
+ elrsr = read_gicreg(ICH_ELRSR_EL2);
write_gicreg(cpu_if->vgic_hcr & ~ICH_HCR_EN, ICH_HCR_EL2);
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 09/27] KVM: arm64: Fix ICH_ELRSR_EL2 sysreg naming
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
We previously incorrectly named the define for this system register.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/arch_gicv3.h | 4 ++--
arch/arm64/include/asm/sysreg.h | 2 +-
virt/kvm/arm/hyp/vgic-v3-sr.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/arm/include/asm/arch_gicv3.h b/arch/arm/include/asm/arch_gicv3.h
index 0bd530702118..bdc87700def2 100644
--- a/arch/arm/include/asm/arch_gicv3.h
+++ b/arch/arm/include/asm/arch_gicv3.h
@@ -54,7 +54,7 @@
#define ICH_VTR __ACCESS_CP15(c12, 4, c11, 1)
#define ICH_MISR __ACCESS_CP15(c12, 4, c11, 2)
#define ICH_EISR __ACCESS_CP15(c12, 4, c11, 3)
-#define ICH_ELSR __ACCESS_CP15(c12, 4, c11, 5)
+#define ICH_ELRSR __ACCESS_CP15(c12, 4, c11, 5)
#define ICH_VMCR __ACCESS_CP15(c12, 4, c11, 7)
#define __LR0(x) __ACCESS_CP15(c12, 4, c12, x)
@@ -151,7 +151,7 @@ CPUIF_MAP(ICH_HCR, ICH_HCR_EL2)
CPUIF_MAP(ICH_VTR, ICH_VTR_EL2)
CPUIF_MAP(ICH_MISR, ICH_MISR_EL2)
CPUIF_MAP(ICH_EISR, ICH_EISR_EL2)
-CPUIF_MAP(ICH_ELSR, ICH_ELSR_EL2)
+CPUIF_MAP(ICH_ELRSR, ICH_ELRSR_EL2)
CPUIF_MAP(ICH_VMCR, ICH_VMCR_EL2)
CPUIF_MAP(ICH_AP0R3, ICH_AP0R3_EL2)
CPUIF_MAP(ICH_AP0R2, ICH_AP0R2_EL2)
diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h
index 72dc4c011014..3e5650903d6d 100644
--- a/arch/arm64/include/asm/sysreg.h
+++ b/arch/arm64/include/asm/sysreg.h
@@ -426,7 +426,7 @@
#define SYS_ICH_VTR_EL2 sys_reg(3, 4, 12, 11, 1)
#define SYS_ICH_MISR_EL2 sys_reg(3, 4, 12, 11, 2)
#define SYS_ICH_EISR_EL2 sys_reg(3, 4, 12, 11, 3)
-#define SYS_ICH_ELSR_EL2 sys_reg(3, 4, 12, 11, 5)
+#define SYS_ICH_ELRSR_EL2 sys_reg(3, 4, 12, 11, 5)
#define SYS_ICH_VMCR_EL2 sys_reg(3, 4, 12, 11, 7)
#define __SYS__LR0_EL2(x) sys_reg(3, 4, 12, 12, x)
diff --git a/virt/kvm/arm/hyp/vgic-v3-sr.c b/virt/kvm/arm/hyp/vgic-v3-sr.c
index 9652c453480f..264d92da3240 100644
--- a/virt/kvm/arm/hyp/vgic-v3-sr.c
+++ b/virt/kvm/arm/hyp/vgic-v3-sr.c
@@ -226,7 +226,7 @@ void __hyp_text __vgic_v3_save_state(struct kvm_vcpu *vcpu)
int i;
u32 elrsr;
- elrsr = read_gicreg(ICH_ELSR_EL2);
+ elrsr = read_gicreg(ICH_ELRSR_EL2);
write_gicreg(cpu_if->vgic_hcr & ~ICH_HCR_EN, ICH_HCR_EL2);
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 10/27] KVM: arm64: Reuse sys_reg() macro when searching the trap table
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
Instead of having an open-coded macro, reuse the sys_reg() macro
that does the exact same thing (the encoding is slightly different,
but the ordering property is the same).
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm64/kvm/sys_regs.c | 19 ++++++-------------
1 file changed, 6 insertions(+), 13 deletions(-)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index e3e37228ae4e..1a5bea4285e4 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -965,6 +965,10 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
return true;
}
+#define reg_to_encoding(x) \
+ sys_reg((u32)(x)->Op0, (u32)(x)->Op1, \
+ (u32)(x)->CRn, (u32)(x)->CRm, (u32)(x)->Op2);
+
/* Silly macro to expand the DBG{BCR,BVR,WVR,WCR}n_EL1 registers in one go */
#define DBG_BCR_BVR_WCR_WVR_EL1(n) \
{ SYS_DESC(SYS_DBGBVRn_EL1(n)), \
@@ -1820,30 +1824,19 @@ static const struct sys_reg_desc *get_target_table(unsigned target,
}
}
-#define reg_to_match_value(x) \
- ({ \
- unsigned long val; \
- val = (x)->Op0 << 14; \
- val |= (x)->Op1 << 11; \
- val |= (x)->CRn << 7; \
- val |= (x)->CRm << 3; \
- val |= (x)->Op2; \
- val; \
- })
-
static int match_sys_reg(const void *key, const void *elt)
{
const unsigned long pval = (unsigned long)key;
const struct sys_reg_desc *r = elt;
- return pval - reg_to_match_value(r);
+ return pval - reg_to_encoding(r);
}
static const struct sys_reg_desc *find_reg(const struct sys_reg_params *params,
const struct sys_reg_desc table[],
unsigned int num)
{
- unsigned long pval = reg_to_match_value(params);
+ unsigned long pval = reg_to_encoding(params);
return bsearch((void *)pval, table, num, sizeof(table[0]), match_sys_reg);
}
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 10/27] KVM: arm64: Reuse sys_reg() macro when searching the trap table
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
Instead of having an open-coded macro, reuse the sys_reg() macro
that does the exact same thing (the encoding is slightly different,
but the ordering property is the same).
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm64/kvm/sys_regs.c | 19 ++++++-------------
1 file changed, 6 insertions(+), 13 deletions(-)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index e3e37228ae4e..1a5bea4285e4 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -965,6 +965,10 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
return true;
}
+#define reg_to_encoding(x) \
+ sys_reg((u32)(x)->Op0, (u32)(x)->Op1, \
+ (u32)(x)->CRn, (u32)(x)->CRm, (u32)(x)->Op2);
+
/* Silly macro to expand the DBG{BCR,BVR,WVR,WCR}n_EL1 registers in one go */
#define DBG_BCR_BVR_WCR_WVR_EL1(n) \
{ SYS_DESC(SYS_DBGBVRn_EL1(n)), \
@@ -1820,30 +1824,19 @@ static const struct sys_reg_desc *get_target_table(unsigned target,
}
}
-#define reg_to_match_value(x) \
- ({ \
- unsigned long val; \
- val = (x)->Op0 << 14; \
- val |= (x)->Op1 << 11; \
- val |= (x)->CRn << 7; \
- val |= (x)->CRm << 3; \
- val |= (x)->Op2; \
- val; \
- })
-
static int match_sys_reg(const void *key, const void *elt)
{
const unsigned long pval = (unsigned long)key;
const struct sys_reg_desc *r = elt;
- return pval - reg_to_match_value(r);
+ return pval - reg_to_encoding(r);
}
static const struct sys_reg_desc *find_reg(const struct sys_reg_params *params,
const struct sys_reg_desc table[],
unsigned int num)
{
- unsigned long pval = reg_to_match_value(params);
+ unsigned long pval = reg_to_encoding(params);
return bsearch((void *)pval, table, num, sizeof(table[0]), match_sys_reg);
}
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 11/27] KVM: arm/arm64: consolidate arch timer trap handlers
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Andre Przywara <andre.przywara@arm.com>
At the moment we have separate system register emulation handlers for
each timer register. Actually they are quite similar, and we rely on
kvm_arm_timer_[gs]et_reg() for the actual emulation anyways, so let's
just merge all of those handlers into one function, which just marshalls
the arguments and then hands off to a set of common accessors.
This makes extending the emulation to include EL2 timers much easier.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
[Fixed 32-bit VM breakage and reduced to reworking existing code]
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
[Fixed 32bit host, general cleanup]
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/kvm/coproc.c | 23 +++---
arch/arm64/include/asm/sysreg.h | 4 +
arch/arm64/kvm/sys_regs.c | 73 ++++++++----------
include/kvm/arm_arch_timer.h | 23 ++++++
virt/kvm/arm/arch_timer.c | 130 +++++++++++++++++++++++++++-----
5 files changed, 187 insertions(+), 66 deletions(-)
diff --git a/arch/arm/kvm/coproc.c b/arch/arm/kvm/coproc.c
index 222c1635bc7a..51863364f8d1 100644
--- a/arch/arm/kvm/coproc.c
+++ b/arch/arm/kvm/coproc.c
@@ -293,15 +293,16 @@ static bool access_cntp_tval(struct kvm_vcpu *vcpu,
const struct coproc_params *p,
const struct coproc_reg *r)
{
- u64 now = kvm_phys_timer_read();
- u64 val;
+ u32 val;
if (p->is_write) {
val = *vcpu_reg(vcpu, p->Rt1);
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL, val + now);
+ kvm_arm_timer_write_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_TVAL, val);
} else {
- val = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
- *vcpu_reg(vcpu, p->Rt1) = val - now;
+ val = kvm_arm_timer_read_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_TVAL);
+ *vcpu_reg(vcpu, p->Rt1) = val;
}
return true;
@@ -315,9 +316,11 @@ static bool access_cntp_ctl(struct kvm_vcpu *vcpu,
if (p->is_write) {
val = *vcpu_reg(vcpu, p->Rt1);
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CTL, val);
+ kvm_arm_timer_write_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CTL, val);
} else {
- val = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CTL);
+ val = kvm_arm_timer_read_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CTL);
*vcpu_reg(vcpu, p->Rt1) = val;
}
@@ -333,9 +336,11 @@ static bool access_cntp_cval(struct kvm_vcpu *vcpu,
if (p->is_write) {
val = (u64)*vcpu_reg(vcpu, p->Rt2) << 32;
val |= *vcpu_reg(vcpu, p->Rt1);
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL, val);
+ kvm_arm_timer_write_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CVAL, val);
} else {
- val = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
+ val = kvm_arm_timer_read_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CVAL);
*vcpu_reg(vcpu, p->Rt1) = val;
*vcpu_reg(vcpu, p->Rt2) = val >> 32;
}
diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h
index 3e5650903d6d..6482e8bcf1b8 100644
--- a/arch/arm64/include/asm/sysreg.h
+++ b/arch/arm64/include/asm/sysreg.h
@@ -392,6 +392,10 @@
#define SYS_CNTP_CTL_EL0 sys_reg(3, 3, 14, 2, 1)
#define SYS_CNTP_CVAL_EL0 sys_reg(3, 3, 14, 2, 2)
+#define SYS_AARCH32_CNTP_TVAL sys_reg(0, 0, 14, 2, 0)
+#define SYS_AARCH32_CNTP_CTL sys_reg(0, 0, 14, 2, 1)
+#define SYS_AARCH32_CNTP_CVAL sys_reg(0, 2, 0, 14, 0)
+
#define __PMEV_op2(n) ((n) & 0x7)
#define __CNTR_CRm(n) (0x8 | (((n) >> 3) & 0x3))
#define SYS_PMEVCNTRn_EL0(n) sys_reg(3, 3, 14, __CNTR_CRm(n), __PMEV_op2(n))
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 1a5bea4285e4..0d348500b24b 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -990,44 +990,38 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
{ SYS_DESC(SYS_PMEVTYPERn_EL0(n)), \
access_pmu_evtyper, reset_unknown, (PMEVTYPER0_EL0 + n), }
-static bool access_cntp_tval(struct kvm_vcpu *vcpu,
- struct sys_reg_params *p,
- const struct sys_reg_desc *r)
+static bool access_arch_timer(struct kvm_vcpu *vcpu,
+ struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
{
- u64 now = kvm_phys_timer_read();
- u64 cval;
+ enum kvm_arch_timers tmr;
+ enum kvm_arch_timer_regs treg;
+ u64 reg = reg_to_encoding(r);
- if (p->is_write) {
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL,
- p->regval + now);
- } else {
- cval = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
- p->regval = cval - now;
+ switch (reg) {
+ case SYS_CNTP_TVAL_EL0:
+ case SYS_AARCH32_CNTP_TVAL:
+ tmr = TIMER_PTIMER;
+ treg = TIMER_REG_TVAL;
+ break;
+ case SYS_CNTP_CTL_EL0:
+ case SYS_AARCH32_CNTP_CTL:
+ tmr = TIMER_PTIMER;
+ treg = TIMER_REG_CTL;
+ break;
+ case SYS_CNTP_CVAL_EL0:
+ case SYS_AARCH32_CNTP_CVAL:
+ tmr = TIMER_PTIMER;
+ treg = TIMER_REG_CVAL;
+ break;
+ default:
+ BUG();
}
- return true;
-}
-
-static bool access_cntp_ctl(struct kvm_vcpu *vcpu,
- struct sys_reg_params *p,
- const struct sys_reg_desc *r)
-{
- if (p->is_write)
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CTL, p->regval);
- else
- p->regval = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CTL);
-
- return true;
-}
-
-static bool access_cntp_cval(struct kvm_vcpu *vcpu,
- struct sys_reg_params *p,
- const struct sys_reg_desc *r)
-{
if (p->is_write)
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL, p->regval);
+ kvm_arm_timer_write_sysreg(vcpu, tmr, treg, p->regval);
else
- p->regval = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
+ p->regval = kvm_arm_timer_read_sysreg(vcpu, tmr, treg);
return true;
}
@@ -1392,9 +1386,9 @@ static const struct sys_reg_desc sys_reg_descs[] = {
{ SYS_DESC(SYS_TPIDR_EL0), NULL, reset_unknown, TPIDR_EL0 },
{ SYS_DESC(SYS_TPIDRRO_EL0), NULL, reset_unknown, TPIDRRO_EL0 },
- { SYS_DESC(SYS_CNTP_TVAL_EL0), access_cntp_tval },
- { SYS_DESC(SYS_CNTP_CTL_EL0), access_cntp_ctl },
- { SYS_DESC(SYS_CNTP_CVAL_EL0), access_cntp_cval },
+ { SYS_DESC(SYS_CNTP_TVAL_EL0), access_arch_timer },
+ { SYS_DESC(SYS_CNTP_CTL_EL0), access_arch_timer },
+ { SYS_DESC(SYS_CNTP_CVAL_EL0), access_arch_timer },
/* PMEVCNTRn_EL0 */
PMU_PMEVCNTR_EL0(0),
@@ -1715,10 +1709,9 @@ static const struct sys_reg_desc cp15_regs[] = {
{ Op1( 0), CRn(13), CRm( 0), Op2( 1), access_vm_reg, NULL, c13_CID },
- /* CNTP_TVAL */
- { Op1( 0), CRn(14), CRm( 2), Op2( 0), access_cntp_tval },
- /* CNTP_CTL */
- { Op1( 0), CRn(14), CRm( 2), Op2( 1), access_cntp_ctl },
+ /* Arch Tmers */
+ { SYS_DESC(SYS_AARCH32_CNTP_TVAL), access_arch_timer },
+ { SYS_DESC(SYS_AARCH32_CNTP_CTL), access_arch_timer },
/* PMEVCNTRn */
PMU_PMEVCNTR(0),
@@ -1795,7 +1788,7 @@ static const struct sys_reg_desc cp15_64_regs[] = {
{ Op1( 1), CRn( 0), CRm( 2), Op2( 0), access_vm_reg, NULL, c2_TTBR1 },
{ Op1( 1), CRn( 0), CRm(12), Op2( 0), access_gic_sgi }, /* ICC_ASGI1R */
{ Op1( 2), CRn( 0), CRm(12), Op2( 0), access_gic_sgi }, /* ICC_SGI0R */
- { Op1( 2), CRn( 0), CRm(14), Op2( 0), access_cntp_cval },
+ { SYS_DESC(SYS_AARCH32_CNTP_CVAL), access_arch_timer },
};
/* Target specific emulation tables */
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index d6e6a45d1d24..d26b7fde9935 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -22,6 +22,19 @@
#include <linux/clocksource.h>
#include <linux/hrtimer.h>
+enum kvm_arch_timers {
+ TIMER_PTIMER,
+ TIMER_VTIMER,
+ NR_KVM_TIMERS
+};
+
+enum kvm_arch_timer_regs {
+ TIMER_REG_CNT,
+ TIMER_REG_CVAL,
+ TIMER_REG_TVAL,
+ TIMER_REG_CTL,
+};
+
struct arch_timer_context {
/* Registers: control register, timer value */
u32 cnt_ctl;
@@ -87,5 +100,15 @@ bool kvm_arch_timer_get_input_level(int vintid);
#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.vtimer)
#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.ptimer)
+#define vcpu_get_timer(v,t) \
+ (t == TIMER_VTIMER ? vcpu_vtimer(v) : vcpu_ptimer(v))
+
+u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg);
+void kvm_arm_timer_write_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg,
+ u64 val);
#endif
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 4986028d9829..f7d377448438 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -25,6 +25,7 @@
#include <clocksource/arm_arch_timer.h>
#include <asm/arch_timer.h>
+#include <asm/kvm_emulate.h>
#include <asm/kvm_hyp.h>
#include <kvm/arm_vgic.h>
@@ -52,6 +53,13 @@ static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx);
static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level,
struct arch_timer_context *timer_ctx);
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx);
+static void kvm_arm_timer_write(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg,
+ u64 val);
+static u64 kvm_arm_timer_read(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg);
u64 kvm_phys_timer_read(void)
{
@@ -628,24 +636,25 @@ static void kvm_timer_init_interrupt(void *info)
int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value)
{
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
-
switch (regid) {
case KVM_REG_ARM_TIMER_CTL:
- vtimer->cnt_ctl = value & ~ARCH_TIMER_CTRL_IT_STAT;
+ kvm_arm_timer_write(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_TIMER_CNT:
update_vtimer_cntvoff(vcpu, kvm_phys_timer_read() - value);
break;
case KVM_REG_ARM_TIMER_CVAL:
- vtimer->cnt_cval = value;
+ kvm_arm_timer_write(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CVAL, value);
break;
case KVM_REG_ARM_PTIMER_CTL:
- ptimer->cnt_ctl = value & ~ARCH_TIMER_CTRL_IT_STAT;
+ kvm_arm_timer_write(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_PTIMER_CVAL:
- ptimer->cnt_cval = value;
+ kvm_arm_timer_write(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CVAL, value);
break;
default:
@@ -672,26 +681,113 @@ static u64 read_timer_ctl(struct arch_timer_context *timer)
u64 kvm_arm_timer_get_reg(struct kvm_vcpu *vcpu, u64 regid)
{
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
-
switch (regid) {
case KVM_REG_ARM_TIMER_CTL:
- return read_timer_ctl(vtimer);
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CTL);
case KVM_REG_ARM_TIMER_CNT:
- return kvm_phys_timer_read() - vtimer->cntvoff;
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CNT);
case KVM_REG_ARM_TIMER_CVAL:
- return vtimer->cnt_cval;
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CVAL);
case KVM_REG_ARM_PTIMER_CTL:
- return read_timer_ctl(ptimer);
- case KVM_REG_ARM_PTIMER_CVAL:
- return ptimer->cnt_cval;
+ return kvm_arm_timer_read(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CTL);
case KVM_REG_ARM_PTIMER_CNT:
- return kvm_phys_timer_read();
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CNT);
+ case KVM_REG_ARM_PTIMER_CVAL:
+ return kvm_arm_timer_read(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CVAL);
}
return (u64)-1;
}
+static u64 kvm_arm_timer_read(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg)
+{
+ u64 val;
+
+ switch (treg) {
+ case TIMER_REG_TVAL:
+ val = kvm_phys_timer_read() - timer->cntvoff - timer->cnt_cval;
+ break;
+
+ case TIMER_REG_CTL:
+ val = read_timer_ctl(timer);
+ break;
+
+ case TIMER_REG_CVAL:
+ val = timer->cnt_cval;
+ break;
+
+ case TIMER_REG_CNT:
+ val = kvm_phys_timer_read() - timer->cntvoff;
+ break;
+
+ default:
+ BUG();
+ }
+
+ return val;
+}
+
+u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg)
+{
+ u64 val;
+
+ preempt_disable();
+ kvm_timer_vcpu_put(vcpu);
+
+ val = kvm_arm_timer_read(vcpu, vcpu_get_timer(vcpu, tmr), treg);
+
+ kvm_timer_vcpu_load(vcpu);
+ preempt_enable();
+
+ return val;
+}
+
+static void kvm_arm_timer_write(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg,
+ u64 val)
+{
+ switch (treg) {
+ case TIMER_REG_TVAL:
+ timer->cnt_cval = val - kvm_phys_timer_read() - timer->cntvoff;
+ break;
+
+ case TIMER_REG_CTL:
+ timer->cnt_ctl = val & ~ARCH_TIMER_CTRL_IT_STAT;
+ break;
+
+ case TIMER_REG_CVAL:
+ timer->cnt_cval = val;
+ break;
+
+ default:
+ BUG();
+ }
+}
+
+void kvm_arm_timer_write_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg,
+ u64 val)
+{
+ preempt_disable();
+ kvm_timer_vcpu_put(vcpu);
+
+ kvm_arm_timer_write(vcpu, vcpu_get_timer(vcpu, tmr), treg, val);
+
+ kvm_timer_vcpu_load(vcpu);
+ preempt_enable();
+}
+
static int kvm_timer_starting_cpu(unsigned int cpu)
{
kvm_timer_init_interrupt(NULL);
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 11/27] KVM: arm/arm64: consolidate arch timer trap handlers
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Andre Przywara <andre.przywara@arm.com>
At the moment we have separate system register emulation handlers for
each timer register. Actually they are quite similar, and we rely on
kvm_arm_timer_[gs]et_reg() for the actual emulation anyways, so let's
just merge all of those handlers into one function, which just marshalls
the arguments and then hands off to a set of common accessors.
