arm64: Support perf event modifiers :G and :H
Subsystems (5)
v9 cover arm64: Support perf event modifiers :G and :H
This patchset provides support for perf event modifiers :G and :H which allows for filtering of PMU events between host and guests when used with KVM.
As the underlying hardware cannot distinguish between guest and host context, the performance counters must be stopped and started upon entry/exit to the guest. This is performed at EL2 in a way that minimizes overhead and improves accuracy of recording events that only occur in the requested context.
This has been tested with VHE and non-VHE kernels with a KVM guest.
Changes from v8:
- Added additional comments
- Fixed bisect build failure
- Renamed cpu_ctxt variable to cpu_data
Changes from v7:
- Added additional patch to encapsulate kvm_cpu_context in kvm_host_data
Changes from v6:
- Move events_host/events_guest out of kvm_cpu_context
Changes from v5:
- Tweak logic in use of kvm_set_pmu_events
Changes from v4:
- Prevent unnecessary write_sysreg calls by improving __pmu_switch_to_xxx logic.
Changes from v3:
- Remove confusing _only suffix from bitfields in kvm_cpu_context
- Remove unnecessary condition when clearing event bits in disable
- Simplify API of KVM accessors
- Prevent unnecessary setting of pmcnten when guest/host events are the same.
Changes from v2:
- Ensured that exclude_kernel works for guest
- Removed unnecessary exclusion of EL2 with exclude_host on !VHE
- Renamed kvm_clr_set_host_pmu_events to reflect args order
- Added additional information to isb patch
Changes from v1:
- Removed unnecessary exclusion of EL1 with exclude_guest on VHE
- Removed unnecessary isb from existing perf_event.c driver
- Folded perf_event.c patches together
- Added additional information to last patch commit message
Andrew Murray (5):
arm64: arm_pmu: remove unnecessary isb instruction
arm64: KVM: encapsulate kvm_cpu_context in kvm_host_data
arm64: KVM: add accessors to track guest/host only counters
arm64: arm_pmu: Add support for exclude_host/exclude_guest attributes
arm64: KVM: Enable support for :G/:H perf event modifiers
arch/arm/include/asm/kvm_host.h | 8 ++++-- arch/arm64/include/asm/kvm_asm.h | 4 +-- arch/arm64/include/asm/kvm_host.h | 42 +++++++++++++++++++++++---- arch/arm64/kernel/asm-offsets.c | 2 +- arch/arm64/kernel/perf_event.c | 54 +++++++++++++++++++++++++++++------ arch/arm64/kvm/hyp/switch.c | 60 +++++++++++++++++++++++++++++++++++++++ virt/kvm/arm/arm.c | 12 ++++---- 7 files changed, 159 insertions(+), 23 deletions(-)
Signature · 1 line
2.7.4
PATCH v9 1/5 arm64: arm_pmu: remove unnecessary isb instruction
The armv8pmu_enable_event_counter function issues an isb instruction after enabling a pair of counters - this doesn't provide any value and is inconsistent with the armv8pmu_disable_event_counter.
In any case armv8pmu_enable_event_counter is always called with the PMU stopped. Starting the PMU with armv8pmu_start results in an isb instruction being issued prior to writing to PMCR_EL0.
Let's remove the unnecessary isb instruction.
--- arch/arm64/kernel/perf_event.c | 1 - 1 file changed, 1 deletion(-)
arch/arm64/kernel/perf_event.c +0 -1
@@ -652,7 +652,6 @@ static inline void armv8pmu_enable_event_counter(struct perf_event *event)armv8pmu_enable_counter (idx);if (armv8pmu_event_is_chained (event))armv8pmu_enable_counter (idx -1 ); -isb (); }static inline int armv8pmu_disable_counter (int idx)
Signature · 1 line
2.7.4
PATCH v9 2/5 arm64: KVM: encapsulate kvm_cpu_context in kvm_host_data
In reply to Andrew MurrayThe virt/arm core allocates a kvm_cpu_context_t percpu, at present this is a typedef to kvm_cpu_context and is used to store host cpu context. The kvm_cpu_context structure is also used elsewhere to hold vcpu context. In order to use the percpu to hold additional future host information we encapsulate kvm_cpu_context in a new structure and rename the typedef and percpu to match.
