Skip to contentSkip to conversation

Add perf support to collect KVM guest statistics from host side

PATCH Superseded [email protected] 2 participants 7 messages Jul 18, 2024 – Jul 19, 2024

Subsystems (4)

v1 of 5 Superseded Latest is v5 →

lore
Share & tools
lore
#0 lore raw

cover Add perf support to collect KVM guest statistics from host side

From: Quan Zhou <[email protected]>

Add basic guest support to RISC-V perf, enabling it to distinguish whether PMU interrupts occur in the host or the guest, and then collect some basic guest information from the host side (guest os callchain is not supported for now).

Based on the x86/arm implementation, tested with kvm-riscv. test env:
- host: qemu-9.0.0
- guest: qemu-9.0.0 --enable-kvm (only start one guest and run top)

-----------------------------------------
1) perf kvm top
./perf kvm --host --guest \
--guestkallsyms=/root/repo/shared/kallsyms \
--guestmodules=/root/repo/shared/modules top

PerfTop: 41 irqs/sec kernel:97.6% us: 0.0% guest kernel: 0.0% guest us: 0.0% exact: 0.0% [250Hz cycles:P], (all, 4 CPUs)
-------------------------------------------------------------------------------

    64.57%  [kernel]        [k] default_idle_call
     3.12%  [kernel]        [k] _raw_spin_unlock_irqrestore
     3.03%  [guest.kernel]  [g] mem_serial_out
     2.61%  [kernel]        [k] handle_softirqs
     2.32%  [kernel]        [k] do_trap_ecall_u
     1.71%  [kernel]        [k] _raw_spin_unlock_irq
     1.26%  [guest.kernel]  [g] do_raw_spin_lock
     1.25%  [kernel]        [k] finish_task_switch.isra.0
     1.16%  [kernel]        [k] do_idle
     0.77%  libc.so.6       [.] ioctl
     0.76%  [kernel]        [k] queue_work_on
     0.69%  [kernel]        [k] __local_bh_enable_ip
     0.67%  [guest.kernel]  [g] __noinstr_text_start
     0.64%  [guest.kernel]  [g] mem_serial_in
     0.41%  libc.so.6       [.] pthread_sigmask
     0.39%  [kernel]        [k] mem_cgroup_uncharge_skmem
     0.39%  [kernel]        [k] __might_resched
     0.39%  [guest.kernel]  [g] _nohz_idle_balance.isra.0
     0.37%  [kernel]        [k] sched_balance_update_blocked_averages
     0.34%  [kernel]        [k] sched_balance_rq

2) perf kvm record
./perf kvm --host --guest \
--guestkallsyms=/root/repo/shared/kallsyms \
--guestmodules=/root/repo/shared/modules record -a sleep 60

[ perf record: Woken up 3 times to write data ]
[ perf record: Captured and wrote 1.292 MB perf.data.kvm (17990 samples) ]

3) perf kvm report (the data shown here is not complete) ./perf kvm --host --guest \
--guestkallsyms=/root/repo/shared/kallsyms \
--guestmodules=/root/repo/shared/modules report -i perf.data.kvm

