Skip to contentSkip to conversation

KVM: x86/pmu: Add *basic* support to enable guest PEBS via DS

PATCH Superseded Like Xu 1 participant 19 messages Mar 4, 2022 – Mar 15, 2022

Subsystems (4)
lore
Share & tools
lore

Like Xu

author
#0 lore raw

v12 cover KVM: x86/pmu: Add *basic* support to enable guest PEBS via DS

Hi,

Out of more accurate profiling results, this feature still has loyal followers and another new rebased version is here. PeterZ had acked the V9 patchset [0] and Paolo had asked for a new version, so please check the changelog and feel free to review, test and comment.

[0] https://lore.kernel.org/kvm/[email protected]/
[1] https://lore.kernel.org/kvm/[email protected]/

---

The guest Precise Event Based Sampling (PEBS) feature can provide an architectural state of the instruction executed after the guest instruction that exactly caused the event. It needs new hardware facility only available on Intel Ice Lake Server platforms. This patch set enables the basic PEBS feature for KVM guests on ICX.

We can use PEBS feature on the Linux guest like native:

   # echo 0 > /proc/sys/kernel/watchdog (on the host)
   # perf record -e instructions:ppp ./br_instr a
   # perf record -c 100000 -e instructions:pp ./br_instr a

To emulate guest PEBS facility for the above perf usages, we need to implement 2 code paths:

1) Fast path

This is when the host assigned physical PMC has an identical index as the virtual PMC (e.g. using physical PMC0 to emulate virtual PMC0). This path is used in most common use cases.

2) Slow path

This is when the host assigned physical PMC has a different index from the virtual PMC (e.g. using physical PMC1 to emulate virtual PMC0) In this case, KVM needs to rewrite the PEBS records to change the applicable counter indexes to the virtual PMC indexes, which would otherwise contain the physical counter index written by PEBS facility, and switch the counter reset values to the offset corresponding to the physical counter indexes in the DS data structure.

The previous version [3] enables both fast path and slow path, which seems a bit more complex as the first step. In this patchset, we want to start with the fast path to get the basic guest PEBS enabled while keeping the slow path disabled. More focused discussion on the slow path [4] is planned to be put to another patchset in the next step.

Compared to later versions in subsequent steps, the functionality to support host-guest PEBS both enabled and the functionality to emulate guest PEBS when the counter is cross-mapped are missing in this patch set (neither of these are typical scenarios).

With the basic support, the guest can retrieve the correct PEBS information from its own PEBS records on the Ice Lake servers. And we expect it should work when migrating to another Ice Lake and no regression about host perf is expected.

Here are the results of pebs test from guest/host for same workload:

perf report on guest:
# Samples: 2K of event 'instructions:ppp', # Event count (approx.): 1473377250 # Overhead Command Shared Object Symbol

   57.74%  br_instr  br_instr           [.] lfsr_cond
   41.40%  br_instr  br_instr           [.] cmp_end
    0.21%  br_instr  [kernel.kallsyms]  [k] __lock_acquire

perf report on host:
# Samples: 2K of event 'instructions:ppp', # Event count (approx.): 1462721386 # Overhead Command Shared Object Symbol

   57.90%  br_instr  br_instr          [.] lfsr_cond
   41.95%  br_instr  br_instr          [.] cmp_end
    0.05%  br_instr  [kernel.vmlinux]  [k] lock_acquire

Conclusion: the profiling results on the guest are similar tothat on the host.

A minimum guest kernel version may be v5.4 or a backport version support Icelake server PEBS.

Please check more details in each commit and feel free to comment.

Previous:
https://lore.kernel.org/kvm/[email protected]/

[3]
https://lore.kernel.org/kvm/[email protected]/
[4]
https://lore.kernel.org/kvm/[email protected]/

V11->V12:
- Apply the new perf interface from tip/perf/core and fix the merge conflict;
- Rename "x86_pmu.pebs_ept" to "x86_pmu.pebs_ept"; (Sean)
- Rebased on the top of kvm/queue (b13a3befc815); (Paolo)

V10->V11:
- Merge perf_guest_info_callbacks static_call to the tip/perf/core;
- Keep use perf_guest_cbs in the kvm/queue context before merge window;
- Fix MSR_IA32_MISC_ENABLE_EMON bit (Liu XiangDong);
- Rebase "Reprogram PEBS event to emulate guest PEBS counter" patch;

V9->V10:
- improve readability in core.c(Peter Z)
- reuse guest_pebs_idxs(Liu XiangDong)

V8 -> V9 Changelog:
-fix a brackets error in xen_guest_state()

V7 -> V8 Changelog:
- fix coding style, add {} for single statement of multiple lines(Peter Z)
- fix coding style in xen_guest_state() (Boris Ostrovsky)
- s/pmu/kvm_pmu/ in intel_guest_get_msrs() (Peter Z)
- put lower cost branch in the first place for x86_pmu_handle_guest_pebs() (Peter Z)

V6 -> V7 Changelog:
- Fix conditions order and call x86_pmu_handle_guest_pebs() unconditionally; (PeterZ)
- Add a new patch to make all that perf_guest_cbs stuff suck less; (PeterZ)
- Document IA32_MISC_ENABLE[7] that that behavior matches bare metal; (Sean & Venkatesh)
- Update commit message for fixed counter mask refactoring;(PeterZ)
- Clarifying comments about {.host and .guest} for intel_guest_get_msrs(); (PeterZ)
- Add pebs_capable to store valid PEBS_COUNTER_MASK value; (PeterZ)
- Add more comments for perf's precise_ip field; (Andi & PeterZ)
- Refactor perf_overflow_handler_t and make it more legible; (PeterZ)
- Use "(unsigned long)cpuc->ds" instead of __this_cpu_read(cpu_hw_events.ds); (PeterZ)
- Keep using "(struct kvm_pmu *)data" to follow K&R; (Andi)

Like Xu (16):
perf/x86/intel: Add EPT-Friendly PEBS for Ice Lake Server perf/x86/intel: Handle guest PEBS overflow PMI for KVM guest perf/x86/core: Pass "struct kvm_pmu *" to determine the guest values
KVM: x86/pmu: Set MSR_IA32_MISC_ENABLE_EMON bit when vPMU is enabled
KVM: x86/pmu: Introduce the ctrl_mask value for fixed counter
KVM: x86/pmu: Add IA32_PEBS_ENABLE MSR emulation for extended PEBS
KVM: x86/pmu: Reprogram PEBS event to emulate guest PEBS counter
KVM: x86/pmu: Adjust precise_ip to emulate Ice Lake guest PDIR counter
KVM: x86/pmu: Add IA32_DS_AREA MSR emulation to support guest DS
KVM: x86/pmu: Add PEBS_DATA_CFG MSR emulation to support adaptive PEBS
KVM: x86: Set PEBS_UNAVAIL in IA32_MISC_ENABLE when PEBS is enabled
KVM: x86/pmu: Move pmc_speculative_in_use() to arch/x86/kvm/pmu.h
KVM: x86/pmu: Disable guest PEBS temporarily in two rare situations
KVM: x86/pmu: Add kvm_pmu_cap to optimize perf_get_x86_pmu_capability
KVM: x86/cpuid: Refactor host/guest CPU model consistency check
KVM: x86/pmu: Expose CPUIDs feature bits PDCM, DS, DTES64

Peter Zijlstra (Intel) (1):
x86/perf/core: Add pebs_capable to store valid PEBS_COUNTER_MASK value

 arch/x86/events/core.c            |   5 +-
 arch/x86/events/intel/core.c      | 157 +++++++++++++++++++++++++-----
 arch/x86/events/perf_event.h      |   6 +-
 arch/x86/include/asm/kvm_host.h   |  16 +++
 arch/x86/include/asm/msr-index.h  |   6 ++
 arch/x86/include/asm/perf_event.h |   5 +-
 arch/x86/kvm/cpuid.c              |  26 ++---
 arch/x86/kvm/cpuid.h              |   5 +
 arch/x86/kvm/pmu.c                |  52 +++++++---
 arch/x86/kvm/pmu.h                |  37 +++++++
 arch/x86/kvm/vmx/capabilities.h   |  28 +++---
 arch/x86/kvm/vmx/pmu_intel.c      | 118 ++++++++++++++++++----
 arch/x86/kvm/vmx/vmx.c            |  24 ++++-
 arch/x86/kvm/vmx/vmx.h            |   2 +-
 arch/x86/kvm/x86.c                |  30 ++++--
 15 files changed, 410 insertions(+), 107 deletions(-)
Signature · 1 line
2.35.1
author
#1 lore raw

PATCH v12 1/17 perf/x86/intel: Add EPT-Friendly PEBS for Ice Lake Server

From: Like Xu <[email protected]>

Add support for EPT-Friendly PEBS, a new CPU feature that enlightens PEBS to translate guest linear address through EPT, and facilitates handling VM-Exits that occur when accessing PEBS records. More information can be found in the December 2021 release of Intel's SDM, Volume 3, 18.9.5 "EPT-Friendly PEBS". This new hardware facility makes sure the guest PEBS records will not be lost, which is available on Intel Ice Lake Server platforms (and later).

KVM will check this field through perf_get_x86_pmu_capability() instead of hard coding the CPU models in the KVM code. If it is supported, the guest PEBS capability will be exposed to the guest. Guest PEBS can be enabled when and only when "EPT-Friendly PEBS" is supported and EPT is enabled.

---
 arch/x86/events/core.c            | 1 +
 arch/x86/events/intel/core.c      | 1 +
 arch/x86/events/perf_event.h      | 3 ++-
 arch/x86/include/asm/perf_event.h | 1 +
 4 files changed, 5 insertions(+), 1 deletion(-)
arch/x86/events/core.c +1 -0
index e686c5e0537b..7e3d0a019444 100644
@@ -2994,5 +2994,6 @@ void perf_get_x86_pmu_capability(struct x86_pmu_capability *cap)
 	cap->bit_width_fixed	= x86_pmu.cntval_bits;
 	cap->events_mask	= (unsigned int)x86_pmu.events_maskl;
 	cap->events_mask_len	= x86_pmu.events_mask_len;
+	cap->pebs_ept		= x86_pmu.pebs_ept;
 }
 EXPORT_SYMBOL_GPL(perf_get_x86_pmu_capability);
arch/x86/events/intel/core.c +1 -0
index a3c7ca876aeb..7723fa6ed65e 100644
@@ -6113,6 +6113,7 @@ __init int intel_pmu_init(void)
 
 	case INTEL_FAM6_ICELAKE_X:
 	case INTEL_FAM6_ICELAKE_D:
+		x86_pmu.pebs_ept = 1;
 		pmem = true;
 		fallthrough;
 	case INTEL_FAM6_ICELAKE_L:
arch/x86/events/perf_event.h +2 -1
index 150261d929b9..0998742760c8 100644
@@ -815,7 +815,8 @@ struct x86_pmu {
 			pebs_prec_dist		:1,
 			pebs_no_tlb		:1,
 			pebs_no_isolation	:1,
-			pebs_block		:1;
+			pebs_block		:1,
+			pebs_ept		:1;
 	int		pebs_record_size;
 	int		pebs_buffer_size;
 	int		max_pebs_events;
arch/x86/include/asm/perf_event.h +1 -0
index 8fc1b5003713..2c9dce37d095 100644
@@ -192,6 +192,7 @@ struct x86_pmu_capability {
 	int		bit_width_fixed;
 	unsigned int	events_mask;
 	int		events_mask_len;
+	unsigned int	pebs_ept	:1;
 };
 
 /*
Signature · 1 line
2.35.1
author
#2 lore raw

PATCH v12 2/17 perf/x86/intel: Handle guest PEBS overflow PMI for KVM guest

In reply to Like Xu

From: Like Xu <[email protected]>

With PEBS virtualization, the guest PEBS records get delivered to the guest DS, and the host pmi handler uses perf_guest_cbs->is_in_guest() to distinguish whether the PMI comes from the guest code like Intel PT.