--- arch/arm/include/asm/kvm_host.h | 8 ++++++-- arch/arm64/include/asm/kvm_asm.h | 4 ++-- arch/arm64/include/asm/kvm_host.h | 14 +++++++++----- arch/arm64/kernel/asm-offsets.c | 2 +- virt/kvm/arm/arm.c | 12 +++++++----- 5 files changed, 25 insertions(+), 15 deletions(-)
arch/arm/include/asm/kvm_host.h +6 -2
@@ -145,7 +145,11 @@ struct kvm_cpu_context {u32 cp15[NR_CP15_REGS ]; }; -typedef struct kvm_cpu_context kvm_cpu_context_t ; +struct kvm_host_data { +struct kvm_cpu_context host_ctxt; +}; + +typedef struct kvm_host_data kvm_host_data_t ;struct kvm_vcpu_arch {struct kvm_cpu_context ctxt;
@@ -163,7 +167,7 @@ struct kvm_vcpu_arch {struct kvm_vcpu_fault_info fault;/* Host FP context */ -kvm_cpu_context_t *host_cpu_context; +struct kvm_cpu_context *host_cpu_context;/* VGIC state */ struct vgic_cpu vgic_cpu;
arch/arm64/include/asm/kvm_asm.h +2 -2
@@ -102,12 +102,12 @@ extern u32 __init_stage2_translation(void); .endm .macro get_host_ctxt reg, tmp - hyp_adr_this_cpu \reg, kvm_host_cpu_state, \tmp + hyp_adr_this_cpu \reg, kvm_host_data, \tmp .endm .macro get_vcpu_ptr vcpu, ctxt get_host_ctxt \ctxt, \vcpu - ldr \vcpu, [\ctxt, #HOST_CONTEXT_VCPU] + ldr \vcpu, [\ctxt, #HOST_DATA_VCPU] kern_hyp_va \vcpu .endm
arch/arm64/include/asm/kvm_host.h +9 -5
@@ -205,7 +205,11 @@ struct kvm_cpu_context {struct kvm_vcpu *__hyp_running_vcpu; }; -typedef struct kvm_cpu_context kvm_cpu_context_t ; +struct kvm_host_data { +struct kvm_cpu_context host_ctxt; +}; + +typedef struct kvm_host_data kvm_host_data_t ;struct kvm_vcpu_arch {struct kvm_cpu_context ctxt;
@@ -241,7 +245,7 @@ struct kvm_vcpu_arch {struct kvm_guest_debug_arch external_debug_state;/* Pointer to host CPU context */ -kvm_cpu_context_t *host_cpu_context; +struct kvm_cpu_context *host_cpu_context;struct thread_info *host_thread_info;/* hyp VA */ struct user_fpsimd_state *host_fpsimd_state;/* hyp VA */
@@ -387,7 +391,7 @@ void kvm_set_sei_esr(struct kvm_vcpu *vcpu, u64 syndrome);struct kvm_vcpu *kvm_mpidr_to_vcpu (struct kvm *kvm,unsigned long mpidr); -DECLARE_PER_CPU (kvm_cpu_context_t, kvm_host_cpu_state); +DECLARE_PER_CPU (kvm_host_data_t, kvm_host_data);void __kvm_enable_ssbs (void );
@@ -400,8 +404,8 @@ static inline void __cpu_init_hyp_mode(phys_addr_t pgd_ptr,* kernel's mapping to the linear mapping, and store it in tpidr_el2 * so that we can use adr_l to access per-cpu variables in EL2. */ -u64 tpidr_el2 = ((u64 )this_cpu_ptr (&kvm_host_cpu_state) - - (u64 )kvm_ksym_ref (kvm_host_cpu_state)); +u64 tpidr_el2 = ((u64 )this_cpu_ptr (&kvm_host_data) - + (u64 )kvm_ksym_ref (kvm_host_data));/* * Call initialization code, and switch to the full blown HYP code.