This makes extending the emulation to include EL2 timers much easier.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
[Fixed 32-bit VM breakage and reduced to reworking existing code]
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
[Fixed 32bit host, general cleanup]
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/kvm/coproc.c | 23 +++---
arch/arm64/include/asm/sysreg.h | 4 +
arch/arm64/kvm/sys_regs.c | 73 ++++++++----------
include/kvm/arm_arch_timer.h | 23 ++++++
virt/kvm/arm/arch_timer.c | 130 +++++++++++++++++++++++++++-----
5 files changed, 187 insertions(+), 66 deletions(-)
diff --git a/arch/arm/kvm/coproc.c b/arch/arm/kvm/coproc.c
index 222c1635bc7a..51863364f8d1 100644
--- a/arch/arm/kvm/coproc.c
+++ b/arch/arm/kvm/coproc.c
@@ -293,15 +293,16 @@ static bool access_cntp_tval(struct kvm_vcpu *vcpu,
const struct coproc_params *p,
const struct coproc_reg *r)
{
- u64 now = kvm_phys_timer_read();
- u64 val;
+ u32 val;
if (p->is_write) {
val = *vcpu_reg(vcpu, p->Rt1);
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL, val + now);
+ kvm_arm_timer_write_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_TVAL, val);
} else {
- val = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
- *vcpu_reg(vcpu, p->Rt1) = val - now;
+ val = kvm_arm_timer_read_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_TVAL);
+ *vcpu_reg(vcpu, p->Rt1) = val;
}
return true;
@@ -315,9 +316,11 @@ static bool access_cntp_ctl(struct kvm_vcpu *vcpu,
if (p->is_write) {
val = *vcpu_reg(vcpu, p->Rt1);
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CTL, val);
+ kvm_arm_timer_write_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CTL, val);
} else {
- val = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CTL);
+ val = kvm_arm_timer_read_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CTL);
*vcpu_reg(vcpu, p->Rt1) = val;
}
@@ -333,9 +336,11 @@ static bool access_cntp_cval(struct kvm_vcpu *vcpu,
if (p->is_write) {
val = (u64)*vcpu_reg(vcpu, p->Rt2) << 32;
val |= *vcpu_reg(vcpu, p->Rt1);
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL, val);
+ kvm_arm_timer_write_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CVAL, val);
} else {
- val = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
+ val = kvm_arm_timer_read_sysreg(vcpu,
+ TIMER_PTIMER, TIMER_REG_CVAL);
*vcpu_reg(vcpu, p->Rt1) = val;
*vcpu_reg(vcpu, p->Rt2) = val >> 32;
}
diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h
index 3e5650903d6d..6482e8bcf1b8 100644
--- a/arch/arm64/include/asm/sysreg.h
+++ b/arch/arm64/include/asm/sysreg.h
@@ -392,6 +392,10 @@
#define SYS_CNTP_CTL_EL0 sys_reg(3, 3, 14, 2, 1)
#define SYS_CNTP_CVAL_EL0 sys_reg(3, 3, 14, 2, 2)
+#define SYS_AARCH32_CNTP_TVAL sys_reg(0, 0, 14, 2, 0)
+#define SYS_AARCH32_CNTP_CTL sys_reg(0, 0, 14, 2, 1)
+#define SYS_AARCH32_CNTP_CVAL sys_reg(0, 2, 0, 14, 0)
+
#define __PMEV_op2(n) ((n) & 0x7)
#define __CNTR_CRm(n) (0x8 | (((n) >> 3) & 0x3))
#define SYS_PMEVCNTRn_EL0(n) sys_reg(3, 3, 14, __CNTR_CRm(n), __PMEV_op2(n))
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 1a5bea4285e4..0d348500b24b 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -990,44 +990,38 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
{ SYS_DESC(SYS_PMEVTYPERn_EL0(n)), \
access_pmu_evtyper, reset_unknown, (PMEVTYPER0_EL0 + n), }
-static bool access_cntp_tval(struct kvm_vcpu *vcpu,
- struct sys_reg_params *p,
- const struct sys_reg_desc *r)
+static bool access_arch_timer(struct kvm_vcpu *vcpu,
+ struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
{
- u64 now = kvm_phys_timer_read();
- u64 cval;
+ enum kvm_arch_timers tmr;
+ enum kvm_arch_timer_regs treg;
+ u64 reg = reg_to_encoding(r);
- if (p->is_write) {
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL,
- p->regval + now);
- } else {
- cval = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
- p->regval = cval - now;
+ switch (reg) {
+ case SYS_CNTP_TVAL_EL0:
+ case SYS_AARCH32_CNTP_TVAL:
+ tmr = TIMER_PTIMER;
+ treg = TIMER_REG_TVAL;
+ break;
+ case SYS_CNTP_CTL_EL0:
+ case SYS_AARCH32_CNTP_CTL:
+ tmr = TIMER_PTIMER;
+ treg = TIMER_REG_CTL;
+ break;
+ case SYS_CNTP_CVAL_EL0:
+ case SYS_AARCH32_CNTP_CVAL:
+ tmr = TIMER_PTIMER;
+ treg = TIMER_REG_CVAL;
+ break;
+ default:
+ BUG();
}
- return true;
-}
-
-static bool access_cntp_ctl(struct kvm_vcpu *vcpu,
- struct sys_reg_params *p,
- const struct sys_reg_desc *r)
-{
- if (p->is_write)
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CTL, p->regval);
- else
- p->regval = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CTL);
-
- return true;
-}
-
-static bool access_cntp_cval(struct kvm_vcpu *vcpu,
- struct sys_reg_params *p,
- const struct sys_reg_desc *r)
-{
if (p->is_write)
- kvm_arm_timer_set_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL, p->regval);
+ kvm_arm_timer_write_sysreg(vcpu, tmr, treg, p->regval);
else
- p->regval = kvm_arm_timer_get_reg(vcpu, KVM_REG_ARM_PTIMER_CVAL);
+ p->regval = kvm_arm_timer_read_sysreg(vcpu, tmr, treg);
return true;
}
@@ -1392,9 +1386,9 @@ static const struct sys_reg_desc sys_reg_descs[] = {
{ SYS_DESC(SYS_TPIDR_EL0), NULL, reset_unknown, TPIDR_EL0 },
{ SYS_DESC(SYS_TPIDRRO_EL0), NULL, reset_unknown, TPIDRRO_EL0 },
- { SYS_DESC(SYS_CNTP_TVAL_EL0), access_cntp_tval },
- { SYS_DESC(SYS_CNTP_CTL_EL0), access_cntp_ctl },
- { SYS_DESC(SYS_CNTP_CVAL_EL0), access_cntp_cval },
+ { SYS_DESC(SYS_CNTP_TVAL_EL0), access_arch_timer },
+ { SYS_DESC(SYS_CNTP_CTL_EL0), access_arch_timer },
+ { SYS_DESC(SYS_CNTP_CVAL_EL0), access_arch_timer },
/* PMEVCNTRn_EL0 */
PMU_PMEVCNTR_EL0(0),
@@ -1715,10 +1709,9 @@ static const struct sys_reg_desc cp15_regs[] = {
{ Op1( 0), CRn(13), CRm( 0), Op2( 1), access_vm_reg, NULL, c13_CID },
- /* CNTP_TVAL */
- { Op1( 0), CRn(14), CRm( 2), Op2( 0), access_cntp_tval },
- /* CNTP_CTL */
- { Op1( 0), CRn(14), CRm( 2), Op2( 1), access_cntp_ctl },
+ /* Arch Tmers */
+ { SYS_DESC(SYS_AARCH32_CNTP_TVAL), access_arch_timer },
+ { SYS_DESC(SYS_AARCH32_CNTP_CTL), access_arch_timer },
/* PMEVCNTRn */
PMU_PMEVCNTR(0),
@@ -1795,7 +1788,7 @@ static const struct sys_reg_desc cp15_64_regs[] = {
{ Op1( 1), CRn( 0), CRm( 2), Op2( 0), access_vm_reg, NULL, c2_TTBR1 },
{ Op1( 1), CRn( 0), CRm(12), Op2( 0), access_gic_sgi }, /* ICC_ASGI1R */
{ Op1( 2), CRn( 0), CRm(12), Op2( 0), access_gic_sgi }, /* ICC_SGI0R */
- { Op1( 2), CRn( 0), CRm(14), Op2( 0), access_cntp_cval },
+ { SYS_DESC(SYS_AARCH32_CNTP_CVAL), access_arch_timer },
};
/* Target specific emulation tables */
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index d6e6a45d1d24..d26b7fde9935 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -22,6 +22,19 @@
#include <linux/clocksource.h>
#include <linux/hrtimer.h>
+enum kvm_arch_timers {
+ TIMER_PTIMER,
+ TIMER_VTIMER,
+ NR_KVM_TIMERS
+};
+
+enum kvm_arch_timer_regs {
+ TIMER_REG_CNT,
+ TIMER_REG_CVAL,
+ TIMER_REG_TVAL,
+ TIMER_REG_CTL,
+};
+
struct arch_timer_context {
/* Registers: control register, timer value */
u32 cnt_ctl;
@@ -87,5 +100,15 @@ bool kvm_arch_timer_get_input_level(int vintid);
#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.vtimer)
#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.ptimer)
+#define vcpu_get_timer(v,t) \
+ (t == TIMER_VTIMER ? vcpu_vtimer(v) : vcpu_ptimer(v))
+
+u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg);
+void kvm_arm_timer_write_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg,
+ u64 val);
#endif
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 4986028d9829..f7d377448438 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -25,6 +25,7 @@
#include <clocksource/arm_arch_timer.h>
#include <asm/arch_timer.h>
+#include <asm/kvm_emulate.h>
#include <asm/kvm_hyp.h>
#include <kvm/arm_vgic.h>
@@ -52,6 +53,13 @@ static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx);
static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level,
struct arch_timer_context *timer_ctx);
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx);
+static void kvm_arm_timer_write(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg,
+ u64 val);
+static u64 kvm_arm_timer_read(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg);
u64 kvm_phys_timer_read(void)
{
@@ -628,24 +636,25 @@ static void kvm_timer_init_interrupt(void *info)
int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value)
{
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
-
switch (regid) {
case KVM_REG_ARM_TIMER_CTL:
- vtimer->cnt_ctl = value & ~ARCH_TIMER_CTRL_IT_STAT;
+ kvm_arm_timer_write(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_TIMER_CNT:
update_vtimer_cntvoff(vcpu, kvm_phys_timer_read() - value);
break;
case KVM_REG_ARM_TIMER_CVAL:
- vtimer->cnt_cval = value;
+ kvm_arm_timer_write(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CVAL, value);
break;
case KVM_REG_ARM_PTIMER_CTL:
- ptimer->cnt_ctl = value & ~ARCH_TIMER_CTRL_IT_STAT;
+ kvm_arm_timer_write(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_PTIMER_CVAL:
- ptimer->cnt_cval = value;
+ kvm_arm_timer_write(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CVAL, value);
break;
default:
@@ -672,26 +681,113 @@ static u64 read_timer_ctl(struct arch_timer_context *timer)
u64 kvm_arm_timer_get_reg(struct kvm_vcpu *vcpu, u64 regid)
{
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
-
switch (regid) {
case KVM_REG_ARM_TIMER_CTL:
- return read_timer_ctl(vtimer);
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CTL);
case KVM_REG_ARM_TIMER_CNT:
- return kvm_phys_timer_read() - vtimer->cntvoff;
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CNT);
case KVM_REG_ARM_TIMER_CVAL:
- return vtimer->cnt_cval;
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CVAL);
case KVM_REG_ARM_PTIMER_CTL:
- return read_timer_ctl(ptimer);
- case KVM_REG_ARM_PTIMER_CVAL:
- return ptimer->cnt_cval;
+ return kvm_arm_timer_read(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CTL);
case KVM_REG_ARM_PTIMER_CNT:
- return kvm_phys_timer_read();
+ return kvm_arm_timer_read(vcpu,
+ vcpu_vtimer(vcpu), TIMER_REG_CNT);
+ case KVM_REG_ARM_PTIMER_CVAL:
+ return kvm_arm_timer_read(vcpu,
+ vcpu_ptimer(vcpu), TIMER_REG_CVAL);
}
return (u64)-1;
}
+static u64 kvm_arm_timer_read(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg)
+{
+ u64 val;
+
+ switch (treg) {
+ case TIMER_REG_TVAL:
+ val = kvm_phys_timer_read() - timer->cntvoff - timer->cnt_cval;
+ break;
+
+ case TIMER_REG_CTL:
+ val = read_timer_ctl(timer);
+ break;
+
+ case TIMER_REG_CVAL:
+ val = timer->cnt_cval;
+ break;
+
+ case TIMER_REG_CNT:
+ val = kvm_phys_timer_read() - timer->cntvoff;
+ break;
+
+ default:
+ BUG();
+ }
+
+ return val;
+}
+
+u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg)
+{
+ u64 val;
+
+ preempt_disable();
+ kvm_timer_vcpu_put(vcpu);
+
+ val = kvm_arm_timer_read(vcpu, vcpu_get_timer(vcpu, tmr), treg);
+
+ kvm_timer_vcpu_load(vcpu);
+ preempt_enable();
+
+ return val;
+}
+
+static void kvm_arm_timer_write(struct kvm_vcpu *vcpu,
+ struct arch_timer_context *timer,
+ enum kvm_arch_timer_regs treg,
+ u64 val)
+{
+ switch (treg) {
+ case TIMER_REG_TVAL:
+ timer->cnt_cval = val - kvm_phys_timer_read() - timer->cntvoff;
+ break;
+
+ case TIMER_REG_CTL:
+ timer->cnt_ctl = val & ~ARCH_TIMER_CTRL_IT_STAT;
+ break;
+
+ case TIMER_REG_CVAL:
+ timer->cnt_cval = val;
+ break;
+
+ default:
+ BUG();
+ }
+}
+
+void kvm_arm_timer_write_sysreg(struct kvm_vcpu *vcpu,
+ enum kvm_arch_timers tmr,
+ enum kvm_arch_timer_regs treg,
+ u64 val)
+{
+ preempt_disable();
+ kvm_timer_vcpu_put(vcpu);
+
+ kvm_arm_timer_write(vcpu, vcpu_get_timer(vcpu, tmr), treg, val);
+
+ kvm_timer_vcpu_load(vcpu);
+ preempt_enable();
+}
+
static int kvm_timer_starting_cpu(unsigned int cpu)
{
kvm_timer_init_interrupt(NULL);
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 12/27] KVM: arm/arm64: timer: Rework data structures for multiple timers
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Christoffer Dall <christoffer.dall@arm.com>
Prepare for having 4 timer data structures (2 for now).
Move loaded to the cpu data structure and not the individual timer
structure, in preparation for assigning the EL1 phys timer as well.
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
include/kvm/arm_arch_timer.h | 41 ++++++++++++-------------
virt/kvm/arm/arch_timer.c | 58 +++++++++++++++++++-----------------
2 files changed, 51 insertions(+), 48 deletions(-)
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index d26b7fde9935..ab835112204d 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -36,6 +36,8 @@ enum kvm_arch_timer_regs {
};
struct arch_timer_context {
+ struct kvm_vcpu *vcpu;
+
/* Registers: control register, timer value */
u32 cnt_ctl;
u64 cnt_cval;
@@ -43,32 +45,31 @@ struct arch_timer_context {
/* Timer IRQ */
struct kvm_irq_level irq;
- /*
- * We have multiple paths which can save/restore the timer state
- * onto the hardware, so we need some way of keeping track of
- * where the latest state is.
- *
- * loaded == true: State is loaded on the hardware registers.
- * loaded == false: State is stored in memory.
- */
- bool loaded;
-
/* Virtual offset */
- u64 cntvoff;
+ u64 cntvoff;
+
+ /* Emulated Timer (may be unused) */
+ struct hrtimer hrtimer;
};
struct arch_timer_cpu {
- struct arch_timer_context vtimer;
- struct arch_timer_context ptimer;
+ struct arch_timer_context timers[NR_KVM_TIMERS];
/* Background timer used when the guest is not running */
struct hrtimer bg_timer;
- /* Physical timer emulation */
- struct hrtimer phys_timer;
-
/* Is the timer enabled */
bool enabled;
+
+ /*
+ * We have multiple paths which can save/restore the timer state
+ * onto the hardware, so we need some way of keeping track of
+ * where the latest state is.
+ *
+ * loaded == true: State is loaded on the hardware registers.
+ * loaded == false: State is stored in memory.
+ */
+ bool loaded;
};
int kvm_timer_hyp_init(bool);
@@ -98,10 +99,10 @@ void kvm_timer_init_vhe(void);
bool kvm_arch_timer_get_input_level(int vintid);
-#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.vtimer)
-#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.ptimer)
-#define vcpu_get_timer(v,t) \
- (t == TIMER_VTIMER ? vcpu_vtimer(v) : vcpu_ptimer(v))
+#define vcpu_timer(v) (&(v)->arch.timer_cpu)
+#define vcpu_get_timer(v,t) (&vcpu_timer(v)->timers[(t)])
+#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.timers[TIMER_VTIMER])
+#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.timers[TIMER_PTIMER])
u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
enum kvm_arch_timers tmr,
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index f7d377448438..471f9fd004c9 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -184,13 +184,11 @@ static enum hrtimer_restart kvm_bg_timer_expire(struct hrtimer *hrt)
static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
{
struct arch_timer_context *ptimer;
- struct arch_timer_cpu *timer;
struct kvm_vcpu *vcpu;
u64 ns;
- timer = container_of(hrt, struct arch_timer_cpu, phys_timer);
- vcpu = container_of(timer, struct kvm_vcpu, arch.timer_cpu);
- ptimer = vcpu_ptimer(vcpu);
+ ptimer = container_of(hrt, struct arch_timer_context, hrtimer);
+ vcpu = ptimer->vcpu;
/*
* Check that the timer has really expired from the guest's
@@ -209,9 +207,10 @@ static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
+ struct arch_timer_cpu *timer = vcpu_timer(timer_ctx->vcpu);
u64 cval, now;
- if (timer_ctx->loaded) {
+ if (timer->loaded) {
u32 cnt_ctl;
/* Only the virtual timer can be loaded so far */
@@ -280,7 +279,6 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level,
/* Schedule the background timer for the emulated timer. */
static void phys_timer_emulate(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
/*
@@ -289,11 +287,11 @@ static void phys_timer_emulate(struct kvm_vcpu *vcpu)
* then we also don't need a soft timer.
*/
if (kvm_timer_should_fire(ptimer) || !kvm_timer_irq_can_fire(ptimer)) {
- soft_timer_cancel(&timer->phys_timer);
+ soft_timer_cancel(&ptimer->hrtimer);
return;
}
- soft_timer_start(&timer->phys_timer, kvm_timer_compute_delta(ptimer));
+ soft_timer_start(&ptimer->hrtimer, kvm_timer_compute_delta(ptimer));
}
/*
@@ -303,7 +301,7 @@ static void phys_timer_emulate(struct kvm_vcpu *vcpu)
*/
static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
bool level;
@@ -329,13 +327,13 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
static void vtimer_save_state(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
unsigned long flags;
local_irq_save(flags);
- if (!vtimer->loaded)
+ if (!timer->loaded)
goto out;
if (timer->enabled) {
@@ -347,7 +345,7 @@ static void vtimer_save_state(struct kvm_vcpu *vcpu)
write_sysreg_el0(0, cntv_ctl);
isb();
- vtimer->loaded = false;
+ timer->loaded = false;
out:
local_irq_restore(flags);
}
@@ -359,7 +357,7 @@ static void vtimer_save_state(struct kvm_vcpu *vcpu)
*/
static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
@@ -379,20 +377,20 @@ static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
static void kvm_timer_unblocking(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
soft_timer_cancel(&timer->bg_timer);
}
static void vtimer_restore_state(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
unsigned long flags;
local_irq_save(flags);
- if (vtimer->loaded)
+ if (timer->loaded)
goto out;
if (timer->enabled) {
@@ -401,7 +399,7 @@ static void vtimer_restore_state(struct kvm_vcpu *vcpu)
write_sysreg_el0(vtimer->cnt_ctl, cntv_ctl);
}
- vtimer->loaded = true;
+ timer->loaded = true;
out:
local_irq_restore(flags);
}
@@ -462,7 +460,7 @@ static void kvm_timer_vcpu_load_nogic(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
@@ -507,7 +505,8 @@ bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
+ struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
if (unlikely(!timer->enabled))
return;
@@ -523,7 +522,7 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
* In any case, we re-schedule the hrtimer for the physical timer when
* coming back to the VCPU thread in kvm_timer_vcpu_load().
*/
- soft_timer_cancel(&timer->phys_timer);
+ soft_timer_cancel(&ptimer->hrtimer);
if (swait_active(kvm_arch_vcpu_wq(vcpu)))
kvm_timer_blocking(vcpu);
@@ -559,7 +558,7 @@ static void unmask_vtimer_irq_user(struct kvm_vcpu *vcpu)
void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
if (unlikely(!timer->enabled))
return;
@@ -570,7 +569,7 @@ void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu)
int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
@@ -611,22 +610,25 @@ static void update_vtimer_cntvoff(struct kvm_vcpu *vcpu, u64 cntvoff)
void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
/* Synchronize cntvoff across all vtimers of a VM. */
update_vtimer_cntvoff(vcpu, kvm_phys_timer_read());
- vcpu_ptimer(vcpu)->cntvoff = 0;
+ ptimer->cntvoff = 0;
hrtimer_init(&timer->bg_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
timer->bg_timer.function = kvm_bg_timer_expire;
- hrtimer_init(&timer->phys_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
- timer->phys_timer.function = kvm_phys_timer_expire;
+ hrtimer_init(&ptimer->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+ ptimer->hrtimer.function = kvm_phys_timer_expire;
vtimer->irq.irq = default_vtimer_irq.irq;
ptimer->irq.irq = default_ptimer_irq.irq;
+
+ vtimer->vcpu = vcpu;
+ ptimer->vcpu = vcpu;
}
static void kvm_timer_init_interrupt(void *info)
@@ -860,7 +862,7 @@ int kvm_timer_hyp_init(bool has_gic)
void kvm_timer_vcpu_terminate(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
soft_timer_cancel(&timer->bg_timer);
}
@@ -904,7 +906,7 @@ bool kvm_arch_timer_get_input_level(int vintid)
int kvm_timer_enable(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
int ret;
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 12/27] KVM: arm/arm64: timer: Rework data structures for multiple timers
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Christoffer Dall <christoffer.dall@arm.com>
Prepare for having 4 timer data structures (2 for now).
Move loaded to the cpu data structure and not the individual timer
structure, in preparation for assigning the EL1 phys timer as well.
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
include/kvm/arm_arch_timer.h | 41 ++++++++++++-------------
virt/kvm/arm/arch_timer.c | 58 +++++++++++++++++++-----------------
2 files changed, 51 insertions(+), 48 deletions(-)
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index d26b7fde9935..ab835112204d 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -36,6 +36,8 @@ enum kvm_arch_timer_regs {
};
struct arch_timer_context {
+ struct kvm_vcpu *vcpu;
+
/* Registers: control register, timer value */
u32 cnt_ctl;
u64 cnt_cval;
@@ -43,32 +45,31 @@ struct arch_timer_context {
/* Timer IRQ */
struct kvm_irq_level irq;
- /*
- * We have multiple paths which can save/restore the timer state
- * onto the hardware, so we need some way of keeping track of
- * where the latest state is.
- *
- * loaded == true: State is loaded on the hardware registers.
- * loaded == false: State is stored in memory.
- */
- bool loaded;
-
/* Virtual offset */
- u64 cntvoff;
+ u64 cntvoff;
+
+ /* Emulated Timer (may be unused) */
+ struct hrtimer hrtimer;
};
struct arch_timer_cpu {
- struct arch_timer_context vtimer;
- struct arch_timer_context ptimer;
+ struct arch_timer_context timers[NR_KVM_TIMERS];
/* Background timer used when the guest is not running */
struct hrtimer bg_timer;
- /* Physical timer emulation */
- struct hrtimer phys_timer;
-
/* Is the timer enabled */
bool enabled;
+
+ /*
+ * We have multiple paths which can save/restore the timer state
+ * onto the hardware, so we need some way of keeping track of
+ * where the latest state is.
+ *
+ * loaded == true: State is loaded on the hardware registers.
+ * loaded == false: State is stored in memory.
+ */
+ bool loaded;
};
int kvm_timer_hyp_init(bool);
@@ -98,10 +99,10 @@ void kvm_timer_init_vhe(void);
bool kvm_arch_timer_get_input_level(int vintid);
-#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.vtimer)
-#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.ptimer)
-#define vcpu_get_timer(v,t) \
- (t == TIMER_VTIMER ? vcpu_vtimer(v) : vcpu_ptimer(v))
+#define vcpu_timer(v) (&(v)->arch.timer_cpu)
+#define vcpu_get_timer(v,t) (&vcpu_timer(v)->timers[(t)])
+#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.timers[TIMER_VTIMER])
+#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.timers[TIMER_PTIMER])
u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
enum kvm_arch_timers tmr,
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index f7d377448438..471f9fd004c9 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -184,13 +184,11 @@ static enum hrtimer_restart kvm_bg_timer_expire(struct hrtimer *hrt)
static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
{
struct arch_timer_context *ptimer;
- struct arch_timer_cpu *timer;
struct kvm_vcpu *vcpu;
u64 ns;
- timer = container_of(hrt, struct arch_timer_cpu, phys_timer);
- vcpu = container_of(timer, struct kvm_vcpu, arch.timer_cpu);
- ptimer = vcpu_ptimer(vcpu);
+ ptimer = container_of(hrt, struct arch_timer_context, hrtimer);
+ vcpu = ptimer->vcpu;
/*
* Check that the timer has really expired from the guest's
@@ -209,9 +207,10 @@ static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
+ struct arch_timer_cpu *timer = vcpu_timer(timer_ctx->vcpu);
u64 cval, now;
- if (timer_ctx->loaded) {
+ if (timer->loaded) {
u32 cnt_ctl;
/* Only the virtual timer can be loaded so far */
@@ -280,7 +279,6 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level,
/* Schedule the background timer for the emulated timer. */
static void phys_timer_emulate(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
/*
@@ -289,11 +287,11 @@ static void phys_timer_emulate(struct kvm_vcpu *vcpu)
* then we also don't need a soft timer.