No matter how many guest PEBS counters are overflowed, only triggering one fake event is enough. The fake event causes the KVM PMI callback to be called, thereby injecting the PEBS overflow PMI into the guest.

KVM may inject the PMI with BUFFER_OVF set, even if the guest DS is empty. That should really be harmless. Thus guest PEBS handler would retrieve the correct information from its own PEBS records buffer.

---
 arch/x86/events/intel/core.c | 42 ++++++++++++++++++++++++++++++++++++
 1 file changed, 42 insertions(+)
arch/x86/events/intel/core.c +42 -0
index 7723fa6ed65e..da4b77f6c6a4 100644
@@ -2831,6 +2831,47 @@ static void intel_pmu_reset(void)
 	local_irq_restore(flags);
 }
 
+/*
+ * We may be running with guest PEBS events created by KVM, and the
+ * PEBS records are logged into the guest's DS and invisible to host.
+ *
+ * In the case of guest PEBS overflow, we only trigger a fake event
+ * to emulate the PEBS overflow PMI for guest PEBS counters in KVM.
+ * The guest will then vm-entry and check the guest DS area to read
+ * the guest PEBS records.
+ *
+ * The contents and other behavior of the guest event do not matter.
+ */
+static void x86_pmu_handle_guest_pebs(struct pt_regs *regs,
+				      struct perf_sample_data *data)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	u64 guest_pebs_idxs = cpuc->pebs_enabled & ~cpuc->intel_ctrl_host_mask;
+	struct perf_event *event = NULL;
+	int bit;
+
+	if (!unlikely(perf_guest_state()))
+		return;
+
+	if (!x86_pmu.pebs_ept || !x86_pmu.pebs_active ||
+	    !guest_pebs_idxs)
+		return;
+
+	for_each_set_bit(bit, (unsigned long *)&guest_pebs_idxs,
+			 INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed) {
+		event = cpuc->events[bit];
+		if (!event->attr.precise_ip)
+			continue;
+
+		perf_sample_data_init(data, 0, event->hw.last_period);
+		if (perf_event_overflow(event, data, regs))
+			x86_pmu_stop(event, 0);
+
+		/* Inject one fake event is enough. */
+		break;
+	}
+}
+
 static int handle_pmi_common(struct pt_regs *regs, u64 status)
 {
 	struct perf_sample_data data;
@@ -2882,6 +2923,7 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)
 		u64 pebs_enabled = cpuc->pebs_enabled;
 
 		handled++;
+		x86_pmu_handle_guest_pebs(regs, &data);
 		x86_pmu.drain_pebs(regs, &data);
 		status &= intel_ctrl | GLOBAL_STATUS_TRACE_TOPAPMI;
 
Signature · 1 line
2.35.1
author
#3 lore raw

PATCH v12 3/17 perf/x86/core: Pass "struct kvm_pmu *" to determine the guest values

In reply to Like Xu

From: Like Xu <[email protected]>

Splitting the logic for determining the guest values is unnecessarily confusing, and potentially fragile. Perf should have full knowledge and control of what values are loaded for the guest.

If we change .guest_get_msrs() to take a struct kvm_pmu pointer, then it can generate the full set of guest values by grabbing guest ds_area and pebs_data_cfg. Alternatively, .guest_get_msrs() could take the desired guest MSR values directly (ds_area and pebs_data_cfg), but kvm_pmu is vendor agnostic, so we don't see any reason to not just pass the pointer.

---
 arch/x86/events/core.c            | 4 ++--
 arch/x86/events/intel/core.c      | 4 ++--
 arch/x86/events/perf_event.h      | 2 +-
 arch/x86/include/asm/perf_event.h | 4 ++--
 arch/x86/kvm/vmx/vmx.c            | 3 ++-
 5 files changed, 9 insertions(+), 8 deletions(-)
arch/x86/events/core.c +2 -2
index 7e3d0a019444..0ce9493cc110 100644
@@ -693,9 +693,9 @@ void x86_pmu_disable_all(void)
 	}
 }
 
-struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr)
+struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr, void *data)
 {
-	return static_call(x86_pmu_guest_get_msrs)(nr);
+	return static_call(x86_pmu_guest_get_msrs)(nr, data);
 }
 EXPORT_SYMBOL_GPL(perf_guest_get_msrs);
 
arch/x86/events/intel/core.c +2 -2
index da4b77f6c6a4..820f9fb9339b 100644
@@ -3951,7 +3951,7 @@ static int intel_pmu_hw_config(struct perf_event *event)
 	return 0;
 }
 
-static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr)
+static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
@@ -3984,7 +3984,7 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr)
 	return arr;
 }
 
-static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr)
+static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr, void *data)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
arch/x86/events/perf_event.h +1 -1
index 0998742760c8..bf23cbe4f6cf 100644
@@ -900,7 +900,7 @@ struct x86_pmu {
 	/*
 	 * Intel host/guest support (KVM)
 	 */
-	struct perf_guest_switch_msr *(*guest_get_msrs)(int *nr);
+	struct perf_guest_switch_msr *(*guest_get_msrs)(int *nr, void *data);
 
 	/*
 	 * Check period value for PERF_EVENT_IOC_PERIOD ioctl.
arch/x86/include/asm/perf_event.h +2 -2
index 2c9dce37d095..729658f00ee6 100644
@@ -492,10 +492,10 @@ static inline void perf_check_microcode(void) { }
 #endif
 
 #if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_INTEL)
-extern struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr);
+extern struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr, void *data);
 extern int x86_perf_get_lbr(struct x86_pmu_lbr *lbr);
 #else
-struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr);
+struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr, void *data);
 static inline int x86_perf_get_lbr(struct x86_pmu_lbr *lbr)
 {
 	return -1;
arch/x86/kvm/vmx/vmx.c +2 -1
index b325f99b2177..8fb29bbfe875 100644
@@ -6710,9 +6710,10 @@ static void atomic_switch_perf_msrs(struct vcpu_vmx *vmx)
 {
 	int i, nr_msrs;
 	struct perf_guest_switch_msr *msrs;
+	struct kvm_pmu *pmu = vcpu_to_pmu(&vmx->vcpu);
 
 	/* Note, nr_msrs may be garbage if perf_guest_get_msrs() returns NULL. */
-	msrs = perf_guest_get_msrs(&nr_msrs);
+	msrs = perf_guest_get_msrs(&nr_msrs, (void *)pmu);
 	if (!msrs)
 		return;
 
Signature · 1 line
2.35.1
author
#4 lore raw

PATCH v12 4/17 KVM: x86/pmu: Set MSR_IA32_MISC_ENABLE_EMON bit when vPMU is enabled

In reply to Like Xu

From: Like Xu <[email protected]>

On Intel platforms, the software can use the IA32_MISC_ENABLE[7] bit to detect whether the processor supports performance monitoring facility.

It depends on the PMU is enabled for the guest, and a software write operation to this available bit will be ignored. The proposal to ignore the toggle in KVM is the way to go and that behavior matches bare metal.

---
 arch/x86/kvm/vmx/pmu_intel.c |  1 +
 arch/x86/kvm/x86.c           | 15 +++++++++++++--
 2 files changed, 14 insertions(+), 2 deletions(-)
arch/x86/kvm/vmx/pmu_intel.c +1 -0
index 4e5b1eeeb77c..61de7b07489c 100644
@@ -496,6 +496,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	if (!pmu->version)
 		return;
 
+	vcpu->arch.ia32_misc_enable_msr |= MSR_IA32_MISC_ENABLE_EMON;
 	perf_get_x86_pmu_capability(&x86_pmu);
 
 	pmu->nr_arch_gp_counters = min_t(int, eax.split.num_counters,
arch/x86/kvm/x86.c +13 -2
index cf17af4d6904..3c7a51468216 100644
@@ -3515,9 +3515,19 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			vcpu->arch.ia32_tsc_adjust_msr = data;
 		}
 		break;
-	case MSR_IA32_MISC_ENABLE:
+	case MSR_IA32_MISC_ENABLE: {
+		u64 old_val = vcpu->arch.ia32_misc_enable_msr;
+		u64 pmu_mask = MSR_IA32_MISC_ENABLE_EMON;
+
+		/*
+		 * For a dummy user space, the order of setting vPMU capabilities and
+		 * initialising MSR_IA32_MISC_ENABLE is not strictly guaranteed, so to
+		 * avoid inconsistent functionality we keep the vPMU bits unchanged here.
+		 */
+		data &= ~pmu_mask;
+		data |= old_val & pmu_mask;
 		if (!kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT) &&
-		    ((vcpu->arch.ia32_misc_enable_msr ^ data) & MSR_IA32_MISC_ENABLE_MWAIT)) {
+		    ((old_val ^ data)  & MSR_IA32_MISC_ENABLE_MWAIT)) {
 			if (!guest_cpuid_has(vcpu, X86_FEATURE_XMM3))
 				return 1;
 			vcpu->arch.ia32_misc_enable_msr = data;
@@ -3526,6 +3536,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			vcpu->arch.ia32_misc_enable_msr = data;
 		}
 		break;
+	}
 	case MSR_IA32_SMBASE:
 		if (!msr_info->host_initiated)
 			return 1;
Signature · 1 line
2.35.1
author
#5 lore raw

PATCH v12 5/17 KVM: x86/pmu: Introduce the ctrl_mask value for fixed counter

In reply to Like Xu

From: Like Xu <[email protected]>

The mask value of fixed counter control register should be dynamic adjusted with the number of fixed counters. This patch introduces a variable that includes the reserved bits of fixed counter control registers. This is a generic code refactoring.