arch/arm64/kernel/asm-offsets.c +1 -1
@@ -142,7 +142,7 @@ int main(void)DEFINE (CPU_FP_REGS , offsetof(struct kvm_regs , fp_regs)); DEFINE(VCPU_FPEXC32_EL2, offsetof(struct kvm_vcpu, arch.ctxt.sys_regs[FPEXC32_EL2])); DEFINE(VCPU_HOST_CONTEXT, offsetof(struct kvm_vcpu, arch.host_cpu_context)); -DEFINE (HOST_CONTEXT_VCPU , offsetof(struct kvm_cpu_context , __hyp_running_vcpu)); + DEFINE(HOST_DATA_VCPU, offsetof(struct kvm_host_data, host_ctxt.__hyp_running_vcpu));#endif #ifdef CONFIG_CPU_PM DEFINE (CPU_SUSPEND_SZ ,sizeof (struct cpu_suspend_ctx ));
virt/kvm/arm/arm.c +7 -5
@@ -56,7 +56,7 @@ __asm__(".arch_extension virt" );#endif -DEFINE_PER_CPU (kvm_cpu_context_t, kvm_host_cpu_state); +DEFINE_PER_CPU (kvm_host_data_t, kvm_host_data); static DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page);/* Per-CPU variable containing the currently running vcpu. */
@@ -361,8 +361,10 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)void kvm_arch_vcpu_load (struct kvm_vcpu *vcpu,int cpu) {int *last_ran; +kvm_host_data_t *cpu_data; last_ran =this_cpu_ptr (vcpu->kvm->arch.last_vcpu_ran); + cpu_data =this_cpu_ptr (&kvm_host_data);/* * We might get preempted before the vCPU actually runs, but
@@ -374,7 +376,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) } vcpu->cpu = cpu; - vcpu->arch.host_cpu_context =this_cpu_ptr (&kvm_host_cpu_state); + vcpu->arch.host_cpu_context = &cpu_data->host_ctxt;kvm_arm_set_running_vcpu (vcpu);kvm_vgic_load (vcpu);
@@ -1547,10 +1549,10 @@ static int init_hyp_mode(void) }for_each_possible_cpu (cpu) { -kvm_cpu_context_t *cpu_ctxt; +kvm_host_data_t *cpu_data; - cpu_ctxt =per_cpu_ptr (&kvm_host_cpu_state, cpu); - err =create_hyp_mappings (cpu_ctxt, cpu_ctxt +1 ,PAGE_HYP ); + cpu_data =per_cpu_ptr (&kvm_host_data, cpu); + err =create_hyp_mappings (cpu_data, cpu_data +1 ,PAGE_HYP );if (err) {kvm_err ("Cannot map host CPU state: %d\n" , err);
Signature · 1 line
2.7.4
Re: [PATCH v9 2/5] arm64: KVM: encapsulate kvm_cpu_context in kvm_host_data
Hi Andrew,
Quoted 47 lines · Andrew Murray
The
virt/armcore allocates akvm_cpu_context_tpercpu, at present this is a typedef tokvm_cpu_contextand is used to store host cpu context. Thekvm_cpu_contextstructure is also used elsewhere to hold vcpu context. In order to use the percpu to hold additional future host information we encapsulatekvm_cpu_contextin a new structure and rename the typedef and percpu to match.Signed-off-by Andrew Murray <[email protected]>--- arch/arm/include/asm/kvm_host.h | 8 ++++++-- arch/arm64/include/asm/kvm_asm.h | 4 ++-- arch/arm64/include/asm/kvm_host.h | 14 +++++++++----- arch/arm64/kernel/asm-offsets.c | 2 +- virt/kvm/arm/arm.c | 12 +++++++----- 5 files changed, 25 insertions(+), 15 deletions(-)arch/arm/include/asm/kvm_host.h +6
-2@@ -145,7 +145,11 @@ struct kvm_cpu_context {u32 cp15[NR_CP15_REGS ]; }; -typedef struct kvm_cpu_context kvm_cpu_context_t ; +struct kvm_host_data { +struct kvm_cpu_context host_ctxt; +}; + +typedef struct kvm_host_data kvm_host_data_t ;struct kvm_vcpu_arch {struct kvm_cpu_context ctxt;@@ -163,7 +167,7 @@ struct kvm_vcpu_arch {struct kvm_vcpu_fault_info fault;/* Host FP context */ -kvm_cpu_context_t *host_cpu_context; +struct kvm_cpu_context *host_cpu_context;/* VGIC state */ struct vgic_cpu vgic_cpu;arch/arm64/include/asm/kvm_asm.h +0
-0
arch/arm64/include/asm/kvm_asm.h extern u32 __init_stage2_translation(void); · line 106
Quoted 6 lines
arch/arm64/include/asm/kvm_asm.h +1
-1@@ -102,12 +102,12 @@ extern u32 __init_stage2_translation(void); .endm .macro get_host_ctxt reg, tmp - hyp_adr_this_cpu \reg, kvm_host_cpu_state, \tmp + hyp_adr_this_cpu \reg, kvm_host_data, \tmp .endm
minor nit: We now load "kvm_host_data" instead of the host_ctxt here, even though they both are same. Do we need to make this explicitly load the address of the host_ctxt member to make sure this doesn't break with future changes to the structure ?