*/
if (kvm_timer_should_fire(ptimer) || !kvm_timer_irq_can_fire(ptimer)) {
- soft_timer_cancel(&timer->phys_timer);
+ soft_timer_cancel(&ptimer->hrtimer);
return;
}
- soft_timer_start(&timer->phys_timer, kvm_timer_compute_delta(ptimer));
+ soft_timer_start(&ptimer->hrtimer, kvm_timer_compute_delta(ptimer));
}
/*
@@ -303,7 +301,7 @@ static void phys_timer_emulate(struct kvm_vcpu *vcpu)
*/
static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
bool level;
@@ -329,13 +327,13 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
static void vtimer_save_state(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
unsigned long flags;
local_irq_save(flags);
- if (!vtimer->loaded)
+ if (!timer->loaded)
goto out;
if (timer->enabled) {
@@ -347,7 +345,7 @@ static void vtimer_save_state(struct kvm_vcpu *vcpu)
write_sysreg_el0(0, cntv_ctl);
isb();
- vtimer->loaded = false;
+ timer->loaded = false;
out:
local_irq_restore(flags);
}
@@ -359,7 +357,7 @@ static void vtimer_save_state(struct kvm_vcpu *vcpu)
*/
static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
@@ -379,20 +377,20 @@ static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
static void kvm_timer_unblocking(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
soft_timer_cancel(&timer->bg_timer);
}
static void vtimer_restore_state(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
unsigned long flags;
local_irq_save(flags);
- if (vtimer->loaded)
+ if (timer->loaded)
goto out;
if (timer->enabled) {
@@ -401,7 +399,7 @@ static void vtimer_restore_state(struct kvm_vcpu *vcpu)
write_sysreg_el0(vtimer->cnt_ctl, cntv_ctl);
}
- vtimer->loaded = true;
+ timer->loaded = true;
out:
local_irq_restore(flags);
}
@@ -462,7 +460,7 @@ static void kvm_timer_vcpu_load_nogic(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
@@ -507,7 +505,8 @@ bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
+ struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
if (unlikely(!timer->enabled))
return;
@@ -523,7 +522,7 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
* In any case, we re-schedule the hrtimer for the physical timer when
* coming back to the VCPU thread in kvm_timer_vcpu_load().
*/
- soft_timer_cancel(&timer->phys_timer);
+ soft_timer_cancel(&ptimer->hrtimer);
if (swait_active(kvm_arch_vcpu_wq(vcpu)))
kvm_timer_blocking(vcpu);
@@ -559,7 +558,7 @@ static void unmask_vtimer_irq_user(struct kvm_vcpu *vcpu)
void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
if (unlikely(!timer->enabled))
return;
@@ -570,7 +569,7 @@ void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu)
int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
@@ -611,22 +610,25 @@ static void update_vtimer_cntvoff(struct kvm_vcpu *vcpu, u64 cntvoff)
void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
/* Synchronize cntvoff across all vtimers of a VM. */
update_vtimer_cntvoff(vcpu, kvm_phys_timer_read());
- vcpu_ptimer(vcpu)->cntvoff = 0;
+ ptimer->cntvoff = 0;
hrtimer_init(&timer->bg_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
timer->bg_timer.function = kvm_bg_timer_expire;
- hrtimer_init(&timer->phys_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
- timer->phys_timer.function = kvm_phys_timer_expire;
+ hrtimer_init(&ptimer->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+ ptimer->hrtimer.function = kvm_phys_timer_expire;
vtimer->irq.irq = default_vtimer_irq.irq;
ptimer->irq.irq = default_ptimer_irq.irq;
+
+ vtimer->vcpu = vcpu;
+ ptimer->vcpu = vcpu;
}
static void kvm_timer_init_interrupt(void *info)
@@ -860,7 +862,7 @@ int kvm_timer_hyp_init(bool has_gic)
void kvm_timer_vcpu_terminate(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
soft_timer_cancel(&timer->bg_timer);
}
@@ -904,7 +906,7 @@ bool kvm_arch_timer_get_input_level(int vintid)
int kvm_timer_enable(struct kvm_vcpu *vcpu)
{
- struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+ struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
int ret;
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 13/27] KVM: arm/arm64: arch_timer: Assign the phys timer on VHE systems
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Christoffer Dall <christoffer.dall@arm.com>
VHE systems don't have to emulate the physical timer, we can simply
assign the EL1 physical timer directly to the VM as the host always
uses the EL2 timers.
In order to minimize the amount of cruft, AArch32 gets definitions for
the physical timer too, but is should be generally unused on this
architecture.
Co-written with Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_hyp.h | 4 +
include/kvm/arm_arch_timer.h | 6 +
virt/kvm/arm/arch_timer.c | 219 +++++++++++++++++++++++++--------
3 files changed, 180 insertions(+), 49 deletions(-)
diff --git a/arch/arm/include/asm/kvm_hyp.h b/arch/arm/include/asm/kvm_hyp.h
index e93a0cac9add..87bcd18df8d5 100644
--- a/arch/arm/include/asm/kvm_hyp.h
+++ b/arch/arm/include/asm/kvm_hyp.h
@@ -40,6 +40,7 @@
#define TTBR1 __ACCESS_CP15_64(1, c2)
#define VTTBR __ACCESS_CP15_64(6, c2)
#define PAR __ACCESS_CP15_64(0, c7)
+#define CNTP_CVAL __ACCESS_CP15_64(2, c14)
#define CNTV_CVAL __ACCESS_CP15_64(3, c14)
#define CNTVOFF __ACCESS_CP15_64(4, c14)
@@ -85,6 +86,7 @@
#define TID_PRIV __ACCESS_CP15(c13, 0, c0, 4)
#define HTPIDR __ACCESS_CP15(c13, 4, c0, 2)
#define CNTKCTL __ACCESS_CP15(c14, 0, c1, 0)
+#define CNTP_CTL __ACCESS_CP15(c14, 0, c2, 1)
#define CNTV_CTL __ACCESS_CP15(c14, 0, c3, 1)
#define CNTHCTL __ACCESS_CP15(c14, 4, c1, 0)
@@ -94,6 +96,8 @@
#define read_sysreg_el0(r) read_sysreg(r##_el0)
#define write_sysreg_el0(v, r) write_sysreg(v, r##_el0)
+#define cntp_ctl_el0 CNTP_CTL
+#define cntp_cval_el0 CNTP_CVAL
#define cntv_ctl_el0 CNTV_CTL
#define cntv_cval_el0 CNTV_CVAL
#define cntvoff_el2 CNTVOFF
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index ab835112204d..6d4a33a9c45a 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -50,6 +50,10 @@ struct arch_timer_context {
/* Emulated Timer (may be unused) */
struct hrtimer hrtimer;
+
+ /* Duplicated state from arch_timer.c for convenience */
+ u32 host_timer_irq;
+ u32 host_timer_irq_flags;
};
struct arch_timer_cpu {
@@ -104,6 +108,8 @@ bool kvm_arch_timer_get_input_level(int vintid);
#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.timers[TIMER_VTIMER])
#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.timers[TIMER_PTIMER])
+#define arch_timer_ctx_index(ctx) ((ctx) - vcpu_timer((ctx)->vcpu)->timers)
+
u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
enum kvm_arch_timers tmr,
enum kvm_arch_timer_regs treg);
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 471f9fd004c9..10c15151c87e 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -35,7 +35,9 @@
static struct timecounter *timecounter;
static unsigned int host_vtimer_irq;
+static unsigned int host_ptimer_irq;
static u32 host_vtimer_irq_flags;
+static u32 host_ptimer_irq_flags;
static DEFINE_STATIC_KEY_FALSE(has_gic_active_state);
@@ -86,20 +88,24 @@ static void soft_timer_cancel(struct hrtimer *hrt)
static irqreturn_t kvm_arch_timer_handler(int irq, void *dev_id)
{
struct kvm_vcpu *vcpu = *(struct kvm_vcpu **)dev_id;
- struct arch_timer_context *vtimer;
+ struct arch_timer_context *ctx;
/*
* We may see a timer interrupt after vcpu_put() has been called which
* sets the CPU's vcpu pointer to NULL, because even though the timer
- * has been disabled in vtimer_save_state(), the hardware interrupt
+ * has been disabled in timer_save_state(), the hardware interrupt
* signal may not have been retired from the interrupt controller yet.
*/
if (!vcpu)
return IRQ_HANDLED;
- vtimer = vcpu_vtimer(vcpu);
- if (kvm_timer_should_fire(vtimer))
- kvm_timer_update_irq(vcpu, true, vtimer);
+ if (irq == host_vtimer_irq)
+ ctx = vcpu_vtimer(vcpu);
+ else
+ ctx = vcpu_ptimer(vcpu);
+
+ if (kvm_timer_should_fire(ctx))
+ kvm_timer_update_irq(vcpu, true, ctx);
if (userspace_irqchip(vcpu->kvm) &&
!static_branch_unlikely(&has_gic_active_state))
@@ -208,13 +214,25 @@ static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
struct arch_timer_cpu *timer = vcpu_timer(timer_ctx->vcpu);
+ enum kvm_arch_timers index = arch_timer_ctx_index(timer_ctx);
u64 cval, now;
if (timer->loaded) {
- u32 cnt_ctl;
+ u32 cnt_ctl = 0;
+
+ switch (index) {
+ case TIMER_VTIMER:
+ cnt_ctl = read_sysreg_el0(cntv_ctl);
+ break;
+ case TIMER_PTIMER:
+ cnt_ctl = read_sysreg_el0(cntp_ctl);
+ break;
+ case NR_KVM_TIMERS:
+ /* GCC is braindead */
+ cnt_ctl = 0;
+ break;
+ }
- /* Only the virtual timer can be loaded so far */
- cnt_ctl = read_sysreg_el0(cntv_ctl);
return (cnt_ctl & ARCH_TIMER_CTRL_ENABLE) &&
(cnt_ctl & ARCH_TIMER_CTRL_IT_STAT) &&
!(cnt_ctl & ARCH_TIMER_CTRL_IT_MASK);
@@ -310,7 +328,7 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
return;
/*
- * The vtimer virtual interrupt is a 'mapped' interrupt, meaning part
+ * If the timer virtual interrupt is a 'mapped' interrupt, part
* of its lifecycle is offloaded to the hardware, and we therefore may
* not have lowered the irq.level value before having to signal a new
* interrupt, but have to signal an interrupt every time the level is
@@ -319,31 +337,55 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
level = kvm_timer_should_fire(vtimer);
kvm_timer_update_irq(vcpu, level, vtimer);
+ if (has_vhe()) {
+ level = kvm_timer_should_fire(ptimer);
+ kvm_timer_update_irq(vcpu, level, ptimer);
+
+ return;
+ }
+
phys_timer_emulate(vcpu);
if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
}
-static void vtimer_save_state(struct kvm_vcpu *vcpu)
+static void timer_save_state(struct arch_timer_context *ctx)
{
- struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct arch_timer_cpu *timer = vcpu_timer(ctx->vcpu);
+ enum kvm_arch_timers index = arch_timer_ctx_index(ctx);
unsigned long flags;
+ if (!timer->enabled)
+ return;
+
local_irq_save(flags);
if (!timer->loaded)
goto out;
- if (timer->enabled) {
- vtimer->cnt_ctl = read_sysreg_el0(cntv_ctl);
- vtimer->cnt_cval = read_sysreg_el0(cntv_cval);
- }
+ switch (index) {
+ case TIMER_VTIMER:
+ ctx->cnt_ctl = read_sysreg_el0(cntv_ctl);
+ ctx->cnt_cval = read_sysreg_el0(cntv_cval);
- /* Disable the virtual timer */
- write_sysreg_el0(0, cntv_ctl);
- isb();
+ /* Disable the timer */
+ write_sysreg_el0(0, cntv_ctl);
+ isb();
+
+ break;
+ case TIMER_PTIMER:
+ ctx->cnt_ctl = read_sysreg_el0(cntp_ctl);
+ ctx->cnt_cval = read_sysreg_el0(cntp_cval);
+
+ /* Disable the timer */
+ write_sysreg_el0(0, cntp_ctl);
+ isb();
+
+ break;
+ case NR_KVM_TIMERS:
+ break; /* GCC is braindead */
+ }
timer->loaded = false;
out:
@@ -382,21 +424,33 @@ static void kvm_timer_unblocking(struct kvm_vcpu *vcpu)
soft_timer_cancel(&timer->bg_timer);
}
-static void vtimer_restore_state(struct kvm_vcpu *vcpu)
+static void timer_restore_state(struct arch_timer_context *ctx)
{
- struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct arch_timer_cpu *timer = vcpu_timer(ctx->vcpu);
+ enum kvm_arch_timers index = arch_timer_ctx_index(ctx);
unsigned long flags;
+ if (!timer->enabled)
+ return;
+
local_irq_save(flags);
if (timer->loaded)
goto out;
- if (timer->enabled) {
- write_sysreg_el0(vtimer->cnt_cval, cntv_cval);
+ switch (index) {
+ case TIMER_VTIMER:
+ write_sysreg_el0(ctx->cnt_cval, cntv_cval);
+ isb();
+ write_sysreg_el0(ctx->cnt_ctl, cntv_ctl);
+ break;
+ case TIMER_PTIMER:
+ write_sysreg_el0(ctx->cnt_cval, cntp_cval);
isb();
- write_sysreg_el0(vtimer->cnt_ctl, cntv_ctl);
+ write_sysreg_el0(ctx->cnt_ctl, cntp_ctl);
+ break;
+ case NR_KVM_TIMERS:
+ break; /* GCC is braindead */
}
timer->loaded = true;
@@ -419,23 +473,23 @@ static void set_cntvoff(u64 cntvoff)
kvm_call_hyp(__kvm_timer_set_cntvoff, low, high);
}
-static inline void set_vtimer_irq_phys_active(struct kvm_vcpu *vcpu, bool active)
+static inline void set_timer_irq_phys_active(struct arch_timer_context *ctx, bool active)
{
int r;
- r = irq_set_irqchip_state(host_vtimer_irq, IRQCHIP_STATE_ACTIVE, active);
+ r = irq_set_irqchip_state(ctx->host_timer_irq, IRQCHIP_STATE_ACTIVE, active);
WARN_ON(r);
}
-static void kvm_timer_vcpu_load_gic(struct kvm_vcpu *vcpu)
+static void kvm_timer_vcpu_load_gic(struct arch_timer_context *ctx)
{
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct kvm_vcpu *vcpu = ctx->vcpu;
bool phys_active;
if (irqchip_in_kernel(vcpu->kvm))
- phys_active = kvm_vgic_map_is_active(vcpu, vtimer->irq.irq);
+ phys_active = kvm_vgic_map_is_active(vcpu, ctx->irq.irq);
else
- phys_active = vtimer->irq.level;
- set_vtimer_irq_phys_active(vcpu, phys_active);
+ phys_active = ctx->irq.level;
+ set_timer_irq_phys_active(ctx, phys_active);
}
static void kvm_timer_vcpu_load_nogic(struct kvm_vcpu *vcpu)
@@ -467,14 +521,22 @@ void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
if (unlikely(!timer->enabled))
return;
- if (static_branch_likely(&has_gic_active_state))
- kvm_timer_vcpu_load_gic(vcpu);
- else
+ if (static_branch_likely(&has_gic_active_state)) {
+ kvm_timer_vcpu_load_gic(vtimer);
+ if (has_vhe())
+ kvm_timer_vcpu_load_gic(ptimer);
+ } else {
kvm_timer_vcpu_load_nogic(vcpu);
+ }
set_cntvoff(vtimer->cntvoff);
- vtimer_restore_state(vcpu);
+ timer_restore_state(vtimer);
+
+ if (has_vhe()) {
+ timer_restore_state(ptimer);
+ return;
+ }
/* Set the background timer for the physical timer emulation. */
phys_timer_emulate(vcpu);
@@ -506,12 +568,17 @@ bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
+ struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
if (unlikely(!timer->enabled))
return;
- vtimer_save_state(vcpu);
+ timer_save_state(vtimer);
+ if (has_vhe()) {
+ timer_save_state(ptimer);
+ return;
+ }
/*
* Cancel the physical timer emulation, because the only case where we
@@ -534,8 +601,7 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
* counter of non-VHE case. For VHE, the virtual counter uses a fixed
* virtual offset of zero, so no need to zero CNTVOFF_EL2 register.
*/
- if (!has_vhe())
- set_cntvoff(0);
+ set_cntvoff(0);
}
/*
@@ -550,7 +616,7 @@ static void unmask_vtimer_irq_user(struct kvm_vcpu *vcpu)
if (!kvm_timer_should_fire(vtimer)) {
kvm_timer_update_irq(vcpu, false, vtimer);
if (static_branch_likely(&has_gic_active_state))
- set_vtimer_irq_phys_active(vcpu, false);
+ set_timer_irq_phys_active(vtimer, false);
else
enable_percpu_irq(host_vtimer_irq, host_vtimer_irq_flags);
}
@@ -625,7 +691,12 @@ void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
ptimer->hrtimer.function = kvm_phys_timer_expire;
vtimer->irq.irq = default_vtimer_irq.irq;
+ vtimer->host_timer_irq = host_vtimer_irq;
+ vtimer->host_timer_irq_flags = host_vtimer_irq_flags;
+
ptimer->irq.irq = default_ptimer_irq.irq;
+ ptimer->host_timer_irq = host_ptimer_irq;
+ ptimer->host_timer_irq_flags = host_ptimer_irq_flags;
vtimer->vcpu = vcpu;
ptimer->vcpu = vcpu;
@@ -634,6 +705,7 @@ void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
static void kvm_timer_init_interrupt(void *info)
{
enable_percpu_irq(host_vtimer_irq, host_vtimer_irq_flags);
+ enable_percpu_irq(host_ptimer_irq, host_ptimer_irq_flags);
}
int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value)
@@ -815,6 +887,8 @@ int kvm_timer_hyp_init(bool has_gic)
return -ENODEV;
}
+ /* First, do the virtual EL1 timer irq */
+
if (info->virtual_irq <= 0) {
kvm_err("kvm_arch_timer: invalid virtual timer IRQ: %d\n",
info->virtual_irq);
@@ -825,15 +899,15 @@ int kvm_timer_hyp_init(bool has_gic)
host_vtimer_irq_flags = irq_get_trigger_type(host_vtimer_irq);
if (host_vtimer_irq_flags != IRQF_TRIGGER_HIGH &&
host_vtimer_irq_flags != IRQF_TRIGGER_LOW) {
- kvm_err("Invalid trigger for IRQ%d, assuming level low\n",
+ kvm_err("Invalid trigger for vtimer IRQ%d, assuming level low\n",
host_vtimer_irq);
host_vtimer_irq_flags = IRQF_TRIGGER_LOW;
}
err = request_percpu_irq(host_vtimer_irq, kvm_arch_timer_handler,
- "kvm guest timer", kvm_get_running_vcpus());
+ "kvm guest vtimer", kvm_get_running_vcpus());
if (err) {
- kvm_err("kvm_arch_timer: can't request interrupt %d (%d)\n",
+ kvm_err("kvm_arch_timer: can't request vtimer interrupt %d (%d)\n",
host_vtimer_irq, err);
return err;
}
@@ -851,6 +925,43 @@ int kvm_timer_hyp_init(bool has_gic)
kvm_debug("virtual timer IRQ%d\n", host_vtimer_irq);
+ /* Now let's do the physical EL1 timer irq */
+
+ if (info->physical_irq > 0) {
+ host_ptimer_irq = info->physical_irq;
+ host_ptimer_irq_flags = irq_get_trigger_type(host_ptimer_irq);
+ if (host_ptimer_irq_flags != IRQF_TRIGGER_HIGH &&
+ host_ptimer_irq_flags != IRQF_TRIGGER_LOW) {
+ kvm_err("Invalid trigger for ptimer IRQ%d, assuming level low\n",
+ host_ptimer_irq);
+ host_ptimer_irq_flags = IRQF_TRIGGER_LOW;
+ }
+
+ err = request_percpu_irq(host_ptimer_irq, kvm_arch_timer_handler,
+ "kvm guest ptimer", kvm_get_running_vcpus());
+ if (err) {
+ kvm_err("kvm_arch_timer: can't request ptimer interrupt %d (%d)\n",
+ host_ptimer_irq, err);
+ return err;
+ }
+
+ if (has_gic) {
+ err = irq_set_vcpu_affinity(host_ptimer_irq,
+ kvm_get_running_vcpus());
+ if (err) {
+ kvm_err("kvm_arch_timer: error setting vcpu affinity\n");
+ goto out_free_irq;
+ }
+ }
+
+ kvm_debug("physical timer IRQ%d\n", host_ptimer_irq);
+ } else if (has_vhe()) {
+ kvm_err("kvm_arch_timer: invalid physical timer IRQ: %d\n",
+ info->physical_irq);
+ err = -ENODEV;
+ goto out_free_irq;
+ }
+
cpuhp_setup_state(CPUHP_AP_KVM_ARM_TIMER_STARTING,
"kvm/arm/timer:starting", kvm_timer_starting_cpu,
kvm_timer_dying_cpu);
@@ -898,8 +1009,10 @@ bool kvm_arch_timer_get_input_level(int vintid)
if (vintid == vcpu_vtimer(vcpu)->irq.irq)
timer = vcpu_vtimer(vcpu);
+ else if (vintid == vcpu_ptimer(vcpu)->irq.irq)
+ timer = vcpu_ptimer(vcpu);
else
- BUG(); /* We only map the vtimer so far */
+ BUG();
return kvm_timer_should_fire(timer);
}
@@ -908,6 +1021,7 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
int ret;
if (timer->enabled)
@@ -930,14 +1044,21 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
if (ret)
return ret;
+ if (has_vhe()) {
+ ret = kvm_vgic_map_phys_irq(vcpu, host_ptimer_irq, ptimer->irq.irq,
+ kvm_arch_timer_get_input_level);
+ if (ret)
+ return ret;
+ }
+
no_vgic:
timer->enabled = 1;
return 0;
}
/*
- * On VHE system, we only need to configure trap on physical timer and counter
- * accesses in EL0 and EL1 once, not for every world switch.
+ * On VHE system, we only need to configure the EL2 timer trap register once,
+ * not for every world switch.
* The host kernel runs at EL2 with HCR_EL2.TGE == 1,
* and this makes those bits have no effect for the host kernel execution.
*/
@@ -948,11 +1069,11 @@ void kvm_timer_init_vhe(void)
u64 val;
/*
- * Disallow physical timer access for the guest.
- * Physical counter access is allowed.
+ * VHE systems allow the guest direct access to the EL1 physical
+ * timer/counter.
*/
val = read_sysreg(cnthctl_el2);
- val &= ~(CNTHCTL_EL1PCEN << cnthctl_shift);
+ val |= (CNTHCTL_EL1PCEN << cnthctl_shift);
val |= (CNTHCTL_EL1PCTEN << cnthctl_shift);
write_sysreg(val, cnthctl_el2);
}
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 13/27] KVM: arm/arm64: arch_timer: Assign the phys timer on VHE systems
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Christoffer Dall <christoffer.dall@arm.com>
VHE systems don't have to emulate the physical timer, we can simply
assign the EL1 physical timer directly to the VM as the host always
uses the EL2 timers.
In order to minimize the amount of cruft, AArch32 gets definitions for
the physical timer too, but is should be generally unused on this
architecture.