---
 arch/x86/include/asm/kvm_host.h | 1 +
 arch/x86/kvm/vmx/pmu_intel.c    | 6 +++++-
 2 files changed, 6 insertions(+), 1 deletion(-)
arch/x86/include/asm/kvm_host.h +1 -0
index c45ab8b5c37f..cf5ce12557f2 100644
@@ -504,6 +504,7 @@ struct kvm_pmu {
 	unsigned nr_arch_fixed_counters;
 	unsigned available_event_types;
 	u64 fixed_ctr_ctrl;
+	u64 fixed_ctr_ctrl_mask;
 	u64 global_ctrl;
 	u64 global_status;
 	u64 counter_bitmask[2];
arch/x86/kvm/vmx/pmu_intel.c +5 -1
index 61de7b07489c..c5f885198c60 100644
@@ -394,7 +394,7 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_CORE_PERF_FIXED_CTR_CTRL:
 		if (pmu->fixed_ctr_ctrl == data)
 			return 0;
-		if (!(data & 0xfffffffffffff444ull)) {
+		if (!(data & pmu->fixed_ctr_ctrl_mask)) {
 			reprogram_fixed_counters(pmu, data);
 			return 0;
 		}
@@ -478,6 +478,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	struct kvm_cpuid_entry2 *entry;
 	union cpuid10_eax eax;
 	union cpuid10_edx edx;
+	int i;
 
 	pmu->nr_arch_gp_counters = 0;
 	pmu->nr_arch_fixed_counters = 0;
@@ -485,6 +486,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	pmu->counter_bitmask[KVM_PMC_FIXED] = 0;
 	pmu->version = 0;
 	pmu->reserved_bits = 0xffffffff00200000ull;
+	pmu->fixed_ctr_ctrl_mask = ~0ull;
 
 	entry = kvm_find_cpuid_entry(vcpu, 0xa, 0);
 	if (!entry || !vcpu->kvm->arch.enable_pmu)
@@ -521,6 +523,8 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		setup_fixed_pmc_eventsel(pmu);
 	}
 
+	for (i = 0; i < pmu->nr_arch_fixed_counters; i++)
+		pmu->fixed_ctr_ctrl_mask &= ~(0xbull << (i * 4));
 	pmu->global_ctrl = ((1ull << pmu->nr_arch_gp_counters) - 1) |
 		(((1ull << pmu->nr_arch_fixed_counters) - 1) << INTEL_PMC_IDX_FIXED);
 	pmu->global_ctrl_mask = ~pmu->global_ctrl;
Signature · 1 line
2.35.1
author
#6 lore raw

PATCH v12 6/17 x86/perf/core: Add pebs_capable to store valid PEBS_COUNTER_MASK value

In reply to Like Xu

From: "Peter Zijlstra (Intel)" <[email protected]>

The value of pebs_counter_mask will be accessed frequently for repeated use in the intel_guest_get_msrs(). So it can be optimized instead of endlessly mucking about with branches.

---
 arch/x86/events/intel/core.c | 14 ++++++--------
 arch/x86/events/perf_event.h |  1 +
 2 files changed, 7 insertions(+), 8 deletions(-)
arch/x86/events/intel/core.c +6 -8
index 820f9fb9339b..7f0bab2f70fd 100644
@@ -2911,10 +2911,7 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status)
 	 * counters from the GLOBAL_STATUS mask and we always process PEBS
 	 * events via drain_pebs().
 	 */
-	if (x86_pmu.flags & PMU_FL_PEBS_ALL)
-		status &= ~cpuc->pebs_enabled;
-	else
-		status &= ~(cpuc->pebs_enabled & PEBS_COUNTER_MASK);
+	status &= ~(cpuc->pebs_enabled & x86_pmu.pebs_capable);
 
 	/*
 	 * PEBS overflow sets bit 62 in the global status register
@@ -3960,10 +3957,7 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 	arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL;
 	arr[0].host = intel_ctrl & ~cpuc->intel_ctrl_guest_mask;
 	arr[0].guest = intel_ctrl & ~cpuc->intel_ctrl_host_mask;
-	if (x86_pmu.flags & PMU_FL_PEBS_ALL)
-		arr[0].guest &= ~cpuc->pebs_enabled;
-	else
-		arr[0].guest &= ~(cpuc->pebs_enabled & PEBS_COUNTER_MASK);
+	arr[0].guest &= ~(cpuc->pebs_enabled & x86_pmu.pebs_capable);
 	*nr = 1;
 
 	if (x86_pmu.pebs && x86_pmu.pebs_no_isolation) {
@@ -5667,6 +5661,7 @@ __init int intel_pmu_init(void)
 	x86_pmu.events_mask_len		= eax.split.mask_length;
 
 	x86_pmu.max_pebs_events		= min_t(unsigned, MAX_PEBS_EVENTS, x86_pmu.num_counters);
+	x86_pmu.pebs_capable		= PEBS_COUNTER_MASK;
 
 	/*
 	 * Quirk: v2 perfmon does not report fixed-purpose events, so
@@ -5851,6 +5846,7 @@ __init int intel_pmu_init(void)
 		x86_pmu.pebs_aliases = NULL;
 		x86_pmu.pebs_prec_dist = true;
 		x86_pmu.lbr_pt_coexist = true;
+		x86_pmu.pebs_capable = ~0ULL;
 		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
 		x86_pmu.flags |= PMU_FL_PEBS_ALL;
 		x86_pmu.get_event_constraints = glp_get_event_constraints;
@@ -6208,6 +6204,7 @@ __init int intel_pmu_init(void)
 		x86_pmu.pebs_aliases = NULL;
 		x86_pmu.pebs_prec_dist = true;
 		x86_pmu.pebs_block = true;
+		x86_pmu.pebs_capable = ~0ULL;
 		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
 		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
 		x86_pmu.flags |= PMU_FL_PEBS_ALL;
@@ -6250,6 +6247,7 @@ __init int intel_pmu_init(void)
 		x86_pmu.pebs_aliases = NULL;
 		x86_pmu.pebs_prec_dist = true;
 		x86_pmu.pebs_block = true;
+		x86_pmu.pebs_capable = ~0ULL;
 		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
 		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
 		x86_pmu.flags |= PMU_FL_PEBS_ALL;
arch/x86/events/perf_event.h +1 -0
index bf23cbe4f6cf..9e1bef9c2b0c 100644
@@ -825,6 +825,7 @@ struct x86_pmu {
 	void		(*pebs_aliases)(struct perf_event *event);
 	unsigned long	large_pebs_flags;
 	u64		rtm_abort_event;
+	u64		pebs_capable;
 
 	/*
 	 * Intel LBR
Signature · 1 line
2.35.1
author
#7 lore raw

PATCH v12 7/17 KVM: x86/pmu: Add IA32_PEBS_ENABLE MSR emulation for extended PEBS

In reply to Like Xu

From: Like Xu <[email protected]>

If IA32_PERF_CAPABILITIES.PEBS_BASELINE [bit 14] is set, the IA32_PEBS_ENABLE MSR exists and all architecturally enumerated fixed and general-purpose counters have corresponding bits in IA32_PEBS_ENABLE that enable generation of PEBS records. The general-purpose counter bits start at bit IA32_PEBS_ENABLE[0], and the fixed counter bits start at bit IA32_PEBS_ENABLE[32].

When guest PEBS is enabled, the IA32_PEBS_ENABLE MSR will be added to the perf_guest_switch_msr() and atomically switched during the VMX transitions just like CORE_PERF_GLOBAL_CTRL MSR.

Based on whether the platform supports x86_pmu.pebs_ept, it has also refactored the way to add more msrs to arr[] in intel_guest_get_msrs() for extensibility.

---
 arch/x86/events/intel/core.c     | 75 ++++++++++++++++++++++++--------
 arch/x86/include/asm/kvm_host.h  |  3 ++
 arch/x86/include/asm/msr-index.h |  6 +++
 arch/x86/kvm/vmx/pmu_intel.c     | 31 +++++++++++++
 4 files changed, 97 insertions(+), 18 deletions(-)
arch/x86/events/intel/core.c +57 -18
index 7f0bab2f70fd..b08f0eb6cfee 100644
@@ -3948,33 +3948,72 @@ static int intel_pmu_hw_config(struct perf_event *event)
 	return 0;
 }
 
+/*
+ * Currently, the only caller of this function is the atomic_switch_perf_msrs().
+ * The host perf conext helps to prepare the values of the real hardware for
+ * a set of msrs that need to be switched atomically in a vmx transaction.
+ *
+ * For example, the pseudocode needed to add a new msr should look like:
+ *
+ * arr[(*nr)++] = (struct perf_guest_switch_msr){
+ *	.msr = the hardware msr address,
+ *	.host = the value the hardware has when it doesn't run a guest,
+ *	.guest = the value the hardware has when it runs a guest,
+ * };
+ *
+ * These values have nothing to do with the emulated values the guest sees
+ * when it uses {RD,WR}MSR, which should be handled by the KVM context,
+ * specifically in the intel_pmu_{get,set}_msr().
+ */
 static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
 	u64 intel_ctrl = hybrid(cpuc->pmu, intel_ctrl);
+	u64 pebs_mask = cpuc->pebs_enabled & x86_pmu.pebs_capable;
+	int global_ctrl, pebs_enable;
 
-	arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL;
-	arr[0].host = intel_ctrl & ~cpuc->intel_ctrl_guest_mask;
-	arr[0].guest = intel_ctrl & ~cpuc->intel_ctrl_host_mask;
-	arr[0].guest &= ~(cpuc->pebs_enabled & x86_pmu.pebs_capable);
-	*nr = 1;
+	*nr = 0;
+	global_ctrl = (*nr)++;
+	arr[global_ctrl] = (struct perf_guest_switch_msr){
+		.msr = MSR_CORE_PERF_GLOBAL_CTRL,
+		.host = intel_ctrl & ~cpuc->intel_ctrl_guest_mask,
+		.guest = intel_ctrl & (~cpuc->intel_ctrl_host_mask | ~pebs_mask),
+	};
 
-	if (x86_pmu.pebs && x86_pmu.pebs_no_isolation) {
-		/*
-		 * If PMU counter has PEBS enabled it is not enough to
-		 * disable counter on a guest entry since PEBS memory
-		 * write can overshoot guest entry and corrupt guest
-		 * memory. Disabling PEBS solves the problem.
-		 *
-		 * Don't do this if the CPU already enforces it.
-		 */
-		arr[1].msr = MSR_IA32_PEBS_ENABLE;
-		arr[1].host = cpuc->pebs_enabled;
-		arr[1].guest = 0;
-		*nr = 2;
+	if (!x86_pmu.pebs)
+		return arr;
+
+	/*
+	 * If PMU counter has PEBS enabled it is not enough to
+	 * disable counter on a guest entry since PEBS memory
+	 * write can overshoot guest entry and corrupt guest
+	 * memory. Disabling PEBS solves the problem.
+	 *
+	 * Don't do this if the CPU already enforces it.
+	 */
+	if (x86_pmu.pebs_no_isolation) {
+		arr[(*nr)++] = (struct perf_guest_switch_msr){
+			.msr = MSR_IA32_PEBS_ENABLE,
+			.host = cpuc->pebs_enabled,
+			.guest = 0,
+		};
+		return arr;
 	}
 
+	if (!x86_pmu.pebs_ept)
+		return arr;
+	pebs_enable = (*nr)++;
+
+	arr[pebs_enable] = (struct perf_guest_switch_msr){
+		.msr = MSR_IA32_PEBS_ENABLE,
+		.host = cpuc->pebs_enabled & ~cpuc->intel_ctrl_guest_mask,
+		.guest = pebs_mask & ~cpuc->intel_ctrl_host_mask,
+	};
+
+	/* Set hw GLOBAL_CTRL bits for PEBS counter when it runs for guest */
+	arr[0].guest |= arr[*nr].guest;
+
 	return arr;
 }
 
arch/x86/include/asm/kvm_host.h +3 -0
index cf5ce12557f2..3bc4c5d79110 100644
@@ -519,6 +519,9 @@ struct kvm_pmu {
 	DECLARE_BITMAP(all_valid_pmc_idx, X86_PMC_IDX_MAX);
 	DECLARE_BITMAP(pmc_in_use, X86_PMC_IDX_MAX);
 
+	u64 pebs_enable;
+	u64 pebs_enable_mask;
+
 	/*
 	 * The gate to release perf_events not marked in
 	 * pmc_in_use only once in a vcpu time slice.
arch/x86/include/asm/msr-index.h +6 -0
index a4a39c3e0f19..ec0207a43f4d 100644
@@ -189,6 +189,12 @@
 #define PERF_CAP_PT_IDX			16
 