i.e,
add \reg, \reg, #HOST_DATA_CONTEXT
Otherwise,
Re: [PATCH v9 2/5] arm64: KVM: encapsulate kvm_cpu_context in kvm_host_data
Quoted 59 lines · Suzuki K Poulose
Hi Andrew,
Quoted 43 lines · Andrew Murray
The
virt/armcore allocates akvm_cpu_context_tpercpu, at present this is a typedef tokvm_cpu_contextand is used to store host cpu context. Thekvm_cpu_contextstructure is also used elsewhere to hold vcpu context. In order to use the percpu to hold additional future host information we encapsulatekvm_cpu_contextin a new structure and rename the typedef and percpu to match.Signed-off-by Andrew Murray <[email protected]>--- arch/arm/include/asm/kvm_host.h | 8 ++++++-- arch/arm64/include/asm/kvm_asm.h | 4 ++-- arch/arm64/include/asm/kvm_host.h | 14 +++++++++----- arch/arm64/kernel/asm-offsets.c | 2 +- virt/kvm/arm/arm.c | 12 +++++++----- 5 files changed, 25 insertions(+), 15 deletions(-)arch/arm/include/asm/kvm_host.h +6
-2@@ -145,7 +145,11 @@ struct kvm_cpu_context {u32 cp15[NR_CP15_REGS ]; }; -typedef struct kvm_cpu_context kvm_cpu_context_t ; +struct kvm_host_data { +struct kvm_cpu_context host_ctxt; +}; + +typedef struct kvm_host_data kvm_host_data_t ;struct kvm_vcpu_arch {struct kvm_cpu_context ctxt;@@ -163,7 +167,7 @@ struct kvm_vcpu_arch {struct kvm_vcpu_fault_info fault;/* Host FP context */ -kvm_cpu_context_t *host_cpu_context; +struct kvm_cpu_context *host_cpu_context;/* VGIC state */ struct vgic_cpu vgic_cpu;arch/arm64/include/asm/kvm_asm.h +0
-0arch/arm64/include/asm/kvm_asm.h extern u32 __init_stage2_translation(void); · line 105
Quoted 5 lines
arch/arm64/include/asm/kvm_asm.h +1
-1@@ -102,12 +102,12 @@ extern u32 __init_stage2_translation(void); .endm .macro get_host_ctxt reg, tmp - hyp_adr_this_cpu \reg, kvm_host_cpu_state, \tmp + hyp_adr_this_cpu \reg, kvm_host_data, \tmp .endmminor nit: We now load "
kvm_host_data" instead of thehost_ctxthere, even though they both are same. Do we need to make this explicitly load the address of thehost_ctxtmember to make sure this doesn't break with future changes to the structure ?
i.e,
add \reg, \reg, #HOST_DATA_CONTEXT
Quoted 3 lines
Otherwise,
Reviewed-by Suzuki K Poulose <[email protected]>
Thanks for spotting this.