Co-written with Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
arch/arm/include/asm/kvm_hyp.h | 4 +
include/kvm/arm_arch_timer.h | 6 +
virt/kvm/arm/arch_timer.c | 219 +++++++++++++++++++++++++--------
3 files changed, 180 insertions(+), 49 deletions(-)
diff --git a/arch/arm/include/asm/kvm_hyp.h b/arch/arm/include/asm/kvm_hyp.h
index e93a0cac9add..87bcd18df8d5 100644
--- a/arch/arm/include/asm/kvm_hyp.h
+++ b/arch/arm/include/asm/kvm_hyp.h
@@ -40,6 +40,7 @@
#define TTBR1 __ACCESS_CP15_64(1, c2)
#define VTTBR __ACCESS_CP15_64(6, c2)
#define PAR __ACCESS_CP15_64(0, c7)
+#define CNTP_CVAL __ACCESS_CP15_64(2, c14)
#define CNTV_CVAL __ACCESS_CP15_64(3, c14)
#define CNTVOFF __ACCESS_CP15_64(4, c14)
@@ -85,6 +86,7 @@
#define TID_PRIV __ACCESS_CP15(c13, 0, c0, 4)
#define HTPIDR __ACCESS_CP15(c13, 4, c0, 2)
#define CNTKCTL __ACCESS_CP15(c14, 0, c1, 0)
+#define CNTP_CTL __ACCESS_CP15(c14, 0, c2, 1)
#define CNTV_CTL __ACCESS_CP15(c14, 0, c3, 1)
#define CNTHCTL __ACCESS_CP15(c14, 4, c1, 0)
@@ -94,6 +96,8 @@
#define read_sysreg_el0(r) read_sysreg(r##_el0)
#define write_sysreg_el0(v, r) write_sysreg(v, r##_el0)
+#define cntp_ctl_el0 CNTP_CTL
+#define cntp_cval_el0 CNTP_CVAL
#define cntv_ctl_el0 CNTV_CTL
#define cntv_cval_el0 CNTV_CVAL
#define cntvoff_el2 CNTVOFF
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index ab835112204d..6d4a33a9c45a 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -50,6 +50,10 @@ struct arch_timer_context {
/* Emulated Timer (may be unused) */
struct hrtimer hrtimer;
+
+ /* Duplicated state from arch_timer.c for convenience */
+ u32 host_timer_irq;
+ u32 host_timer_irq_flags;
};
struct arch_timer_cpu {
@@ -104,6 +108,8 @@ bool kvm_arch_timer_get_input_level(int vintid);
#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.timers[TIMER_VTIMER])
#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.timers[TIMER_PTIMER])
+#define arch_timer_ctx_index(ctx) ((ctx) - vcpu_timer((ctx)->vcpu)->timers)
+
u64 kvm_arm_timer_read_sysreg(struct kvm_vcpu *vcpu,
enum kvm_arch_timers tmr,
enum kvm_arch_timer_regs treg);
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 471f9fd004c9..10c15151c87e 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -35,7 +35,9 @@
static struct timecounter *timecounter;
static unsigned int host_vtimer_irq;
+static unsigned int host_ptimer_irq;
static u32 host_vtimer_irq_flags;
+static u32 host_ptimer_irq_flags;
static DEFINE_STATIC_KEY_FALSE(has_gic_active_state);
@@ -86,20 +88,24 @@ static void soft_timer_cancel(struct hrtimer *hrt)
static irqreturn_t kvm_arch_timer_handler(int irq, void *dev_id)
{
struct kvm_vcpu *vcpu = *(struct kvm_vcpu **)dev_id;
- struct arch_timer_context *vtimer;
+ struct arch_timer_context *ctx;
/*
* We may see a timer interrupt after vcpu_put() has been called which
* sets the CPU's vcpu pointer to NULL, because even though the timer
- * has been disabled in vtimer_save_state(), the hardware interrupt
+ * has been disabled in timer_save_state(), the hardware interrupt
* signal may not have been retired from the interrupt controller yet.
*/
if (!vcpu)
return IRQ_HANDLED;
- vtimer = vcpu_vtimer(vcpu);
- if (kvm_timer_should_fire(vtimer))
- kvm_timer_update_irq(vcpu, true, vtimer);
+ if (irq == host_vtimer_irq)
+ ctx = vcpu_vtimer(vcpu);
+ else
+ ctx = vcpu_ptimer(vcpu);
+
+ if (kvm_timer_should_fire(ctx))
+ kvm_timer_update_irq(vcpu, true, ctx);
if (userspace_irqchip(vcpu->kvm) &&
!static_branch_unlikely(&has_gic_active_state))
@@ -208,13 +214,25 @@ static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
struct arch_timer_cpu *timer = vcpu_timer(timer_ctx->vcpu);
+ enum kvm_arch_timers index = arch_timer_ctx_index(timer_ctx);
u64 cval, now;
if (timer->loaded) {
- u32 cnt_ctl;
+ u32 cnt_ctl = 0;
+
+ switch (index) {
+ case TIMER_VTIMER:
+ cnt_ctl = read_sysreg_el0(cntv_ctl);
+ break;
+ case TIMER_PTIMER:
+ cnt_ctl = read_sysreg_el0(cntp_ctl);
+ break;
+ case NR_KVM_TIMERS:
+ /* GCC is braindead */
+ cnt_ctl = 0;
+ break;
+ }
- /* Only the virtual timer can be loaded so far */
- cnt_ctl = read_sysreg_el0(cntv_ctl);
return (cnt_ctl & ARCH_TIMER_CTRL_ENABLE) &&
(cnt_ctl & ARCH_TIMER_CTRL_IT_STAT) &&
!(cnt_ctl & ARCH_TIMER_CTRL_IT_MASK);
@@ -310,7 +328,7 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
return;
/*
- * The vtimer virtual interrupt is a 'mapped' interrupt, meaning part
+ * If the timer virtual interrupt is a 'mapped' interrupt, part
* of its lifecycle is offloaded to the hardware, and we therefore may
* not have lowered the irq.level value before having to signal a new
* interrupt, but have to signal an interrupt every time the level is
@@ -319,31 +337,55 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
level = kvm_timer_should_fire(vtimer);
kvm_timer_update_irq(vcpu, level, vtimer);
+ if (has_vhe()) {
+ level = kvm_timer_should_fire(ptimer);
+ kvm_timer_update_irq(vcpu, level, ptimer);
+
+ return;
+ }
+
phys_timer_emulate(vcpu);
if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
}
-static void vtimer_save_state(struct kvm_vcpu *vcpu)
+static void timer_save_state(struct arch_timer_context *ctx)
{
- struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct arch_timer_cpu *timer = vcpu_timer(ctx->vcpu);
+ enum kvm_arch_timers index = arch_timer_ctx_index(ctx);
unsigned long flags;
+ if (!timer->enabled)
+ return;
+
local_irq_save(flags);
if (!timer->loaded)
goto out;
- if (timer->enabled) {
- vtimer->cnt_ctl = read_sysreg_el0(cntv_ctl);
- vtimer->cnt_cval = read_sysreg_el0(cntv_cval);
- }
+ switch (index) {
+ case TIMER_VTIMER:
+ ctx->cnt_ctl = read_sysreg_el0(cntv_ctl);
+ ctx->cnt_cval = read_sysreg_el0(cntv_cval);
- /* Disable the virtual timer */
- write_sysreg_el0(0, cntv_ctl);
- isb();
+ /* Disable the timer */
+ write_sysreg_el0(0, cntv_ctl);
+ isb();
+
+ break;
+ case TIMER_PTIMER:
+ ctx->cnt_ctl = read_sysreg_el0(cntp_ctl);
+ ctx->cnt_cval = read_sysreg_el0(cntp_cval);
+
+ /* Disable the timer */
+ write_sysreg_el0(0, cntp_ctl);
+ isb();
+
+ break;
+ case NR_KVM_TIMERS:
+ break; /* GCC is braindead */
+ }
timer->loaded = false;
out:
@@ -382,21 +424,33 @@ static void kvm_timer_unblocking(struct kvm_vcpu *vcpu)
soft_timer_cancel(&timer->bg_timer);
}
-static void vtimer_restore_state(struct kvm_vcpu *vcpu)
+static void timer_restore_state(struct arch_timer_context *ctx)
{
- struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct arch_timer_cpu *timer = vcpu_timer(ctx->vcpu);
+ enum kvm_arch_timers index = arch_timer_ctx_index(ctx);
unsigned long flags;
+ if (!timer->enabled)
+ return;
+
local_irq_save(flags);
if (timer->loaded)
goto out;
- if (timer->enabled) {
- write_sysreg_el0(vtimer->cnt_cval, cntv_cval);
+ switch (index) {
+ case TIMER_VTIMER:
+ write_sysreg_el0(ctx->cnt_cval, cntv_cval);
+ isb();
+ write_sysreg_el0(ctx->cnt_ctl, cntv_ctl);
+ break;
+ case TIMER_PTIMER:
+ write_sysreg_el0(ctx->cnt_cval, cntp_cval);
isb();
- write_sysreg_el0(vtimer->cnt_ctl, cntv_ctl);
+ write_sysreg_el0(ctx->cnt_ctl, cntp_ctl);
+ break;
+ case NR_KVM_TIMERS:
+ break; /* GCC is braindead */
}
timer->loaded = true;
@@ -419,23 +473,23 @@ static void set_cntvoff(u64 cntvoff)
kvm_call_hyp(__kvm_timer_set_cntvoff, low, high);
}
-static inline void set_vtimer_irq_phys_active(struct kvm_vcpu *vcpu, bool active)
+static inline void set_timer_irq_phys_active(struct arch_timer_context *ctx, bool active)
{
int r;
- r = irq_set_irqchip_state(host_vtimer_irq, IRQCHIP_STATE_ACTIVE, active);
+ r = irq_set_irqchip_state(ctx->host_timer_irq, IRQCHIP_STATE_ACTIVE, active);
WARN_ON(r);
}
-static void kvm_timer_vcpu_load_gic(struct kvm_vcpu *vcpu)
+static void kvm_timer_vcpu_load_gic(struct arch_timer_context *ctx)
{
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct kvm_vcpu *vcpu = ctx->vcpu;
bool phys_active;
if (irqchip_in_kernel(vcpu->kvm))
- phys_active = kvm_vgic_map_is_active(vcpu, vtimer->irq.irq);
+ phys_active = kvm_vgic_map_is_active(vcpu, ctx->irq.irq);
else
- phys_active = vtimer->irq.level;
- set_vtimer_irq_phys_active(vcpu, phys_active);
+ phys_active = ctx->irq.level;
+ set_timer_irq_phys_active(ctx, phys_active);
}
static void kvm_timer_vcpu_load_nogic(struct kvm_vcpu *vcpu)
@@ -467,14 +521,22 @@ void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
if (unlikely(!timer->enabled))
return;
- if (static_branch_likely(&has_gic_active_state))
- kvm_timer_vcpu_load_gic(vcpu);
- else
+ if (static_branch_likely(&has_gic_active_state)) {
+ kvm_timer_vcpu_load_gic(vtimer);
+ if (has_vhe())
+ kvm_timer_vcpu_load_gic(ptimer);
+ } else {
kvm_timer_vcpu_load_nogic(vcpu);
+ }
set_cntvoff(vtimer->cntvoff);
- vtimer_restore_state(vcpu);
+ timer_restore_state(vtimer);
+
+ if (has_vhe()) {
+ timer_restore_state(ptimer);
+ return;
+ }
/* Set the background timer for the physical timer emulation. */
phys_timer_emulate(vcpu);
@@ -506,12 +568,17 @@ bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
+ struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
if (unlikely(!timer->enabled))
return;
- vtimer_save_state(vcpu);
+ timer_save_state(vtimer);
+ if (has_vhe()) {
+ timer_save_state(ptimer);
+ return;
+ }
/*
* Cancel the physical timer emulation, because the only case where we
@@ -534,8 +601,7 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
* counter of non-VHE case. For VHE, the virtual counter uses a fixed
* virtual offset of zero, so no need to zero CNTVOFF_EL2 register.
*/
- if (!has_vhe())
- set_cntvoff(0);
+ set_cntvoff(0);
}
/*
@@ -550,7 +616,7 @@ static void unmask_vtimer_irq_user(struct kvm_vcpu *vcpu)
if (!kvm_timer_should_fire(vtimer)) {
kvm_timer_update_irq(vcpu, false, vtimer);
if (static_branch_likely(&has_gic_active_state))
- set_vtimer_irq_phys_active(vcpu, false);
+ set_timer_irq_phys_active(vtimer, false);
else
enable_percpu_irq(host_vtimer_irq, host_vtimer_irq_flags);
}
@@ -625,7 +691,12 @@ void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
ptimer->hrtimer.function = kvm_phys_timer_expire;
vtimer->irq.irq = default_vtimer_irq.irq;
+ vtimer->host_timer_irq = host_vtimer_irq;
+ vtimer->host_timer_irq_flags = host_vtimer_irq_flags;
+
ptimer->irq.irq = default_ptimer_irq.irq;
+ ptimer->host_timer_irq = host_ptimer_irq;
+ ptimer->host_timer_irq_flags = host_ptimer_irq_flags;
vtimer->vcpu = vcpu;
ptimer->vcpu = vcpu;
@@ -634,6 +705,7 @@ void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
static void kvm_timer_init_interrupt(void *info)
{
enable_percpu_irq(host_vtimer_irq, host_vtimer_irq_flags);
+ enable_percpu_irq(host_ptimer_irq, host_ptimer_irq_flags);
}
int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value)
@@ -815,6 +887,8 @@ int kvm_timer_hyp_init(bool has_gic)
return -ENODEV;
}
+ /* First, do the virtual EL1 timer irq */
+
if (info->virtual_irq <= 0) {
kvm_err("kvm_arch_timer: invalid virtual timer IRQ: %d\n",
info->virtual_irq);
@@ -825,15 +899,15 @@ int kvm_timer_hyp_init(bool has_gic)
host_vtimer_irq_flags = irq_get_trigger_type(host_vtimer_irq);
if (host_vtimer_irq_flags != IRQF_TRIGGER_HIGH &&
host_vtimer_irq_flags != IRQF_TRIGGER_LOW) {
- kvm_err("Invalid trigger for IRQ%d, assuming level low\n",
+ kvm_err("Invalid trigger for vtimer IRQ%d, assuming level low\n",
host_vtimer_irq);
host_vtimer_irq_flags = IRQF_TRIGGER_LOW;
}
err = request_percpu_irq(host_vtimer_irq, kvm_arch_timer_handler,
- "kvm guest timer", kvm_get_running_vcpus());
+ "kvm guest vtimer", kvm_get_running_vcpus());
if (err) {
- kvm_err("kvm_arch_timer: can't request interrupt %d (%d)\n",
+ kvm_err("kvm_arch_timer: can't request vtimer interrupt %d (%d)\n",
host_vtimer_irq, err);
return err;
}
@@ -851,6 +925,43 @@ int kvm_timer_hyp_init(bool has_gic)
kvm_debug("virtual timer IRQ%d\n", host_vtimer_irq);
+ /* Now let's do the physical EL1 timer irq */
+
+ if (info->physical_irq > 0) {
+ host_ptimer_irq = info->physical_irq;
+ host_ptimer_irq_flags = irq_get_trigger_type(host_ptimer_irq);
+ if (host_ptimer_irq_flags != IRQF_TRIGGER_HIGH &&
+ host_ptimer_irq_flags != IRQF_TRIGGER_LOW) {
+ kvm_err("Invalid trigger for ptimer IRQ%d, assuming level low\n",
+ host_ptimer_irq);
+ host_ptimer_irq_flags = IRQF_TRIGGER_LOW;
+ }
+
+ err = request_percpu_irq(host_ptimer_irq, kvm_arch_timer_handler,
+ "kvm guest ptimer", kvm_get_running_vcpus());
+ if (err) {
+ kvm_err("kvm_arch_timer: can't request ptimer interrupt %d (%d)\n",
+ host_ptimer_irq, err);
+ return err;
+ }
+
+ if (has_gic) {
+ err = irq_set_vcpu_affinity(host_ptimer_irq,
+ kvm_get_running_vcpus());
+ if (err) {
+ kvm_err("kvm_arch_timer: error setting vcpu affinity\n");
+ goto out_free_irq;
+ }
+ }
+
+ kvm_debug("physical timer IRQ%d\n", host_ptimer_irq);
+ } else if (has_vhe()) {
+ kvm_err("kvm_arch_timer: invalid physical timer IRQ: %d\n",
+ info->physical_irq);
+ err = -ENODEV;
+ goto out_free_irq;
+ }
+
cpuhp_setup_state(CPUHP_AP_KVM_ARM_TIMER_STARTING,
"kvm/arm/timer:starting", kvm_timer_starting_cpu,
kvm_timer_dying_cpu);
@@ -898,8 +1009,10 @@ bool kvm_arch_timer_get_input_level(int vintid)
if (vintid == vcpu_vtimer(vcpu)->irq.irq)
timer = vcpu_vtimer(vcpu);
+ else if (vintid == vcpu_ptimer(vcpu)->irq.irq)
+ timer = vcpu_ptimer(vcpu);
else
- BUG(); /* We only map the vtimer so far */
+ BUG();
return kvm_timer_should_fire(timer);
}
@@ -908,6 +1021,7 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
+ struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
int ret;
if (timer->enabled)
@@ -930,14 +1044,21 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
if (ret)
return ret;
+ if (has_vhe()) {
+ ret = kvm_vgic_map_phys_irq(vcpu, host_ptimer_irq, ptimer->irq.irq,
+ kvm_arch_timer_get_input_level);
+ if (ret)
+ return ret;
+ }
+
no_vgic:
timer->enabled = 1;
return 0;
}
/*
- * On VHE system, we only need to configure trap on physical timer and counter
- * accesses in EL0 and EL1 once, not for every world switch.
+ * On VHE system, we only need to configure the EL2 timer trap register once,
+ * not for every world switch.
* The host kernel runs at EL2 with HCR_EL2.TGE == 1,
* and this makes those bits have no effect for the host kernel execution.
*/
@@ -948,11 +1069,11 @@ void kvm_timer_init_vhe(void)
u64 val;
/*
- * Disallow physical timer access for the guest.
- * Physical counter access is allowed.
+ * VHE systems allow the guest direct access to the EL1 physical
+ * timer/counter.
*/
val = read_sysreg(cnthctl_el2);
- val &= ~(CNTHCTL_EL1PCEN << cnthctl_shift);
+ val |= (CNTHCTL_EL1PCEN << cnthctl_shift);
val |= (CNTHCTL_EL1PCTEN << cnthctl_shift);
write_sysreg(val, cnthctl_el2);
}
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 14/27] KVM: arm/arm64: Rework the timer code to use a timer_map
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Christoffer Dall <christoffer.dall@arm.com>
We are currently emulating two timers in two different ways. When we
add support for nested virtualization in the future, we are going to be
emulating either two timers in two diffferent ways, or four timers in a
single way.
We need a unified data structure to keep track of how we map virtual
state to physical state and we need to cleanup some of the timer code to
operate more independently on a struct arch_timer_context instead of
trying to consider the global state of the VCPU and recomputing all
state.
Co-written with Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
include/kvm/arm_arch_timer.h | 23 +--
virt/kvm/arm/arch_timer.c | 295 +++++++++++++++++++----------------
virt/kvm/arm/trace.h | 105 +++++++++++++
3 files changed, 278 insertions(+), 145 deletions(-)
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index 6d4a33a9c45a..05a18dd265b5 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -51,11 +51,24 @@ struct arch_timer_context {
/* Emulated Timer (may be unused) */
struct hrtimer hrtimer;
+ /*
+ * We have multiple paths which can save/restore the timer state onto
+ * the hardware, so we need some way of keeping track of where the
+ * latest state is.
+ */
+ bool loaded;
+
/* Duplicated state from arch_timer.c for convenience */
u32 host_timer_irq;
u32 host_timer_irq_flags;
};
+struct timer_map {
+ struct arch_timer_context *direct_vtimer;
+ struct arch_timer_context *direct_ptimer;
+ struct arch_timer_context *emul_ptimer;
+};
+
struct arch_timer_cpu {
struct arch_timer_context timers[NR_KVM_TIMERS];
@@ -64,16 +77,6 @@ struct arch_timer_cpu {
/* Is the timer enabled */
bool enabled;
-
- /*
- * We have multiple paths which can save/restore the timer state
- * onto the hardware, so we need some way of keeping track of
- * where the latest state is.
- *
- * loaded == true: State is loaded on the hardware registers.
- * loaded == false: State is stored in memory.
- */
- bool loaded;
};
int kvm_timer_hyp_init(bool);
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 10c15151c87e..17f9de73cc8a 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -68,6 +68,21 @@ u64 kvm_phys_timer_read(void)
return timecounter->cc->read(timecounter->cc);
}
+static void get_timer_map(struct kvm_vcpu *vcpu, struct timer_map *map)
+{
+ if (has_vhe()) {
+ map->direct_vtimer = vcpu_vtimer(vcpu);
+ map->direct_ptimer = vcpu_ptimer(vcpu);
+ map->emul_ptimer = NULL;
+ } else {
+ map->direct_vtimer = vcpu_vtimer(vcpu);
+ map->direct_ptimer = NULL;
+ map->emul_ptimer = vcpu_ptimer(vcpu);
+ }
+
+ trace_kvm_get_timer_map(vcpu->vcpu_id, map);
+}
+
static inline bool userspace_irqchip(struct kvm *kvm)
{
return static_branch_unlikely(&userspace_irqchip_in_use) &&
@@ -89,6 +104,7 @@ static irqreturn_t kvm_arch_timer_handler(int irq, void *dev_id)
{
struct kvm_vcpu *vcpu = *(struct kvm_vcpu **)dev_id;
struct arch_timer_context *ctx;
+ struct timer_map map;
/*
* We may see a timer interrupt after vcpu_put() has been called which
@@ -99,10 +115,12 @@ static irqreturn_t kvm_arch_timer_handler(int irq, void *dev_id)
if (!vcpu)
return IRQ_HANDLED;
+ get_timer_map(vcpu, &map);
+
if (irq == host_vtimer_irq)
- ctx = vcpu_vtimer(vcpu);
+ ctx = map.direct_vtimer;
else
- ctx = vcpu_ptimer(vcpu);
+ ctx = map.direct_ptimer;
if (kvm_timer_should_fire(ctx))
kvm_timer_update_irq(vcpu, true, ctx);
@@ -136,7 +154,9 @@ static u64 kvm_timer_compute_delta(struct arch_timer_context *timer_ctx)
static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx)
{
- return !(timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_IT_MASK) &&
+ WARN_ON(timer_ctx && timer_ctx->loaded);
+ return timer_ctx &&
+ !(timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_IT_MASK) &&
(timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_ENABLE);
}
@@ -146,21 +166,22 @@ static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx)
*/
static u64 kvm_timer_earliest_exp(struct kvm_vcpu *vcpu)
{
- u64 min_virt = ULLONG_MAX, min_phys = ULLONG_MAX;
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ u64 min_delta = ULLONG_MAX;
+ int i;
- if (kvm_timer_irq_can_fire(vtimer))
- min_virt = kvm_timer_compute_delta(vtimer);
+ for (i = 0; i < NR_KVM_TIMERS; i++) {
+ struct arch_timer_context *ctx = &vcpu->arch.timer_cpu.timers[i];
- if (kvm_timer_irq_can_fire(ptimer))
- min_phys = kvm_timer_compute_delta(ptimer);
+ WARN(ctx->loaded, "timer %d loaded\n", i);
+ if (kvm_timer_irq_can_fire(ctx))
+ min_delta = min(min_delta, kvm_timer_compute_delta(ctx));
+ }
/* If none of timers can fire, then return 0 */
- if ((min_virt == ULLONG_MAX) && (min_phys == ULLONG_MAX))
+ if (min_delta == ULLONG_MAX)
return 0;
- return min(min_virt, min_phys);
+ return min_delta;
}
static enum hrtimer_restart kvm_bg_timer_expire(struct hrtimer *hrt)
@@ -187,37 +208,45 @@ static enum hrtimer_restart kvm_bg_timer_expire(struct hrtimer *hrt)
return HRTIMER_NORESTART;
}
-static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
+static enum hrtimer_restart kvm_hrtimer_expire(struct hrtimer *hrt)
{
- struct arch_timer_context *ptimer;
+ struct arch_timer_context *ctx;
struct kvm_vcpu *vcpu;
u64 ns;
- ptimer = container_of(hrt, struct arch_timer_context, hrtimer);
- vcpu = ptimer->vcpu;
+ ctx = container_of(hrt, struct arch_timer_context, hrtimer);
+ vcpu = ctx->vcpu;
+
+ trace_kvm_timer_hrtimer_expire(ctx);
/*
* Check that the timer has really expired from the guest's
* PoV (NTP on the host may have forced it to expire
* early). If not ready, schedule for a later time.
