aboutsummaryrefslogtreecommitdiff
path: root/arch/x86/kvm/pmu.c
diff options
context:
space:
mode:
authorWei Huang <wehuang@redhat.com>2015-06-19 15:45:05 +0200
committerPaolo Bonzini <pbonzini@redhat.com>2015-06-23 14:12:14 +0200
commit25462f7f5295e2d3e9c2b31761ac95f0b3c8562f (patch)
treeed5d7df52d131213a898c34d4596d60e997c485c /arch/x86/kvm/pmu.c
parent41aac14a8dee66a720894e5979c2372c0d5afd34 (diff)
downloadlinux-stericsson-25462f7f5295e2d3e9c2b31761ac95f0b3c8562f.tar.gz
KVM: x86/vPMU: Define kvm_pmu_ops to support vPMU function dispatch
This patch defines a new function pointer struct (kvm_pmu_ops) to support vPMU for both Intel and AMD. The functions pointers defined in this new struct will be linked with Intel and AMD functions later. In the meanwhile the struct that maps from event_sel bits to PERF_TYPE_HARDWARE events is renamed and moved from Intel specific code to kvm_host.h as a common struct. Reviewed-by: Joerg Roedel <jroedel@suse.de> Tested-by: Joerg Roedel <jroedel@suse.de> Signed-off-by: Wei Huang <wei@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Diffstat (limited to 'arch/x86/kvm/pmu.c')
-rw-r--r--arch/x86/kvm/pmu.c383
1 files changed, 47 insertions, 336 deletions
diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
index bd5dbd9ce0e3..31aa2c85dc97 100644
--- a/arch/x86/kvm/pmu.c
+++ b/arch/x86/kvm/pmu.c
@@ -1,11 +1,12 @@
/*
* Kernel-based Virtual Machine -- Performance Monitoring Unit support
*
- * Copyright 2011 Red Hat, Inc. and/or its affiliates.
+ * Copyright 2015 Red Hat, Inc. and/or its affiliates.
*
* Authors:
* Avi Kivity <avi@redhat.com>
* Gleb Natapov <gleb@redhat.com>
+ * Wei Huang <wei@redhat.com>
*
* This work is licensed under the terms of the GNU GPL, version 2. See
* the COPYING file in the top-level directory.
@@ -21,67 +22,30 @@
#include "lapic.h"
#include "pmu.h"
-static struct kvm_event_hw_type_mapping arch_events[] = {
- /* Index must match CPUID 0x0A.EBX bit vector */
- [0] = { 0x3c, 0x00, PERF_COUNT_HW_CPU_CYCLES },
- [1] = { 0xc0, 0x00, PERF_COUNT_HW_INSTRUCTIONS },
- [2] = { 0x3c, 0x01, PERF_COUNT_HW_BUS_CYCLES },
- [3] = { 0x2e, 0x4f, PERF_COUNT_HW_CACHE_REFERENCES },
- [4] = { 0x2e, 0x41, PERF_COUNT_HW_CACHE_MISSES },
- [5] = { 0xc4, 0x00, PERF_COUNT_HW_BRANCH_INSTRUCTIONS },
- [6] = { 0xc5, 0x00, PERF_COUNT_HW_BRANCH_MISSES },
- [7] = { 0x00, 0x30, PERF_COUNT_HW_REF_CPU_CYCLES },
-};
-
-/* mapping between fixed pmc index and arch_events array */
-static int fixed_pmc_events[] = {1, 0, 7};
-
-static bool pmc_is_gp(struct kvm_pmc *pmc)
-{
- return pmc->type == KVM_PMC_GP;
-}
-
-static inline u64 pmc_bitmask(struct kvm_pmc *pmc)
-{
- struct kvm_pmu *pmu = pmc_to_pmu(pmc);
-
- return pmu->counter_bitmask[pmc->type];
-}
-
-static inline bool pmc_is_enabled(struct kvm_pmc *pmc)
-{
- struct kvm_pmu *pmu = pmc_to_pmu(pmc);
- return test_bit(pmc->idx, (unsigned long *)&pmu->global_ctrl);
-}
-
-static inline struct kvm_pmc *get_gp_pmc(struct kvm_pmu *pmu, u32 msr,
- u32 base)
-{
- if (msr >= base && msr < base + pmu->nr_arch_gp_counters)
- return &pmu->gp_counters[msr - base];
- return NULL;
-}
-
-static inline struct kvm_pmc *get_fixed_pmc(struct kvm_pmu *pmu, u32 msr)
-{
- int base = MSR_CORE_PERF_FIXED_CTR0;
- if (msr >= base && msr < base + pmu->nr_arch_fixed_counters)
- return &pmu->fixed_counters[msr - base];
- return NULL;
-}
-
-static inline struct kvm_pmc *get_fixed_pmc_idx(struct kvm_pmu *pmu, int idx)
-{
- return get_fixed_pmc(pmu, MSR_CORE_PERF_FIXED_CTR0 + idx);
-}
-
-static struct kvm_pmc *global_idx_to_pmc(struct kvm_pmu *pmu, int idx)
-{
- if (idx < INTEL_PMC_IDX_FIXED)
- return get_gp_pmc(pmu, MSR_P6_EVNTSEL0 + idx, MSR_P6_EVNTSEL0);
- else
- return get_fixed_pmc_idx(pmu, idx - INTEL_PMC_IDX_FIXED);
-}
+/* NOTE:
+ * - Each perf counter is defined as "struct kvm_pmc";
+ * - There are two types of perf counters: general purpose (gp) and fixed.