Are you happy for me to add the Reviewed-by tag if I squash in the following?
arch/arm64/include/asm/kvm_asm.h +2 -1
@@ -103,11 +103,12 @@ extern u32 __init_stage2_translation(void); .macro get_host_ctxt reg, tmp hyp_adr_this_cpu \reg, kvm_host_data, \tmp + add \reg, \reg, #HOST_DATA_CONTEXT .endm .macro get_vcpu_ptr vcpu, ctxt get_host_ctxt \ctxt, \vcpu - ldr \vcpu, [\ctxt, #HOST_DATA_VCPU] + ldr \vcpu, [\ctxt, #HOST_CONTEXT_VCPU] kern_hyp_va \vcpu .endm
arch/arm64/kernel/asm-offsets.c +2 -1
@@ -142,7 +142,8 @@ int main(void)DEFINE (CPU_FP_REGS , offsetof(struct kvm_regs , fp_regs)); DEFINE(VCPU_FPEXC32_EL2, offsetof(struct kvm_vcpu, arch.ctxt.sys_regs[FPEXC32_EL2])); DEFINE(VCPU_HOST_CONTEXT, offsetof(struct kvm_vcpu, arch.host_cpu_context)); - DEFINE(HOST_DATA_VCPU, offsetof(struct kvm_host_data, host_ctxt.__hyp_running_vcpu)); +DEFINE (HOST_CONTEXT_VCPU , offsetof(struct kvm_cpu_context , __hyp_running_vcpu)); +DEFINE (HOST_DATA_CONTEXT , offsetof(struct kvm_host_data , host_ctxt));#endif #ifdef CONFIG_CPU_PM DEFINE (CPU_SUSPEND_SZ ,sizeof (struct cpu_suspend_ctx ));
Thanks,
Andrew Murray
Suzuki K Poulose
reviewerRe: [PATCH v9 2/5] arm64: KVM: encapsulate kvm_cpu_context in kvm_host_data
Quoted 60 lines · Andrew Murray
Quoted 59 lines · Suzuki K Poulose
Hi Andrew,
Quoted 43 lines · Andrew Murray
The
virt/armcore allocates akvm_cpu_context_tpercpu, at present this is a typedef tokvm_cpu_contextand is used to store host cpu context. Thekvm_cpu_contextstructure is also used elsewhere to hold vcpu context. In order to use the percpu to hold additional future host information we encapsulatekvm_cpu_contextin a new structure and rename the typedef and percpu to match.Signed-off-by Andrew Murray <[email protected]>--- arch/arm/include/asm/kvm_host.h | 8 ++++++-- arch/arm64/include/asm/kvm_asm.h | 4 ++-- arch/arm64/include/asm/kvm_host.h | 14 +++++++++----- arch/arm64/kernel/asm-offsets.c | 2 +- virt/kvm/arm/arm.c | 12 +++++++----- 5 files changed, 25 insertions(+), 15 deletions(-)arch/arm/include/asm/kvm_host.h +6
-2@@ -145,7 +145,11 @@ struct kvm_cpu_context {u32 cp15[NR_CP15_REGS ]; }; -typedef struct kvm_cpu_context kvm_cpu_context_t ; +struct kvm_host_data { +struct kvm_cpu_context host_ctxt; +}; + +typedef struct kvm_host_data kvm_host_data_t ;struct kvm_vcpu_arch {struct kvm_cpu_context ctxt;@@ -163,7 +167,7 @@ struct kvm_vcpu_arch {struct kvm_vcpu_fault_info fault;/* Host FP context */ -kvm_cpu_context_t *host_cpu_context; +struct kvm_cpu_context *host_cpu_context;/* VGIC state */ struct vgic_cpu vgic_cpu;arch/arm64/include/asm/kvm_asm.h +0
-0arch/arm64/include/asm/kvm_asm.h extern u32 __init_stage2_translation(void); · line 105
Quoted 5 lines
arch/arm64/include/asm/kvm_asm.h +1
-1@@ -102,12 +102,12 @@ extern u32 __init_stage2_translation(void); .endm .macro get_host_ctxt reg, tmp - hyp_adr_this_cpu \reg, kvm_host_cpu_state, \tmp + hyp_adr_this_cpu \reg, kvm_host_data, \tmp .endmminor nit: We now load "
kvm_host_data" instead of thehost_ctxthere, even though they both are same. Do we need to make this explicitly load the address of thehost_ctxtmember to make sure this doesn't break with future changes to the structure ?
i.e,
add \reg, \reg, #HOST_DATA_CONTEXT
Quoted 7 lines
Quoted 3 lines
Otherwise,
Reviewed-by Suzuki K Poulose <[email protected]>Thanks for spotting this.