*/
- ns = kvm_timer_compute_delta(ptimer);
+ ns = kvm_timer_compute_delta(ctx);
if (unlikely(ns)) {
hrtimer_forward_now(hrt, ns_to_ktime(ns));
return HRTIMER_RESTART;
}
- kvm_timer_update_irq(vcpu, true, ptimer);
+ kvm_timer_update_irq(vcpu, true, ctx);
return HRTIMER_NORESTART;
}
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
- struct arch_timer_cpu *timer = vcpu_timer(timer_ctx->vcpu);
- enum kvm_arch_timers index = arch_timer_ctx_index(timer_ctx);
+ struct arch_timer_cpu *timer;
+ enum kvm_arch_timers index;
u64 cval, now;
- if (timer->loaded) {
+ if (!timer_ctx)
+ return false;
+
+ timer = vcpu_timer(timer_ctx->vcpu);
+ index = arch_timer_ctx_index(timer_ctx);
+
+ if (timer_ctx->loaded) {
u32 cnt_ctl = 0;
switch (index) {
@@ -249,13 +278,13 @@ static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
bool kvm_timer_is_pending(struct kvm_vcpu *vcpu)
{
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
- if (kvm_timer_should_fire(vtimer))
- return true;
+ get_timer_map(vcpu, &map);
- return kvm_timer_should_fire(ptimer);
+ return kvm_timer_should_fire(map.direct_vtimer) ||
+ kvm_timer_should_fire(map.direct_ptimer) ||
+ kvm_timer_should_fire(map.emul_ptimer);
}
/*
@@ -294,60 +323,28 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level,
}
}
-/* Schedule the background timer for the emulated timer. */
-static void phys_timer_emulate(struct kvm_vcpu *vcpu)
+static void timer_emulate(struct arch_timer_context *ctx)
{
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ bool should_fire = kvm_timer_should_fire(ctx);
- /*
- * If the timer can fire now, we don't need to have a soft timer
- * scheduled for the future. If the timer cannot fire at all,
- * then we also don't need a soft timer.
- */
- if (kvm_timer_should_fire(ptimer) || !kvm_timer_irq_can_fire(ptimer)) {
- soft_timer_cancel(&ptimer->hrtimer);
- return;
- }
-
- soft_timer_start(&ptimer->hrtimer, kvm_timer_compute_delta(ptimer));
-}
-
-/*
- * Check if there was a change in the timer state, so that we should either
- * raise or lower the line level to the GIC or schedule a background timer to
- * emulate the physical timer.
- */
-static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
-{
- struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
- bool level;
+ trace_kvm_timer_emulate(ctx, should_fire);
- if (unlikely(!timer->enabled))
+ if (should_fire) {
+ kvm_timer_update_irq(ctx->vcpu, true, ctx);
return;
+ }
/*
- * If the timer virtual interrupt is a 'mapped' interrupt, part
- * of its lifecycle is offloaded to the hardware, and we therefore may
- * not have lowered the irq.level value before having to signal a new
- * interrupt, but have to signal an interrupt every time the level is
- * asserted.
+ * If the timer can fire now, we don't need to have a soft timer
+ * scheduled for the future. If the timer cannot fire at all,
+ * then we also don't need a soft timer.
*/
- level = kvm_timer_should_fire(vtimer);
- kvm_timer_update_irq(vcpu, level, vtimer);
-
- if (has_vhe()) {
- level = kvm_timer_should_fire(ptimer);
- kvm_timer_update_irq(vcpu, level, ptimer);
-
+ if (!kvm_timer_irq_can_fire(ctx)) {
+ soft_timer_cancel(&ctx->hrtimer);
return;
}
- phys_timer_emulate(vcpu);
-
- if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
- kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
+ soft_timer_start(&ctx->hrtimer, kvm_timer_compute_delta(ctx));
}
static void timer_save_state(struct arch_timer_context *ctx)
@@ -361,7 +358,7 @@ static void timer_save_state(struct arch_timer_context *ctx)
local_irq_save(flags);
- if (!timer->loaded)
+ if (!ctx->loaded)
goto out;
switch (index) {
@@ -384,10 +381,12 @@ static void timer_save_state(struct arch_timer_context *ctx)
break;
case NR_KVM_TIMERS:
- break; /* GCC is braindead */
+ BUG();
}
- timer->loaded = false;
+ trace_kvm_timer_save_state(ctx);
+
+ ctx->loaded = false;
out:
local_irq_restore(flags);
}
@@ -400,14 +399,17 @@ static void timer_save_state(struct arch_timer_context *ctx)
static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
+
+ get_timer_map(vcpu, &map);
/*
- * If both timers are not capable of raising interrupts (disabled or
+ * If no timers are capable of raising interrupts (disabled or
* masked), then there's no more work for us to do.
*/
- if (!kvm_timer_irq_can_fire(vtimer) && !kvm_timer_irq_can_fire(ptimer))
+ if (!kvm_timer_irq_can_fire(map.direct_vtimer) &&
+ !kvm_timer_irq_can_fire(map.direct_ptimer) &&
+ !kvm_timer_irq_can_fire(map.emul_ptimer))
return;
/*
@@ -435,7 +437,7 @@ static void timer_restore_state(struct arch_timer_context *ctx)
local_irq_save(flags);
- if (timer->loaded)
+ if (ctx->loaded)
goto out;
switch (index) {
@@ -450,10 +452,12 @@ static void timer_restore_state(struct arch_timer_context *ctx)
write_sysreg_el0(ctx->cnt_ctl, cntp_ctl);
break;
case NR_KVM_TIMERS:
- break; /* GCC is braindead */
+ BUG();
}
- timer->loaded = true;
+ trace_kvm_timer_restore_state(ctx);
+
+ ctx->loaded = true;
out:
local_irq_restore(flags);
}
@@ -515,37 +519,31 @@ static void kvm_timer_vcpu_load_nogic(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
if (unlikely(!timer->enabled))
return;
+ get_timer_map(vcpu, &map);
+
if (static_branch_likely(&has_gic_active_state)) {
- kvm_timer_vcpu_load_gic(vtimer);
- if (has_vhe())
- kvm_timer_vcpu_load_gic(ptimer);
+ kvm_timer_vcpu_load_gic(map.direct_vtimer);
+ if (map.direct_ptimer)
+ kvm_timer_vcpu_load_gic(map.direct_ptimer);
} else {
kvm_timer_vcpu_load_nogic(vcpu);
}
- set_cntvoff(vtimer->cntvoff);
-
- timer_restore_state(vtimer);
-
- if (has_vhe()) {
- timer_restore_state(ptimer);
- return;
- }
-
- /* Set the background timer for the physical timer emulation. */
- phys_timer_emulate(vcpu);
+ set_cntvoff(map.direct_vtimer->cntvoff);
kvm_timer_unblocking(vcpu);
- /* If the timer fired while we weren't running, inject it now */
- if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
- kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
+ timer_restore_state(map.direct_vtimer);
+ if (map.direct_ptimer)
+ timer_restore_state(map.direct_ptimer);
+
+ if (map.emul_ptimer)
+ timer_emulate(map.emul_ptimer);
}
bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
@@ -568,20 +566,19 @@ bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
if (unlikely(!timer->enabled))
return;
- timer_save_state(vtimer);
- if (has_vhe()) {
- timer_save_state(ptimer);
- return;
- }
+ get_timer_map(vcpu, &map);
+
+ timer_save_state(map.direct_vtimer);
+ if (map.direct_ptimer)
+ timer_save_state(map.direct_ptimer);
/*
- * Cancel the physical timer emulation, because the only case where we
+ * Cancel soft timer emulation, because the only case where we
* need it after a vcpu_put is in the context of a sleeping VCPU, and
* in that case we already factor in the deadline for the physical
* timer when scheduling the bg_timer.
@@ -589,7 +586,8 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
* In any case, we re-schedule the hrtimer for the physical timer when
* coming back to the VCPU thread in kvm_timer_vcpu_load().
*/
- soft_timer_cancel(&ptimer->hrtimer);
+ if (map.emul_ptimer)
+ soft_timer_cancel(&map.emul_ptimer->hrtimer);
if (swait_active(kvm_arch_vcpu_wq(vcpu)))
kvm_timer_blocking(vcpu);
@@ -636,8 +634,9 @@ void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu)
int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
+
+ get_timer_map(vcpu, &map);
/*
* The bits in CNTV_CTL are architecturally reset to UNKNOWN for ARMv8
@@ -645,12 +644,22 @@ int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu)
* resets the timer to be disabled and unmasked and is compliant with
* the ARMv7 architecture.
*/
- vtimer->cnt_ctl = 0;
- ptimer->cnt_ctl = 0;
- kvm_timer_update_state(vcpu);
+ vcpu_vtimer(vcpu)->cnt_ctl = 0;
+ vcpu_ptimer(vcpu)->cnt_ctl = 0;
+
+ if (timer->enabled) {
+ kvm_timer_update_irq(vcpu, false, vcpu_vtimer(vcpu));
+ kvm_timer_update_irq(vcpu, false, vcpu_ptimer(vcpu));
+
+ if (irqchip_in_kernel(vcpu->kvm)) {
+ kvm_vgic_reset_mapped_irq(vcpu, map.direct_vtimer->irq.irq);
+ if (map.direct_ptimer)
+ kvm_vgic_reset_mapped_irq(vcpu, map.direct_ptimer->irq.irq);
+ }
+ }
- if (timer->enabled && irqchip_in_kernel(vcpu->kvm))
- kvm_vgic_reset_mapped_irq(vcpu, vtimer->irq.irq);
+ if (map.emul_ptimer)
+ soft_timer_cancel(&map.emul_ptimer->hrtimer);
return 0;
}
@@ -687,15 +696,18 @@ void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
hrtimer_init(&timer->bg_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
timer->bg_timer.function = kvm_bg_timer_expire;
+ hrtimer_init(&vtimer->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
hrtimer_init(&ptimer->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
- ptimer->hrtimer.function = kvm_phys_timer_expire;
+ vtimer->hrtimer.function = kvm_hrtimer_expire;
+ ptimer->hrtimer.function = kvm_hrtimer_expire;
vtimer->irq.irq = default_vtimer_irq.irq;
- vtimer->host_timer_irq = host_vtimer_irq;
- vtimer->host_timer_irq_flags = host_vtimer_irq_flags;
-
ptimer->irq.irq = default_ptimer_irq.irq;
+
+ vtimer->host_timer_irq = host_vtimer_irq;
ptimer->host_timer_irq = host_ptimer_irq;
+
+ vtimer->host_timer_irq_flags = host_vtimer_irq_flags;
ptimer->host_timer_irq_flags = host_ptimer_irq_flags;
vtimer->vcpu = vcpu;
@@ -710,32 +722,39 @@ static void kvm_timer_init_interrupt(void *info)
int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value)
{
+ struct arch_timer_context *timer;
+ bool level;
+
switch (regid) {
case KVM_REG_ARM_TIMER_CTL:
- kvm_arm_timer_write(vcpu,
- vcpu_vtimer(vcpu), TIMER_REG_CTL, value);
+ timer = vcpu_vtimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_TIMER_CNT:
+ timer = vcpu_vtimer(vcpu);
update_vtimer_cntvoff(vcpu, kvm_phys_timer_read() - value);
break;
case KVM_REG_ARM_TIMER_CVAL:
- kvm_arm_timer_write(vcpu,
- vcpu_vtimer(vcpu), TIMER_REG_CVAL, value);
+ timer = vcpu_vtimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CVAL, value);
break;
case KVM_REG_ARM_PTIMER_CTL:
- kvm_arm_timer_write(vcpu,
- vcpu_ptimer(vcpu), TIMER_REG_CTL, value);
+ timer = vcpu_ptimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_PTIMER_CVAL:
- kvm_arm_timer_write(vcpu,
- vcpu_ptimer(vcpu), TIMER_REG_CVAL, value);
+ timer = vcpu_ptimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CVAL, value);
break;
default:
return -1;
}
- kvm_timer_update_state(vcpu);
+ level = kvm_timer_should_fire(timer);
+ kvm_timer_update_irq(vcpu, level, timer);
+ timer_emulate(timer);
+
return 0;
}
@@ -1020,8 +1039,7 @@ bool kvm_arch_timer_get_input_level(int vintid)
int kvm_timer_enable(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
int ret;
if (timer->enabled)
@@ -1039,18 +1057,25 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
return -EINVAL;
}
- ret = kvm_vgic_map_phys_irq(vcpu, host_vtimer_irq, vtimer->irq.irq,
+ get_timer_map(vcpu, &map);
+
+ ret = kvm_vgic_map_phys_irq(vcpu,
+ map.direct_vtimer->host_timer_irq,
+ map.direct_vtimer->irq.irq,
kvm_arch_timer_get_input_level);
if (ret)
return ret;
- if (has_vhe()) {
- ret = kvm_vgic_map_phys_irq(vcpu, host_ptimer_irq, ptimer->irq.irq,
+ if (map.direct_ptimer) {
+ ret = kvm_vgic_map_phys_irq(vcpu,
+ map.direct_ptimer->host_timer_irq,
+ map.direct_ptimer->irq.irq,
kvm_arch_timer_get_input_level);
- if (ret)
- return ret;
}
+ if (ret)
+ return ret;
+
no_vgic:
timer->enabled = 1;
return 0;
diff --git a/virt/kvm/arm/trace.h b/virt/kvm/arm/trace.h
index 3828beab93f2..54bb059243b9 100644
--- a/virt/kvm/arm/trace.h
+++ b/virt/kvm/arm/trace.h
@@ -2,6 +2,7 @@
#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ)
#define _TRACE_KVM_H
+#include <kvm/arm_arch_timer.h>
#include <linux/tracepoint.h>
#undef TRACE_SYSTEM
@@ -262,6 +263,110 @@ TRACE_EVENT(kvm_timer_update_irq,
__entry->vcpu_id, __entry->irq, __entry->level)
);
+TRACE_EVENT(kvm_get_timer_map,
+ TP_PROTO(unsigned long vcpu_id, struct timer_map *map),
+ TP_ARGS(vcpu_id, map),
+
+ TP_STRUCT__entry(
+ __field( unsigned long, vcpu_id )
+ __field( int, direct_vtimer )
+ __field( int, direct_ptimer )
+ __field( int, emul_ptimer )
+ ),
+
+ TP_fast_assign(
+ __entry->vcpu_id = vcpu_id;
+ __entry->direct_vtimer = arch_timer_ctx_index(map->direct_vtimer);
+ __entry->direct_ptimer =
+ (map->direct_ptimer) ? arch_timer_ctx_index(map->direct_ptimer) : -1;
+ __entry->emul_ptimer =
+ (map->emul_ptimer) ? arch_timer_ctx_index(map->emul_ptimer) : -1;
+ ),
+
+ TP_printk("VCPU: %ld, dv: %d, dp: %d, ep: %d",
+ __entry->vcpu_id,
+ __entry->direct_vtimer,
+ __entry->direct_ptimer,
+ __entry->emul_ptimer)
+);
+
+TRACE_EVENT(kvm_timer_save_state,
+ TP_PROTO(struct arch_timer_context *ctx),
+ TP_ARGS(ctx),
+
+ TP_STRUCT__entry(
+ __field( unsigned long, ctl )
+ __field( unsigned long long, cval )
+ __field( int, timer_idx )
+ ),
+
+ TP_fast_assign(
+ __entry->ctl = ctx->cnt_ctl;
+ __entry->cval = ctx->cnt_cval;
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ ),
+
+ TP_printk(" CTL: %#08lx CVAL: %#16llx arch_timer_ctx_index: %d",
+ __entry->ctl,
+ __entry->cval,
+ __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_restore_state,
+ TP_PROTO(struct arch_timer_context *ctx),
+ TP_ARGS(ctx),
+
+ TP_STRUCT__entry(
+ __field( unsigned long, ctl )
+ __field( unsigned long long, cval )
+ __field( int, timer_idx )
+ ),
+
+ TP_fast_assign(
+ __entry->ctl = ctx->cnt_ctl;
+ __entry->cval = ctx->cnt_cval;
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ ),
+
+ TP_printk("CTL: %#08lx CVAL: %#16llx arch_timer_ctx_index: %d",
+ __entry->ctl,
+ __entry->cval,
+ __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_hrtimer_expire,
+ TP_PROTO(struct arch_timer_context *ctx),
+ TP_ARGS(ctx),
+
+ TP_STRUCT__entry(
+ __field( int, timer_idx )
+ ),
+
+ TP_fast_assign(
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ ),
+
+ TP_printk("arch_timer_ctx_index: %d", __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_emulate,
+ TP_PROTO(struct arch_timer_context *ctx, bool should_fire),
+ TP_ARGS(ctx, should_fire),
+
+ TP_STRUCT__entry(
+ __field( int, timer_idx )
+ __field( bool, should_fire )
+ ),
+
+ TP_fast_assign(
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ __entry->should_fire = should_fire;
+ ),
+
+ TP_printk("arch_timer_ctx_index: %d (should_fire: %d)",
+ __entry->timer_idx, __entry->should_fire)
+);
+
#endif /* _TRACE_KVM_H */
#undef TRACE_INCLUDE_PATH
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 14/27] KVM: arm/arm64: Rework the timer code to use a timer_map
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Christoffer Dall <christoffer.dall@arm.com>
We are currently emulating two timers in two different ways. When we
add support for nested virtualization in the future, we are going to be
emulating either two timers in two diffferent ways, or four timers in a
single way.
We need a unified data structure to keep track of how we map virtual
state to physical state and we need to cleanup some of the timer code to
operate more independently on a struct arch_timer_context instead of
trying to consider the global state of the VCPU and recomputing all
state.
Co-written with Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
---
include/kvm/arm_arch_timer.h | 23 +--
virt/kvm/arm/arch_timer.c | 295 +++++++++++++++++++----------------
virt/kvm/arm/trace.h | 105 +++++++++++++
3 files changed, 278 insertions(+), 145 deletions(-)
diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h
index 6d4a33a9c45a..05a18dd265b5 100644
--- a/include/kvm/arm_arch_timer.h
+++ b/include/kvm/arm_arch_timer.h
@@ -51,11 +51,24 @@ struct arch_timer_context {
/* Emulated Timer (may be unused) */
struct hrtimer hrtimer;
+ /*
+ * We have multiple paths which can save/restore the timer state onto
+ * the hardware, so we need some way of keeping track of where the
+ * latest state is.
+ */
+ bool loaded;
+
/* Duplicated state from arch_timer.c for convenience */
u32 host_timer_irq;
u32 host_timer_irq_flags;
};
+struct timer_map {
+ struct arch_timer_context *direct_vtimer;
+ struct arch_timer_context *direct_ptimer;
+ struct arch_timer_context *emul_ptimer;
+};
+
struct arch_timer_cpu {
struct arch_timer_context timers[NR_KVM_TIMERS];
@@ -64,16 +77,6 @@ struct arch_timer_cpu {
/* Is the timer enabled */
bool enabled;
-
- /*
- * We have multiple paths which can save/restore the timer state
- * onto the hardware, so we need some way of keeping track of
- * where the latest state is.
- *
- * loaded == true: State is loaded on the hardware registers.
- * loaded == false: State is stored in memory.
- */
- bool loaded;
};
int kvm_timer_hyp_init(bool);
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 10c15151c87e..17f9de73cc8a 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -68,6 +68,21 @@ u64 kvm_phys_timer_read(void)
return timecounter->cc->read(timecounter->cc);
}
+static void get_timer_map(struct kvm_vcpu *vcpu, struct timer_map *map)
+{
+ if (has_vhe()) {
+ map->direct_vtimer = vcpu_vtimer(vcpu);
+ map->direct_ptimer = vcpu_ptimer(vcpu);
+ map->emul_ptimer = NULL;
+ } else {
+ map->direct_vtimer = vcpu_vtimer(vcpu);
+ map->direct_ptimer = NULL;
+ map->emul_ptimer = vcpu_ptimer(vcpu);
+ }
+
+ trace_kvm_get_timer_map(vcpu->vcpu_id, map);
+}
+
static inline bool userspace_irqchip(struct kvm *kvm)
{
return static_branch_unlikely(&userspace_irqchip_in_use) &&
@@ -89,6 +104,7 @@ static irqreturn_t kvm_arch_timer_handler(int irq, void *dev_id)
{
struct kvm_vcpu *vcpu = *(struct kvm_vcpu **)dev_id;
struct arch_timer_context *ctx;
+ struct timer_map map;
/*
* We may see a timer interrupt after vcpu_put() has been called which
@@ -99,10 +115,12 @@ static irqreturn_t kvm_arch_timer_handler(int irq, void *dev_id)
if (!vcpu)
return IRQ_HANDLED;
+ get_timer_map(vcpu, &map);
+
if (irq == host_vtimer_irq)
- ctx = vcpu_vtimer(vcpu);
+ ctx = map.direct_vtimer;
else
- ctx = vcpu_ptimer(vcpu);
+ ctx = map.direct_ptimer;
if (kvm_timer_should_fire(ctx))
kvm_timer_update_irq(vcpu, true, ctx);
@@ -136,7 +154,9 @@ static u64 kvm_timer_compute_delta(struct arch_timer_context *timer_ctx)
static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx)
{
- return !(timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_IT_MASK) &&
+ WARN_ON(timer_ctx && timer_ctx->loaded);
+ return timer_ctx &&
+ !(timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_IT_MASK) &&
(timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_ENABLE);
}
@@ -146,21 +166,22 @@ static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx)
*/
static u64 kvm_timer_earliest_exp(struct kvm_vcpu *vcpu)
{
- u64 min_virt = ULLONG_MAX, min_phys = ULLONG_MAX;
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ u64 min_delta = ULLONG_MAX;
+ int i;
- if (kvm_timer_irq_can_fire(vtimer))
- min_virt = kvm_timer_compute_delta(vtimer);
+ for (i = 0; i < NR_KVM_TIMERS; i++) {
+ struct arch_timer_context *ctx = &vcpu->arch.timer_cpu.timers[i];
- if (kvm_timer_irq_can_fire(ptimer))
- min_phys = kvm_timer_compute_delta(ptimer);
+ WARN(ctx->loaded, "timer %d loaded\n", i);
+ if (kvm_timer_irq_can_fire(ctx))
+ min_delta = min(min_delta, kvm_timer_compute_delta(ctx));
+ }
/* If none of timers can fire, then return 0 */
- if ((min_virt == ULLONG_MAX) && (min_phys == ULLONG_MAX))
+ if (min_delta == ULLONG_MAX)
return 0;
- return min(min_virt, min_phys);
+ return min_delta;
}
static enum hrtimer_restart kvm_bg_timer_expire(struct hrtimer *hrt)
@@ -187,37 +208,45 @@ static enum hrtimer_restart kvm_bg_timer_expire(struct hrtimer *hrt)
return HRTIMER_NORESTART;
}
-static enum hrtimer_restart kvm_phys_timer_expire(struct hrtimer *hrt)
+static enum hrtimer_restart kvm_hrtimer_expire(struct hrtimer *hrt)
{
- struct arch_timer_context *ptimer;
+ struct arch_timer_context *ctx;
struct kvm_vcpu *vcpu;
u64 ns;
- ptimer = container_of(hrt, struct arch_timer_context, hrtimer);
- vcpu = ptimer->vcpu;
+ ctx = container_of(hrt, struct arch_timer_context, hrtimer);
+ vcpu = ctx->vcpu;
+
+ trace_kvm_timer_hrtimer_expire(ctx);
/*
* Check that the timer has really expired from the guest's
* PoV (NTP on the host may have forced it to expire
* early). If not ready, schedule for a later time.