+ * gp counters are stored in gp_counters[] and fixed counters are stored
+ * in fixed_counters[] respectively. Both of them are part of "struct
+ * kvm_pmu";
+ * - pmu.c understands the difference between gp counters and fixed counters.
+ * However AMD doesn't support fixed-counters;
+ * - There are three types of index to access perf counters (PMC):
+ * 1. MSR (named msr): For example Intel has MSR_IA32_PERFCTRn and AMD
+ * has MSR_K7_PERFCTRn.
+ * 2. MSR Index (named idx): This normally is used by RDPMC instruction.
+ * For instance AMD RDPMC instruction uses 0000_0003h in ECX to access
+ * C001_0007h (MSR_K7_PERCTR3). Intel has a similar mechanism, except
+ * that it also supports fixed counters. idx can be used to as index to
+ * gp and fixed counters.
+ * 3. Global PMC Index (named pmc): pmc is an index specific to PMU
+ * code. Each pmc, stored in kvm_pmc.idx field, is unique across
+ * all perf counters (both gp and fixed). The mapping relationship
+ * between pmc and perf counters is as the following:
+ * * Intel: [0 .. INTEL_PMC_MAX_GENERIC-1] <=> gp counters
+ * [INTEL_PMC_IDX_FIXED .. INTEL_PMC_IDX_FIXED + 2] <=> fixed
+ * * AMD: [0 .. AMD64_NUM_COUNTERS-1] <=> gp counters
+ */
static void kvm_pmi_trigger_fn(struct irq_work *irq_work)
{
@@ -132,30 +96,6 @@ static void kvm_perf_overflow_intr(struct perf_event *perf_event,
}
}
-static u64 pmc_read_counter(struct kvm_pmc *pmc)
-{
- u64 counter, enabled, running;
-
- counter = pmc->counter;
-
- if (pmc->perf_event)
- counter += perf_event_read_value(pmc->perf_event,
- &enabled, &running);
-
- /* FIXME: Scaling needed? */
-
- return counter & pmc_bitmask(pmc);
-}
-
-static void pmc_stop_counter(struct kvm_pmc *pmc)
-{
- if (pmc->perf_event) {
- pmc->counter = pmc_read_counter(pmc);
- perf_event_release_kernel(pmc->perf_event);
- pmc->perf_event = NULL;
- }
-}
-
static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
unsigned config, bool exclude_user,
bool exclude_kernel, bool intr,
@@ -193,24 +133,7 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
clear_bit(pmc->idx, (unsigned long*)&pmc_to_pmu(pmc)->reprogram_pmi);
}
-static unsigned find_arch_event(struct kvm_pmu *pmu, u8 event_select,
- u8 unit_mask)
-{
- int i;
-
- for (i = 0; i < ARRAY_SIZE(arch_events); i++)
- if (arch_events[i].eventsel == event_select
- && arch_events[i].unit_mask == unit_mask
- && (pmu->available_event_types & (1 << i)))
- break;
-
- if (i == ARRAY_SIZE(arch_events))
- return PERF_COUNT_HW_MAX;
-
- return arch_events[i].event_type;
-}
-
-static void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
+void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
{
unsigned config, type = PERF_TYPE_RAW;
u8 event_select, unit_mask;
@@ -233,8 +156,9 @@ static void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
ARCH_PERFMON_EVENTSEL_CMASK |
HSW_IN_TX |
HSW_IN_TX_CHECKPOINTED))) {
- config = find_arch_event(pmc_to_pmu(pmc), event_select,
- unit_mask);
+ config = kvm_x86_ops->pmu_ops->find_arch_event(pmc_to_pmu(pmc),
+ event_select,
+ unit_mask);
if (config != PERF_COUNT_HW_MAX)
type = PERF_TYPE_HARDWARE;
}
@@ -249,8 +173,9 @@ static void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