 #define MSR_PEBS_LD_LAT_THRESHOLD	0x000003f6
+#define PERF_CAP_PEBS_TRAP             BIT_ULL(6)
+#define PERF_CAP_ARCH_REG              BIT_ULL(7)
+#define PERF_CAP_PEBS_FORMAT           0xf00
+#define PERF_CAP_PEBS_BASELINE         BIT_ULL(14)
+#define PERF_CAP_PEBS_MASK	(PERF_CAP_PEBS_TRAP | PERF_CAP_ARCH_REG | \
+				 PERF_CAP_PEBS_FORMAT | PERF_CAP_PEBS_BASELINE)
 
 #define MSR_IA32_RTIT_CTL		0x00000570
 #define RTIT_CTL_TRACEEN		BIT(0)
arch/x86/kvm/vmx/pmu_intel.c +31 -0
index c5f885198c60..9f1b25d7b966 100644
@@ -214,6 +214,9 @@ static bool intel_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 	case MSR_CORE_PERF_GLOBAL_OVF_CTRL:
 		ret = pmu->version > 1;
 		break;
+	case MSR_IA32_PEBS_ENABLE:
+		ret = vcpu->arch.perf_capabilities & PERF_CAP_PEBS_FORMAT;
+		break;
 	default:
 		ret = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0) ||
 			get_gp_pmc(pmu, msr, MSR_P6_EVNTSEL0) ||
@@ -361,6 +364,9 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_CORE_PERF_GLOBAL_OVF_CTRL:
 		msr_info->data = 0;
 		return 0;
+	case MSR_IA32_PEBS_ENABLE:
+		msr_info->data = pmu->pebs_enable;
+		return 0;
 	default:
 		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
 		    (pmc = get_gp_pmc(pmu, msr, MSR_IA32_PMC0))) {
@@ -420,6 +426,14 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			return 0;
 		}
 		break;
+	case MSR_IA32_PEBS_ENABLE:
+		if (pmu->pebs_enable == data)
+			return 0;
+		if (!(data & pmu->pebs_enable_mask)) {
+			pmu->pebs_enable = data;
+			return 0;
+		}
+		break;
 	default:
 		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
 		    (pmc = get_gp_pmc(pmu, msr, MSR_IA32_PMC0))) {
@@ -487,6 +501,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	pmu->version = 0;
 	pmu->reserved_bits = 0xffffffff00200000ull;
 	pmu->fixed_ctr_ctrl_mask = ~0ull;
+	pmu->pebs_enable_mask = ~0ull;
 
 	entry = kvm_find_cpuid_entry(vcpu, 0xa, 0);
 	if (!entry || !vcpu->kvm->arch.enable_pmu)
@@ -556,6 +571,22 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 
 	if (lbr_desc->records.nr)
 		bitmap_set(pmu->all_valid_pmc_idx, INTEL_PMC_IDX_FIXED_VLBR, 1);
+
+	if (vcpu->arch.perf_capabilities & PERF_CAP_PEBS_FORMAT) {
+		if (vcpu->arch.perf_capabilities & PERF_CAP_PEBS_BASELINE) {
+			pmu->pebs_enable_mask = ~pmu->global_ctrl;
+			pmu->reserved_bits &= ~ICL_EVENTSEL_ADAPTIVE;
+			for (i = 0; i < pmu->nr_arch_fixed_counters; i++) {
+				pmu->fixed_ctr_ctrl_mask &=
+					~(1ULL << (INTEL_PMC_IDX_FIXED + i * 4));
+			}
+		} else {
+			pmu->pebs_enable_mask =
+				~((1ull << pmu->nr_arch_gp_counters) - 1);
+		}
+	} else {
+		vcpu->arch.perf_capabilities &= ~PERF_CAP_PEBS_MASK;
+	}
 }
 
 static void intel_pmu_init(struct kvm_vcpu *vcpu)
Signature · 1 line
2.35.1
author
#8 lore raw

PATCH v12 8/17 KVM: x86/pmu: Reprogram PEBS event to emulate guest PEBS counter

In reply to Like Xu

From: Like Xu <[email protected]>

When a guest counter is configured as a PEBS counter through IA32_PEBS_ENABLE, a guest PEBS event will be reprogrammed by configuring a non-zero precision level in the perf_event_attr.

The guest PEBS overflow PMI bit would be set in the guest GLOBAL_STATUS MSR when PEBS facility generates a PEBS overflow PMI based on guest IA32_DS_AREA MSR.

Even with the same counter index and the same event code and mask, guest PEBS events will not be reused for non-PEBS events.

---
 arch/x86/kvm/pmu.c | 36 +++++++++++++++++++++++++++++++++---
 1 file changed, 33 insertions(+), 3 deletions(-)
arch/x86/kvm/pmu.c +33 -3
index b1a02993782b..51a218f53fed 100644
@@ -60,15 +60,22 @@ static void kvm_pmi_trigger_fn(struct irq_work *irq_work)
 static inline void __kvm_perf_overflow(struct kvm_pmc *pmc, bool in_pmi)
 {
 	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
+	bool skip_pmi = false;
 
 	/* Ignore counters that have been reprogrammed already. */
 	if (test_and_set_bit(pmc->idx, pmu->reprogram_pmi))
 		return;
 
-	__set_bit(pmc->idx, (unsigned long *)&pmu->global_status);
+	if (pmc->perf_event && pmc->perf_event->attr.precise_ip) {
+		/* Indicate PEBS overflow PMI to guest. */
+		skip_pmi = __test_and_set_bit(GLOBAL_STATUS_BUFFER_OVF_BIT,
+					      (unsigned long *)&pmu->global_status);
+	} else {
+		__set_bit(pmc->idx, (unsigned long *)&pmu->global_status);
+	}
 	kvm_make_request(KVM_REQ_PMU, pmc->vcpu);
 
-	if (!pmc->intr)
+	if (!pmc->intr || skip_pmi)
 		return;
 
 	/*
@@ -99,6 +106,7 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
 				  bool exclude_kernel, bool intr,
 				  bool in_tx, bool in_tx_cp)
 {
+	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
 	struct perf_event *event;
 	struct perf_event_attr attr = {
 		.type = type,
@@ -110,6 +118,7 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
 		.exclude_kernel = exclude_kernel,
 		.config = config,
 	};
+	bool pebs = test_bit(pmc->idx, (unsigned long *)&pmu->pebs_enable);
 
 	if (type == PERF_TYPE_HARDWARE && config >= PERF_COUNT_HW_MAX)
 		return;
@@ -127,6 +136,23 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
 		attr.sample_period = 0;
 		attr.config |= HSW_IN_TX_CHECKPOINTED;
 	}
+	if (pebs) {
+		/*
+		 * The non-zero precision level of guest event makes the ordinary
+		 * guest event becomes a guest PEBS event and triggers the host
+		 * PEBS PMI handler to determine whether the PEBS overflow PMI
+		 * comes from the host counters or the guest.
+		 *
+		 * For most PEBS hardware events, the difference in the software
+		 * precision levels of guest and host PEBS events will not affect
+		 * the accuracy of the PEBS profiling result, because the "event IP"
+		 * in the PEBS record is calibrated on the guest side.
+		 *
+		 * On Icelake everything is fine. Other hardware (GLC+, TNT+) that
+		 * could possibly care here is unsupported and needs changes.
+		 */
+		attr.precise_ip = 1;
+	}
 
 	event = perf_event_create_kernel_counter(&attr, -1, current,
 						 kvm_perf_overflow, pmc);
@@ -140,7 +166,7 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
 	pmc_to_pmu(pmc)->event_count++;
 	clear_bit(pmc->idx, pmc_to_pmu(pmc)->reprogram_pmi);
 	pmc->is_paused = false;
-	pmc->intr = intr;
+	pmc->intr = intr || pebs;
 }
 
 static void pmc_pause_counter(struct kvm_pmc *pmc)
@@ -166,6 +192,10 @@ static bool pmc_resume_counter(struct kvm_pmc *pmc)
 			      get_sample_period(pmc, pmc->counter)))
 		return false;
 
+	if (!test_bit(pmc->idx, (unsigned long *)&pmc_to_pmu(pmc)->pebs_enable) &&
+	    pmc->perf_event->attr.precise_ip)
+		return false;
+
 	/* reuse perf_event to serve as pmc_reprogram_counter() does*/
 	perf_event_enable(pmc->perf_event);
 	pmc->is_paused = false;
Signature · 1 line
2.35.1
author
#9 lore raw

PATCH v12 9/17 KVM: x86/pmu: Adjust precise_ip to emulate Ice Lake guest PDIR counter

In reply to Like Xu

From: Like Xu <[email protected]>

The PEBS-PDIR facility on Ice Lake server is supported on IA31_FIXED0 only. If the guest configures counter 32 and PEBS is enabled, the PEBS-PDIR facility is supposed to be used, in which case KVM adjusts attr.precise_ip to 3 and request host perf to assign the exactly requested counter or fail.

The CPU model check is also required since some platforms may place the PEBS-PDIR facility in another counter index.

---
 arch/x86/events/intel/core.c | 2 +-
 arch/x86/kvm/pmu.c           | 2 ++
 arch/x86/kvm/pmu.h           | 7 +++++++
 3 files changed, 10 insertions(+), 1 deletion(-)
arch/x86/events/intel/core.c +1 -1
index b08f0eb6cfee..f88b3be88061 100644
@@ -4003,8 +4003,8 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 
 	if (!x86_pmu.pebs_ept)
 		return arr;
-	pebs_enable = (*nr)++;
 
+	pebs_enable = (*nr)++;
 	arr[pebs_enable] = (struct perf_guest_switch_msr){
 		.msr = MSR_IA32_PEBS_ENABLE,
 		.host = cpuc->pebs_enabled & ~cpuc->intel_ctrl_guest_mask,
arch/x86/kvm/pmu.c +2 -0
index 51a218f53fed..f16dfd7431d1 100644
@@ -152,6 +152,8 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
 		 * could possibly care here is unsupported and needs changes.
 		 */
 		attr.precise_ip = 1;
+		if (x86_match_cpu(vmx_icl_pebs_cpu) && pmc->idx == 32)
+			attr.precise_ip = 3;
 	}
 
 	event = perf_event_create_kernel_counter(&attr, -1, current,
arch/x86/kvm/pmu.h +7 -0
index 7a7b8d5b775e..28c3a826f169 100644
@@ -4,6 +4,8 @@
 
 #include <linux/nospec.h>
 
+#include <asm/cpu_device_id.h>
+
 #define vcpu_to_pmu(vcpu) (&(vcpu)->arch.pmu)
 #define pmu_to_vcpu(pmu)  (container_of((pmu), struct kvm_vcpu, arch.pmu))
 #define pmc_to_pmu(pmc)   (&(pmc)->vcpu->arch.pmu)
@@ -16,6 +18,11 @@
 #define VMWARE_BACKDOOR_PMC_APPARENT_TIME	0x10002
 
 #define MAX_FIXED_COUNTERS	3
+static const struct x86_cpu_id vmx_icl_pebs_cpu[] = {
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_D, NULL),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X, NULL),
+	{}
+};
 
 struct kvm_event_hw_type_mapping {
 	u8 eventsel;
Signature · 1 line
2.35.1
author
#10 lore raw

PATCH v12 10/17 KVM: x86/pmu: Add IA32_DS_AREA MSR emulation to support guest DS

In reply to Like Xu

From: Like Xu <[email protected]>

When CPUID.01H:EDX.DS[21] is set, the IA32_DS_AREA MSR exists and points to the linear address of the first byte of the DS buffer management area, which is used to manage the PEBS records.