# Total Lost Samples: 0
#
# Samples: 17K of event 'cycles:P'
# Event count (approx.): 269968947184
#

# Overhead  Command          Shared Object            Symbol                                        
# ........  ...............  .......................  ..............................................
#
    61.86%  swapper          [kernel.kallsyms]        [k] default_idle_call
     2.93%  :6463            [guest.kernel.kallsyms]  [g] do_raw_spin_lock
     2.82%  :6462            [guest.kernel.kallsyms]  [g] mem_serial_out
     2.11%  sshd             [kernel.kallsyms]        [k] _raw_spin_unlock_irqrestore
     1.78%  :6462            [guest.kernel.kallsyms]  [g] do_raw_spin_lock
     1.37%  swapper          [kernel.kallsyms]        [k] handle_softirqs
     1.36%  swapper          [kernel.kallsyms]        [k] do_idle
     1.21%  sshd             [kernel.kallsyms]        [k] do_trap_ecall_u
     1.21%  sshd             [kernel.kallsyms]        [k] _raw_spin_unlock_irq
     1.11%  qemu-system-ris  [kernel.kallsyms]        [k] do_trap_ecall_u
     0.93%  qemu-system-ris  libc.so.6                [.] ioctl
     0.89%  sshd             [kernel.kallsyms]        [k] __local_bh_enable_ip
     0.77%  qemu-system-ris  [kernel.kallsyms]        [k] _raw_spin_unlock_irqrestore
     0.68%  qemu-system-ris  [kernel.kallsyms]        [k] queue_work_on
     0.65%  sshd             [kernel.kallsyms]        [k] handle_softirqs
     0.44%  :6462            [guest.kernel.kallsyms]  [g] mem_serial_in
     0.42%  sshd             libc.so.6                [.] pthread_sigmask
     0.34%  :6462            [guest.kernel.kallsyms]  [g] serial8250_tx_chars
     0.30%  swapper          [kernel.kallsyms]        [k] finish_task_switch.isra.0
     0.29%  swapper          [kernel.kallsyms]        [k] sched_balance_rq
     0.29%  sshd             [kernel.kallsyms]        [k] __might_resched
     0.26%  swapper          [kernel.kallsyms]        [k] tick_nohz_idle_exit
     0.26%  swapper          [kernel.kallsyms]        [k] sched_balance_update_blocked_averages
     0.26%  swapper          [kernel.kallsyms]        [k] _nohz_idle_balance.isra.0
     0.24%  qemu-system-ris  [kernel.kallsyms]        [k] finish_task_switch.isra.0
     0.23%  :6462            [guest.kernel.kallsyms]  [g] __noinstr_text_start

Quan Zhou (2):
riscv: perf: add guest vs host distinction
riscv: KVM: add basic support for host vs guest profiling

 arch/riscv/include/asm/kvm_host.h   |  6 +++++
 arch/riscv/include/asm/perf_event.h |  7 ++++++
 arch/riscv/kernel/perf_callchain.c  | 38 +++++++++++++++++++++++++++++
 arch/riscv/kvm/Kconfig              |  1 +
 arch/riscv/kvm/main.c               | 12 +++++++--
 arch/riscv/kvm/vcpu.c               |  7 ++++++
 6 files changed, 69 insertions(+), 2 deletions(-)

base-commit: 0c3836482481200ead7b416ca80c68a29cfdaabd

Signature · 1 line
2.34.1
#1 lore raw

PATCH 1/2 riscv: perf: add guest vs host distinction

From: Quan Zhou <[email protected]>

Introduce basic guest support in perf, enabling it to distinguish between PMU interrupts in the host or guest, and collect fundamental information.

---
 arch/riscv/include/asm/perf_event.h |  7 ++++++
 arch/riscv/kernel/perf_callchain.c  | 38 +++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+)
arch/riscv/include/asm/perf_event.h +7 -0
index 665bbc9b2f84..5866d028aee5 100644
@@ -8,13 +8,20 @@
 #ifndef _ASM_RISCV_PERF_EVENT_H
 #define _ASM_RISCV_PERF_EVENT_H
 
+#ifdef CONFIG_PERF_EVENTS
 #include <linux/perf_event.h>
 #define perf_arch_bpf_user_pt_regs(regs) (struct user_regs_struct *)regs
 
+extern unsigned long perf_instruction_pointer(struct pt_regs *regs);
+extern unsigned long perf_misc_flags(struct pt_regs *regs);
+#define perf_misc_flags(regs) perf_misc_flags(regs)
+
 #define perf_arch_fetch_caller_regs(regs, __ip) { \
 	(regs)->epc = (__ip); \
 	(regs)->s0 = (unsigned long) __builtin_frame_address(0); \
 	(regs)->sp = current_stack_pointer; \
 	(regs)->status = SR_PP; \
 }
+#endif
+
 #endif /* _ASM_RISCV_PERF_EVENT_H */
arch/riscv/kernel/perf_callchain.c +38 -0
index 3348a61de7d9..c673dc6d9bd2 100644
@@ -58,6 +58,11 @@ void perf_callchain_user(struct perf_callchain_entry_ctx *entry,
 {
 	unsigned long fp = 0;
 
+	if (perf_guest_state()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
 	fp = regs->s0;
 	perf_callchain_store(entry, regs->epc);
 