(eventsel & HSW_IN_TX),
(eventsel & HSW_IN_TX_CHECKPOINTED));
}
+EXPORT_SYMBOL_GPL(reprogram_gp_counter);
-static void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 ctrl, int idx)
+void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 ctrl, int idx)
{
unsigned en_field = ctrl & 0x3;
bool pmi = ctrl & 0x8;
@@ -261,38 +186,16 @@ static void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 ctrl, int idx)
return;
pmc_reprogram_counter(pmc, PERF_TYPE_HARDWARE,
- arch_events[fixed_pmc_events[idx]].event_type,
+ kvm_x86_ops->pmu_ops->find_fixed_event(idx),
!(en_field & 0x2), /* exclude user */
!(en_field & 0x1), /* exclude kernel */
pmi, false, false);
}
+EXPORT_SYMBOL_GPL(reprogram_fixed_counter);
-static inline u8 fixed_ctrl_field(u64 ctrl, int idx)
-{
- return (ctrl >> (idx * 4)) & 0xf;
-}
-
-static void reprogram_fixed_counters(struct kvm_pmu *pmu, u64 data)
+void reprogram_counter(struct kvm_pmu *pmu, int pmc_idx)
{
- int i;
-
- for (i = 0; i < pmu->nr_arch_fixed_counters; i++) {
- u8 old_ctrl = fixed_ctrl_field(pmu->fixed_ctr_ctrl, i);
- u8 new_ctrl = fixed_ctrl_field(data, i);
- struct kvm_pmc *pmc = get_fixed_pmc_idx(pmu, i);
-
- if (old_ctrl == new_ctrl)
- continue;
-
- reprogram_fixed_counter(pmc, new_ctrl, i);
- }
-
- pmu->fixed_ctr_ctrl = data;
-}
-
-static void reprogram_counter(struct kvm_pmu *pmu, int pmc_idx)
-{
- struct kvm_pmc *pmc = global_idx_to_pmc(pmu, pmc_idx);
+ struct kvm_pmc *pmc = kvm_x86_ops->pmu_ops->pmc_idx_to_pmc(pmu, pmc_idx);
if (!pmc)
return;
@@ -306,17 +209,7 @@ static void reprogram_counter(struct kvm_pmu *pmu, int pmc_idx)
reprogram_fixed_counter(pmc, ctrl, idx);
}
}
-
-static void global_ctrl_changed(struct kvm_pmu *pmu, u64 data)
-{
- int bit;
- u64 diff = pmu->global_ctrl ^ data;
-
- pmu->global_ctrl = data;
-
- for_each_set_bit(bit, (unsigned long *)&diff, X86_PMC_IDX_MAX)
- reprogram_counter(pmu, bit);
-}
+EXPORT_SYMBOL_GPL(reprogram_counter);
void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
{
@@ -327,7 +220,7 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
bitmask = pmu->reprogram_pmi;
for_each_set_bit(bit, (unsigned long *)&bitmask, X86_PMC_IDX_MAX) {
- struct kvm_pmc *pmc = global_idx_to_pmc(pmu, bit);
+ struct kvm_pmc *pmc = kvm_x86_ops->pmu_ops->pmc_idx_to_pmc(pmu, bit);
if (unlikely(!pmc || !pmc->perf_event)) {
clear_bit(bit, (unsigned long *)&pmu->reprogram_pmi);
@@ -341,28 +234,7 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
/* check if idx is a valid index to access PMU */
int kvm_pmu_is_valid_msr_idx(struct kvm_vcpu *vcpu, unsigned idx)
{
- struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
- bool fixed = idx & (1u << 30);
- idx &= ~(3u << 30);
- return (!fixed && idx >= pmu->nr_arch_gp_counters) ||
- (fixed && idx >= pmu->nr_arch_fixed_counters);
-}
-
-static struct kvm_pmc *kvm_pmu_msr_idx_to_pmc(struct kvm_vcpu *vcpu,
- unsigned idx)
-{
- struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
- bool fixed = idx & (1u << 30);
- struct kvm_pmc *counters;
-
- idx &= ~(3u << 30);
- if (!fixed && idx >= pmu->nr_arch_gp_counters)
- return NULL;
- if (fixed && idx >= pmu->nr_arch_fixed_counters)