Are you happy for me to add the Reviewed-by tag if I squash in the following?
Yes, that looks fine to me.
Cheers
Suzuki
PATCH v9 3/5 arm64: KVM: add accessors to track guest/host only counters
In reply to Andrew MurrayIn order to effeciently enable/disable guest/host only perf counters at guest entry/exit we add bitfields to kvm_cpu_context for guest and host events as well as accessors for updating them.
--- arch/arm64/include/asm/kvm_host.h | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+)
arch/arm64/include/asm/kvm_host.h +28 -0
@@ -205,8 +205,14 @@ struct kvm_cpu_context {struct kvm_vcpu *__hyp_running_vcpu; }; +struct kvm_pmu_events { +u32 events_host; +u32 events_guest; +}; +struct kvm_host_data {struct kvm_cpu_context host_ctxt; +struct kvm_pmu_events pmu_events; };typedef struct kvm_host_data kvm_host_data_t ;
@@ -471,11 +477,33 @@ void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu);void kvm_arch_vcpu_ctxsync_fp (struct kvm_vcpu *vcpu);void kvm_arch_vcpu_put_fp (struct kvm_vcpu *vcpu); +#define KVM_PMU_EVENTS_HOST 1 +#define KVM_PMU_EVENTS_GUEST 2 +#ifdef CONFIG_KVM /* Avoid conflicts with core headers if CONFIG_KVM=n */ static inline int kvm_arch_vcpu_run_pid_change (struct kvm_vcpu *vcpu) {return kvm_arch_vcpu_run_map_fp (vcpu); } +static inline void kvm_set_pmu_events (u32 set,int flags) +{ +struct kvm_host_data *ctx =this_cpu_ptr (&kvm_host_data); + +if (flags &KVM_PMU_EVENTS_HOST ) + ctx->pmu_events.events_host |= set; +if (flags &KVM_PMU_EVENTS_GUEST ) + ctx->pmu_events.events_guest |= set; +} +static inline void kvm_clr_pmu_events (u32 clr) +{ +struct kvm_host_data *ctx =this_cpu_ptr (&kvm_host_data); + + ctx->pmu_events.events_host &= ~clr; + ctx->pmu_events.events_guest &= ~clr; +} +#else +static inline void kvm_set_pmu_events (u32 set,int flags) {} +static inline void kvm_clr_pmu_events (u32 clr) {}#endif static inline void kvm_arm_vhe_guest_enter (void )
Signature · 1 line
2.7.4
PATCH v9 4/5 arm64: arm_pmu: Add support for exclude_host/exclude_guest attributes
In reply to Andrew MurrayAdd support for the :G and :H attributes in perf by handling the exclude_host/exclude_guest event attributes.
We notify KVM of counters that we wish to be enabled or disabled on guest entry/exit and thus defer from starting or stopping :G events as per the events exclude_host attribute.
With both VHE and non-VHE we switch the counters between host/guest at EL2. We are able to eliminate counters counting host events on the boundaries of guest entry/exit when using :G by filtering out EL2 for exclude_host. However when using :H unless exclude_hv is set on non-VHE then there is a small blackout window at the guest entry/exit where host events are not captured.