@@ -74,5 +79,38 @@ static bool fill_callchain(void *entry, unsigned long pc)
 void perf_callchain_kernel(struct perf_callchain_entry_ctx *entry,
 			   struct pt_regs *regs)
 {
+	if (perf_guest_state()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
 	walk_stackframe(NULL, regs, fill_callchain, entry);
 }
+
+unsigned long perf_instruction_pointer(struct pt_regs *regs)
+{
+	if (perf_guest_state())
+		return perf_guest_get_ip();
+
+	return instruction_pointer(regs);
+}
+
+unsigned long perf_misc_flags(struct pt_regs *regs)
+{
+	unsigned int guest_state = perf_guest_state();
+	int misc = 0;
+
+	if (guest_state) {
+		if (guest_state & PERF_GUEST_USER)
+			misc |= PERF_RECORD_MISC_GUEST_USER;
+		else
+			misc |= PERF_RECORD_MISC_GUEST_KERNEL;
+	} else {
+		if (user_mode(regs))
+			misc |= PERF_RECORD_MISC_USER;
+		else
+			misc |= PERF_RECORD_MISC_KERNEL;
+	}
+
+	return misc;
+}
Signature · 1 line
2.34.1
#2 lore raw

Re: [PATCH 1/2] riscv: perf: add guest vs host distinction

Quoted 19 lines · zhouquan at iscas.ac.cn

From: Quan Zhou <[email protected]>

Introduce basic guest support in perf, enabling it to distinguish between PMU interrupts in the host or guest, and collect fundamental information.

---
 arch/riscv/include/asm/perf_event.h |  7 ++++++
 arch/riscv/kernel/perf_callchain.c  | 38 +++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+)
arch/riscv/include/asm/perf_event.h +0 -0
index 665bbc9b2f84..5866d028aee5 100644
@@ -8,13 +8,20 @@
 #ifndef _ASM_RISCV_PERF_EVENT_H
 #define _ASM_RISCV_PERF_EVENT_H

arch/riscv/include/asm/perf_event.h line 19

Quoted 10 lines
arch/riscv/include/asm/perf_event.h +5 -0
 
+#ifdef CONFIG_PERF_EVENTS
 #include <linux/perf_event.h>
 #define perf_arch_bpf_user_pt_regs(regs) (struct user_regs_struct *)regs
 
+extern unsigned long perf_instruction_pointer(struct pt_regs *regs);
+extern unsigned long perf_misc_flags(struct pt_regs *regs);
+#define perf_misc_flags(regs) perf_misc_flags(regs)
+
 #define perf_arch_fetch_caller_regs(regs, __ip) { \

Arm has this outside the #ifdef CONFIG_PERF_EVENTS, but it doesn't look like it should be.

Quoted 38 lines
arch/riscv/include/asm/perf_event.h +2 -0
 	(regs)->epc = (__ip); \
 	(regs)->s0 = (unsigned long) __builtin_frame_address(0); \
 	(regs)->sp = current_stack_pointer; \
 	(regs)->status = SR_PP; \
 }
+#endif
+
 #endif /* _ASM_RISCV_PERF_EVENT_H */
arch/riscv/kernel/perf_callchain.c +13 -0
index 3348a61de7d9..c673dc6d9bd2 100644
@@ -58,6 +58,11 @@ void perf_callchain_user(struct perf_callchain_entry_ctx *entry,
 {
 	unsigned long fp = 0;
 
+	if (perf_guest_state()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
 	fp = regs->s0;
 	perf_callchain_store(entry, regs->epc);
 
@@ -74,5 +79,38 @@ static bool fill_callchain(void *entry, unsigned long pc)
 void perf_callchain_kernel(struct perf_callchain_entry_ctx *entry,
 			   struct pt_regs *regs)
 {
+	if (perf_guest_state()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
 	walk_stackframe(NULL, regs, fill_callchain, entry);
 }
+
+unsigned long perf_instruction_pointer(struct pt_regs *regs)
+{

arch/riscv/kernel/perf_callchain.c static bool fill_callchain(void *entry, unsigned long pc) · line 101

Quoted 10 lines
arch/riscv/kernel/perf_callchain.c +10 -0
+	if (perf_guest_state())
+		return perf_guest_get_ip();
+
+	return instruction_pointer(regs);
+}
+
+unsigned long perf_misc_flags(struct pt_regs *regs)
+{
+	unsigned int guest_state = perf_guest_state();
+	int misc = 0;

Should use unsigned long for misc.