When guest PEBS is enabled, the MSR_IA32_DS_AREA MSR will be added to the perf_guest_switch_msr() and switched during the VMX transitions just like CORE_PERF_GLOBAL_CTRL MSR. The WRMSR to IA32_DS_AREA MSR brings a #GP(0) if the source register contains a non-canonical address.

---
 arch/x86/events/intel/core.c    | 10 +++++++++-
 arch/x86/include/asm/kvm_host.h |  1 +
 arch/x86/kvm/vmx/pmu_intel.c    | 11 +++++++++++
 3 files changed, 21 insertions(+), 1 deletion(-)
arch/x86/events/intel/core.c +9 -1
index f88b3be88061..1e303539f205 100644
@@ -21,6 +21,7 @@
 #include <asm/intel_pt.h>
 #include <asm/apic.h>
 #include <asm/cpu_device_id.h>
+#include <asm/kvm_host.h>
 
 #include "../perf_event.h"
 
@@ -3969,6 +3970,7 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+	struct kvm_pmu *kvm_pmu = (struct kvm_pmu *)data;
 	u64 intel_ctrl = hybrid(cpuc->pmu, intel_ctrl);
 	u64 pebs_mask = cpuc->pebs_enabled & x86_pmu.pebs_capable;
 	int global_ctrl, pebs_enable;
@@ -4001,9 +4003,15 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 		return arr;
 	}
 
-	if (!x86_pmu.pebs_ept)
+	if (!kvm_pmu || !x86_pmu.pebs_ept)
 		return arr;
 
+	arr[(*nr)++] = (struct perf_guest_switch_msr){
+		.msr = MSR_IA32_DS_AREA,
+		.host = (unsigned long)cpuc->ds,
+		.guest = kvm_pmu->ds_area,
+	};
+
 	pebs_enable = (*nr)++;
 	arr[pebs_enable] = (struct perf_guest_switch_msr){
 		.msr = MSR_IA32_PEBS_ENABLE,
arch/x86/include/asm/kvm_host.h +1 -0
index 3bc4c5d79110..c6ccb8aea407 100644
@@ -519,6 +519,7 @@ struct kvm_pmu {
 	DECLARE_BITMAP(all_valid_pmc_idx, X86_PMC_IDX_MAX);
 	DECLARE_BITMAP(pmc_in_use, X86_PMC_IDX_MAX);
 
+	u64 ds_area;
 	u64 pebs_enable;
 	u64 pebs_enable_mask;
 
arch/x86/kvm/vmx/pmu_intel.c +11 -0
index 9f1b25d7b966..b0c3f13e392b 100644
@@ -217,6 +217,9 @@ static bool intel_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 	case MSR_IA32_PEBS_ENABLE:
 		ret = vcpu->arch.perf_capabilities & PERF_CAP_PEBS_FORMAT;
 		break;
+	case MSR_IA32_DS_AREA:
+		ret = guest_cpuid_has(vcpu, X86_FEATURE_DS);
+		break;
 	default:
 		ret = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0) ||
 			get_gp_pmc(pmu, msr, MSR_P6_EVNTSEL0) ||
@@ -367,6 +370,9 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_IA32_PEBS_ENABLE:
 		msr_info->data = pmu->pebs_enable;
 		return 0;
+	case MSR_IA32_DS_AREA:
+		msr_info->data = pmu->ds_area;
+		return 0;
 	default:
 		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
 		    (pmc = get_gp_pmc(pmu, msr, MSR_IA32_PMC0))) {
@@ -434,6 +440,11 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			return 0;
 		}
 		break;
+	case MSR_IA32_DS_AREA:
+		if (is_noncanonical_address(data, vcpu))
+			return 1;
+		pmu->ds_area = data;
+		return 0;
 	default:
 		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
 		    (pmc = get_gp_pmc(pmu, msr, MSR_IA32_PMC0))) {
Signature · 1 line
2.35.1
author
#11 lore raw

PATCH v12 11/17 KVM: x86/pmu: Add PEBS_DATA_CFG MSR emulation to support adaptive PEBS

In reply to Like Xu

From: Like Xu <[email protected]>

If IA32_PERF_CAPABILITIES.PEBS_BASELINE [bit 14] is set, the adaptive PEBS is supported. The PEBS_DATA_CFG MSR and adaptive record enable bits (IA32_PERFEVTSELx.Adaptive_Record and IA32_FIXED_CTR_CTRL. FCx_Adaptive_Record) are also supported.

Adaptive PEBS provides software the capability to configure the PEBS records to capture only the data of interest, keeping the record size compact. An overflow of PMCx results in generation of an adaptive PEBS record with state information based on the selections specified in MSR_PEBS_DATA_CFG.By default, the record only contain the Basic group.

When guest adaptive PEBS is enabled, the IA32_PEBS_ENABLE MSR will be added to the perf_guest_switch_msr() and switched during the VMX transitions just like CORE_PERF_GLOBAL_CTRL MSR.

According to Intel SDM, software is recommended to PEBS Baseline when the following is true. IA32_PERF_CAPABILITIES.PEBS_BASELINE[14] && IA32_PERF_CAPABILITIES.PEBS_FMT[11:8] ≥ 4.

---
 arch/x86/events/intel/core.c    |  8 ++++++++
 arch/x86/include/asm/kvm_host.h |  2 ++
 arch/x86/kvm/vmx/pmu_intel.c    | 20 +++++++++++++++++++-
 3 files changed, 29 insertions(+), 1 deletion(-)
arch/x86/events/intel/core.c +8 -0
index 1e303539f205..6d6a8bac5ae2 100644
@@ -4012,6 +4012,14 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 		.guest = kvm_pmu->ds_area,
 	};
 
+	if (x86_pmu.intel_cap.pebs_baseline) {
+		arr[(*nr)++] = (struct perf_guest_switch_msr){
+			.msr = MSR_PEBS_DATA_CFG,
+			.host = cpuc->pebs_data_cfg,
+			.guest = kvm_pmu->pebs_data_cfg,
+		};
+	}
+
 	pebs_enable = (*nr)++;
 	arr[pebs_enable] = (struct perf_guest_switch_msr){
 		.msr = MSR_IA32_PEBS_ENABLE,
arch/x86/include/asm/kvm_host.h +2 -0
index c6ccb8aea407..549477700b4d 100644
@@ -522,6 +522,8 @@ struct kvm_pmu {
 	u64 ds_area;
 	u64 pebs_enable;
 	u64 pebs_enable_mask;
+	u64 pebs_data_cfg;
+	u64 pebs_data_cfg_mask;
 
 	/*
 	 * The gate to release perf_events not marked in
arch/x86/kvm/vmx/pmu_intel.c +19 -1
index b0c3f13e392b..3cfc90d5c4f4 100644
@@ -205,6 +205,7 @@ static bool intel_pmu_is_valid_lbr_msr(struct kvm_vcpu *vcpu, u32 index)
 static bool intel_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 {
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+	u64 perf_capabilities = vcpu->arch.perf_capabilities;
 	int ret;
 
 	switch (msr) {
@@ -215,11 +216,15 @@ static bool intel_is_valid_msr(struct kvm_vcpu *vcpu, u32 msr)
 		ret = pmu->version > 1;
 		break;
 	case MSR_IA32_PEBS_ENABLE:
-		ret = vcpu->arch.perf_capabilities & PERF_CAP_PEBS_FORMAT;
+		ret = perf_capabilities & PERF_CAP_PEBS_FORMAT;
 		break;
 	case MSR_IA32_DS_AREA:
 		ret = guest_cpuid_has(vcpu, X86_FEATURE_DS);
 		break;
+	case MSR_PEBS_DATA_CFG:
+		ret = (perf_capabilities & PERF_CAP_PEBS_BASELINE) &&
+			((perf_capabilities & PERF_CAP_PEBS_FORMAT) > 3);
+		break;
 	default:
 		ret = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0) ||
 			get_gp_pmc(pmu, msr, MSR_P6_EVNTSEL0) ||
@@ -373,6 +378,9 @@ static int intel_pmu_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_IA32_DS_AREA:
 		msr_info->data = pmu->ds_area;
 		return 0;
+	case MSR_PEBS_DATA_CFG:
+		msr_info->data = pmu->pebs_data_cfg;
+		return 0;
 	default:
 		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
 		    (pmc = get_gp_pmc(pmu, msr, MSR_IA32_PMC0))) {
@@ -445,6 +453,14 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			return 1;
 		pmu->ds_area = data;
 		return 0;
+	case MSR_PEBS_DATA_CFG:
+		if (pmu->pebs_data_cfg == data)
+			return 0;
+		if (!(data & pmu->pebs_data_cfg_mask)) {
+			pmu->pebs_data_cfg = data;
+			return 0;
+		}
+		break;
 	default:
 		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
 		    (pmc = get_gp_pmc(pmu, msr, MSR_IA32_PMC0))) {
@@ -513,6 +529,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	pmu->reserved_bits = 0xffffffff00200000ull;
 	pmu->fixed_ctr_ctrl_mask = ~0ull;
 	pmu->pebs_enable_mask = ~0ull;
+	pmu->pebs_data_cfg_mask = ~0ull;
 
 	entry = kvm_find_cpuid_entry(vcpu, 0xa, 0);
 	if (!entry || !vcpu->kvm->arch.enable_pmu)
@@ -591,6 +608,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 				pmu->fixed_ctr_ctrl_mask &=
 					~(1ULL << (INTEL_PMC_IDX_FIXED + i * 4));
 			}
+			pmu->pebs_data_cfg_mask = ~0xff00000full;
 		} else {
 			pmu->pebs_enable_mask =
 				~((1ull << pmu->nr_arch_gp_counters) - 1);
Signature · 1 line
2.35.1
author
#12 lore raw

PATCH v12 12/17 KVM: x86: Set PEBS_UNAVAIL in IA32_MISC_ENABLE when PEBS is enabled

In reply to Like Xu

From: Like Xu <[email protected]>

The bit 12 represents "Processor Event Based Sampling Unavailable (RO)" :
1 = PEBS is not supported.
0 = PEBS is supported.

A write to this PEBS_UNAVL available bit will bring #GP(0) when guest PEBS is enabled. Some PEBS drivers in guest may care about this bit.

---
 arch/x86/kvm/vmx/pmu_intel.c | 2 ++
 arch/x86/kvm/x86.c           | 8 +++++++-
 2 files changed, 9 insertions(+), 1 deletion(-)
arch/x86/kvm/vmx/pmu_intel.c +2 -0
index 3cfc90d5c4f4..a1fadfa3fbec 100644
@@ -601,6 +601,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		bitmap_set(pmu->all_valid_pmc_idx, INTEL_PMC_IDX_FIXED_VLBR, 1);
 
 	if (vcpu->arch.perf_capabilities & PERF_CAP_PEBS_FORMAT) {
+		vcpu->arch.ia32_misc_enable_msr &= ~MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL;
 		if (vcpu->arch.perf_capabilities & PERF_CAP_PEBS_BASELINE) {
 			pmu->pebs_enable_mask = ~pmu->global_ctrl;
 			pmu->reserved_bits &= ~ICL_EVENTSEL_ADAPTIVE;
@@ -614,6 +615,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 				~((1ull << pmu->nr_arch_gp_counters) - 1);
 		}
 	} else {
+		vcpu->arch.ia32_misc_enable_msr |= MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL;
 		vcpu->arch.perf_capabilities &= ~PERF_CAP_PEBS_MASK;
 	}
 }
arch/x86/kvm/x86.c +7 -1
index 3c7a51468216..41b25412bad4 100644
@@ -3517,7 +3517,13 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		break;
 	case MSR_IA32_MISC_ENABLE: {
 		u64 old_val = vcpu->arch.ia32_misc_enable_msr;
-		u64 pmu_mask = MSR_IA32_MISC_ENABLE_EMON;
+		u64 pmu_mask = MSR_IA32_MISC_ENABLE_EMON |
+			MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL;
+
+		/* RO bits */
+		if (!msr_info->host_initiated &&
+		    ((old_val ^ data) & MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL))
+			return 1;
 
 		/*
 		 * For a dummy user space, the order of setting vPMU capabilities and
Signature · 1 line
2.35.1
author
#13 lore raw

PATCH v12 13/17 KVM: x86/pmu: Move pmc_speculative_in_use() to arch/x86/kvm/pmu.h

In reply to Like Xu

From: Like Xu <[email protected]>

It allows this inline function to be reused by more callers in more files, such as pmu_intel.c.