- return NULL;
- counters = fixed ? pmu->fixed_counters : pmu->gp_counters;
-
- return &counters[idx];
+ return kvm_x86_ops->pmu_ops->is_valid_msr_idx(vcpu, idx);
}
int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
@@ -371,7 +243,7 @@ int kvm_pmu_rdpmc(struct kvm_vcpu *vcpu, unsigned idx, u64 *data)
struct kvm_pmc *pmc;
u64 ctr_val;
- pmc = kvm_pmu_msr_idx_to_pmc(vcpu, idx);
+ pmc = kvm_x86_ops->pmu_ops->msr_idx_to_pmc(vcpu, idx);
if (!pmc)
return 1;
@@ -391,111 +263,17 @@ void kvm_pmu_deliver_pmi(struct kvm_vcpu *vcpu)
bool kvm_pmu_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
{
- struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
- int ret;
-
- switch (msr) {
- case MSR_CORE_PERF_FIXED_CTR_CTRL:
- case MSR_CORE_PERF_GLOBAL_STATUS:
- case MSR_CORE_PERF_GLOBAL_CTRL:
- case MSR_CORE_PERF_GLOBAL_OVF_CTRL:
- ret = pmu->version > 1;
- break;
- default:
- ret = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)
- || get_gp_pmc(pmu, msr, MSR_P6_EVNTSEL0)
- || get_fixed_pmc(pmu, msr);
- break;
- }
- return ret;
+ return kvm_x86_ops->pmu_ops->is_valid_msr(vcpu, msr);
}
-int kvm_pmu_get_msr(struct kvm_vcpu *vcpu, u32 index, u64 *data)
+int kvm_pmu_get_msr(struct kvm_vcpu *vcpu, u32 msr, u64 *data)
{
- struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
- struct kvm_pmc *pmc;
-
- switch (index) {
- case MSR_CORE_PERF_FIXED_CTR_CTRL:
- *data = pmu->fixed_ctr_ctrl;
- return 0;
- case MSR_CORE_PERF_GLOBAL_STATUS:
- *data = pmu->global_status;
- return 0;
- case MSR_CORE_PERF_GLOBAL_CTRL:
- *data = pmu->global_ctrl;
- return 0;
- case MSR_CORE_PERF_GLOBAL_OVF_CTRL:
- *data = pmu->global_ovf_ctrl;
- return 0;
- default:
- if ((pmc = get_gp_pmc(pmu, index, MSR_IA32_PERFCTR0)) ||
- (pmc = get_fixed_pmc(pmu, index))) {
- *data = pmc_read_counter(pmc);
- return 0;
- } else if ((pmc = get_gp_pmc(pmu, index, MSR_P6_EVNTSEL0))) {
- *data = pmc->eventsel;
- return 0;
- }
- }
- return 1;
+ return kvm_x86_ops->pmu_ops->get_msr(vcpu, msr, data);
}
int kvm_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
{
- struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
- struct kvm_pmc *pmc;
- u32 index = msr_info->index;
- u64 data = msr_info->data;
-
- switch (index) {
- case MSR_CORE_PERF_FIXED_CTR_CTRL:
- if (pmu->fixed_ctr_ctrl == data)
- return 0;
- if (!(data & 0xfffffffffffff444ull)) {
- reprogram_fixed_counters(pmu, data);
- return 0;
- }
- break;
- case MSR_CORE_PERF_GLOBAL_STATUS:
- if (msr_info->host_initiated) {
- pmu->global_status = data;
- return 0;
- }
- break; /* RO MSR */
- case MSR_CORE_PERF_GLOBAL_CTRL:
- if (pmu->global_ctrl == data)
- return 0;
- if (!(data & pmu->global_ctrl_mask)) {
- global_ctrl_changed(pmu, data);
- return 0;
- }
- break;
- case MSR_CORE_PERF_GLOBAL_OVF_CTRL:
- if (!(data & (pmu->global_ctrl_mask & ~(3ull<<62)))) {
- if (!msr_info->host_initiated)
- pmu->global_status &= ~data;
- pmu->global_ovf_ctrl = data;
- return 0;
- }
- break;
- default:
- if ((pmc = get_gp_pmc(pmu, index, MSR_IA32_PERFCTR0)) ||
- (pmc = get_fixed_pmc(pmu, index))) {
- if (!msr_info->host_initiated)
- data = (s64)(s32)data;
- pmc->counter += data - pmc_read_counter(pmc);