Quoted 18 lines
arch/riscv/kernel/perf_callchain.c +15 -0
+
+	if (guest_state) {
+		if (guest_state & PERF_GUEST_USER)
+			misc |= PERF_RECORD_MISC_GUEST_USER;
+		else
+			misc |= PERF_RECORD_MISC_GUEST_KERNEL;
+	} else {
+		if (user_mode(regs))
+			misc |= PERF_RECORD_MISC_USER;
+		else
+			misc |= PERF_RECORD_MISC_KERNEL;
+	}
+
+	return misc;
+}
Signature · 1 line
2.34.1

Thanks,
drew

author
#3 lore raw

Re: [PATCH 1/2] riscv: perf: add guest vs host distinction

Quoted 25 lines · Andrew Jones
Quoted 24 lines · zhouquan at iscas.ac.cn

From: Quan Zhou <[email protected]>

Introduce basic guest support in perf, enabling it to distinguish between PMU interrupts in the host or guest, and collect fundamental information.

---
  arch/riscv/include/asm/perf_event.h |  7 ++++++
  arch/riscv/kernel/perf_callchain.c  | 38 +++++++++++++++++++++++++++++
  2 files changed, 45 insertions(+)
arch/riscv/include/asm/perf_event.h +1 -0
index 665bbc9b2f84..5866d028aee5 100644
@@ -8,13 +8,20 @@
  #ifndef _ASM_RISCV_PERF_EVENT_H
  #define _ASM_RISCV_PERF_EVENT_H
 
+#ifdef CONFIG_PERF_EVENTS
  #include <linux/perf_event.h>
  #define perf_arch_bpf_user_pt_regs(regs) (struct user_regs_struct *)regs
 
Quoted 9 lines

arch/riscv/include/asm/perf_event.h line 19

Quoted 5 lines
arch/riscv/include/asm/perf_event.h +4 -0
+extern unsigned long perf_instruction_pointer(struct pt_regs *regs);
+extern unsigned long perf_misc_flags(struct pt_regs *regs);
+#define perf_misc_flags(regs) perf_misc_flags(regs)
+
  #define perf_arch_fetch_caller_regs(regs, __ip) { \

Arm has this outside the #ifdef CONFIG_PERF_EVENTS, but it doesn't look like it should be.

Yes, Arm makes perf_arch_fetch_caller_regs independent of CONFIG_PERF_EVENTS. What is the rationale behind this? I'm not clear on this point. It's reasonable to have them inside for riscv, right?

Quoted 42 lines
Quoted 42 lines
arch/riscv/include/asm/perf_event.h +2 -0
  	(regs)->epc = (__ip); \
  	(regs)->s0 = (unsigned long) __builtin_frame_address(0); \
  	(regs)->sp = current_stack_pointer; \
  	(regs)->status = SR_PP; \
  }
+#endif
+
  #endif /* _ASM_RISCV_PERF_EVENT_H */
arch/riscv/kernel/perf_callchain.c +17 -0
index 3348a61de7d9..c673dc6d9bd2 100644
@@ -58,6 +58,11 @@ void perf_callchain_user(struct perf_callchain_entry_ctx *entry,
  {
  	unsigned long fp = 0;
 
+	if (perf_guest_state()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
  	fp = regs->s0;
  	perf_callchain_store(entry, regs->epc);
 
@@ -74,5 +79,38 @@ static bool fill_callchain(void *entry, unsigned long pc)
  void perf_callchain_kernel(struct perf_callchain_entry_ctx *entry,
  			   struct pt_regs *regs)
  {
+	if (perf_guest_state()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
  	walk_stackframe(NULL, regs, fill_callchain, entry);
  }
+
+unsigned long perf_instruction_pointer(struct pt_regs *regs)
+{
+	if (perf_guest_state())
+		return perf_guest_get_ip();
+
+	return instruction_pointer(regs);
Quoted 9 lines

arch/riscv/kernel/perf_callchain.c static bool fill_callchain(void *entry, unsigned long pc) · line 101

Quoted 6 lines
arch/riscv/kernel/perf_callchain.c +6 -0
+}
+
+unsigned long perf_misc_flags(struct pt_regs *regs)
+{
+	unsigned int guest_state = perf_guest_state();
+	int misc = 0;

Should use unsigned long for misc.