--- arch/arm64/kernel/perf_event.c | 53 ++++++++++++++++++++++++++++++++++++------ 1 file changed, 46 insertions(+), 7 deletions(-)
arch/arm64/kernel/perf_event.c +46 -7
@@ -26,6 +26,7 @@#include <linux/acpi.h> #include <linux/clocksource.h> +#include <linux/kvm_host.h> #include <linux/of.h> #include <linux/perf/arm_pmu.h> #include <linux/platform_device.h>
@@ -647,11 +648,27 @@ static inline int armv8pmu_enable_counter(int idx)static inline void armv8pmu_enable_event_counter (struct perf_event *event) { +struct perf_event_attr *attr = &event->attr;int idx = event->hw.idx; +int flags =0 ; +u32 counter_bits =BIT (ARMV8_IDX_TO_COUNTER (idx)); -armv8pmu_enable_counter (idx);if (armv8pmu_event_is_chained (event)) -armv8pmu_enable_counter (idx -1 ); + counter_bits |=BIT (ARMV8_IDX_TO_COUNTER (idx -1 )); + +if (!attr->exclude_host) + flags |=KVM_PMU_EVENTS_HOST ; +if (!attr->exclude_guest) + flags |=KVM_PMU_EVENTS_GUEST ; + +kvm_set_pmu_events (counter_bits, flags); + +/* We rely on the hypervisor switch code to enable guest counters */ +if (!attr->exclude_host) { +armv8pmu_enable_counter (idx); +if (armv8pmu_event_is_chained (event)) +armv8pmu_enable_counter (idx -1 ); + } }static inline int armv8pmu_disable_counter (int idx)
@@ -664,11 +681,21 @@ static inline int armv8pmu_disable_counter(int idx)static inline void armv8pmu_disable_event_counter (struct perf_event *event) {struct hw_perf_event *hwc = &event->hw; +struct perf_event_attr *attr = &event->attr;int idx = hwc->idx; +u32 counter_bits =BIT (ARMV8_IDX_TO_COUNTER (idx));if (armv8pmu_event_is_chained (event)) -armv8pmu_disable_counter (idx -1 ); -armv8pmu_disable_counter (idx); + counter_bits |=BIT (ARMV8_IDX_TO_COUNTER (idx -1 )); + +kvm_clr_pmu_events (counter_bits); + +/* We rely on the hypervisor switch code to disable guest counters */ +if (!attr->exclude_host) { +if (armv8pmu_event_is_chained (event)) +armv8pmu_disable_counter (idx -1 ); +armv8pmu_disable_counter (idx); + } }static inline int armv8pmu_enable_intens (int idx)
@@ -943,16 +970,25 @@ static int armv8pmu_set_event_filter(struct hw_perf_event *event,* Therefore we ignore exclude_hv in this configuration, since * there's no hypervisor to sample anyway. This is consistent * with other architectures (x86 and Power). +* +* To eliminate counting host events on the boundaries of +* guest entry/exit we ensure EL2 is not included in hyp mode +* with !exclude_host. */ if (is_kernel_in_hyp_mode ()) { -if (!attr->exclude_kernel) +if (!attr->exclude_kernel && !attr->exclude_host) config_base |=ARMV8_PMU_INCLUDE_EL2 ; }else { -if (attr->exclude_kernel) - config_base |=ARMV8_PMU_EXCLUDE_EL1 ;if (!attr->exclude_hv) config_base |=ARMV8_PMU_INCLUDE_EL2 ; } + +/* +* Filter out !VHE kernels and guest kernels +*/ +if (attr->exclude_kernel) + config_base |=ARMV8_PMU_EXCLUDE_EL1 ; +if (attr->exclude_user) config_base |=ARMV8_PMU_EXCLUDE_EL0 ;
@@ -976,6 +1012,9 @@ static void armv8pmu_reset(void *info)armv8pmu_disable_intens (idx); } +/* Clear the counters we flip at guest entry/exit */ +kvm_clr_pmu_events (U32_MAX ); +/* * Initialize & Reset PMNC. Request overflow interrupt for * 64 bit cycle counter but cheat in armv8pmu_write_counter().
Signature · 1 line
2.7.4
PATCH v9 5/5 arm64: KVM: Enable support for :G/:H perf event modifiers
In reply to Andrew MurrayEnable/disable event counters as appropriate when entering and exiting the guest to enable support for guest or host only event counting.