---
 arch/x86/kvm/pmu.c | 11 -----------
 arch/x86/kvm/pmu.h | 11 +++++++++++
 2 files changed, 11 insertions(+), 11 deletions(-)
arch/x86/kvm/pmu.c +0 -11
index f16dfd7431d1..917d2cbb9ede 100644
@@ -479,17 +479,6 @@ void kvm_pmu_init(struct kvm_vcpu *vcpu)
 	kvm_pmu_refresh(vcpu);
 }
 
-static inline bool pmc_speculative_in_use(struct kvm_pmc *pmc)
-{
-	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
-
-	if (pmc_is_fixed(pmc))
-		return fixed_ctrl_field(pmu->fixed_ctr_ctrl,
-			pmc->idx - INTEL_PMC_IDX_FIXED) & 0x3;
-
-	return pmc->eventsel & ARCH_PERFMON_EVENTSEL_ENABLE;
-}
-
 /* Release perf_events for vPMCs that have been unused for a full time slice.  */
 void kvm_pmu_cleanup(struct kvm_vcpu *vcpu)
 {
arch/x86/kvm/pmu.h +11 -0
index 28c3a826f169..386c8120d4ee 100644
@@ -147,6 +147,17 @@ static inline u64 get_sample_period(struct kvm_pmc *pmc, u64 counter_value)
 	return sample_period;
 }
 
+static inline bool pmc_speculative_in_use(struct kvm_pmc *pmc)
+{
+	struct kvm_pmu *pmu = pmc_to_pmu(pmc);
+
+	if (pmc_is_fixed(pmc))
+		return fixed_ctrl_field(pmu->fixed_ctr_ctrl,
+					pmc->idx - INTEL_PMC_IDX_FIXED) & 0x3;
+
+	return pmc->eventsel & ARCH_PERFMON_EVENTSEL_ENABLE;
+}
+
 void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel);
 void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 ctrl, int fixed_idx);
 void reprogram_counter(struct kvm_pmu *pmu, int pmc_idx);
Signature · 1 line
2.35.1
author
#14 lore raw

PATCH v12 14/17 KVM: x86/pmu: Disable guest PEBS temporarily in two rare situations

In reply to Like Xu

From: Like Xu <[email protected]>

The guest PEBS will be disabled when some users try to perf KVM and its user-space through the same PEBS facility OR when the host perf doesn't schedule the guest PEBS counter in a one-to-one mapping manner (neither of these are typical scenarios).

The PEBS records in the guest DS buffer are still accurate and the above two restrictions will be checked before each vm-entry only if guest PEBS is deemed to be enabled.

---
 arch/x86/events/intel/core.c    | 11 +++++++++--
 arch/x86/include/asm/kvm_host.h |  9 +++++++++
 arch/x86/kvm/vmx/pmu_intel.c    | 20 ++++++++++++++++++++
 arch/x86/kvm/vmx/vmx.c          |  4 ++++
 arch/x86/kvm/vmx/vmx.h          |  1 +
 5 files changed, 43 insertions(+), 2 deletions(-)
arch/x86/events/intel/core.c +9 -2
index 6d6a8bac5ae2..c064339bdfe3 100644
@@ -4027,8 +4027,15 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 		.guest = pebs_mask & ~cpuc->intel_ctrl_host_mask,
 	};
 
-	/* Set hw GLOBAL_CTRL bits for PEBS counter when it runs for guest */
-	arr[0].guest |= arr[*nr].guest;
+	if (arr[pebs_enable].host) {
+		/* Disable guest PEBS if host PEBS is enabled. */
+		arr[pebs_enable].guest = 0;
+	} else {
+		/* Disable guest PEBS for cross-mapped PEBS counters. */
+		arr[pebs_enable].guest &= ~kvm_pmu->host_cross_mapped_mask;
+		/* Set hw GLOBAL_CTRL bits for PEBS counter when it runs for guest */
+		arr[global_ctrl].guest |= arr[pebs_enable].guest;
+	}
 
 	return arr;
 }
arch/x86/include/asm/kvm_host.h +9 -0
index 549477700b4d..ea4e64f4339f 100644
@@ -525,6 +525,15 @@ struct kvm_pmu {
 	u64 pebs_data_cfg;
 	u64 pebs_data_cfg_mask;
 
+	/*
+	 * If a guest counter is cross-mapped to host counter with different
+	 * index, its PEBS capability will be temporarily disabled.
+	 *
+	 * The user should make sure that this mask is updated
+	 * after disabling interrupts and before perf_guest_get_msrs();
+	 */
+	u64 host_cross_mapped_mask;
+
 	/*
 	 * The gate to release perf_events not marked in
 	 * pmc_in_use only once in a vcpu time slice.
arch/x86/kvm/vmx/pmu_intel.c +20 -0
index a1fadfa3fbec..e877e1d0a147 100644
@@ -782,6 +782,26 @@ static void intel_pmu_cleanup(struct kvm_vcpu *vcpu)
 		intel_pmu_release_guest_lbr_event(vcpu);
 }
 
+void intel_pmu_cross_mapped_check(struct kvm_pmu *pmu)
+{
+	struct kvm_pmc *pmc = NULL;
+	int bit;
+
+	for_each_set_bit(bit, (unsigned long *)&pmu->global_ctrl,
+			 X86_PMC_IDX_MAX) {
+		pmc = intel_pmc_idx_to_pmc(pmu, bit);
+
+		if (!pmc || !pmc_speculative_in_use(pmc) ||
+		    !intel_pmc_is_enabled(pmc))
+			continue;
+
+		if (pmc->perf_event && (pmc->idx != pmc->perf_event->hw.idx)) {
+			pmu->host_cross_mapped_mask |=
+				BIT_ULL(pmc->perf_event->hw.idx);
+		}
+	}
+}
+
 struct kvm_pmu_ops intel_pmu_ops = {
 	.pmc_perf_hw_id = intel_pmc_perf_hw_id,
 	.pmc_is_enabled = intel_pmc_is_enabled,
arch/x86/kvm/vmx/vmx.c +4 -0
index 8fb29bbfe875..84635d6950c9 100644
@@ -6712,6 +6712,10 @@ static void atomic_switch_perf_msrs(struct vcpu_vmx *vmx)
 	struct perf_guest_switch_msr *msrs;
 	struct kvm_pmu *pmu = vcpu_to_pmu(&vmx->vcpu);
 
+	pmu->host_cross_mapped_mask = 0;
+	if (pmu->pebs_enable & pmu->global_ctrl)
+		intel_pmu_cross_mapped_check(pmu);
+
 	/* Note, nr_msrs may be garbage if perf_guest_get_msrs() returns NULL. */
 	msrs = perf_guest_get_msrs(&nr_msrs, (void *)pmu);
 	if (!msrs)
arch/x86/kvm/vmx/vmx.h +1 -0
index 7f2c82e7f38f..2abb1cb63616 100644
@@ -94,6 +94,7 @@ union vmx_exit_reason {
 #define vcpu_to_lbr_desc(vcpu) (&to_vmx(vcpu)->lbr_desc)
 #define vcpu_to_lbr_records(vcpu) (&to_vmx(vcpu)->lbr_desc.records)
 
+void intel_pmu_cross_mapped_check(struct kvm_pmu *pmu);
 bool intel_pmu_lbr_is_compatible(struct kvm_vcpu *vcpu);
 bool intel_pmu_lbr_is_enabled(struct kvm_vcpu *vcpu);
 
Signature · 1 line
2.35.1
author
#15 lore raw

PATCH v12 15/17 KVM: x86/pmu: Add kvm_pmu_cap to optimize perf_get_x86_pmu_capability

In reply to Like Xu

From: Like Xu <[email protected]>

The information obtained from the interface perf_get_x86_pmu_capability() doesn't change, so an exported "struct x86_pmu_capability" is introduced for all guests in the KVM, and it's initialized before hardware_setup().

---
 arch/x86/kvm/cpuid.c         | 26 ++++++++------------------
 arch/x86/kvm/pmu.c           |  3 +++
 arch/x86/kvm/pmu.h           | 19 +++++++++++++++++++
 arch/x86/kvm/vmx/pmu_intel.c | 17 ++++++++---------
 arch/x86/kvm/x86.c           |  9 ++++-----
 5 files changed, 42 insertions(+), 32 deletions(-)
arch/x86/kvm/cpuid.c +8 -18
index ff756cdc31ce..896e70f6bf7d 100644
@@ -866,33 +866,23 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 	case 9:
 		break;
 	case 0xa: { /* Architectural Performance Monitoring */
-		struct x86_pmu_capability cap;
 		union cpuid10_eax eax;
 		union cpuid10_edx edx;
 
-		perf_get_x86_pmu_capability(&cap);
+		eax.split.version_id = kvm_pmu_cap.version;
+		eax.split.num_counters = kvm_pmu_cap.num_counters_gp;
+		eax.split.bit_width = kvm_pmu_cap.bit_width_gp;
+		eax.split.mask_length = kvm_pmu_cap.events_mask_len;
+		edx.split.num_counters_fixed = kvm_pmu_cap.num_counters_fixed;
+		edx.split.bit_width_fixed = kvm_pmu_cap.bit_width_fixed;
 
-		/*
-		 * The guest architecture pmu is only supported if the architecture
-		 * pmu exists on the host and the module parameters allow it.
-		 */
-		if (!cap.version || !enable_pmu)
-			memset(&cap, 0, sizeof(cap));
-
-		eax.split.version_id = min(cap.version, 2);
-		eax.split.num_counters = cap.num_counters_gp;
-		eax.split.bit_width = cap.bit_width_gp;
-		eax.split.mask_length = cap.events_mask_len;
-
-		edx.split.num_counters_fixed = min(cap.num_counters_fixed, MAX_FIXED_COUNTERS);
-		edx.split.bit_width_fixed = cap.bit_width_fixed;
-		if (cap.version)
+		if (kvm_pmu_cap.version)
 			edx.split.anythread_deprecated = 1;
 		edx.split.reserved1 = 0;
 		edx.split.reserved2 = 0;
 
 		entry->eax = eax.full;
-		entry->ebx = cap.events_mask;
+		entry->ebx = kvm_pmu_cap.events_mask;
 		entry->ecx = 0;
 		entry->edx = edx.full;
 		break;
arch/x86/kvm/pmu.c +3 -0
index 917d2cbb9ede..883a52ced637 100644
@@ -21,6 +21,9 @@
 #include "lapic.h"
 #include "pmu.h"
 