*/
- ns = kvm_timer_compute_delta(ptimer);
+ ns = kvm_timer_compute_delta(ctx);
if (unlikely(ns)) {
hrtimer_forward_now(hrt, ns_to_ktime(ns));
return HRTIMER_RESTART;
}
- kvm_timer_update_irq(vcpu, true, ptimer);
+ kvm_timer_update_irq(vcpu, true, ctx);
return HRTIMER_NORESTART;
}
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
- struct arch_timer_cpu *timer = vcpu_timer(timer_ctx->vcpu);
- enum kvm_arch_timers index = arch_timer_ctx_index(timer_ctx);
+ struct arch_timer_cpu *timer;
+ enum kvm_arch_timers index;
u64 cval, now;
- if (timer->loaded) {
+ if (!timer_ctx)
+ return false;
+
+ timer = vcpu_timer(timer_ctx->vcpu);
+ index = arch_timer_ctx_index(timer_ctx);
+
+ if (timer_ctx->loaded) {
u32 cnt_ctl = 0;
switch (index) {
@@ -249,13 +278,13 @@ static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
bool kvm_timer_is_pending(struct kvm_vcpu *vcpu)
{
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
- if (kvm_timer_should_fire(vtimer))
- return true;
+ get_timer_map(vcpu, &map);
- return kvm_timer_should_fire(ptimer);
+ return kvm_timer_should_fire(map.direct_vtimer) ||
+ kvm_timer_should_fire(map.direct_ptimer) ||
+ kvm_timer_should_fire(map.emul_ptimer);
}
/*
@@ -294,60 +323,28 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level,
}
}
-/* Schedule the background timer for the emulated timer. */
-static void phys_timer_emulate(struct kvm_vcpu *vcpu)
+static void timer_emulate(struct arch_timer_context *ctx)
{
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ bool should_fire = kvm_timer_should_fire(ctx);
- /*
- * If the timer can fire now, we don't need to have a soft timer
- * scheduled for the future. If the timer cannot fire at all,
- * then we also don't need a soft timer.
- */
- if (kvm_timer_should_fire(ptimer) || !kvm_timer_irq_can_fire(ptimer)) {
- soft_timer_cancel(&ptimer->hrtimer);
- return;
- }
-
- soft_timer_start(&ptimer->hrtimer, kvm_timer_compute_delta(ptimer));
-}
-
-/*
- * Check if there was a change in the timer state, so that we should either
- * raise or lower the line level to the GIC or schedule a background timer to
- * emulate the physical timer.
- */
-static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
-{
- struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
- bool level;
+ trace_kvm_timer_emulate(ctx, should_fire);
- if (unlikely(!timer->enabled))
+ if (should_fire) {
+ kvm_timer_update_irq(ctx->vcpu, true, ctx);
return;
+ }
/*
- * If the timer virtual interrupt is a 'mapped' interrupt, part
- * of its lifecycle is offloaded to the hardware, and we therefore may
- * not have lowered the irq.level value before having to signal a new
- * interrupt, but have to signal an interrupt every time the level is
- * asserted.
+ * If the timer can fire now, we don't need to have a soft timer
+ * scheduled for the future. If the timer cannot fire at all,
+ * then we also don't need a soft timer.
*/
- level = kvm_timer_should_fire(vtimer);
- kvm_timer_update_irq(vcpu, level, vtimer);
-
- if (has_vhe()) {
- level = kvm_timer_should_fire(ptimer);
- kvm_timer_update_irq(vcpu, level, ptimer);
-
+ if (!kvm_timer_irq_can_fire(ctx)) {
+ soft_timer_cancel(&ctx->hrtimer);
return;
}
- phys_timer_emulate(vcpu);
-
- if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
- kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
+ soft_timer_start(&ctx->hrtimer, kvm_timer_compute_delta(ctx));
}
static void timer_save_state(struct arch_timer_context *ctx)
@@ -361,7 +358,7 @@ static void timer_save_state(struct arch_timer_context *ctx)
local_irq_save(flags);
- if (!timer->loaded)
+ if (!ctx->loaded)
goto out;
switch (index) {
@@ -384,10 +381,12 @@ static void timer_save_state(struct arch_timer_context *ctx)
break;
case NR_KVM_TIMERS:
- break; /* GCC is braindead */
+ BUG();
}
- timer->loaded = false;
+ trace_kvm_timer_save_state(ctx);
+
+ ctx->loaded = false;
out:
local_irq_restore(flags);
}
@@ -400,14 +399,17 @@ static void timer_save_state(struct arch_timer_context *ctx)
static void kvm_timer_blocking(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
+
+ get_timer_map(vcpu, &map);
/*
- * If both timers are not capable of raising interrupts (disabled or
+ * If no timers are capable of raising interrupts (disabled or
* masked), then there's no more work for us to do.
*/
- if (!kvm_timer_irq_can_fire(vtimer) && !kvm_timer_irq_can_fire(ptimer))
+ if (!kvm_timer_irq_can_fire(map.direct_vtimer) &&
+ !kvm_timer_irq_can_fire(map.direct_ptimer) &&
+ !kvm_timer_irq_can_fire(map.emul_ptimer))
return;
/*
@@ -435,7 +437,7 @@ static void timer_restore_state(struct arch_timer_context *ctx)
local_irq_save(flags);
- if (timer->loaded)
+ if (ctx->loaded)
goto out;
switch (index) {
@@ -450,10 +452,12 @@ static void timer_restore_state(struct arch_timer_context *ctx)
write_sysreg_el0(ctx->cnt_ctl, cntp_ctl);
break;
case NR_KVM_TIMERS:
- break; /* GCC is braindead */
+ BUG();
}
- timer->loaded = true;
+ trace_kvm_timer_restore_state(ctx);
+
+ ctx->loaded = true;
out:
local_irq_restore(flags);
}
@@ -515,37 +519,31 @@ static void kvm_timer_vcpu_load_nogic(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_load(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
if (unlikely(!timer->enabled))
return;
+ get_timer_map(vcpu, &map);
+
if (static_branch_likely(&has_gic_active_state)) {
- kvm_timer_vcpu_load_gic(vtimer);
- if (has_vhe())
- kvm_timer_vcpu_load_gic(ptimer);
+ kvm_timer_vcpu_load_gic(map.direct_vtimer);
+ if (map.direct_ptimer)
+ kvm_timer_vcpu_load_gic(map.direct_ptimer);
} else {
kvm_timer_vcpu_load_nogic(vcpu);
}
- set_cntvoff(vtimer->cntvoff);
-
- timer_restore_state(vtimer);
-
- if (has_vhe()) {
- timer_restore_state(ptimer);
- return;
- }
-
- /* Set the background timer for the physical timer emulation. */
- phys_timer_emulate(vcpu);
+ set_cntvoff(map.direct_vtimer->cntvoff);
kvm_timer_unblocking(vcpu);
- /* If the timer fired while we weren't running, inject it now */
- if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
- kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
+ timer_restore_state(map.direct_vtimer);
+ if (map.direct_ptimer)
+ timer_restore_state(map.direct_ptimer);
+
+ if (map.emul_ptimer)
+ timer_emulate(map.emul_ptimer);
}
bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
@@ -568,20 +566,19 @@ bool kvm_timer_should_notify_user(struct kvm_vcpu *vcpu)
void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
if (unlikely(!timer->enabled))
return;
- timer_save_state(vtimer);
- if (has_vhe()) {
- timer_save_state(ptimer);
- return;
- }
+ get_timer_map(vcpu, &map);
+
+ timer_save_state(map.direct_vtimer);
+ if (map.direct_ptimer)
+ timer_save_state(map.direct_ptimer);
/*
- * Cancel the physical timer emulation, because the only case where we
+ * Cancel soft timer emulation, because the only case where we
* need it after a vcpu_put is in the context of a sleeping VCPU, and
* in that case we already factor in the deadline for the physical
* timer when scheduling the bg_timer.
@@ -589,7 +586,8 @@ void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
* In any case, we re-schedule the hrtimer for the physical timer when
* coming back to the VCPU thread in kvm_timer_vcpu_load().
*/
- soft_timer_cancel(&ptimer->hrtimer);
+ if (map.emul_ptimer)
+ soft_timer_cancel(&map.emul_ptimer->hrtimer);
if (swait_active(kvm_arch_vcpu_wq(vcpu)))
kvm_timer_blocking(vcpu);
@@ -636,8 +634,9 @@ void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu)
int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
+
+ get_timer_map(vcpu, &map);
/*
* The bits in CNTV_CTL are architecturally reset to UNKNOWN for ARMv8
@@ -645,12 +644,22 @@ int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu)
* resets the timer to be disabled and unmasked and is compliant with
* the ARMv7 architecture.
*/
- vtimer->cnt_ctl = 0;
- ptimer->cnt_ctl = 0;
- kvm_timer_update_state(vcpu);
+ vcpu_vtimer(vcpu)->cnt_ctl = 0;
+ vcpu_ptimer(vcpu)->cnt_ctl = 0;
+
+ if (timer->enabled) {
+ kvm_timer_update_irq(vcpu, false, vcpu_vtimer(vcpu));
+ kvm_timer_update_irq(vcpu, false, vcpu_ptimer(vcpu));
+
+ if (irqchip_in_kernel(vcpu->kvm)) {
+ kvm_vgic_reset_mapped_irq(vcpu, map.direct_vtimer->irq.irq);
+ if (map.direct_ptimer)
+ kvm_vgic_reset_mapped_irq(vcpu, map.direct_ptimer->irq.irq);
+ }
+ }
- if (timer->enabled && irqchip_in_kernel(vcpu->kvm))
- kvm_vgic_reset_mapped_irq(vcpu, vtimer->irq.irq);
+ if (map.emul_ptimer)
+ soft_timer_cancel(&map.emul_ptimer->hrtimer);
return 0;
}
@@ -687,15 +696,18 @@ void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu)
hrtimer_init(&timer->bg_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
timer->bg_timer.function = kvm_bg_timer_expire;
+ hrtimer_init(&vtimer->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
hrtimer_init(&ptimer->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
- ptimer->hrtimer.function = kvm_phys_timer_expire;
+ vtimer->hrtimer.function = kvm_hrtimer_expire;
+ ptimer->hrtimer.function = kvm_hrtimer_expire;
vtimer->irq.irq = default_vtimer_irq.irq;
- vtimer->host_timer_irq = host_vtimer_irq;
- vtimer->host_timer_irq_flags = host_vtimer_irq_flags;
-
ptimer->irq.irq = default_ptimer_irq.irq;
+
+ vtimer->host_timer_irq = host_vtimer_irq;
ptimer->host_timer_irq = host_ptimer_irq;
+
+ vtimer->host_timer_irq_flags = host_vtimer_irq_flags;
ptimer->host_timer_irq_flags = host_ptimer_irq_flags;
vtimer->vcpu = vcpu;
@@ -710,32 +722,39 @@ static void kvm_timer_init_interrupt(void *info)
int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value)
{
+ struct arch_timer_context *timer;
+ bool level;
+
switch (regid) {
case KVM_REG_ARM_TIMER_CTL:
- kvm_arm_timer_write(vcpu,
- vcpu_vtimer(vcpu), TIMER_REG_CTL, value);
+ timer = vcpu_vtimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_TIMER_CNT:
+ timer = vcpu_vtimer(vcpu);
update_vtimer_cntvoff(vcpu, kvm_phys_timer_read() - value);
break;
case KVM_REG_ARM_TIMER_CVAL:
- kvm_arm_timer_write(vcpu,
- vcpu_vtimer(vcpu), TIMER_REG_CVAL, value);
+ timer = vcpu_vtimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CVAL, value);
break;
case KVM_REG_ARM_PTIMER_CTL:
- kvm_arm_timer_write(vcpu,
- vcpu_ptimer(vcpu), TIMER_REG_CTL, value);
+ timer = vcpu_ptimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CTL, value);
break;
case KVM_REG_ARM_PTIMER_CVAL:
- kvm_arm_timer_write(vcpu,
- vcpu_ptimer(vcpu), TIMER_REG_CVAL, value);
+ timer = vcpu_ptimer(vcpu);
+ kvm_arm_timer_write(vcpu, timer, TIMER_REG_CVAL, value);
break;
default:
return -1;
}
- kvm_timer_update_state(vcpu);
+ level = kvm_timer_should_fire(timer);
+ kvm_timer_update_irq(vcpu, level, timer);
+ timer_emulate(timer);
+
return 0;
}
@@ -1020,8 +1039,7 @@ bool kvm_arch_timer_get_input_level(int vintid)
int kvm_timer_enable(struct kvm_vcpu *vcpu)
{
struct arch_timer_cpu *timer = vcpu_timer(vcpu);
- struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
- struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+ struct timer_map map;
int ret;
if (timer->enabled)
@@ -1039,18 +1057,25 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
return -EINVAL;
}
- ret = kvm_vgic_map_phys_irq(vcpu, host_vtimer_irq, vtimer->irq.irq,
+ get_timer_map(vcpu, &map);
+
+ ret = kvm_vgic_map_phys_irq(vcpu,
+ map.direct_vtimer->host_timer_irq,
+ map.direct_vtimer->irq.irq,
kvm_arch_timer_get_input_level);
if (ret)
return ret;
- if (has_vhe()) {
- ret = kvm_vgic_map_phys_irq(vcpu, host_ptimer_irq, ptimer->irq.irq,
+ if (map.direct_ptimer) {
+ ret = kvm_vgic_map_phys_irq(vcpu,
+ map.direct_ptimer->host_timer_irq,
+ map.direct_ptimer->irq.irq,
kvm_arch_timer_get_input_level);
- if (ret)
- return ret;
}
+ if (ret)
+ return ret;
+
no_vgic:
timer->enabled = 1;
return 0;
diff --git a/virt/kvm/arm/trace.h b/virt/kvm/arm/trace.h
index 3828beab93f2..54bb059243b9 100644
--- a/virt/kvm/arm/trace.h
+++ b/virt/kvm/arm/trace.h
@@ -2,6 +2,7 @@
#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ)
#define _TRACE_KVM_H
+#include <kvm/arm_arch_timer.h>
#include <linux/tracepoint.h>
#undef TRACE_SYSTEM
@@ -262,6 +263,110 @@ TRACE_EVENT(kvm_timer_update_irq,
__entry->vcpu_id, __entry->irq, __entry->level)
);
+TRACE_EVENT(kvm_get_timer_map,
+ TP_PROTO(unsigned long vcpu_id, struct timer_map *map),
+ TP_ARGS(vcpu_id, map),
+
+ TP_STRUCT__entry(
+ __field( unsigned long, vcpu_id )
+ __field( int, direct_vtimer )
+ __field( int, direct_ptimer )
+ __field( int, emul_ptimer )
+ ),
+
+ TP_fast_assign(
+ __entry->vcpu_id = vcpu_id;
+ __entry->direct_vtimer = arch_timer_ctx_index(map->direct_vtimer);
+ __entry->direct_ptimer =
+ (map->direct_ptimer) ? arch_timer_ctx_index(map->direct_ptimer) : -1;
+ __entry->emul_ptimer =
+ (map->emul_ptimer) ? arch_timer_ctx_index(map->emul_ptimer) : -1;
+ ),
+
+ TP_printk("VCPU: %ld, dv: %d, dp: %d, ep: %d",
+ __entry->vcpu_id,
+ __entry->direct_vtimer,
+ __entry->direct_ptimer,
+ __entry->emul_ptimer)
+);
+
+TRACE_EVENT(kvm_timer_save_state,
+ TP_PROTO(struct arch_timer_context *ctx),
+ TP_ARGS(ctx),
+
+ TP_STRUCT__entry(
+ __field( unsigned long, ctl )
+ __field( unsigned long long, cval )
+ __field( int, timer_idx )
+ ),
+
+ TP_fast_assign(
+ __entry->ctl = ctx->cnt_ctl;
+ __entry->cval = ctx->cnt_cval;
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ ),
+
+ TP_printk(" CTL: %#08lx CVAL: %#16llx arch_timer_ctx_index: %d",
+ __entry->ctl,
+ __entry->cval,
+ __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_restore_state,
+ TP_PROTO(struct arch_timer_context *ctx),
+ TP_ARGS(ctx),
+
+ TP_STRUCT__entry(
+ __field( unsigned long, ctl )
+ __field( unsigned long long, cval )
+ __field( int, timer_idx )
+ ),
+
+ TP_fast_assign(
+ __entry->ctl = ctx->cnt_ctl;
+ __entry->cval = ctx->cnt_cval;
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ ),
+
+ TP_printk("CTL: %#08lx CVAL: %#16llx arch_timer_ctx_index: %d",
+ __entry->ctl,
+ __entry->cval,
+ __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_hrtimer_expire,
+ TP_PROTO(struct arch_timer_context *ctx),
+ TP_ARGS(ctx),
+
+ TP_STRUCT__entry(
+ __field( int, timer_idx )
+ ),
+
+ TP_fast_assign(
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ ),
+
+ TP_printk("arch_timer_ctx_index: %d", __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_emulate,
+ TP_PROTO(struct arch_timer_context *ctx, bool should_fire),
+ TP_ARGS(ctx, should_fire),
+
+ TP_STRUCT__entry(
+ __field( int, timer_idx )
+ __field( bool, should_fire )
+ ),
+
+ TP_fast_assign(
+ __entry->timer_idx = arch_timer_ctx_index(ctx);
+ __entry->should_fire = should_fire;
+ ),
+
+ TP_printk("arch_timer_ctx_index: %d (should_fire: %d)",
+ __entry->timer_idx, __entry->should_fire)
+);
+
#endif /* _TRACE_KVM_H */
#undef TRACE_INCLUDE_PATH
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 15/27] KVM: arm/arm64: Move kvm_is_write_fault to header file
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Christoffer Dall <christoffer.dall@arm.com>
Move this little function to the header files for arm/arm64 so other
code can make use of it directly.
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/include/asm/kvm_emulate.h | 8 ++++++++
arch/arm64/include/asm/kvm_emulate.h | 8 ++++++++
virt/kvm/arm/mmu.c | 8 --------
3 files changed, 16 insertions(+), 8 deletions(-)
diff --git a/arch/arm/include/asm/kvm_emulate.h b/arch/arm/include/asm/kvm_emulate.h
index 77121b713bef..8927cae7c966 100644
--- a/arch/arm/include/asm/kvm_emulate.h
+++ b/arch/arm/include/asm/kvm_emulate.h
@@ -265,6 +265,14 @@ static inline bool kvm_vcpu_dabt_isextabt(struct kvm_vcpu *vcpu)
}
}
+static inline bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
+{
+ if (kvm_vcpu_trap_is_iabt(vcpu))
+ return false;
+
+ return kvm_vcpu_dabt_iswrite(vcpu);
+}
+
static inline u32 kvm_vcpu_hvc_get_imm(struct kvm_vcpu *vcpu)
{
return kvm_vcpu_get_hsr(vcpu) & HSR_HVC_IMM_MASK;
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index 506386a3edde..a0d1ce9ae12b 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -331,6 +331,14 @@ static inline int kvm_vcpu_sys_get_rt(struct kvm_vcpu *vcpu)
return ESR_ELx_SYS64_ISS_RT(esr);
}
+static inline bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
+{
+ if (kvm_vcpu_trap_is_iabt(vcpu))
+ return false;
+
+ return kvm_vcpu_dabt_iswrite(vcpu);
+}
+
static inline unsigned long kvm_vcpu_get_mpidr_aff(struct kvm_vcpu *vcpu)
{
return vcpu_read_sys_reg(vcpu, MPIDR_EL1) & MPIDR_HWID_BITMASK;
diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index f8dda452ea24..e3e5a26b4845 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -1403,14 +1403,6 @@ static bool transparent_hugepage_adjust(kvm_pfn_t *pfnp, phys_addr_t *ipap)
return false;
}
-static bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
-{
- if (kvm_vcpu_trap_is_iabt(vcpu))
- return false;
-
- return kvm_vcpu_dabt_iswrite(vcpu);
-}
-
/**
* stage2_wp_ptes - write protect PMD range
* @pmd: pointer to pmd entry
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 15/27] KVM: arm/arm64: Move kvm_is_write_fault to header file
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Christoffer Dall <christoffer.dall@arm.com>
Move this little function to the header files for arm/arm64 so other
code can make use of it directly.
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/include/asm/kvm_emulate.h | 8 ++++++++
arch/arm64/include/asm/kvm_emulate.h | 8 ++++++++
virt/kvm/arm/mmu.c | 8 --------
3 files changed, 16 insertions(+), 8 deletions(-)
diff --git a/arch/arm/include/asm/kvm_emulate.h b/arch/arm/include/asm/kvm_emulate.h
index 77121b713bef..8927cae7c966 100644
--- a/arch/arm/include/asm/kvm_emulate.h
+++ b/arch/arm/include/asm/kvm_emulate.h
@@ -265,6 +265,14 @@ static inline bool kvm_vcpu_dabt_isextabt(struct kvm_vcpu *vcpu)
}
}
+static inline bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
+{
+ if (kvm_vcpu_trap_is_iabt(vcpu))
+ return false;
+
+ return kvm_vcpu_dabt_iswrite(vcpu);
+}
+
static inline u32 kvm_vcpu_hvc_get_imm(struct kvm_vcpu *vcpu)
{
return kvm_vcpu_get_hsr(vcpu) & HSR_HVC_IMM_MASK;
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index 506386a3edde..a0d1ce9ae12b 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -331,6 +331,14 @@ static inline int kvm_vcpu_sys_get_rt(struct kvm_vcpu *vcpu)
return ESR_ELx_SYS64_ISS_RT(esr);
}
+static inline bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
+{
+ if (kvm_vcpu_trap_is_iabt(vcpu))
+ return false;
+
+ return kvm_vcpu_dabt_iswrite(vcpu);
+}
+
static inline unsigned long kvm_vcpu_get_mpidr_aff(struct kvm_vcpu *vcpu)
{
return vcpu_read_sys_reg(vcpu, MPIDR_EL1) & MPIDR_HWID_BITMASK;
diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index f8dda452ea24..e3e5a26b4845 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -1403,14 +1403,6 @@ static bool transparent_hugepage_adjust(kvm_pfn_t *pfnp, phys_addr_t *ipap)
return false;
}
-static bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
-{
- if (kvm_vcpu_trap_is_iabt(vcpu))
- return false;
-
- return kvm_vcpu_dabt_iswrite(vcpu);
-}
-
/**
* stage2_wp_ptes - write protect PMD range
* @pmd: pointer to pmd entry
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 16/27] arm64: KVM: Expose sanitised cache type register to guest
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
We currently permit CPUs in the same system to deviate in the exact
topology of the caches, and we subsequently hide this fact from user
space by exposing a sanitised value of the cache type register CTR_EL0.
However, guests running under KVM see the bare value of CTR_EL0, which
could potentially result in issues with, e.g., JITs or other pieces of
code that are sensitive to misreported cache line sizes.
So let's start trapping cache ID instructions if there is a mismatch,
and expose the sanitised version of CTR_EL0 to guests. Note that CTR_EL0
is treated as an invariant to KVM user space, so update that part as well.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/include/asm/kvm_emulate.h | 3 ++
arch/arm64/include/asm/sysreg.h | 1 +
arch/arm64/kvm/sys_regs.c | 59 +++++++++++++++++++++++++++-
3 files changed, 61 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index a0d1ce9ae12b..9acccb1e3746 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -77,6 +77,9 @@ static inline void vcpu_reset_hcr(struct kvm_vcpu *vcpu)
*/
if (!vcpu_el1_is_32bit(vcpu))
vcpu->arch.hcr_el2 |= HCR_TID3;
+
+ if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE))
+ vcpu->arch.hcr_el2 |= HCR_TID2;
}
static inline unsigned long *vcpu_hcr(struct kvm_vcpu *vcpu)
diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h
index 6482e8bcf1b8..5b267dec6194 100644
--- a/arch/arm64/include/asm/sysreg.h
+++ b/arch/arm64/include/asm/sysreg.h
@@ -361,6 +361,7 @@
#define SYS_CNTKCTL_EL1 sys_reg(3, 0, 14, 1, 0)
+#define SYS_CCSIDR_EL1 sys_reg(3, 1, 0, 0, 0)
#define SYS_CLIDR_EL1 sys_reg(3, 1, 0, 0, 1)
#define SYS_AIDR_EL1 sys_reg(3, 1, 0, 0, 7)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 0d348500b24b..45a07cfc57ed 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1146,6 +1146,49 @@ static int set_raz_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
return __set_id_reg(rd, uaddr, true);
}
+static bool access_ctr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ if (p->is_write)
+ return write_to_read_only(vcpu, p, r);
+
+ p->regval = read_sanitised_ftr_reg(SYS_CTR_EL0);
+ return true;
+}
+
+static bool access_clidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ if (p->is_write)
+ return write_to_read_only(vcpu, p, r);
+
+ p->regval = read_sysreg(clidr_el1);
+ return true;
+}
+
+static bool access_csselr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ if (p->is_write)
+ vcpu_write_sys_reg(vcpu, p->regval, r->reg);
+ else
+ p->regval = vcpu_read_sys_reg(vcpu, r->reg);
+ return true;
+}
+
+static bool access_ccsidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ u32 csselr;
+
+ if (p->is_write)
+ return write_to_read_only(vcpu, p, r);
+
+ csselr = vcpu_read_sys_reg(vcpu, CSSELR_EL1);
+ p->regval = get_ccsidr(csselr);
+ return true;
+}
+
/* sys_reg_desc initialiser for known cpufeature ID registers */
#define ID_SANITISED(name) { \
SYS_DESC(SYS_##name), \
@@ -1363,7 +1406,10 @@ static const struct sys_reg_desc sys_reg_descs[] = {
{ SYS_DESC(SYS_CNTKCTL_EL1), NULL, reset_val, CNTKCTL_EL1, 0},
- { SYS_DESC(SYS_CSSELR_EL1), NULL, reset_unknown, CSSELR_EL1 },
+ { SYS_DESC(SYS_CCSIDR_EL1), access_ccsidr },
+ { SYS_DESC(SYS_CLIDR_EL1), access_clidr },
+ { SYS_DESC(SYS_CSSELR_EL1), access_csselr, reset_unknown, CSSELR_EL1 },
+ { SYS_DESC(SYS_CTR_EL0), access_ctr },
{ SYS_DESC(SYS_PMCR_EL0), access_pmcr, reset_pmcr, },
{ SYS_DESC(SYS_PMCNTENSET_EL0), access_pmcnten, reset_unknown, PMCNTENSET_EL0 },
@@ -1663,6 +1709,7 @@ static const struct sys_reg_desc cp14_64_regs[] = {
* register).
