* [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv @ 2024-04-15 3:11 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: kvm-riscv Changes from v1->v2: - Rebased on Linux 6.9-rc3. 'perf kvm stat report/record' generates a statistical analysis of KVM events and can be used to analyze guest exit reasons. This patch tries to add stat support on riscv. Map the return value of trace_kvm_exit() to the specific cause of the exception, and export it to userspace. It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" and "kvm:kvm_exit", and reports statistical data which includes events handles time, samples, and so on. Simple tests go below: # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" Lowering default frequency rate from 4000 to 2500. Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. [ perf record: Woken up 18 times to write data ] [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) # ./perf kvm report 31K kvm:kvm_entry 31K kvm:kvm_exit # ./perf kvm stat record -a [ perf record: Woken up 3 times to write data ] [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] # ./perf kvm stat report --event=vmexit Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 Shenlin Liang (2): RISCV: KVM: add tracepoints for entry and exit events perf kvm/riscv: Port perf kvm stat to RISC-V arch/riscv/kvm/trace.h | 67 ++++++++++++++++ arch/riscv/kvm/vcpu.c | 7 ++ tools/perf/arch/riscv/Makefile | 1 + tools/perf/arch/riscv/util/Build | 1 + tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ 6 files changed, 195 insertions(+) create mode 100644 arch/riscv/kvm/trace.h create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h -- 2.37.2 ^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv @ 2024-04-15 3:11 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users Cc: Shenlin Liang Changes from v1->v2: - Rebased on Linux 6.9-rc3. 'perf kvm stat report/record' generates a statistical analysis of KVM events and can be used to analyze guest exit reasons. This patch tries to add stat support on riscv. Map the return value of trace_kvm_exit() to the specific cause of the exception, and export it to userspace. It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" and "kvm:kvm_exit", and reports statistical data which includes events handles time, samples, and so on. Simple tests go below: # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" Lowering default frequency rate from 4000 to 2500. Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. [ perf record: Woken up 18 times to write data ] [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) # ./perf kvm report 31K kvm:kvm_entry 31K kvm:kvm_exit # ./perf kvm stat record -a [ perf record: Woken up 3 times to write data ] [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] # ./perf kvm stat report --event=vmexit Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 Shenlin Liang (2): RISCV: KVM: add tracepoints for entry and exit events perf kvm/riscv: Port perf kvm stat to RISC-V arch/riscv/kvm/trace.h | 67 ++++++++++++++++ arch/riscv/kvm/vcpu.c | 7 ++ tools/perf/arch/riscv/Makefile | 1 + tools/perf/arch/riscv/util/Build | 1 + tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ 6 files changed, 195 insertions(+) create mode 100644 arch/riscv/kvm/trace.h create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h -- 2.37.2 _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv @ 2024-04-15 3:11 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users Cc: Shenlin Liang Changes from v1->v2: - Rebased on Linux 6.9-rc3. 'perf kvm stat report/record' generates a statistical analysis of KVM events and can be used to analyze guest exit reasons. This patch tries to add stat support on riscv. Map the return value of trace_kvm_exit() to the specific cause of the exception, and export it to userspace. It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" and "kvm:kvm_exit", and reports statistical data which includes events handles time, samples, and so on. Simple tests go below: # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" Lowering default frequency rate from 4000 to 2500. Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. [ perf record: Woken up 18 times to write data ] [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) # ./perf kvm report 31K kvm:kvm_entry 31K kvm:kvm_exit # ./perf kvm stat record -a [ perf record: Woken up 3 times to write data ] [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] # ./perf kvm stat report --event=vmexit Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 Shenlin Liang (2): RISCV: KVM: add tracepoints for entry and exit events perf kvm/riscv: Port perf kvm stat to RISC-V arch/riscv/kvm/trace.h | 67 ++++++++++++++++ arch/riscv/kvm/vcpu.c | 7 ++ tools/perf/arch/riscv/Makefile | 1 + tools/perf/arch/riscv/util/Build | 1 + tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ 6 files changed, 195 insertions(+) create mode 100644 arch/riscv/kvm/trace.h create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h -- 2.37.2 ^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v2 1/2] RISCV: KVM: add tracepoints for entry and exit events 2024-04-15 3:11 ` Shenlin Liang (?) @ 2024-04-15 3:11 ` Shenlin Liang -1 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: kvm-riscv Like other architectures, RISCV KVM also needs to add these event tracepoints to count the number of times kvm guest entry/exit. Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> --- arch/riscv/kvm/trace.h | 67 ++++++++++++++++++++++++++++++++++++++++++ arch/riscv/kvm/vcpu.c | 7 +++++ 2 files changed, 74 insertions(+) create mode 100644 arch/riscv/kvm/trace.h diff --git a/arch/riscv/kvm/trace.h b/arch/riscv/kvm/trace.h new file mode 100644 index 000000000000..3d54175d805c --- /dev/null +++ b/arch/riscv/kvm/trace.h @@ -0,0 +1,67 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Tracepoints for RISC-V KVM + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_KVM_H + +#include <linux/tracepoint.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM kvm + +TRACE_EVENT(kvm_entry, + TP_PROTO(struct kvm_vcpu *vcpu), + TP_ARGS(vcpu), + + TP_STRUCT__entry( + __field(unsigned long, pc) + ), + + TP_fast_assign( + __entry->pc = vcpu->arch.guest_context.sepc; + ), + + TP_printk("PC: 0x016%lx", __entry->pc) +); + +TRACE_EVENT(kvm_exit, + TP_PROTO(struct kvm_cpu_trap *trap), + TP_ARGS(trap), + + TP_STRUCT__entry( + __field(unsigned long, sepc) + __field(unsigned long, scause) + __field(unsigned long, stval) + __field(unsigned long, htval) + __field(unsigned long, htinst) + ), + + TP_fast_assign( + __entry->sepc = trap->sepc; + __entry->scause = trap->scause; + __entry->stval = trap->stval; + __entry->htval = trap->htval; + __entry->htinst = trap->htinst; + ), + + TP_printk("SEPC:0x%lx, SCAUSE:0x%lx, STVAL:0x%lx, HTVAL:0x%lx, HTINST:0x%lx", + __entry->sepc, + __entry->scause, + __entry->stval, + __entry->htval, + __entry->htinst) +); + +#endif /* _TRACE_RSICV_KVM_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c index b5ca9f2e98ac..f4e27004ceb8 100644 --- a/arch/riscv/kvm/vcpu.c +++ b/arch/riscv/kvm/vcpu.c @@ -21,6 +21,9 @@ #include <asm/cacheflush.h> #include <asm/kvm_vcpu_vector.h> +#define CREATE_TRACE_POINTS +#include "trace.h" + const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = { KVM_GENERIC_VCPU_STATS(), STATS_DESC_COUNTER(VCPU, ecall_exit_stat), @@ -782,6 +785,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) */ kvm_riscv_local_tlb_sanitize(vcpu); + trace_kvm_entry(vcpu); + guest_timing_enter_irqoff(); kvm_riscv_vcpu_enter_exit(vcpu); @@ -820,6 +825,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) local_irq_enable(); + trace_kvm_exit(&trap); + preempt_enable(); kvm_vcpu_srcu_read_lock(vcpu); -- 2.37.2 ^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH v2 1/2] RISCV: KVM: add tracepoints for entry and exit events @ 2024-04-15 3:11 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users Cc: Shenlin Liang Like other architectures, RISCV KVM also needs to add these event tracepoints to count the number of times kvm guest entry/exit. Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> --- arch/riscv/kvm/trace.h | 67 ++++++++++++++++++++++++++++++++++++++++++ arch/riscv/kvm/vcpu.c | 7 +++++ 2 files changed, 74 insertions(+) create mode 100644 arch/riscv/kvm/trace.h diff --git a/arch/riscv/kvm/trace.h b/arch/riscv/kvm/trace.h new file mode 100644 index 000000000000..3d54175d805c --- /dev/null +++ b/arch/riscv/kvm/trace.h @@ -0,0 +1,67 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Tracepoints for RISC-V KVM + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_KVM_H + +#include <linux/tracepoint.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM kvm + +TRACE_EVENT(kvm_entry, + TP_PROTO(struct kvm_vcpu *vcpu), + TP_ARGS(vcpu), + + TP_STRUCT__entry( + __field(unsigned long, pc) + ), + + TP_fast_assign( + __entry->pc = vcpu->arch.guest_context.sepc; + ), + + TP_printk("PC: 0x016%lx", __entry->pc) +); + +TRACE_EVENT(kvm_exit, + TP_PROTO(struct kvm_cpu_trap *trap), + TP_ARGS(trap), + + TP_STRUCT__entry( + __field(unsigned long, sepc) + __field(unsigned long, scause) + __field(unsigned long, stval) + __field(unsigned long, htval) + __field(unsigned long, htinst) + ), + + TP_fast_assign( + __entry->sepc = trap->sepc; + __entry->scause = trap->scause; + __entry->stval = trap->stval; + __entry->htval = trap->htval; + __entry->htinst = trap->htinst; + ), + + TP_printk("SEPC:0x%lx, SCAUSE:0x%lx, STVAL:0x%lx, HTVAL:0x%lx, HTINST:0x%lx", + __entry->sepc, + __entry->scause, + __entry->stval, + __entry->htval, + __entry->htinst) +); + +#endif /* _TRACE_RSICV_KVM_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c index b5ca9f2e98ac..f4e27004ceb8 100644 --- a/arch/riscv/kvm/vcpu.c +++ b/arch/riscv/kvm/vcpu.c @@ -21,6 +21,9 @@ #include <asm/cacheflush.h> #include <asm/kvm_vcpu_vector.h> +#define CREATE_TRACE_POINTS +#include "trace.h" + const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = { KVM_GENERIC_VCPU_STATS(), STATS_DESC_COUNTER(VCPU, ecall_exit_stat), @@ -782,6 +785,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) */ kvm_riscv_local_tlb_sanitize(vcpu); + trace_kvm_entry(vcpu); + guest_timing_enter_irqoff(); kvm_riscv_vcpu_enter_exit(vcpu); @@ -820,6 +825,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) local_irq_enable(); + trace_kvm_exit(&trap); + preempt_enable(); kvm_vcpu_srcu_read_lock(vcpu); -- 2.37.2 _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH v2 1/2] RISCV: KVM: add tracepoints for entry and exit events @ 2024-04-15 3:11 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users Cc: Shenlin Liang Like other architectures, RISCV KVM also needs to add these event tracepoints to count the number of times kvm guest entry/exit. Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> --- arch/riscv/kvm/trace.h | 67 ++++++++++++++++++++++++++++++++++++++++++ arch/riscv/kvm/vcpu.c | 7 +++++ 2 files changed, 74 insertions(+) create mode 100644 arch/riscv/kvm/trace.h diff --git a/arch/riscv/kvm/trace.h b/arch/riscv/kvm/trace.h new file mode 100644 index 000000000000..3d54175d805c --- /dev/null +++ b/arch/riscv/kvm/trace.h @@ -0,0 +1,67 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Tracepoints for RISC-V KVM + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_KVM_H + +#include <linux/tracepoint.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM kvm + +TRACE_EVENT(kvm_entry, + TP_PROTO(struct kvm_vcpu *vcpu), + TP_ARGS(vcpu), + + TP_STRUCT__entry( + __field(unsigned long, pc) + ), + + TP_fast_assign( + __entry->pc = vcpu->arch.guest_context.sepc; + ), + + TP_printk("PC: 0x016%lx", __entry->pc) +); + +TRACE_EVENT(kvm_exit, + TP_PROTO(struct kvm_cpu_trap *trap), + TP_ARGS(trap), + + TP_STRUCT__entry( + __field(unsigned long, sepc) + __field(unsigned long, scause) + __field(unsigned long, stval) + __field(unsigned long, htval) + __field(unsigned long, htinst) + ), + + TP_fast_assign( + __entry->sepc = trap->sepc; + __entry->scause = trap->scause; + __entry->stval = trap->stval; + __entry->htval = trap->htval; + __entry->htinst = trap->htinst; + ), + + TP_printk("SEPC:0x%lx, SCAUSE:0x%lx, STVAL:0x%lx, HTVAL:0x%lx, HTINST:0x%lx", + __entry->sepc, + __entry->scause, + __entry->stval, + __entry->htval, + __entry->htinst) +); + +#endif /* _TRACE_RSICV_KVM_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c index b5ca9f2e98ac..f4e27004ceb8 100644 --- a/arch/riscv/kvm/vcpu.c +++ b/arch/riscv/kvm/vcpu.c @@ -21,6 +21,9 @@ #include <asm/cacheflush.h> #include <asm/kvm_vcpu_vector.h> +#define CREATE_TRACE_POINTS +#include "trace.h" + const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = { KVM_GENERIC_VCPU_STATS(), STATS_DESC_COUNTER(VCPU, ecall_exit_stat), @@ -782,6 +785,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) */ kvm_riscv_local_tlb_sanitize(vcpu); + trace_kvm_entry(vcpu); + guest_timing_enter_irqoff(); kvm_riscv_vcpu_enter_exit(vcpu); @@ -820,6 +825,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) local_irq_enable(); + trace_kvm_exit(&trap); + preempt_enable(); kvm_vcpu_srcu_read_lock(vcpu); -- 2.37.2 ^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH v2 1/2] RISCV: KVM: add tracepoints for entry and exit events 2024-04-15 3:11 ` Shenlin Liang (?) @ 2024-04-16 3:44 ` Anup Patel -1 siblings, 0 replies; 21+ messages in thread From: Anup Patel @ 2024-04-16 3:44 UTC (permalink / raw) To: kvm-riscv On Mon, Apr 15, 2024 at 8:45?AM Shenlin Liang <liangshenlin@eswincomputing.com> wrote: > > Like other architectures, RISCV KVM also needs to add these event > tracepoints to count the number of times kvm guest entry/exit. > > Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> LGTM. Reviewed-by: Anup Patel <anup@brainfault.org> Regards, Anup > --- > arch/riscv/kvm/trace.h | 67 ++++++++++++++++++++++++++++++++++++++++++ > arch/riscv/kvm/vcpu.c | 7 +++++ > 2 files changed, 74 insertions(+) > create mode 100644 arch/riscv/kvm/trace.h > > diff --git a/arch/riscv/kvm/trace.h b/arch/riscv/kvm/trace.h > new file mode 100644 > index 000000000000..3d54175d805c > --- /dev/null > +++ b/arch/riscv/kvm/trace.h > @@ -0,0 +1,67 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Tracepoints for RISC-V KVM > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ) > +#define _TRACE_KVM_H > + > +#include <linux/tracepoint.h> > + > +#undef TRACE_SYSTEM > +#define TRACE_SYSTEM kvm > + > +TRACE_EVENT(kvm_entry, > + TP_PROTO(struct kvm_vcpu *vcpu), > + TP_ARGS(vcpu), > + > + TP_STRUCT__entry( > + __field(unsigned long, pc) > + ), > + > + TP_fast_assign( > + __entry->pc = vcpu->arch.guest_context.sepc; > + ), > + > + TP_printk("PC: 0x016%lx", __entry->pc) > +); > + > +TRACE_EVENT(kvm_exit, > + TP_PROTO(struct kvm_cpu_trap *trap), > + TP_ARGS(trap), > + > + TP_STRUCT__entry( > + __field(unsigned long, sepc) > + __field(unsigned long, scause) > + __field(unsigned long, stval) > + __field(unsigned long, htval) > + __field(unsigned long, htinst) > + ), > + > + TP_fast_assign( > + __entry->sepc = trap->sepc; > + __entry->scause = trap->scause; > + __entry->stval = trap->stval; > + __entry->htval = trap->htval; > + __entry->htinst = trap->htinst; > + ), > + > + TP_printk("SEPC:0x%lx, SCAUSE:0x%lx, STVAL:0x%lx, HTVAL:0x%lx, HTINST:0x%lx", > + __entry->sepc, > + __entry->scause, > + __entry->stval, > + __entry->htval, > + __entry->htinst) > +); > + > +#endif /* _TRACE_RSICV_KVM_H */ > + > +#undef TRACE_INCLUDE_PATH > +#define TRACE_INCLUDE_PATH . > +#undef TRACE_INCLUDE_FILE > +#define TRACE_INCLUDE_FILE trace > + > +/* This part must be outside protection */ > +#include <trace/define_trace.h> > diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c > index b5ca9f2e98ac..f4e27004ceb8 100644 > --- a/arch/riscv/kvm/vcpu.c > +++ b/arch/riscv/kvm/vcpu.c > @@ -21,6 +21,9 @@ > #include <asm/cacheflush.h> > #include <asm/kvm_vcpu_vector.h> > > +#define CREATE_TRACE_POINTS > +#include "trace.h" > + > const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = { > KVM_GENERIC_VCPU_STATS(), > STATS_DESC_COUNTER(VCPU, ecall_exit_stat), > @@ -782,6 +785,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) > */ > kvm_riscv_local_tlb_sanitize(vcpu); > > + trace_kvm_entry(vcpu); > + > guest_timing_enter_irqoff(); > > kvm_riscv_vcpu_enter_exit(vcpu); > @@ -820,6 +825,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) > > local_irq_enable(); > > + trace_kvm_exit(&trap); > + > preempt_enable(); > > kvm_vcpu_srcu_read_lock(vcpu); > -- > 2.37.2 > ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 1/2] RISCV: KVM: add tracepoints for entry