- return 0;
- } else if ((pmc = get_gp_pmc(pmu, index, MSR_P6_EVNTSEL0))) {
- if (data == pmc->eventsel)
- return 0;
- if (!(data & pmu->reserved_bits)) {
- reprogram_gp_counter(pmc, data);
- return 0;
- }
- }
- }
- return 1;
+ return kvm_x86_ops->pmu_ops->set_msr(vcpu, msr_info);
}
/* refresh PMU settings. This function generally is called when underlying
@@ -504,90 +282,23 @@ int kvm_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
*/
void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
{
- struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
- struct kvm_cpuid_entry2 *entry;
- union cpuid10_eax eax;
- union cpuid10_edx edx;
-
- pmu->nr_arch_gp_counters = 0;
- pmu->nr_arch_fixed_counters = 0;
- pmu->counter_bitmask[KVM_PMC_GP] = 0;
- pmu->counter_bitmask[KVM_PMC_FIXED] = 0;
- pmu->version = 0;
- pmu->reserved_bits = 0xffffffff00200000ull;
-
- entry = kvm_find_cpuid_entry(vcpu, 0xa, 0);
- if (!entry)
- return;
- eax.full = entry->eax;
- edx.full = entry->edx;
-
- pmu->version = eax.split.version_id;
- if (!pmu->version)
- return;
-
- pmu->nr_arch_gp_counters = min_t(int, eax.split.num_counters,
- INTEL_PMC_MAX_GENERIC);
- pmu->counter_bitmask[KVM_PMC_GP] = ((u64)1 << eax.split.bit_width) - 1;
- pmu->available_event_types = ~entry->ebx &
- ((1ull << eax.split.mask_length) - 1);
-
- if (pmu->version == 1) {
- pmu->nr_arch_fixed_counters = 0;
- } else {
- pmu->nr_arch_fixed_counters =
- min_t(int, edx.split.num_counters_fixed,
- INTEL_PMC_MAX_FIXED);
- pmu->counter_bitmask[KVM_PMC_FIXED] =
- ((u64)1 << edx.split.bit_width_fixed) - 1;
- }
-
- pmu->global_ctrl = ((1 << pmu->nr_arch_gp_counters) - 1) |
- (((1ull << pmu->nr_arch_fixed_counters) - 1) << INTEL_PMC_IDX_FIXED);
- pmu->global_ctrl_mask = ~pmu->global_ctrl;
-
- entry = kvm_find_cpuid_entry(vcpu, 7, 0);
- if (entry &&
- (boot_cpu_has(X86_FEATURE_HLE) || boot_cpu_has(X86_FEATURE_RTM)) &&
- (entry->ebx & (X86_FEATURE_HLE|X86_FEATURE_RTM)))
- pmu->reserved_bits ^= HSW_IN_TX|HSW_IN_TX_CHECKPOINTED;
+ kvm_x86_ops->pmu_ops->refresh(vcpu);
}
void kvm_pmu_reset(struct kvm_vcpu *vcpu)
{
struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
- int i;
irq_work_sync(&pmu->irq_work);
- for (i = 0; i < INTEL_PMC_MAX_GENERIC; i++) {
- struct kvm_pmc *pmc = &pmu->gp_counters[i];
- pmc_stop_counter(pmc);
- pmc->counter = pmc->eventsel = 0;
- }
-
- for (i = 0; i < INTEL_PMC_MAX_FIXED; i++)
- pmc_stop_counter(&pmu->fixed_counters[i]);
-
- pmu->fixed_ctr_ctrl = pmu->global_ctrl = pmu->global_status =
- pmu->global_ovf_ctrl = 0;
+ kvm_x86_ops->pmu_ops->reset(vcpu);
}
void kvm_pmu_init(struct kvm_vcpu *vcpu)
{
- int i;
struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
memset(pmu, 0, sizeof(*pmu));
- for (i = 0; i < INTEL_PMC_MAX_GENERIC; i++) {
- pmu->gp_counters[i].type = KVM_PMC_GP;
- pmu->gp_counters[i].vcpu = vcpu;
- pmu->gp_counters[i].idx = i;
- }
- for (i = 0; i < INTEL_PMC_MAX_FIXED; i++) {
- pmu->fixed_counters[i].type = KVM_PMC_FIXED;
- pmu->fixed_counters[i].vcpu = vcpu;
- pmu->fixed_counters[i].idx = i + INTEL_PMC_IDX_FIXED;
- }
+ kvm_x86_ops->pmu_ops->init(vcpu);
init_irq_work(&pmu->irq_work, kvm_pmi_trigger_fn);
kvm_pmu_refresh(vcpu);
}