*/
static const struct sys_reg_desc cp15_regs[] = {
+ { Op1( 0), CRn( 0), CRm( 0), Op2( 1), access_ctr },
{ Op1( 0), CRn( 1), CRm( 0), Op2( 0), access_vm_reg, NULL, c1_SCTLR },
{ Op1( 0), CRn( 2), CRm( 0), Op2( 0), access_vm_reg, NULL, c2_TTBR0 },
{ Op1( 0), CRn( 2), CRm( 0), Op2( 1), access_vm_reg, NULL, c2_TTBR1 },
@@ -1779,6 +1826,10 @@ static const struct sys_reg_desc cp15_regs[] = {
PMU_PMEVTYPER(30),
/* PMCCFILTR */
{ Op1(0), CRn(14), CRm(15), Op2(7), access_pmu_evtyper },
+
+ { Op1(1), CRn( 0), CRm( 0), Op2(0), access_ccsidr },
+ { Op1(1), CRn( 0), CRm( 0), Op2(1), access_clidr },
+ { Op1(2), CRn( 0), CRm( 0), Op2(0), access_csselr, NULL, c0_CSSELR },
};
static const struct sys_reg_desc cp15_64_regs[] = {
@@ -2192,11 +2243,15 @@ static const struct sys_reg_desc *index_to_sys_reg_desc(struct kvm_vcpu *vcpu,
}
FUNCTION_INVARIANT(midr_el1)
-FUNCTION_INVARIANT(ctr_el0)
FUNCTION_INVARIANT(revidr_el1)
FUNCTION_INVARIANT(clidr_el1)
FUNCTION_INVARIANT(aidr_el1)
+static void get_ctr_el0(struct kvm_vcpu *v, const struct sys_reg_desc *r)
+{
+ ((struct sys_reg_desc *)r)->val = read_sanitised_ftr_reg(SYS_CTR_EL0);
+}
+
/* ->val is filled in by kvm_sys_reg_table_init() */
static struct sys_reg_desc invariant_sys_regs[] = {
{ SYS_DESC(SYS_MIDR_EL1), NULL, get_midr_el1 },
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 16/27] arm64: KVM: Expose sanitised cache type register to guest
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
We currently permit CPUs in the same system to deviate in the exact
topology of the caches, and we subsequently hide this fact from user
space by exposing a sanitised value of the cache type register CTR_EL0.
However, guests running under KVM see the bare value of CTR_EL0, which
could potentially result in issues with, e.g., JITs or other pieces of
code that are sensitive to misreported cache line sizes.
So let's start trapping cache ID instructions if there is a mismatch,
and expose the sanitised version of CTR_EL0 to guests. Note that CTR_EL0
is treated as an invariant to KVM user space, so update that part as well.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/include/asm/kvm_emulate.h | 3 ++
arch/arm64/include/asm/sysreg.h | 1 +
arch/arm64/kvm/sys_regs.c | 59 +++++++++++++++++++++++++++-
3 files changed, 61 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index a0d1ce9ae12b..9acccb1e3746 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -77,6 +77,9 @@ static inline void vcpu_reset_hcr(struct kvm_vcpu *vcpu)
*/
if (!vcpu_el1_is_32bit(vcpu))
vcpu->arch.hcr_el2 |= HCR_TID3;
+
+ if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE))
+ vcpu->arch.hcr_el2 |= HCR_TID2;
}
static inline unsigned long *vcpu_hcr(struct kvm_vcpu *vcpu)
diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h
index 6482e8bcf1b8..5b267dec6194 100644
--- a/arch/arm64/include/asm/sysreg.h
+++ b/arch/arm64/include/asm/sysreg.h
@@ -361,6 +361,7 @@
#define SYS_CNTKCTL_EL1 sys_reg(3, 0, 14, 1, 0)
+#define SYS_CCSIDR_EL1 sys_reg(3, 1, 0, 0, 0)
#define SYS_CLIDR_EL1 sys_reg(3, 1, 0, 0, 1)
#define SYS_AIDR_EL1 sys_reg(3, 1, 0, 0, 7)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 0d348500b24b..45a07cfc57ed 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1146,6 +1146,49 @@ static int set_raz_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd,
return __set_id_reg(rd, uaddr, true);
}
+static bool access_ctr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ if (p->is_write)
+ return write_to_read_only(vcpu, p, r);
+
+ p->regval = read_sanitised_ftr_reg(SYS_CTR_EL0);
+ return true;
+}
+
+static bool access_clidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ if (p->is_write)
+ return write_to_read_only(vcpu, p, r);
+
+ p->regval = read_sysreg(clidr_el1);
+ return true;
+}
+
+static bool access_csselr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ if (p->is_write)
+ vcpu_write_sys_reg(vcpu, p->regval, r->reg);
+ else
+ p->regval = vcpu_read_sys_reg(vcpu, r->reg);
+ return true;
+}
+
+static bool access_ccsidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
+ const struct sys_reg_desc *r)
+{
+ u32 csselr;
+
+ if (p->is_write)
+ return write_to_read_only(vcpu, p, r);
+
+ csselr = vcpu_read_sys_reg(vcpu, CSSELR_EL1);
+ p->regval = get_ccsidr(csselr);
+ return true;
+}
+
/* sys_reg_desc initialiser for known cpufeature ID registers */
#define ID_SANITISED(name) { \
SYS_DESC(SYS_##name), \
@@ -1363,7 +1406,10 @@ static const struct sys_reg_desc sys_reg_descs[] = {
{ SYS_DESC(SYS_CNTKCTL_EL1), NULL, reset_val, CNTKCTL_EL1, 0},
- { SYS_DESC(SYS_CSSELR_EL1), NULL, reset_unknown, CSSELR_EL1 },
+ { SYS_DESC(SYS_CCSIDR_EL1), access_ccsidr },
+ { SYS_DESC(SYS_CLIDR_EL1), access_clidr },
+ { SYS_DESC(SYS_CSSELR_EL1), access_csselr, reset_unknown, CSSELR_EL1 },
+ { SYS_DESC(SYS_CTR_EL0), access_ctr },
{ SYS_DESC(SYS_PMCR_EL0), access_pmcr, reset_pmcr, },
{ SYS_DESC(SYS_PMCNTENSET_EL0), access_pmcnten, reset_unknown, PMCNTENSET_EL0 },
@@ -1663,6 +1709,7 @@ static const struct sys_reg_desc cp14_64_regs[] = {
* register).
*/
static const struct sys_reg_desc cp15_regs[] = {
+ { Op1( 0), CRn( 0), CRm( 0), Op2( 1), access_ctr },
{ Op1( 0), CRn( 1), CRm( 0), Op2( 0), access_vm_reg, NULL, c1_SCTLR },
{ Op1( 0), CRn( 2), CRm( 0), Op2( 0), access_vm_reg, NULL, c2_TTBR0 },
{ Op1( 0), CRn( 2), CRm( 0), Op2( 1), access_vm_reg, NULL, c2_TTBR1 },
@@ -1779,6 +1826,10 @@ static const struct sys_reg_desc cp15_regs[] = {
PMU_PMEVTYPER(30),
/* PMCCFILTR */
{ Op1(0), CRn(14), CRm(15), Op2(7), access_pmu_evtyper },
+
+ { Op1(1), CRn( 0), CRm( 0), Op2(0), access_ccsidr },
+ { Op1(1), CRn( 0), CRm( 0), Op2(1), access_clidr },
+ { Op1(2), CRn( 0), CRm( 0), Op2(0), access_csselr, NULL, c0_CSSELR },
};
static const struct sys_reg_desc cp15_64_regs[] = {
@@ -2192,11 +2243,15 @@ static const struct sys_reg_desc *index_to_sys_reg_desc(struct kvm_vcpu *vcpu,
}
FUNCTION_INVARIANT(midr_el1)
-FUNCTION_INVARIANT(ctr_el0)
FUNCTION_INVARIANT(revidr_el1)
FUNCTION_INVARIANT(clidr_el1)
FUNCTION_INVARIANT(aidr_el1)
+static void get_ctr_el0(struct kvm_vcpu *v, const struct sys_reg_desc *r)
+{
+ ((struct sys_reg_desc *)r)->val = read_sanitised_ftr_reg(SYS_CTR_EL0);
+}
+
/* ->val is filled in by kvm_sys_reg_table_init() */
static struct sys_reg_desc invariant_sys_regs[] = {
{ SYS_DESC(SYS_MIDR_EL1), NULL, get_midr_el1 },
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 17/27] arm64: KVM: Describe data or unified caches as having 1 set and 1 way
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
On SMP ARM systems, cache maintenance by set/way should only ever be
done in the context of onlining or offlining CPUs, which is typically
done by bare metal firmware and never in a virtual machine. For this
reason, we trap set/way cache maintenance operations and replace them
with conditional flushing of the entire guest address space.
Due to this trapping, the set/way arguments passed into the set/way
ops are completely ignored, and thus irrelevant. This also means that
the set/way geometry is equally irrelevant, and we can simply report
it as 1 set and 1 way, so that legacy 32-bit ARM system software (i.e.,
the kind that only receives odd fixes) doesn't take a performance hit
due to the trapping when iterating over the cachelines.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/include/asm/kvm_emulate.h | 3 ++-
arch/arm64/kvm/sys_regs.c | 15 +++++++++++++++
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index 9acccb1e3746..d3842791e1c4 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -78,7 +78,8 @@ static inline void vcpu_reset_hcr(struct kvm_vcpu *vcpu)
if (!vcpu_el1_is_32bit(vcpu))
vcpu->arch.hcr_el2 |= HCR_TID3;
- if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE))
+ if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE) ||
+ vcpu_el1_is_32bit(vcpu))
vcpu->arch.hcr_el2 |= HCR_TID2;
}
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 45a07cfc57ed..81a342679e60 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1186,6 +1186,21 @@ static bool access_ccsidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
csselr = vcpu_read_sys_reg(vcpu, CSSELR_EL1);
p->regval = get_ccsidr(csselr);
+
+ /*
+ * Guests should not be doing cache operations by set/way at all, and
+ * for this reason, we trap them and attempt to infer the intent, so
+ * that we can flush the entire guest's address space at the appropriate
+ * time.
+ * To prevent this trapping from causing performance problems, let's
+ * expose the geometry of all data and unified caches (which are
+ * guaranteed to be PIPT and thus non-aliasing) as 1 set and 1 way.
+ * [If guests should attempt to infer aliasing properties from the
+ * geometry (which is not permitted by the architecture), they would
+ * only do so for virtually indexed caches.]
+ */
+ if (!(csselr & 1)) // data or unified cache
+ p->regval &= ~GENMASK(27, 3);
return true;
}
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 17/27] arm64: KVM: Describe data or unified caches as having 1 set and 1 way
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
On SMP ARM systems, cache maintenance by set/way should only ever be
done in the context of onlining or offlining CPUs, which is typically
done by bare metal firmware and never in a virtual machine. For this
reason, we trap set/way cache maintenance operations and replace them
with conditional flushing of the entire guest address space.
Due to this trapping, the set/way arguments passed into the set/way
ops are completely ignored, and thus irrelevant. This also means that
the set/way geometry is equally irrelevant, and we can simply report
it as 1 set and 1 way, so that legacy 32-bit ARM system software (i.e.,
the kind that only receives odd fixes) doesn't take a performance hit
due to the trapping when iterating over the cachelines.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/include/asm/kvm_emulate.h | 3 ++-
arch/arm64/kvm/sys_regs.c | 15 +++++++++++++++
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index 9acccb1e3746..d3842791e1c4 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -78,7 +78,8 @@ static inline void vcpu_reset_hcr(struct kvm_vcpu *vcpu)
if (!vcpu_el1_is_32bit(vcpu))
vcpu->arch.hcr_el2 |= HCR_TID3;
- if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE))
+ if (cpus_have_const_cap(ARM64_MISMATCHED_CACHE_TYPE) ||
+ vcpu_el1_is_32bit(vcpu))
vcpu->arch.hcr_el2 |= HCR_TID2;
}
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 45a07cfc57ed..81a342679e60 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1186,6 +1186,21 @@ static bool access_ccsidr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
csselr = vcpu_read_sys_reg(vcpu, CSSELR_EL1);
p->regval = get_ccsidr(csselr);
+
+ /*
+ * Guests should not be doing cache operations by set/way at all, and
+ * for this reason, we trap them and attempt to infer the intent, so
+ * that we can flush the entire guest's address space at the appropriate
+ * time.
+ * To prevent this trapping from causing performance problems, let's
+ * expose the geometry of all data and unified caches (which are
+ * guaranteed to be PIPT and thus non-aliasing) as 1 set and 1 way.
+ * [If guests should attempt to infer aliasing properties from the
+ * geometry (which is not permitted by the architecture), they would
+ * only do so for virtually indexed caches.]
+ */
+ if (!(csselr & 1)) // data or unified cache
+ p->regval &= ~GENMASK(27, 3);
return true;
}
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 18/27] KVM: arm/arm64: arch_timer: Mark physical interrupt active when a virtual interrupt is pending
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
When a guest gets scheduled, KVM performs a "load" operation,
which for the timer includes evaluating the virtual "active" state
of the interrupt, and replicating it on the physical side. This
ensures that the deactivation in the guest will also take place
in the physical GIC distributor.
If the interrupt is not yet active, we flag it as inactive on the
physical side. This means that on restoring the timer registers,
if the timer has expired, we'll immediately take an interrupt.
That's absolutely fine, as the interrupt will then be flagged as
active on the physical side. What this assumes though is that we'll
enter the guest right after having taken the interrupt, and that
the guest will quickly ACK the interrupt, making it active at on
the virtual side.
It turns out that quite often, this assumption doesn't really hold.
The guest may be preempted on the back on this interrupt, either
from kernel space or whilst running at EL1 when a host interrupt
fires. When this happens, we repeat the whole sequence on the
next load (interrupt marked as inactive, timer registers restored,
interrupt fires). And if it takes a really long time for a guest
to activate the interrupt (as it does with nested virt), we end-up
with many such events in quick succession, leading to the guest only
making very slow progress.
This can also be seen with the number of virtual timer interrupt on the
host being far greater than the same number in the guest.
An easy way to fix this is to evaluate the timer state when performing
the "load" operation, just like we do when the interrupt actually fires.
If the timer has a pending virtual interrupt at this stage, then we
can safely flag the physical interrupt as being active, which prevents
spurious exits.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/arch_timer.c | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 17f9de73cc8a..af8f2f1d01cc 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -487,12 +487,21 @@ static inline void set_timer_irq_phys_active(struct arch_timer_context *ctx, boo
static void kvm_timer_vcpu_load_gic(struct arch_timer_context *ctx)
{
struct kvm_vcpu *vcpu = ctx->vcpu;
- bool phys_active;
+ bool phys_active = false;
+
+ /*
+ * Update the timer output so that it is likely to match the
+ * state we're about to restore. If the timer expires between
+ * this point and the register restoration, we'll take the
+ * interrupt anyway.
+ */
+ kvm_timer_update_irq(ctx->vcpu, kvm_timer_should_fire(ctx), ctx);
if (irqchip_in_kernel(vcpu->kvm))
phys_active = kvm_vgic_map_is_active(vcpu, ctx->irq.irq);
- else
- phys_active = ctx->irq.level;
+
+ phys_active |= ctx->irq.level;
+
set_timer_irq_phys_active(ctx, phys_active);
}
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 18/27] KVM: arm/arm64: arch_timer: Mark physical interrupt active when a virtual interrupt is pending
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
When a guest gets scheduled, KVM performs a "load" operation,
which for the timer includes evaluating the virtual "active" state
of the interrupt, and replicating it on the physical side. This
ensures that the deactivation in the guest will also take place
in the physical GIC distributor.
If the interrupt is not yet active, we flag it as inactive on the
physical side. This means that on restoring the timer registers,
if the timer has expired, we'll immediately take an interrupt.
That's absolutely fine, as the interrupt will then be flagged as
active on the physical side. What this assumes though is that we'll
enter the guest right after having taken the interrupt, and that
the guest will quickly ACK the interrupt, making it active at on
the virtual side.
It turns out that quite often, this assumption doesn't really hold.
The guest may be preempted on the back on this interrupt, either
from kernel space or whilst running at EL1 when a host interrupt
fires. When this happens, we repeat the whole sequence on the
next load (interrupt marked as inactive, timer registers restored,
interrupt fires). And if it takes a really long time for a guest
to activate the interrupt (as it does with nested virt), we end-up
with many such events in quick succession, leading to the guest only
making very slow progress.
This can also be seen with the number of virtual timer interrupt on the
host being far greater than the same number in the guest.
An easy way to fix this is to evaluate the timer state when performing
the "load" operation, just like we do when the interrupt actually fires.
If the timer has a pending virtual interrupt at this stage, then we
can safely flag the physical interrupt as being active, which prevents
spurious exits.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/arch_timer.c | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 17f9de73cc8a..af8f2f1d01cc 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -487,12 +487,21 @@ static inline void set_timer_irq_phys_active(struct arch_timer_context *ctx, boo
static void kvm_timer_vcpu_load_gic(struct arch_timer_context *ctx)
{
struct kvm_vcpu *vcpu = ctx->vcpu;
- bool phys_active;
+ bool phys_active = false;
+
+ /*
+ * Update the timer output so that it is likely to match the
+ * state we're about to restore. If the timer expires between
+ * this point and the register restoration, we'll take the
+ * interrupt anyway.
+ */
+ kvm_timer_update_irq(ctx->vcpu, kvm_timer_should_fire(ctx), ctx);
if (irqchip_in_kernel(vcpu->kvm))
phys_active = kvm_vgic_map_is_active(vcpu, ctx->irq.irq);
- else
- phys_active = ctx->irq.level;
+
+ phys_active |= ctx->irq.level;
+
set_timer_irq_phys_active(ctx, phys_active);
}
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 19/27] KVM: arm/arm64: Fix TRACE_INCLUDE_PATH
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Masahiro Yamada <yamada.masahiro@socionext.com>
As the comment block in include/trace/define_trace.h says,
TRACE_INCLUDE_PATH should be a relative path to the define_trace.h
../../virt/kvm/arm is the correct relative path.
../../../virt/kvm/arm is working by coincidence because the top
Makefile adds -I$(srctree)/arch/$(SRCARCH)/include as a header
search path, but we should not rely on it.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/trace.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/virt/kvm/arm/trace.h b/virt/kvm/arm/trace.h
index 54bb059243b9..204d210d01c2 100644
--- a/virt/kvm/arm/trace.h
+++ b/virt/kvm/arm/trace.h
@@ -370,7 +370,7 @@ TRACE_EVENT(kvm_timer_emulate,
#endif /* _TRACE_KVM_H */
#undef TRACE_INCLUDE_PATH
-#define TRACE_INCLUDE_PATH ../../../virt/kvm/arm
+#define TRACE_INCLUDE_PATH ../../virt/kvm/arm
#undef TRACE_INCLUDE_FILE
#define TRACE_INCLUDE_FILE trace
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 19/27] KVM: arm/arm64: Fix TRACE_INCLUDE_PATH
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Masahiro Yamada <yamada.masahiro@socionext.com>
As the comment block in include/trace/define_trace.h says,
TRACE_INCLUDE_PATH should be a relative path to the define_trace.h
../../virt/kvm/arm is the correct relative path.
../../../virt/kvm/arm is working by coincidence because the top
Makefile adds -I$(srctree)/arch/$(SRCARCH)/include as a header
search path, but we should not rely on it.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/trace.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/virt/kvm/arm/trace.h b/virt/kvm/arm/trace.h
index 54bb059243b9..204d210d01c2 100644
--- a/virt/kvm/arm/trace.h
+++ b/virt/kvm/arm/trace.h
@@ -370,7 +370,7 @@ TRACE_EVENT(kvm_timer_emulate,
#endif /* _TRACE_KVM_H */
#undef TRACE_INCLUDE_PATH
-#define TRACE_INCLUDE_PATH ../../../virt/kvm/arm
+#define TRACE_INCLUDE_PATH ../../virt/kvm/arm
#undef TRACE_INCLUDE_FILE
#define TRACE_INCLUDE_FILE trace
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 20/27] KVM: arm/arm64: Remove -I. header search paths
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Masahiro Yamada <yamada.masahiro@socionext.com>
The header search path -I. in kernel Makefiles is very suspicious;
it allows the compiler to search for headers in the top of $(srctree),
where obviously no header file exists.
I was able to build without these extra header search paths.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/kvm/Makefile | 3 +--
arch/arm64/kvm/Makefile | 2 --
2 files changed, 1 insertion(+), 4 deletions(-)
diff --git a/arch/arm/kvm/Makefile b/arch/arm/kvm/Makefile
index 48de846f2246..bca775e222c3 100644
--- a/arch/arm/kvm/Makefile
+++ b/arch/arm/kvm/Makefile
@@ -9,8 +9,7 @@ ifeq ($(plus_virt),+virt)
endif
ccflags-y += -Iarch/arm/kvm -Ivirt/kvm/arm/vgic
-CFLAGS_arm.o := -I. $(plus_virt_def)
-CFLAGS_mmu.o := -I.
+CFLAGS_arm.o := $(plus_virt_def)
AFLAGS_init.o := -Wa,-march=armv7-a$(plus_virt)
AFLAGS_interrupts.o := -Wa,-march=armv7-a$(plus_virt)
diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile
index 0f2a135ba15b..3089b3159135 100644
--- a/arch/arm64/kvm/Makefile
+++ b/arch/arm64/kvm/Makefile
@@ -4,8 +4,6 @@
#
ccflags-y += -Iarch/arm64/kvm -Ivirt/kvm/arm/vgic
-CFLAGS_arm.o := -I.
-CFLAGS_mmu.o := -I.
KVM=../../../virt/kvm
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 20/27] KVM: arm/arm64: Remove -I. header search paths
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Masahiro Yamada <yamada.masahiro@socionext.com>
The header search path -I. in kernel Makefiles is very suspicious;
it allows the compiler to search for headers in the top of $(srctree),
where obviously no header file exists.
I was able to build without these extra header search paths.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/kvm/Makefile | 3 +--
arch/arm64/kvm/Makefile | 2 --
2 files changed, 1 insertion(+), 4 deletions(-)
diff --git a/arch/arm/kvm/Makefile b/arch/arm/kvm/Makefile
index 48de846f2246..bca775e222c3 100644
--- a/arch/arm/kvm/Makefile
+++ b/arch/arm/kvm/Makefile
@@ -9,8 +9,7 @@ ifeq ($(plus_virt),+virt)
endif
ccflags-y += -Iarch/arm/kvm -Ivirt/kvm/arm/vgic
-CFLAGS_arm.o := -I. $(plus_virt_def)
-CFLAGS_mmu.o := -I.