and exit events @ 2024-04-16 3:44 ` Anup Patel 0 siblings, 0 replies; 21+ messages in thread From: Anup Patel @ 2024-04-16 3:44 UTC (permalink / raw) To: Shenlin Liang Cc: atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users On Mon, Apr 15, 2024 at 8:45 AM Shenlin Liang <liangshenlin@eswincomputing.com> wrote: > > Like other architectures, RISCV KVM also needs to add these event > tracepoints to count the number of times kvm guest entry/exit. > > Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> LGTM. Reviewed-by: Anup Patel <anup@brainfault.org> Regards, Anup > --- > arch/riscv/kvm/trace.h | 67 ++++++++++++++++++++++++++++++++++++++++++ > arch/riscv/kvm/vcpu.c | 7 +++++ > 2 files changed, 74 insertions(+) > create mode 100644 arch/riscv/kvm/trace.h > > diff --git a/arch/riscv/kvm/trace.h b/arch/riscv/kvm/trace.h > new file mode 100644 > index 000000000000..3d54175d805c > --- /dev/null > +++ b/arch/riscv/kvm/trace.h > @@ -0,0 +1,67 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Tracepoints for RISC-V KVM > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ) > +#define _TRACE_KVM_H > + > +#include <linux/tracepoint.h> > + > +#undef TRACE_SYSTEM > +#define TRACE_SYSTEM kvm > + > +TRACE_EVENT(kvm_entry, > + TP_PROTO(struct kvm_vcpu *vcpu), > + TP_ARGS(vcpu), > + > + TP_STRUCT__entry( > + __field(unsigned long, pc) > + ), > + > + TP_fast_assign( > + __entry->pc = vcpu->arch.guest_context.sepc; > + ), > + > + TP_printk("PC: 0x016%lx", __entry->pc) > +); > + > +TRACE_EVENT(kvm_exit, > + TP_PROTO(struct kvm_cpu_trap *trap), > + TP_ARGS(trap), > + > + TP_STRUCT__entry( > + __field(unsigned long, sepc) > + __field(unsigned long, scause) > + __field(unsigned long, stval) > + __field(unsigned long, htval) > + __field(unsigned long, htinst) > + ), > + > + TP_fast_assign( > + __entry->sepc = trap->sepc; > + __entry->scause = trap->scause; > + __entry->stval = trap->stval; > + __entry->htval = trap->htval; > + __entry->htinst = trap->htinst; > + ), > + > + TP_printk("SEPC:0x%lx, SCAUSE:0x%lx, STVAL:0x%lx, HTVAL:0x%lx, HTINST:0x%lx", > + __entry->sepc, > + __entry->scause, > + __entry->stval, > + __entry->htval, > + __entry->htinst) > +); > + > +#endif /* _TRACE_RSICV_KVM_H */ > + > +#undef TRACE_INCLUDE_PATH > +#define TRACE_INCLUDE_PATH . > +#undef TRACE_INCLUDE_FILE > +#define TRACE_INCLUDE_FILE trace > + > +/* This part must be outside protection */ > +#include <trace/define_trace.h> > diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c > index b5ca9f2e98ac..f4e27004ceb8 100644 > --- a/arch/riscv/kvm/vcpu.c > +++ b/arch/riscv/kvm/vcpu.c > @@ -21,6 +21,9 @@ > #include <asm/cacheflush.h> > #include <asm/kvm_vcpu_vector.h> > > +#define CREATE_TRACE_POINTS > +#include "trace.h" > + > const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = { > KVM_GENERIC_VCPU_STATS(), > STATS_DESC_COUNTER(VCPU, ecall_exit_stat), > @@ -782,6 +785,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) > */ > kvm_riscv_local_tlb_sanitize(vcpu); > > + trace_kvm_entry(vcpu); > + > guest_timing_enter_irqoff(); > > kvm_riscv_vcpu_enter_exit(vcpu); > @@ -820,6 +825,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) > > local_irq_enable(); > > + trace_kvm_exit(&trap); > + > preempt_enable(); > > kvm_vcpu_srcu_read_lock(vcpu); > -- > 2.37.2 > _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 1/2] RISCV: KVM: add tracepoints for entry and exit events @ 2024-04-16 3:44 ` Anup Patel 0 siblings, 0 replies; 21+ messages in thread From: Anup Patel @ 2024-04-16 3:44 UTC (permalink / raw) To: Shenlin Liang Cc: atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users On Mon, Apr 15, 2024 at 8:45 AM Shenlin Liang <liangshenlin@eswincomputing.com> wrote: > > Like other architectures, RISCV KVM also needs to add these event > tracepoints to count the number of times kvm guest entry/exit. > > Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> LGTM. Reviewed-by: Anup Patel <anup@brainfault.org> Regards, Anup > --- > arch/riscv/kvm/trace.h | 67 ++++++++++++++++++++++++++++++++++++++++++ > arch/riscv/kvm/vcpu.c | 7 +++++ > 2 files changed, 74 insertions(+) > create mode 100644 