+struct x86_pmu_capability __read_mostly kvm_pmu_cap;
+EXPORT_SYMBOL_GPL(kvm_pmu_cap);
+
 /* This is enough to filter the vast majority of currently defined events. */
 #define KVM_PMU_EVENT_FILTER_MAX_EVENTS 300
 
arch/x86/kvm/pmu.h +19 -0
index 386c8120d4ee..e717955a5a19 100644
@@ -158,6 +158,24 @@ static inline bool pmc_speculative_in_use(struct kvm_pmc *pmc)
 	return pmc->eventsel & ARCH_PERFMON_EVENTSEL_ENABLE;
 }
 
+extern struct x86_pmu_capability kvm_pmu_cap;
+
+static inline void kvm_init_pmu_capability(void)
+{
+	perf_get_x86_pmu_capability(&kvm_pmu_cap);
+
+	/*
+	 * Only support guest architectural pmu on
+	 * a host with architectural pmu.
+	 */
+	if (!kvm_pmu_cap.version)
+		memset(&kvm_pmu_cap, 0, sizeof(kvm_pmu_cap));
+
+	kvm_pmu_cap.version = min(kvm_pmu_cap.version, 2);
+	kvm_pmu_cap.num_counters_fixed = min(kvm_pmu_cap.num_counters_fixed,
+					     MAX_FIXED_COUNTERS);
+}
+
 void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel);
 void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 ctrl, int fixed_idx);
 void reprogram_counter(struct kvm_pmu *pmu, int pmc_idx);
@@ -176,6 +194,7 @@ void kvm_pmu_cleanup(struct kvm_vcpu *vcpu);
 void kvm_pmu_destroy(struct kvm_vcpu *vcpu);
 int kvm_vm_ioctl_set_pmu_event_filter(struct kvm *kvm, void __user *argp);
 void kvm_pmu_trigger_event(struct kvm_vcpu *vcpu, u64 perf_hw_id);
+void kvm_init_pmu_capability(void);
 
 bool is_vmware_backdoor_pmc(u32 pmc_idx);
 
arch/x86/kvm/vmx/pmu_intel.c +8 -9
index e877e1d0a147..7707ec01fb1f 100644
@@ -514,8 +514,6 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 {
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	struct lbr_desc *lbr_desc = vcpu_to_lbr_desc(vcpu);
-
-	struct x86_pmu_capability x86_pmu;
 	struct kvm_cpuid_entry2 *entry;
 	union cpuid10_eax eax;
 	union cpuid10_edx edx;
@@ -542,13 +540,14 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		return;
 
 	vcpu->arch.ia32_misc_enable_msr |= MSR_IA32_MISC_ENABLE_EMON;
-	perf_get_x86_pmu_capability(&x86_pmu);
 
 	pmu->nr_arch_gp_counters = min_t(int, eax.split.num_counters,
-					 x86_pmu.num_counters_gp);
-	eax.split.bit_width = min_t(int, eax.split.bit_width, x86_pmu.bit_width_gp);
+					 kvm_pmu_cap.num_counters_gp);
+	eax.split.bit_width = min_t(int, eax.split.bit_width,
+				    kvm_pmu_cap.bit_width_gp);
 	pmu->counter_bitmask[KVM_PMC_GP] = ((u64)1 << eax.split.bit_width) - 1;
-	eax.split.mask_length = min_t(int, eax.split.mask_length, x86_pmu.events_mask_len);
+	eax.split.mask_length = min_t(int, eax.split.mask_length,
+				      kvm_pmu_cap.events_mask_len);
 	pmu->available_event_types = ~entry->ebx &
 					((1ull << eax.split.mask_length) - 1);
 
@@ -558,9 +557,9 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		pmu->nr_arch_fixed_counters =
 			min3(ARRAY_SIZE(fixed_pmc_events),
 			     (size_t) edx.split.num_counters_fixed,
-			     (size_t) x86_pmu.num_counters_fixed);
-		edx.split.bit_width_fixed = min_t(int,
-			edx.split.bit_width_fixed, x86_pmu.bit_width_fixed);
+			     (size_t) kvm_pmu_cap.num_counters_fixed);
+		edx.split.bit_width_fixed = min_t(int, edx.split.bit_width_fixed,
+						  kvm_pmu_cap.bit_width_fixed);
 		pmu->counter_bitmask[KVM_PMC_FIXED] =
 			((u64)1 << edx.split.bit_width_fixed) - 1;
 		setup_fixed_pmc_eventsel(pmu);
arch/x86/kvm/x86.c +4 -5
index 41b25412bad4..174fd582e51e 100644
@@ -6569,15 +6569,12 @@ long kvm_arch_vm_ioctl(struct file *filp,
 
 static void kvm_init_msr_list(void)
 {
-	struct x86_pmu_capability x86_pmu;
 	u32 dummy[2];
 	unsigned i;
 
 	BUILD_BUG_ON_MSG(INTEL_PMC_MAX_FIXED != 4,
 			 "Please update the fixed PMCs in msrs_to_saved_all[]");
 
-	perf_get_x86_pmu_capability(&x86_pmu);
-
 	num_msrs_to_save = 0;
 	num_emulated_msrs = 0;
 	num_msr_based_features = 0;
@@ -6629,12 +6626,12 @@ static void kvm_init_msr_list(void)
 			break;
 		case MSR_ARCH_PERFMON_PERFCTR0 ... MSR_ARCH_PERFMON_PERFCTR0 + 17:
 			if (msrs_to_save_all[i] - MSR_ARCH_PERFMON_PERFCTR0 >=
-			    min(INTEL_PMC_MAX_GENERIC, x86_pmu.num_counters_gp))
+			    min(INTEL_PMC_MAX_GENERIC, kvm_pmu_cap.num_counters_gp))
 				continue;
 			break;
 		case MSR_ARCH_PERFMON_EVENTSEL0 ... MSR_ARCH_PERFMON_EVENTSEL0 + 17:
 			if (msrs_to_save_all[i] - MSR_ARCH_PERFMON_EVENTSEL0 >=
-			    min(INTEL_PMC_MAX_GENERIC, x86_pmu.num_counters_gp))
+			    min(INTEL_PMC_MAX_GENERIC, kvm_pmu_cap.num_counters_gp))
 				continue;
 			break;
 		case MSR_IA32_XFD:
@@ -11513,6 +11510,8 @@ int kvm_arch_hardware_setup(void *opaque)
 	if (boot_cpu_has(X86_FEATURE_XSAVES))
 		rdmsrl(MSR_IA32_XSS, host_xss);
 
+	kvm_init_pmu_capability();
+
 	r = ops->hardware_setup();
 	if (r != 0)
 		return r;
Signature · 1 line
2.35.1
author
#16 lore raw

PATCH v12 16/17 KVM: x86/cpuid: Refactor host/guest CPU model consistency check

In reply to Like Xu

From: Like Xu <[email protected]>

For the same purpose, the leagcy intel_pmu_lbr_is_compatible() can be renamed for reuse by more callers, and remove the comment about LBR use case can be deleted by the way.

---
 arch/x86/kvm/cpuid.h         |  5 +++++
 arch/x86/kvm/vmx/pmu_intel.c | 12 +-----------
 arch/x86/kvm/vmx/vmx.c       |  2 +-
 arch/x86/kvm/vmx/vmx.h       |  1 -
 4 files changed, 7 insertions(+), 13 deletions(-)
arch/x86/kvm/cpuid.h +5 -0
index 8a770b481d9d..ac72aabba981 100644
@@ -145,6 +145,11 @@ static inline int guest_cpuid_model(struct kvm_vcpu *vcpu)
 	return x86_model(best->eax);
 }
 
+static inline bool cpuid_model_is_consistent(struct kvm_vcpu *vcpu)
+{
+	return boot_cpu_data.x86_model == guest_cpuid_model(vcpu);
+}
+
 static inline int guest_cpuid_stepping(struct kvm_vcpu *vcpu)
 {
 	struct kvm_cpuid_entry2 *best;
arch/x86/kvm/vmx/pmu_intel.c +1 -11
index 7707ec01fb1f..45148fa4df36 100644
@@ -167,16 +167,6 @@ static inline struct kvm_pmc *get_fw_gp_pmc(struct kvm_pmu *pmu, u32 msr)
 	return get_gp_pmc(pmu, msr, MSR_IA32_PMC0);
 }
 
-bool intel_pmu_lbr_is_compatible(struct kvm_vcpu *vcpu)
-{
-	/*
-	 * As a first step, a guest could only enable LBR feature if its
-	 * cpu model is the same as the host because the LBR registers
-	 * would be pass-through to the guest and they're model specific.
-	 */
-	return boot_cpu_data.x86_model == guest_cpuid_model(vcpu);
-}
-
 bool intel_pmu_lbr_is_enabled(struct kvm_vcpu *vcpu)
 {
 	struct x86_pmu_lbr *lbr = vcpu_to_lbr_records(vcpu);
@@ -591,7 +581,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	nested_vmx_pmu_refresh(vcpu,
 			       intel_is_valid_msr(vcpu, MSR_CORE_PERF_GLOBAL_CTRL));
 
-	if (intel_pmu_lbr_is_compatible(vcpu))
+	if (cpuid_model_is_consistent(vcpu))
 		x86_perf_get_lbr(&lbr_desc->records);
 	else
 		lbr_desc->records.nr = 0;
arch/x86/kvm/vmx/vmx.c +1 -1
index 84635d6950c9..7ae0a82a2a78 100644
@@ -2244,7 +2244,7 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			if ((data & PMU_CAP_LBR_FMT) !=
 			    (vmx_get_perf_capabilities() & PMU_CAP_LBR_FMT))
 				return 1;
-			if (!intel_pmu_lbr_is_compatible(vcpu))
+			if (!cpuid_model_is_consistent(vcpu))
 				return 1;
 		}
 		ret = kvm_set_msr_common(vcpu, msr_info);
arch/x86/kvm/vmx/vmx.h +0 -1
index 2abb1cb63616..3f33883f6fa9 100644
@@ -95,7 +95,6 @@ union vmx_exit_reason {
 #define vcpu_to_lbr_records(vcpu) (&to_vmx(vcpu)->lbr_desc.records)
 
 void intel_pmu_cross_mapped_check(struct kvm_pmu *pmu);
-bool intel_pmu_lbr_is_compatible(struct kvm_vcpu *vcpu);
 bool intel_pmu_lbr_is_enabled(struct kvm_vcpu *vcpu);
 
 int intel_pmu_create_guest_lbr_event(struct kvm_vcpu *vcpu);
Signature · 1 line
2.35.1
author
#17 lore raw

PATCH v12 17/17 KVM: x86/pmu: Expose CPUIDs feature bits PDCM, DS, DTES64

In reply to Like Xu

From: Like Xu <[email protected]>

The CPUID features PDCM, DS and DTES64 are required for PEBS feature. KVM would expose CPUID feature PDCM, DS and DTES64 to guest when PEBS is supported in the KVM on the Ice Lake server platforms.