+CFLAGS_arm.o := $(plus_virt_def)
AFLAGS_init.o := -Wa,-march=armv7-a$(plus_virt)
AFLAGS_interrupts.o := -Wa,-march=armv7-a$(plus_virt)
diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile
index 0f2a135ba15b..3089b3159135 100644
--- a/arch/arm64/kvm/Makefile
+++ b/arch/arm64/kvm/Makefile
@@ -4,8 +4,6 @@
#
ccflags-y += -Iarch/arm64/kvm -Ivirt/kvm/arm/vgic
-CFLAGS_arm.o := -I.
-CFLAGS_mmu.o := -I.
KVM=../../../virt/kvm
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 21/27] KVM: arm/arm64: Prefix header search paths with $(srctree)/
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:25 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Masahiro Yamada <yamada.masahiro@socionext.com>
Currently, the Kbuild core manipulates header search paths in a crazy
way [1].
To fix this mess, I want all Makefiles to add explicit $(srctree)/ to
the search paths in the srctree. Some Makefiles are already written in
that way, but not all. The goal of this work is to make the notation
consistent, and finally get rid of the gross hacks.
Having whitespaces after -I does not matter since commit 48f6e3cf5bc6
("kbuild: do not drop -I without parameter").
[1]: https://patchwork.kernel.org/patch/9632347/
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/kvm/Makefile | 2 +-
arch/arm64/kvm/Makefile | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm/kvm/Makefile b/arch/arm/kvm/Makefile
index bca775e222c3..531e59f5be9c 100644
--- a/arch/arm/kvm/Makefile
+++ b/arch/arm/kvm/Makefile
@@ -8,7 +8,7 @@ ifeq ($(plus_virt),+virt)
plus_virt_def := -DREQUIRES_VIRT=1
endif
-ccflags-y += -Iarch/arm/kvm -Ivirt/kvm/arm/vgic
+ccflags-y += -I $(srctree)/$(src) -I $(srctree)/virt/kvm/arm/vgic
CFLAGS_arm.o := $(plus_virt_def)
AFLAGS_init.o := -Wa,-march=armv7-a$(plus_virt)
diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile
index 3089b3159135..690e033a91c0 100644
--- a/arch/arm64/kvm/Makefile
+++ b/arch/arm64/kvm/Makefile
@@ -3,7 +3,7 @@
# Makefile for Kernel-based Virtual Machine module
#
-ccflags-y += -Iarch/arm64/kvm -Ivirt/kvm/arm/vgic
+ccflags-y += -I $(srctree)/$(src) -I $(srctree)/virt/kvm/arm/vgic
KVM=../../../virt/kvm
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 21/27] KVM: arm/arm64: Prefix header search paths with $(srctree)/
@ 2019-02-22 16:25 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:25 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Masahiro Yamada <yamada.masahiro@socionext.com>
Currently, the Kbuild core manipulates header search paths in a crazy
way [1].
To fix this mess, I want all Makefiles to add explicit $(srctree)/ to
the search paths in the srctree. Some Makefiles are already written in
that way, but not all. The goal of this work is to make the notation
consistent, and finally get rid of the gross hacks.
Having whitespaces after -I does not matter since commit 48f6e3cf5bc6
("kbuild: do not drop -I without parameter").
[1]: https://patchwork.kernel.org/patch/9632347/
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/kvm/Makefile | 2 +-
arch/arm64/kvm/Makefile | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm/kvm/Makefile b/arch/arm/kvm/Makefile
index bca775e222c3..531e59f5be9c 100644
--- a/arch/arm/kvm/Makefile
+++ b/arch/arm/kvm/Makefile
@@ -8,7 +8,7 @@ ifeq ($(plus_virt),+virt)
plus_virt_def := -DREQUIRES_VIRT=1
endif
-ccflags-y += -Iarch/arm/kvm -Ivirt/kvm/arm/vgic
+ccflags-y += -I $(srctree)/$(src) -I $(srctree)/virt/kvm/arm/vgic
CFLAGS_arm.o := $(plus_virt_def)
AFLAGS_init.o := -Wa,-march=armv7-a$(plus_virt)
diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile
index 3089b3159135..690e033a91c0 100644
--- a/arch/arm64/kvm/Makefile
+++ b/arch/arm64/kvm/Makefile
@@ -3,7 +3,7 @@
# Makefile for Kernel-based Virtual Machine module
#
-ccflags-y += -Iarch/arm64/kvm -Ivirt/kvm/arm/vgic
+ccflags-y += -I $(srctree)/$(src) -I $(srctree)/virt/kvm/arm/vgic
KVM=../../../virt/kvm
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 22/27] KVM: arm/arm64: Update MAINTAINERS entries
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:26 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
For historical reasons, KVM/arm and KVM/arm64 have had different
entries in the MAINTAINER file. This makes little sense, as they are
maintained together.
On top of that, we have a bunch of talented people helping with
the reviewing, and they deserve to be mentioned in the consolidated
entry.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Acked-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Acked-by: James Morse <james.morse@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
MAINTAINERS | 18 +++++++-----------
1 file changed, 7 insertions(+), 11 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 51029a425dbe..5d68f050d155 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8297,29 +8297,25 @@ S: Maintained
F: arch/x86/include/asm/svm.h
F: arch/x86/kvm/svm.c
-KERNEL VIRTUAL MACHINE FOR ARM (KVM/arm)
+KERNEL VIRTUAL MACHINE FOR ARM/ARM64 (KVM/arm, KVM/arm64)
M: Christoffer Dall <christoffer.dall@arm.com>
M: Marc Zyngier <marc.zyngier@arm.com>
+R: James Morse <james.morse@arm.com>
+R: Julien Thierry <julien.thierry@arm.com>
+R: Suzuki K Pouloze <suzuki.poulose@arm.com>
L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
L: kvmarm@lists.cs.columbia.edu
W: http://systems.cs.columbia.edu/projects/kvm-arm
T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git
-S: Supported
+S: Maintained
F: arch/arm/include/uapi/asm/kvm*
F: arch/arm/include/asm/kvm*
F: arch/arm/kvm/
-F: virt/kvm/arm/
-F: include/kvm/arm_*
-
-KERNEL VIRTUAL MACHINE FOR ARM64 (KVM/arm64)
-M: Christoffer Dall <christoffer.dall@arm.com>
-M: Marc Zyngier <marc.zyngier@arm.com>
-L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
-L: kvmarm@lists.cs.columbia.edu
-S: Maintained
F: arch/arm64/include/uapi/asm/kvm*
F: arch/arm64/include/asm/kvm*
F: arch/arm64/kvm/
+F: virt/kvm/arm/
+F: include/kvm/arm_*
KERNEL VIRTUAL MACHINE FOR MIPS (KVM/mips)
M: James Hogan <jhogan@kernel.org>
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 22/27] KVM: arm/arm64: Update MAINTAINERS entries
@ 2019-02-22 16:26 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
For historical reasons, KVM/arm and KVM/arm64 have had different
entries in the MAINTAINER file. This makes little sense, as they are
maintained together.
On top of that, we have a bunch of talented people helping with
the reviewing, and they deserve to be mentioned in the consolidated
entry.
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Acked-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Acked-by: James Morse <james.morse@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
MAINTAINERS | 18 +++++++-----------
1 file changed, 7 insertions(+), 11 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 51029a425dbe..5d68f050d155 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8297,29 +8297,25 @@ S: Maintained
F: arch/x86/include/asm/svm.h
F: arch/x86/kvm/svm.c
-KERNEL VIRTUAL MACHINE FOR ARM (KVM/arm)
+KERNEL VIRTUAL MACHINE FOR ARM/ARM64 (KVM/arm, KVM/arm64)
M: Christoffer Dall <christoffer.dall@arm.com>
M: Marc Zyngier <marc.zyngier@arm.com>
+R: James Morse <james.morse@arm.com>
+R: Julien Thierry <julien.thierry@arm.com>
+R: Suzuki K Pouloze <suzuki.poulose@arm.com>
L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
L: kvmarm@lists.cs.columbia.edu
W: http://systems.cs.columbia.edu/projects/kvm-arm
T: git git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git
-S: Supported
+S: Maintained
F: arch/arm/include/uapi/asm/kvm*
F: arch/arm/include/asm/kvm*
F: arch/arm/kvm/
-F: virt/kvm/arm/
-F: include/kvm/arm_*
-
-KERNEL VIRTUAL MACHINE FOR ARM64 (KVM/arm64)
-M: Christoffer Dall <christoffer.dall@arm.com>
-M: Marc Zyngier <marc.zyngier@arm.com>
-L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
-L: kvmarm@lists.cs.columbia.edu
-S: Maintained
F: arch/arm64/include/uapi/asm/kvm*
F: arch/arm64/include/asm/kvm*
F: arch/arm64/kvm/
+F: virt/kvm/arm/
+F: include/kvm/arm_*
KERNEL VIRTUAL MACHINE FOR MIPS (KVM/mips)
M: James Hogan <jhogan@kernel.org>
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 23/27] KVM: arm/arm64: fix spelling mistake: "auxilary" -> "auxiliary"
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:26 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Colin Ian King <colin.king@canonical.com>
There is a spelling mistake in a kvm_err error message. Fix it.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/arm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 9fbdb9e1c51f..8de55041e7ba 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -1553,7 +1553,7 @@ static int init_hyp_mode(void)
err = hyp_map_aux_data();
if (err)
- kvm_err("Cannot map host auxilary data: %d\n", err);
+ kvm_err("Cannot map host auxiliary data: %d\n", err);
return 0;
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 23/27] KVM: arm/arm64: fix spelling mistake: "auxilary" -> "auxiliary"
@ 2019-02-22 16:26 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Colin Ian King <colin.king@canonical.com>
There is a spelling mistake in a kvm_err error message. Fix it.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/arm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 9fbdb9e1c51f..8de55041e7ba 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -1553,7 +1553,7 @@ static int init_hyp_mode(void)
err = hyp_map_aux_data();
if (err)
- kvm_err("Cannot map host auxilary data: %d\n", err);
+ kvm_err("Cannot map host auxiliary data: %d\n", err);
return 0;
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 24/27] KVM: arm64: Fix comment for KVM_PHYS_SHIFT
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:26 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Zenghui Yu <yuzenghui@huawei.com>
Since Suzuki K Poulose's work on Dynamic IPA support, KVM_PHYS_SHIFT will
be used only when machine_type's bits[7:0] equal to 0 (by default). Thus
the outdated comment should be fixed.
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/include/asm/kvm_mmu.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index c423c8c4fc39..b0742a16c6c9 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -138,7 +138,8 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
})
/*
- * We currently only support a 40bit IPA.
+ * We currently support using a VM-specified IPA size. For backward
+ * compatibility, the default IPA size is fixed to 40bits.
*/
#define KVM_PHYS_SHIFT (40)
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 24/27] KVM: arm64: Fix comment for KVM_PHYS_SHIFT
@ 2019-02-22 16:26 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Zenghui Yu <yuzenghui@huawei.com>
Since Suzuki K Poulose's work on Dynamic IPA support, KVM_PHYS_SHIFT will
be used only when machine_type's bits[7:0] equal to 0 (by default). Thus
the outdated comment should be fixed.
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/include/asm/kvm_mmu.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h
index c423c8c4fc39..b0742a16c6c9 100644
--- a/arch/arm64/include/asm/kvm_mmu.h
+++ b/arch/arm64/include/asm/kvm_mmu.h
@@ -138,7 +138,8 @@ static inline unsigned long __kern_hyp_va(unsigned long v)
})
/*
- * We currently only support a 40bit IPA.
+ * We currently support using a VM-specified IPA size. For backward
+ * compatibility, the default IPA size is fixed to 40bits.
*/
#define KVM_PHYS_SHIFT (40)
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 25/27] KVM: arm/arm64: Remove unused gpa_end variable
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:26 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Shaokun Zhang <zhangshaokun@hisilicon.com>
The 'gpa_end' local variable is never used and let's remove it.
Cc: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/mmu.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index e3e5a26b4845..aae68da2717f 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -1597,14 +1597,13 @@ static void kvm_send_hwpoison_signal(unsigned long address,
static bool fault_supports_stage2_pmd_mappings(struct kvm_memory_slot *memslot,
unsigned long hva)
{
- gpa_t gpa_start, gpa_end;
+ gpa_t gpa_start;
hva_t uaddr_start, uaddr_end;
size_t size;
size = memslot->npages * PAGE_SIZE;
gpa_start = memslot->base_gfn << PAGE_SHIFT;
- gpa_end = gpa_start + size;
uaddr_start = memslot->userspace_addr;
uaddr_end = uaddr_start + size;
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 25/27] KVM: arm/arm64: Remove unused gpa_end variable
@ 2019-02-22 16:26 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Shaokun Zhang <zhangshaokun@hisilicon.com>
The 'gpa_end' local variable is never used and let's remove it.
Cc: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/mmu.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index e3e5a26b4845..aae68da2717f 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -1597,14 +1597,13 @@ static void kvm_send_hwpoison_signal(unsigned long address,
static bool fault_supports_stage2_pmd_mappings(struct kvm_memory_slot *memslot,
unsigned long hva)
{
- gpa_t gpa_start, gpa_end;
+ gpa_t gpa_start;
hva_t uaddr_start, uaddr_end;
size_t size;
size = memslot->npages * PAGE_SIZE;
gpa_start = memslot->base_gfn << PAGE_SHIFT;
- gpa_end = gpa_start + size;
uaddr_start = memslot->userspace_addr;
uaddr_end = uaddr_start + size;
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 26/27] KVM: arm/arm64: Remove unused timer variable
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:26 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Shaokun Zhang <zhangshaokun@hisilicon.com>
The 'timer' local variable became unused after commit bee038a67487
("KVM: arm/arm64: Rework the timer code to use a timer_map").
Remove it to avoid [-Wunused-but-set-variable] warning.
Cc: Christoffer Dall <christoffer.dall@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Suzuki K Pouloze <suzuki.poulose@arm.com>
Reviewed-by: Julien Thierry <julien.thierry@arm.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/arch_timer.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index af8f2f1d01cc..3417f2dbc366 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -236,14 +236,12 @@ static enum hrtimer_restart kvm_hrtimer_expire(struct hrtimer *hrt)
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
- struct arch_timer_cpu *timer;
enum kvm_arch_timers index;
u64 cval, now;
if (!timer_ctx)
return false;
- timer = vcpu_timer(timer_ctx->vcpu);
index = arch_timer_ctx_index(timer_ctx);
if (timer_ctx->loaded) {
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 26/27] KVM: arm/arm64: Remove unused timer variable
@ 2019-02-22 16:26 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Shaokun Zhang <zhangshaokun@hisilicon.com>
The 'timer' local variable became unused after commit bee038a67487
("KVM: arm/arm64: Rework the timer code to use a timer_map").
Remove it to avoid [-Wunused-but-set-variable] warning.
Cc: Christoffer Dall <christoffer.dall@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Suzuki K Pouloze <suzuki.poulose@arm.com>
Reviewed-by: Julien Thierry <julien.thierry@arm.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
virt/kvm/arm/arch_timer.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index af8f2f1d01cc..3417f2dbc366 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -236,14 +236,12 @@ static enum hrtimer_restart kvm_hrtimer_expire(struct hrtimer *hrt)
static bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx)
{
- struct arch_timer_cpu *timer;
enum kvm_arch_timers index;
u64 cval, now;
if (!timer_ctx)
return false;
- timer = vcpu_timer(timer_ctx->vcpu);
index = arch_timer_ctx_index(timer_ctx);
if (timer_ctx->loaded) {
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 27/27] arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:26 ` Marc Zyngier
-1 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
From: Dave Martin <Dave.Martin@arm.com>
Due to what looks like a typo dating back to the original addition
of FPEXC32_EL2 handling, KVM currently initialises this register to
an architecturally invalid value.
As a result, the VECITR field (RES1) in bits [10:8] is initialised
with 0, and the two reserved (RES0) bits [6:5] are initialised with
1. (In the Common VFP Subarchitecture as specified by ARMv7-A,
these two bits were IMP DEF. ARMv8-A removes them.)
This patch changes the reset value from 0x70 to 0x700, which
reflects the architectural constraints and is presumably what was
originally intended.
Cc: <stable@vger.kernel.org> # 4.12.x-
Cc: Christoffer Dall <christoffer.dall@arm.com>
Fixes: 62a89c44954f ("arm64: KVM: 32bit handling of coprocessor traps")
Signed-off-by: Dave Martin <Dave.Martin@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/kvm/sys_regs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 81a342679e60..a398d04274b7 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1523,7 +1523,7 @@ static const struct sys_reg_desc sys_reg_descs[] = {
{ SYS_DESC(SYS_DACR32_EL2), NULL, reset_unknown, DACR32_EL2 },
{ SYS_DESC(SYS_IFSR32_EL2), NULL, reset_unknown, IFSR32_EL2 },
- { SYS_DESC(SYS_FPEXC32_EL2), NULL, reset_val, FPEXC32_EL2, 0x70 },
+ { SYS_DESC(SYS_FPEXC32_EL2), NULL, reset_val, FPEXC32_EL2, 0x700 },
};
static bool trap_dbgidr(struct kvm_vcpu *vcpu,
--
2.20.1
^ permalink raw reply related [flat|nested] 58+ messages in thread
* [PATCH 27/27] arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2
@ 2019-02-22 16:26 ` Marc Zyngier
0 siblings, 0 replies; 58+ messages in thread
From: Marc Zyngier @ 2019-02-22 16:26 UTC (permalink / raw)
To: Paolo Bonzini, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
From: Dave Martin <Dave.Martin@arm.com>
Due to what looks like a typo dating back to the original addition
of FPEXC32_EL2 handling, KVM currently initialises this register to
an architecturally invalid value.
As a result, the VECITR field (RES1) in bits [10:8] is initialised
with 0, and the two reserved (RES0) bits [6:5] are initialised with
1. (In the Common VFP Subarchitecture as specified by ARMv7-A,
these two bits were IMP DEF. ARMv8-A removes them.)
This patch changes the reset value from 0x70 to 0x700, which
reflects the architectural constraints and is presumably what was
originally intended.
Cc: <stable@vger.kernel.org> # 4.12.x-
Cc: Christoffer Dall <christoffer.dall@arm.com>
Fixes: 62a89c44954f ("arm64: KVM: 32bit handling of coprocessor traps")
Signed-off-by: Dave Martin <Dave.Martin@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm64/kvm/sys_regs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 81a342679e60..a398d04274b7 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1523,7 +1523,7 @@ static const struct sys_reg_desc sys_reg_descs[] = {
{ SYS_DESC(SYS_DACR32_EL2), NULL, reset_unknown, DACR32_EL2 },
{ SYS_DESC(SYS_IFSR32_EL2), NULL, reset_unknown, IFSR32_EL2 },
- { SYS_DESC(SYS_FPEXC32_EL2), NULL, reset_val, FPEXC32_EL2, 0x70 },
+ { SYS_DESC(SYS_FPEXC32_EL2), NULL, reset_val, FPEXC32_EL2, 0x700 },
};
static bool trap_dbgidr(struct kvm_vcpu *vcpu,
--
2.20.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 58+ messages in thread
* Re: [GIT PULL] KVM/arm updates for Linux v5.1
2019-02-22 16:25 ` Marc Zyngier
@ 2019-02-22 16:45 ` Paolo Bonzini
-1 siblings, 0 replies; 58+ messages in thread
From: Paolo Bonzini @ 2019-02-22 16:45 UTC (permalink / raw)
To: Marc Zyngier, Radim Krčmář
Cc: kvm, Ard Biesheuvel, Andre Przywara, Daniel Lezcano, kvmarm,
Shaokun Zhang, Masahiro Yamada, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin
On 22/02/19 17:25, Marc Zyngier wrote:
> git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git tags/kvmarm-for-v5.1
Pulled, thanks.
Paolo
^ permalink raw reply [flat|nested] 58+ messages in thread
* Re: [GIT PULL] KVM/arm updates for Linux v5.1
@ 2019-02-22 16:45 ` Paolo Bonzini
0 siblings, 0 replies; 58+ messages in thread
From: Paolo Bonzini @ 2019-02-22 16:45 UTC (permalink / raw)
To: Marc Zyngier, Radim Krčmář
Cc: Julien Thierry, kvm, Ard Biesheuvel, Andre Przywara,
Daniel Lezcano, Christoffer Dall, kvmarm, Shaokun Zhang,
Masahiro Yamada, James Morse, linux-arm-kernel, Zenghui Yu,
Colin Ian King, Dave Martin, Suzuki K Poulose
On 22/02/19 17:25, Marc Zyngier wrote:
> git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git tags/kvmarm-for-v5.1
Pulled, thanks.
Paolo
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 58+ messages in thread
end of thread, other threads:[~2019-02-22 17:11 UTC | newest]
Thread overview: 58+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-02-22 16:25 [GIT PULL] KVM/arm updates for Linux v5.1 Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 01/27] clocksource/arm_arch_timer: Store physical timer IRQ number for KVM on VHE Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 02/27] arm/arm64: KVM: Introduce kvm_call_hyp_ret() Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 03/27] arm64: KVM: Allow for direct call of HYP functions when using VHE Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 04/27] arm64: KVM: Drop VHE-specific HYP call stub Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 05/27] ARM: KVM: Teach some form of type-safety to kvm_call_hyp Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 06/27] arm/arm64: KVM: Statically configure the host's view of MPIDR Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 07/27] KVM: arm/arm64: Factor out VMID into struct kvm_vmid Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 08/27] KVM: arm/arm64: Simplify bg_timer programming Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 09/27] KVM: arm64: Fix ICH_ELRSR_EL2 sysreg naming Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 10/27] KVM: arm64: Reuse sys_reg() macro when searching the trap table Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 11/27] KVM: arm/arm64: consolidate arch timer trap handlers Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 12/27] KVM: arm/arm64: timer: Rework data structures for multiple timers Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 13/27] KVM: arm/arm64: arch_timer: Assign the phys timer on VHE systems Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 14/27] KVM: arm/arm64: Rework the timer code to use a timer_map Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 15/27] KVM: arm/arm64: Move kvm_is_write_fault to header file Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 16/27] arm64: KVM: Expose sanitised cache type register to guest Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 17/27] arm64: KVM: Describe data or unified caches as having 1 set and 1 way Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 18/27] KVM: arm/arm64: arch_timer: Mark physical interrupt active when a virtual interrupt is pending Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 19/27] KVM: arm/arm64: Fix TRACE_INCLUDE_PATH Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 20/27] KVM: arm/arm64: Remove -I. header search paths Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:25 ` [PATCH 21/27] KVM: arm/arm64: Prefix header search paths with $(srctree)/ Marc Zyngier
2019-02-22 16:25 ` Marc Zyngier
2019-02-22 16:26 ` [PATCH 22/27] KVM: arm/arm64: Update MAINTAINERS entries Marc Zyngier
2019-02-22 16:26 ` Marc Zyngier
2019-02-22 16:26 ` [PATCH 23/27] KVM: arm/arm64: fix spelling mistake: "auxilary" -> "auxiliary" Marc Zyngier
2019-02-22 16:26 ` Marc Zyngier
2019-02-22 16:26 ` [PATCH 24/27] KVM: arm64: Fix comment for KVM_PHYS_SHIFT Marc Zyngier
2019-02-22 16:26 ` Marc Zyngier
2019-02-22 16:26 ` [PATCH 25/27] KVM: arm/arm64: Remove unused gpa_end variable Marc Zyngier
2019-02-22 16:26 ` Marc Zyngier
2019-02-22 16:26 ` [PATCH 26/27] KVM: arm/arm64: Remove unused timer variable Marc Zyngier
2019-02-22 16:26 ` Marc Zyngier
2019-02-22 16:26 ` [PATCH 27/27] arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2 Marc Zyngier
2019-02-22 16:26 ` Marc Zyngier
2019-02-22 16:45 ` [GIT PULL] KVM/arm updates for Linux v5.1 Paolo Bonzini
2019-02-22 16:45 ` Paolo Bonzini
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.