arch/riscv/kvm/trace.h > > diff --git a/arch/riscv/kvm/trace.h b/arch/riscv/kvm/trace.h > new file mode 100644 > index 000000000000..3d54175d805c > --- /dev/null > +++ b/arch/riscv/kvm/trace.h > @@ -0,0 +1,67 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Tracepoints for RISC-V KVM > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ) > +#define _TRACE_KVM_H > + > +#include <linux/tracepoint.h> > + > +#undef TRACE_SYSTEM > +#define TRACE_SYSTEM kvm > + > +TRACE_EVENT(kvm_entry, > + TP_PROTO(struct kvm_vcpu *vcpu), > + TP_ARGS(vcpu), > + > + TP_STRUCT__entry( > + __field(unsigned long, pc) > + ), > + > + TP_fast_assign( > + __entry->pc = vcpu->arch.guest_context.sepc; > + ), > + > + TP_printk("PC: 0x016%lx", __entry->pc) > +); > + > +TRACE_EVENT(kvm_exit, > + TP_PROTO(struct kvm_cpu_trap *trap), > + TP_ARGS(trap), > + > + TP_STRUCT__entry( > + __field(unsigned long, sepc) > + __field(unsigned long, scause) > + __field(unsigned long, stval) > + __field(unsigned long, htval) > + __field(unsigned long, htinst) > + ), > + > + TP_fast_assign( > + __entry->sepc = trap->sepc; > + __entry->scause = trap->scause; > + __entry->stval = trap->stval; > + __entry->htval = trap->htval; > + __entry->htinst = trap->htinst; > + ), > + > + TP_printk("SEPC:0x%lx, SCAUSE:0x%lx, STVAL:0x%lx, HTVAL:0x%lx, HTINST:0x%lx", > + __entry->sepc, > + __entry->scause, > + __entry->stval, > + __entry->htval, > + __entry->htinst) > +); > + > +#endif /* _TRACE_RSICV_KVM_H */ > + > +#undef TRACE_INCLUDE_PATH > +#define TRACE_INCLUDE_PATH . > +#undef TRACE_INCLUDE_FILE > +#define TRACE_INCLUDE_FILE trace > + > +/* This part must be outside protection */ > +#include <trace/define_trace.h> > diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c > index b5ca9f2e98ac..f4e27004ceb8 100644 > --- a/arch/riscv/kvm/vcpu.c > +++ b/arch/riscv/kvm/vcpu.c > @@ -21,6 +21,9 @@ > #include <asm/cacheflush.h> > #include <asm/kvm_vcpu_vector.h> > > +#define CREATE_TRACE_POINTS > +#include "trace.h" > + > const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = { > KVM_GENERIC_VCPU_STATS(), > STATS_DESC_COUNTER(VCPU, ecall_exit_stat), > @@ -782,6 +785,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) > */ > kvm_riscv_local_tlb_sanitize(vcpu); > > + trace_kvm_entry(vcpu); > + > guest_timing_enter_irqoff(); > > kvm_riscv_vcpu_enter_exit(vcpu); > @@ -820,6 +825,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) > > local_irq_enable(); > > + trace_kvm_exit(&trap); > + > preempt_enable(); > > kvm_vcpu_srcu_read_lock(vcpu); > -- > 2.37.2 > ^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v2 2/2] perf kvm/riscv: Port perf kvm stat to RISC-V 2024-04-15 3:11 ` Shenlin Liang (?) @ 2024-04-15 3:11 ` Shenlin Liang -1 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: kvm-riscv 'perf kvm stat report/record' generates a statistical analysis of KVM events and can be used to analyze guest exit reasons. "report" reports statistical analysis of guest exit events. To record kvm events on the host: # perf kvm stat record -a To report kvm VM EXIT events: # perf kvm stat report --event=vmexit Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> --- tools/perf/arch/riscv/Makefile | 1 + tools/perf/arch/riscv/util/Build | 1 + tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ 4 files changed, 121 insertions(+) create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h diff --git a/tools/perf/arch/riscv/Makefile b/tools/perf/arch/riscv/Makefile index a8d25d005207..e1e445615536 100644 --- a/tools/perf/arch/riscv/Makefile +++ b/tools/perf/arch/riscv/Makefile @@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1 endif PERF_HAVE_ARCH_REGS_QUERY_REGISTER_OFFSET := 1 PERF_HAVE_JITDUMP := 1 +HAVE_KVM_STAT_SUPPORT := 1 \ No newline at end of file diff --git a/tools/perf/arch/riscv/util/Build b/tools/perf/arch/riscv/util/Build index 603dbb5ae4dc..d72b04f8d32b 100644 --- a/tools/perf/arch/riscv/util/Build +++ b/tools/perf/arch/riscv/util/Build @@ -1,5 +1,6 @@ perf-y += perf_regs.o perf-y += header.