---
 arch/x86/kvm/vmx/capabilities.h | 28 +++++++++++++++++-----------
 arch/x86/kvm/vmx/vmx.c          | 15 +++++++++++++++
 2 files changed, 32 insertions(+), 11 deletions(-)
arch/x86/kvm/vmx/capabilities.h +17 -11
index 3f430e218375..0e3929ddf9c8 100644
@@ -6,6 +6,7 @@
 
 #include "lapic.h"
 #include "x86.h"
+#include "pmu.h"
 
 extern bool __read_mostly enable_vpid;
 extern bool __read_mostly flexpriority_enabled;
@@ -385,23 +386,28 @@ static inline bool vmx_pt_mode_is_host_guest(void)
 	return pt_mode == PT_MODE_HOST_GUEST;
 }
 
+static inline bool vmx_pebs_supported(void)
+{
+	return boot_cpu_has(X86_FEATURE_PEBS) && kvm_pmu_cap.pebs_ept;
+}
+
 static inline u64 vmx_get_perf_capabilities(void)
 {
-	u64 perf_cap = 0;
-
-	if (!enable_pmu)
-		return perf_cap;
+	u64 perf_cap = PMU_CAP_FW_WRITES;
+	u64 host_perf_cap = 0;
 
 	if (boot_cpu_has(X86_FEATURE_PDCM))
-		rdmsrl(MSR_IA32_PERF_CAPABILITIES, perf_cap);
+		rdmsrl(MSR_IA32_PERF_CAPABILITIES, host_perf_cap);
 
-	perf_cap &= PMU_CAP_LBR_FMT;
+	perf_cap |= host_perf_cap & PMU_CAP_LBR_FMT;
 
-	/*
-	 * Since counters are virtualized, KVM would support full
-	 * width counting unconditionally, even if the host lacks it.
-	 */
-	return PMU_CAP_FW_WRITES | perf_cap;
+	if (vmx_pebs_supported()) {
+		perf_cap |= host_perf_cap & PERF_CAP_PEBS_MASK;
+		if ((perf_cap & PERF_CAP_PEBS_FORMAT) < 4)
+			perf_cap &= ~PERF_CAP_PEBS_BASELINE;
+	}
+
+	return perf_cap;
 }
 
 static inline u64 vmx_supported_debugctl(void)
arch/x86/kvm/vmx/vmx.c +15 -0
index 7ae0a82a2a78..89899d1a865d 100644
@@ -2247,6 +2247,17 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			if (!cpuid_model_is_consistent(vcpu))
 				return 1;
 		}
+		if (data & PERF_CAP_PEBS_FORMAT) {
+			if ((data & PERF_CAP_PEBS_MASK) !=
+			    (vmx_get_perf_capabilities() & PERF_CAP_PEBS_MASK))
+				return 1;
+			if (!guest_cpuid_has(vcpu, X86_FEATURE_DS))
+				return 1;
+			if (!guest_cpuid_has(vcpu, X86_FEATURE_DTES64))
+				return 1;
+			if (!cpuid_model_is_consistent(vcpu))
+				return 1;
+		}
 		ret = kvm_set_msr_common(vcpu, msr_info);
 		break;
 
@@ -7416,6 +7427,10 @@ static __init void vmx_set_cpu_caps(void)
 		kvm_cpu_cap_clear(X86_FEATURE_INVPCID);
 	if (vmx_pt_mode_is_host_guest())
 		kvm_cpu_cap_check_and_set(X86_FEATURE_INTEL_PT);
+	if (vmx_pebs_supported()) {
+		kvm_cpu_cap_check_and_set(X86_FEATURE_DS);
+		kvm_cpu_cap_check_and_set(X86_FEATURE_DTES64);
+	}
 
 	if (!enable_sgx) {
 		kvm_cpu_cap_clear(X86_FEATURE_SGX);
Signature · 1 line
2.35.1
author
#18 lore raw
In reply to Like Xu

Knock knock, do we have any more comments on this patch set, Jim ?

Quoted 159 lines · Like Xu

Hi,

Out of more accurate profiling results, this feature still has loyal followers and another new rebased version is here. PeterZ had acked the V9 patchset [0] and Paolo had asked for a new version, so please check the changelog and feel free to review, test and comment.

[0] https://lore.kernel.org/kvm/[email protected]/
[1] https://lore.kernel.org/kvm/[email protected]/

---

The guest Precise Event Based Sampling (PEBS) feature can provide an architectural state of the instruction executed after the guest instruction that exactly caused the event. It needs new hardware facility only available on Intel Ice Lake Server platforms. This patch set enables the basic PEBS feature for KVM guests on ICX.

We can use PEBS feature on the Linux guest like native:

    # echo 0 > /proc/sys/kernel/watchdog (on the host)
    # perf record -e instructions:ppp ./br_instr a
    # perf record -c 100000 -e instructions:pp ./br_instr a

To emulate guest PEBS facility for the above perf usages, we need to implement 2 code paths:

1) Fast path

This is when the host assigned physical PMC has an identical index as the virtual PMC (e.g. using physical PMC0 to emulate virtual PMC0). This path is used in most common use cases.

2) Slow path

This is when the host assigned physical PMC has a different index from the virtual PMC (e.g. using physical PMC1 to emulate virtual PMC0) In this case, KVM needs to rewrite the PEBS records to change the applicable counter indexes to the virtual PMC indexes, which would otherwise contain the physical counter index written by PEBS facility, and switch the counter reset values to the offset corresponding to the physical counter indexes in the DS data structure.

The previous version [3] enables both fast path and slow path, which seems a bit more complex as the first step. In this patchset, we want to start with the fast path to get the basic guest PEBS enabled while keeping the slow path disabled. More focused discussion on the slow path [4] is planned to be put to another patchset in the next step.

Compared to later versions in subsequent steps, the functionality to support host-guest PEBS both enabled and the functionality to emulate guest PEBS when the counter is cross-mapped are missing in this patch set (neither of these are typical scenarios).

With the basic support, the guest can retrieve the correct PEBS information from its own PEBS records on the Ice Lake servers. And we expect it should work when migrating to another Ice Lake and no regression about host perf is expected.

Here are the results of pebs test from guest/host for same workload:

perf report on guest:
# Samples: 2K of event 'instructions:ppp', # Event count (approx.): 1473377250 # Overhead Command Shared Object Symbol

    57.74%  br_instr  br_instr           [.] lfsr_cond
    41.40%  br_instr  br_instr           [.] cmp_end
     0.21%  br_instr  [kernel.kallsyms]  [k] __lock_acquire

perf report on host:
# Samples: 2K of event 'instructions:ppp', # Event count (approx.): 1462721386 # Overhead Command Shared Object Symbol

    57.90%  br_instr  br_instr          [.] lfsr_cond
    41.95%  br_instr  br_instr          [.] cmp_end
     0.05%  br_instr  [kernel.vmlinux]  [k] lock_acquire

Conclusion: the profiling results on the guest are similar tothat on the host.

A minimum guest kernel version may be v5.4 or a backport version support Icelake server PEBS.

Please check more details in each commit and feel free to comment.

Previous:
https://lore.kernel.org/kvm/[email protected]/

[3]
https://lore.kernel.org/kvm/[email protected]/
[4]
https://lore.kernel.org/kvm/[email protected]/

V11->V12:
- Apply the new perf interface from tip/perf/core and fix the merge conflict;
- Rename "x86_pmu.pebs_ept" to "x86_pmu.pebs_ept"; (Sean)
- Rebased on the top of kvm/queue (b13a3befc815); (Paolo)

V10->V11:
- Merge perf_guest_info_callbacks static_call to the tip/perf/core;
- Keep use perf_guest_cbs in the kvm/queue context before merge window;
- Fix MSR_IA32_MISC_ENABLE_EMON bit (Liu XiangDong);
- Rebase "Reprogram PEBS event to emulate guest PEBS counter" patch;

V9->V10:
- improve readability in core.c(Peter Z)
- reuse guest_pebs_idxs(Liu XiangDong)

V8 -> V9 Changelog:
-fix a brackets error in xen_guest_state()

V7 -> V8 Changelog:
- fix coding style, add {} for single statement of multiple lines(Peter Z)
- fix coding style in xen_guest_state() (Boris Ostrovsky)
- s/pmu/kvm_pmu/ in intel_guest_get_msrs() (Peter Z)
- put lower cost branch in the first place for x86_pmu_handle_guest_pebs() (Peter Z)

V6 -> V7 Changelog:
- Fix conditions order and call x86_pmu_handle_guest_pebs() unconditionally; (PeterZ)
- Add a new patch to make all that perf_guest_cbs stuff suck less; (PeterZ)
- Document IA32_MISC_ENABLE[7] that that behavior matches bare metal; (Sean & Venkatesh)
- Update commit message for fixed counter mask refactoring;(PeterZ)
- Clarifying comments about {.host and .guest} for intel_guest_get_msrs(); (PeterZ)
- Add pebs_capable to store valid PEBS_COUNTER_MASK value; (PeterZ)
- Add more comments for perf's precise_ip field; (Andi & PeterZ)
- Refactor perf_overflow_handler_t and make it more legible; (PeterZ)
- Use "(unsigned long)cpuc->ds" instead of __this_cpu_read(cpu_hw_events.ds); (PeterZ)
- Keep using "(struct kvm_pmu *)data" to follow K&R; (Andi)

Like Xu (16):
perf/x86/intel: Add EPT-Friendly PEBS for Ice Lake Server perf/x86/intel: Handle guest PEBS overflow PMI for KVM guest perf/x86/core: Pass "struct kvm_pmu *" to determine the guest values
KVM: x86/pmu: Set MSR_IA32_MISC_ENABLE_EMON bit when vPMU is enabled
KVM: x86/pmu: Introduce the ctrl_mask value for fixed counter
KVM: x86/pmu: Add IA32_PEBS_ENABLE MSR emulation for extended PEBS
KVM: x86/pmu: Reprogram PEBS event to emulate guest PEBS counter
KVM: x86/pmu: Adjust precise_ip to emulate Ice Lake guest PDIR counter
KVM: x86/pmu: Add IA32_DS_AREA MSR emulation to support guest DS
KVM: x86/pmu: Add PEBS_DATA_CFG MSR emulation to support adaptive PEBS
KVM: x86: Set PEBS_UNAVAIL in IA32_MISC_ENABLE when PEBS is enabled
KVM: x86/pmu: Move pmc_speculative_in_use() to arch/x86/kvm/pmu.h
KVM: x86/pmu: Disable guest PEBS temporarily in two rare situations
KVM: x86/pmu: Add kvm_pmu_cap to optimize perf_get_x86_pmu_capability
KVM: x86/cpuid: Refactor host/guest CPU model consistency check
KVM: x86/pmu: Expose CPUIDs feature bits PDCM, DS, DTES64

Peter Zijlstra (Intel) (1):
x86/perf/core: Add pebs_capable to store valid PEBS_COUNTER_MASK value

  arch/x86/events/core.c            |   5 +-
  arch/x86/events/intel/core.c      | 157 +++++++++++++++++++++++++-----
  arch/x86/events/perf_event.h      |   6 +-
  arch/x86/include/asm/kvm_host.h   |  16 +++
  arch/x86/include/asm/msr-index.h  |   6 ++
  arch/x86/include/asm/perf_event.h |   5 +-
  arch/x86/kvm/cpuid.c              |  26 ++---
  arch/x86/kvm/cpuid.h              |   5 +
  arch/x86/kvm/pmu.c                |  52 +++++++---
  arch/x86/kvm/pmu.h                |  37 +++++++
  arch/x86/kvm/vmx/capabilities.h   |  28 +++---
  arch/x86/kvm/vmx/pmu_intel.c      | 118 ++++++++++++++++++----
  arch/x86/kvm/vmx/vmx.c            |  24 ++++-
  arch/x86/kvm/vmx/vmx.h            |   2 +-
  arch/x86/kvm/x86.c                |  30 ++++--
  15 files changed, 410 insertions(+), 107 deletions(-)