Okay, I'll fix it.

Thanks,
Quan

Quoted 21 lines
Quoted 18 lines
arch/riscv/kernel/perf_callchain.c +15 -0
+
+	if (guest_state) {
+		if (guest_state & PERF_GUEST_USER)
+			misc |= PERF_RECORD_MISC_GUEST_USER;
+		else
+			misc |= PERF_RECORD_MISC_GUEST_KERNEL;
+	} else {
+		if (user_mode(regs))
+			misc |= PERF_RECORD_MISC_USER;
+		else
+			misc |= PERF_RECORD_MISC_KERNEL;
+	}
+
+	return misc;
+}
Signature · 1 line
2.34.1

Thanks,
drew

#4 lore raw

PATCH 2/2 riscv: KVM: add basic support for host vs guest profiling

In reply to [email protected]

From: Quan Zhou <[email protected]>

For the information collected on the host side, we need to identify which data originates from the guest and record these events separately. This can be achieved by having KVM register perf callbacks.

---
 arch/riscv/include/asm/kvm_host.h |  6 ++++++
 arch/riscv/kvm/Kconfig            |  1 +
 arch/riscv/kvm/main.c             | 12 ++++++++++--
 arch/riscv/kvm/vcpu.c             |  7 +++++++
 4 files changed, 24 insertions(+), 2 deletions(-)
arch/riscv/include/asm/kvm_host.h +6 -0
index d96281278586..b7bbe1c0c5dd 100644
@@ -285,6 +285,12 @@ struct kvm_vcpu_arch {
 	} sta;
 };
 
+/* TODO: A more explicit approach might be needed here than this simple one */
+static inline bool kvm_arch_pmi_in_guest(struct kvm_vcpu *vcpu)
+{
+	return IS_ENABLED(CONFIG_GUEST_PERF_EVENTS) && !!vcpu;
+}
+
 static inline void kvm_arch_sync_events(struct kvm *kvm) {}
 static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
 
arch/riscv/kvm/Kconfig +1 -0
index 26d1727f0550..0c3cbb0915ff 100644
@@ -32,6 +32,7 @@ config KVM
 	select KVM_XFER_TO_GUEST_WORK
 	select KVM_GENERIC_MMU_NOTIFIER
 	select SCHED_INFO
+	select GUEST_PERF_EVENTS if PERF_EVENTS
 	help
 	  Support hosting virtualized guest machines.
 
arch/riscv/kvm/main.c +10 -2
index bab2ec34cd87..734b48d8f6dd 100644
@@ -51,6 +51,12 @@ void kvm_arch_hardware_disable(void)
 	csr_write(CSR_HIDELEG, 0);
 }
 
+static void kvm_riscv_teardown(void)
+{
+	kvm_riscv_aia_exit();
+	kvm_unregister_perf_callbacks();
+}
+
 static int __init riscv_kvm_init(void)
 {
 	int rc;
@@ -105,9 +111,11 @@ static int __init riscv_kvm_init(void)
 		kvm_info("AIA available with %d guest external interrupts\n",
 			 kvm_riscv_aia_nr_hgei);
 
+	kvm_register_perf_callbacks(NULL);
+
 	rc = kvm_init(sizeof(struct kvm_vcpu), 0, THIS_MODULE);
 	if (rc) {
-		kvm_riscv_aia_exit();
+		kvm_riscv_teardown();
 		return rc;
 	}
 
@@ -117,7 +125,7 @@ module_init(riscv_kvm_init);
 
 static void __exit riscv_kvm_exit(void)
 {
-	kvm_riscv_aia_exit();
+	kvm_riscv_teardown();
 
 	kvm_exit();
 }
arch/riscv/kvm/vcpu.c +7 -0
index 17e21df36cc1..c9d291865141 100644
@@ -222,6 +222,13 @@ bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
 	return (vcpu->arch.guest_context.sstatus & SR_SPP) ? true : false;
 }
 
+#ifdef CONFIG_GUEST_PERF_EVENTS
+unsigned long kvm_arch_vcpu_get_ip(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.guest_context.sepc;
+}
+#endif
+
 vm_fault_t kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 {
 	return VM_FAULT_SIGBUS;
Signature · 1 line
2.34.1
Reviewed-by on v1 2/2
#5 lore raw

Re: [PATCH 2/2] riscv: KVM: add basic support for host vs guest profiling

Quoted 20 lines · zhouquan at iscas.ac.cn

From: Quan Zhou <[email protected]>

For the information collected on the host side, we need to identify which data originates from the guest and record these events separately. This can be achieved by having KVM register perf callbacks.