For both VHE and non-VHE we switch the counters between host/guest at EL2. EL2 is filtered out by the PMU when we are using the :G modifier.
The PMU may be on when we change which counters are enabled however we avoid adding an isb as we instead rely on existing context synchronisation events: the isb in kvm_arm_vhe_guest_exit for VHE and the eret from the hvc in kvm_call_hyp.
--- arch/arm64/kvm/hyp/switch.c | 60 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 60 insertions(+)
arch/arm64/kvm/hyp/switch.c +60 -0
@@ -373,6 +373,54 @@ static bool __hyp_text __hyp_switch_fpsimd(struct kvm_vcpu *vcpu)return true; } +static bool __hyp_text __pmu_switch_to_guest(struct kvm_cpu_context *host_ctxt) +{ +struct kvm_host_data *host; +struct kvm_pmu_events *pmu; +u32 clr, set; + + host = container_of(host_ctxt, struct kvm_host_data, host_ctxt); + pmu = &host->pmu_events; + +/* We can potentially avoid a sysreg write by only changing bits that +* differ between the guest/host. E.g. where events are enabled in +* both guest and host +*/ + clr = pmu->events_host & ~pmu->events_guest; + set = pmu->events_guest & ~pmu->events_host; + +if (clr) +write_sysreg (clr, pmcntenclr_el0); + +if (set) +write_sysreg (set, pmcntenset_el0); + +return (clr || set); +} + +static void __hyp_text __pmu_switch_to_host(struct kvm_cpu_context *host_ctxt) +{ +struct kvm_host_data *host; +struct kvm_pmu_events *pmu; +u32 clr, set; + + host = container_of(host_ctxt, struct kvm_host_data, host_ctxt); + pmu = &host->pmu_events; + +/* We can potentially avoid a sysreg write by only changing bits that +* differ between the guest/host. E.g. where events are enabled in +* both guest and host +*/ + clr = pmu->events_guest & ~pmu->events_host; + set = pmu->events_host & ~pmu->events_guest; + +if (clr) +write_sysreg (clr, pmcntenclr_el0); + +if (set) +write_sysreg (set, pmcntenset_el0); +} +/* * Return true when we were able to fixup the guest exit and should return to * the guest, false when we should restore the host state and return to the
@@ -488,12 +536,15 @@ int kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu) {struct kvm_cpu_context *host_ctxt;struct kvm_cpu_context *guest_ctxt; +bool pmu_switch_needed;u64 exit_code; host_ctxt = vcpu->arch.host_cpu_context; host_ctxt->__hyp_running_vcpu = vcpu; guest_ctxt = &vcpu->arch.ctxt; + pmu_switch_needed =__pmu_switch_to_guest (host_ctxt); +sysreg_save_host_state_vhe (host_ctxt);__activate_traps (vcpu);
@@ -524,6 +575,9 @@ int kvm_vcpu_run_vhe(struct kvm_vcpu *vcpu)__debug_switch_to_host (vcpu); +if (pmu_switch_needed) +__pmu_switch_to_host (host_ctxt); +return exit_code; }
@@ -532,6 +586,7 @@ int __hyp_text __kvm_vcpu_run_nvhe(struct kvm_vcpu *vcpu) {struct kvm_cpu_context *host_ctxt;struct kvm_cpu_context *guest_ctxt; +bool pmu_switch_needed;u64 exit_code; vcpu =kern_hyp_va (vcpu);
@@ -540,6 +595,8 @@ int __hyp_text __kvm_vcpu_run_nvhe(struct kvm_vcpu *vcpu) host_ctxt->__hyp_running_vcpu = vcpu; guest_ctxt = &vcpu->arch.ctxt; + pmu_switch_needed =__pmu_switch_to_guest (host_ctxt); +__sysreg_save_state_nvhe (host_ctxt);__activate_traps (vcpu);
@@ -586,6 +643,9 @@ int __hyp_text __kvm_vcpu_run_nvhe(struct kvm_vcpu *vcpu)*/ __debug_switch_to_host (vcpu); +if (pmu_switch_needed) +__pmu_switch_to_host (host_ctxt); +return exit_code; }
Signature · 1 line
2.7.4