---
 arch/riscv/include/asm/kvm_host.h |  6 ++++++
 arch/riscv/kvm/Kconfig            |  1 +
 arch/riscv/kvm/main.c             | 12 ++++++++++--
 arch/riscv/kvm/vcpu.c             |  7 +++++++
 4 files changed, 24 insertions(+), 2 deletions(-)
arch/riscv/include/asm/kvm_host.h +0 -0
index d96281278586..b7bbe1c0c5dd 100644

arch/riscv/include/asm/kvm_host.h struct kvm_vcpu_arch { · line 288

Quoted 4 lines
arch/riscv/include/asm/kvm_host.h +1 -0
@@ -285,6 +285,12 @@ struct kvm_vcpu_arch {
 	} sta;
 };
 
+/* TODO: A more explicit approach might be needed here than this simple one */

Can you elaborate on this concern?

Quoted 80 lines
arch/riscv/include/asm/kvm_host.h +5 -0
+static inline bool kvm_arch_pmi_in_guest(struct kvm_vcpu *vcpu)
+{
+	return IS_ENABLED(CONFIG_GUEST_PERF_EVENTS) && !!vcpu;
+}
+
 static inline void kvm_arch_sync_events(struct kvm *kvm) {}
 static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
 
arch/riscv/kvm/Kconfig +1 -0
index 26d1727f0550..0c3cbb0915ff 100644
@@ -32,6 +32,7 @@ config KVM
 	select KVM_XFER_TO_GUEST_WORK
 	select KVM_GENERIC_MMU_NOTIFIER
 	select SCHED_INFO
+	select GUEST_PERF_EVENTS if PERF_EVENTS
 	help
 	  Support hosting virtualized guest machines.
 
arch/riscv/kvm/main.c +10 -2
index bab2ec34cd87..734b48d8f6dd 100644
@@ -51,6 +51,12 @@ void kvm_arch_hardware_disable(void)
 	csr_write(CSR_HIDELEG, 0);
 }
 
+static void kvm_riscv_teardown(void)
+{
+	kvm_riscv_aia_exit();
+	kvm_unregister_perf_callbacks();
+}
+
 static int __init riscv_kvm_init(void)
 {
 	int rc;
@@ -105,9 +111,11 @@ static int __init riscv_kvm_init(void)
 		kvm_info("AIA available with %d guest external interrupts\n",
 			 kvm_riscv_aia_nr_hgei);
 
+	kvm_register_perf_callbacks(NULL);
+
 	rc = kvm_init(sizeof(struct kvm_vcpu), 0, THIS_MODULE);
 	if (rc) {
-		kvm_riscv_aia_exit();
+		kvm_riscv_teardown();
 		return rc;
 	}
 
@@ -117,7 +125,7 @@ module_init(riscv_kvm_init);
 
 static void __exit riscv_kvm_exit(void)
 {
-	kvm_riscv_aia_exit();
+	kvm_riscv_teardown();
 
 	kvm_exit();
 }
arch/riscv/kvm/vcpu.c +7 -0
index 17e21df36cc1..c9d291865141 100644
@@ -222,6 +222,13 @@ bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
 	return (vcpu->arch.guest_context.sstatus & SR_SPP) ? true : false;
 }
 
+#ifdef CONFIG_GUEST_PERF_EVENTS
+unsigned long kvm_arch_vcpu_get_ip(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.guest_context.sepc;
+}
+#endif
+
 vm_fault_t kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 {
 	return VM_FAULT_SIGBUS;
Signature · 1 line
2.34.1

Otherwise,

author
#6 lore raw

Re: [PATCH 2/2] riscv: KVM: add basic support for host vs guest profiling

Quoted 24 lines · Andrew Jones
Quoted 23 lines · zhouquan at iscas.ac.cn

From: Quan Zhou <[email protected]>

For the information collected on the host side, we need to identify which data originates from the guest and record these events separately. This can be achieved by having KVM register perf callbacks.

---
  arch/riscv/include/asm/kvm_host.h |  6 ++++++
  arch/riscv/kvm/Kconfig            |  1 +
  arch/riscv/kvm/main.c             | 12 ++++++++++--
  arch/riscv/kvm/vcpu.c             |  7 +++++++
  4 files changed, 24 insertions(+), 2 deletions(-)
arch/riscv/include/asm/kvm_host.h +0 -0
index d96281278586..b7bbe1c0c5dd 100644
@@ -285,6 +285,12 @@ struct kvm_vcpu_arch {
  	} sta;
  };
 
Quoted 4 lines

arch/riscv/include/asm/kvm_host.h struct kvm_vcpu_arch { · line 288

Quoted 1 line
arch/riscv/include/asm/kvm_host.h +1 -0
+/* TODO: A more explicit approach might be needed here than this simple one */

Can you elaborate on this concern?

Can we determine if a PMU interrupt occurs in the guest solely based on whether the vcpu is loaded?

In the kvm_arch_vcpu_ioctl_run while loop, when the guest exits, host-side interrupts should be handled in local_irq_{enable/disable}. Should we add a flag here (referencing x86) to further determine if "this is a host interrupt triggered while the guest was running, rather than when the host is running". Doing so might further reduce the likelihood of perf incorrectly identifying "PMI in host" as "PMI in guest," but it may not completely eliminate it.

Have I misunderstood any points here?

Thanks,
Quan

Quoted 84 lines
Quoted 80 lines
arch/riscv/include/asm/kvm_host.h +5 -0
+static inline bool kvm_arch_pmi_in_guest(struct kvm_vcpu *vcpu)
+{
+	return IS_ENABLED(CONFIG_GUEST_PERF_EVENTS) && !!vcpu;
+}
+
  static inline void kvm_arch_sync_events(struct kvm *kvm) {}
  static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {}
 
arch/riscv/kvm/Kconfig +1 -0
index 26d1727f0550..0c3cbb0915ff 100644
@@ -32,6 +32,7 @@ config KVM
  	select KVM_XFER_TO_GUEST_WORK
  	select KVM_GENERIC_MMU_NOTIFIER
  	select SCHED_INFO
+	select GUEST_PERF_EVENTS if PERF_EVENTS
  	help
  	  Support hosting virtualized guest machines.
 
arch/riscv/kvm/main.c +10 -2
index bab2ec34cd87..734b48d8f6dd 100644
@@ -51,6 +51,12 @@ void kvm_arch_hardware_disable(void)
  	csr_write(CSR_HIDELEG, 0);
  }
 
+static void kvm_riscv_teardown(void)
+{
+	kvm_riscv_aia_exit();
+	kvm_unregister_perf_callbacks();
+}
+
  static int __init riscv_kvm_init(void)
  {
  	int rc;
@@ -105,9 +111,11 @@ static int __init riscv_kvm_init(void)
  		kvm_info("AIA available with %d guest external interrupts\n",
  			 kvm_riscv_aia_nr_hgei);
 
+	kvm_register_perf_callbacks(NULL);
+
  	rc = kvm_init(sizeof(struct kvm_vcpu), 0, THIS_MODULE);
  	if (rc) {
-		kvm_riscv_aia_exit();
+		kvm_riscv_teardown();
  		return rc;
  	}
 
@@ -117,7 +125,7 @@ module_init(riscv_kvm_init);
 
  static void __exit riscv_kvm_exit(void)
  {
-	kvm_riscv_aia_exit();
+	kvm_riscv_teardown();
 
  	kvm_exit();
  }
arch/riscv/kvm/vcpu.c +7 -0
index 17e21df36cc1..c9d291865141 100644
@@ -222,6 +222,13 @@ bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
  	return (vcpu->arch.guest_context.sstatus & SR_SPP) ? true : false;
  }
 
+#ifdef CONFIG_GUEST_PERF_EVENTS
+unsigned long kvm_arch_vcpu_get_ip(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.guest_context.sepc;
+}
+#endif
+
  vm_fault_t kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
  {
  	return VM_FAULT_SIGBUS;
Signature · 1 line
2.34.1

Otherwise,