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o perf-$(CONFIG_DWARF) += dwarf-regs.o perf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o diff --git a/tools/perf/arch/riscv/util/kvm-stat.c b/tools/perf/arch/riscv/util/kvm-stat.c new file mode 100644 index 000000000000..db7183e0f09f --- /dev/null +++ b/tools/perf/arch/riscv/util/kvm-stat.c @@ -0,0 +1,78 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Arch specific functions for perf kvm stat. + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#include <errno.h> +#include <memory.h> +#include "../../../util/evsel.h" +#include "../../../util/kvm-stat.h" +#include "riscv_exception_types.h" +#include "debug.h" + +define_exit_reasons_table(riscv_exit_reasons, kvm_riscv_exception_class); + +const char *kvm_exit_reason = "scause"; +const char *kvm_entry_trace = "kvm:kvm_entry"; +const char *kvm_exit_trace = "kvm:kvm_exit"; + +const char *kvm_events_tp[] = { + "kvm:kvm_entry", + "kvm:kvm_exit", + NULL, +}; + +static void event_get_key(struct evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + key->info = 0; + key->key = evsel__intval(evsel, sample, kvm_exit_reason); + key->key = (int)key->key; + key->exit_reasons = riscv_exit_reasons; +} + +static bool event_begin(struct evsel *evsel, + struct perf_sample *sample __maybe_unused, + struct event_key *key __maybe_unused) +{ + return evsel__name_is(evsel, kvm_entry_trace); +} + +static bool event_end(struct evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + if (evsel__name_is(evsel, kvm_exit_trace)) { + event_get_key(evsel, sample, key); + return true; + } + return false; +} + +static struct kvm_events_ops exit_events = { + .is_begin_event = event_begin, + .is_end_event = event_end, + .decode_key = exit_event_decode_key, + .name = "VM-EXIT" +}; + +struct kvm_reg_events_ops kvm_reg_events_ops[] = { + { + .name = "vmexit", + .ops = &exit_events, + }, + { NULL, NULL }, +}; + +const char * const kvm_skip_events[] = { + NULL, +}; + +int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid __maybe_unused) +{ + kvm->exit_reasons_isa = "riscv64"; + return 0; +} diff --git a/tools/perf/arch/riscv/util/riscv_exception_types.h b/tools/perf/arch/riscv/util/riscv_exception_types.h new file mode 100644 index 000000000000..2e42150f72b2 --- /dev/null +++ b/tools/perf/arch/riscv/util/riscv_exception_types.h @@ -0,0 +1,41 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * RISCV exception types + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#ifndef ARCH_PERF_RISCV_EXCEPTION_TYPES_H +#define ARCH_PERF_RISCV_EXCEPTION_TYPES_H + +#define EXC_INST_MISALIGNED 0 +#define EXC_INST_ACCESS 1 +#define EXC_INST_ILLEGAL 2 +#define EXC_BREAKPOINT 3 +#define EXC_LOAD_MISALIGNED 4 +#define EXC_LOAD_ACCESS 5 +#define EXC_STORE_MISALIGNED 6 +#define EXC_STORE_ACCESS 7 +#define EXC_SYSCALL 8 +#define EXC_HYPERVISOR_SYSCALL 9 +#define EXC_SUPERVISOR_SYSCALL 10 +#define EXC_INST_PAGE_FAULT 12 +#define EXC_LOAD_PAGE_FAULT 13 +#define EXC_STORE_PAGE_FAULT 15 +#define EXC_INST_GUEST_PAGE_FAULT 20 +#define EXC_LOAD_GUEST_PAGE_FAULT 21 +#define EXC_VIRTUAL_INST_FAULT 22 +#define EXC_STORE_GUEST_PAGE_FAULT 23 + +#define EXC(x) {EXC_##x, #x } + +#define kvm_riscv_exception_class \ + (EXC(INST_MISALIGNED), EXC(INST_ACCESS), EXC(INST_ILLEGAL), \ + EXC(BREAKPOINT), EXC(LOAD_MISALIGNED), EXC(LOAD_ACCESS), \ + EXC(STORE_MISALIGNED), EXC(STORE_ACCESS), EXC(SYSCALL), \ + EXC(HYPERVISOR_SYSCALL), EXC(SUPERVISOR_SYSCALL), \ + EXC(INST_PAGE_FAULT), EXC(LOAD_PAGE_FAULT), EXC(STORE_PAGE_FAULT), \ + EXC(INST_GUEST_PAGE_FAULT), EXC(LOAD_GUEST_PAGE_FAULT), \ + EXC(VIRTUAL_INST_FAULT), EXC(STORE_GUEST_PAGE_FAULT)) + +#endif /* ARCH_PERF_RISCV_EXCEPTION_TYPES_H */ -- 2.37.2 ^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH v2 2/2] perf kvm/riscv: Port perf kvm stat to RISC-V @ 2024-04-15 3:11 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users Cc: Shenlin Liang 'perf kvm stat report/record' generates a statistical analysis of KVM events and can be used to analyze guest exit reasons. "report" reports statistical analysis of guest exit events. To record kvm events on the host: # perf kvm stat record -a To report kvm VM EXIT events: # perf kvm stat report --event=vmexit Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> --- tools/perf/arch/riscv/Makefile | 1 + tools/perf/arch/riscv/util/Build | 1 + tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ 4 files changed, 121 insertions(+) create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h diff --git a/tools/perf/arch/riscv/Makefile b/tools/perf/arch/riscv/Makefile index a8d25d005207..e1e445615536 100644 --- a/tools/perf/arch/riscv/Makefile +++ b/tools/perf/arch/riscv/Makefile @@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1 endif PERF_HAVE_ARCH_REGS_QUERY_REGISTER_OFFSET := 1 PERF_HAVE_JITDUMP := 1 +HAVE_KVM_STAT_SUPPORT := 1 \ No newline at end of file diff --git a/tools/perf/arch/riscv/util/Build b/tools/perf/arch/riscv/util/Build index 603dbb5ae4dc..d72b04f8d32b 100644 --- a/tools/perf/arch/riscv/util/Build +++ b/tools/perf/arch/riscv/util/Build @@ -1,5 +1,6 @@ perf-y += perf_regs.o perf-y += header.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o perf-$(CONFIG_DWARF) += dwarf-regs.o perf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o diff --git a/tools/perf/arch/riscv/util/kvm-stat.c b/tools/perf/arch/riscv/util/kvm-stat.c new file mode 100644 index 000000000000..db7183e0f09f --- /dev/null +++ b/tools/perf/arch/riscv/util/kvm-stat.c @@ -0,0 +1,78 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Arch specific functions for perf kvm stat. + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#include <errno.h> +#include <memory.h> +#include "../../../util/evsel.h" +#include "../../../util/kvm-stat.h" +#include "riscv_exception_types.h" +#include "debug.h" + +define_exit_reasons_table(riscv_exit_reasons, kvm_riscv_exception_class); + +const char *kvm_exit_reason = "scause"; +const char *kvm_entry_trace = "kvm:kvm_entry"; +const char *kvm_exit_trace = "kvm:kvm_exit"; + +const char *kvm_events_tp[] = { + "kvm:kvm_entry", + "kvm:kvm_exit", + NULL, +}; + +static void event_get_key(struct evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + key->info = 0; + key->key = evsel__intval(evsel, sample, kvm_exit_reason); + key->key = (int)key->key; + key->exit_reasons = riscv_exit_reasons; +} + +static bool event_begin(struct evsel *evsel, + struct perf_sample *sample __maybe_unused, + struct event_key *key __maybe_unused) +{ + return evsel__name_is(evsel, kvm_entry_trace); +} + +static bool event_end(struct evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + if (evsel__name_is(evsel, kvm_exit_trace)) { + event_get_key(evsel, sample, key); + return true; + } + return false; +} + +static struct kvm_events_ops exit_events = { + .is_begin_event = event_begin, + .is_end_event = event_end, + .decode_key = exit_event_decode_key, + .name = "VM-EXIT" +}; + +struct kvm_reg_events_ops kvm_reg_events_ops[] = { + { + .name = "vmexit", + .ops = &exit_events, + }, + { NULL, NULL }, +}; + +const char * const kvm_skip_events[] = { + NULL, +}; + +int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid __maybe_unused) +{ + kvm->exit_reasons_isa = "riscv64"; + return 0; +} diff --git a/tools/perf/arch/riscv/util/riscv_exception_types.h b/tools/perf/arch/riscv/util/riscv_exception_types.h new file mode 100644 index 000000000000..2e42150f72b2 --- /dev/null +++ b/tools/perf/arch/riscv/util/riscv_exception_types.h @@ -0,0 +1,41 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * RISCV exception types + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#ifndef ARCH_PERF_RISCV_EXCEPTION_TYPES_H +#define ARCH_PERF_RISCV_EXCEPTION_TYPES_H + +#define EXC_INST_MISALIGNED 0 +#define EXC_INST_ACCESS 1 +#define EXC_INST_ILLEGAL 2 +#define EXC_BREAKPOINT 3 +#define EXC_LOAD_MISALIGNED 4 +#define EXC_LOAD_ACCESS 5 +#define EXC_STORE_MISALIGNED 6 +#define EXC_STORE_ACCESS 7 +#define EXC_SYSCALL 8 +#define EXC_HYPERVISOR_SYSCALL 9 +#define EXC_SUPERVISOR_SYSCALL 10 +#define EXC_INST_PAGE_FAULT 12 +#define EXC_LOAD_PAGE_FAULT 13 +#define EXC_STORE_PAGE_FAULT 15 +#define EXC_INST_GUEST_PAGE_FAULT 20 +#define EXC_LOAD_GUEST_PAGE_FAULT 21 +#define EXC_VIRTUAL_INST_FAULT 22 +#define EXC_STORE_GUEST_PAGE_FAULT 23 + +#define EXC(x) {EXC_##x, #x } + +#define kvm_riscv_exception_class \ + (EXC(INST_MISALIGNED), EXC(INST_ACCESS), EXC(INST_ILLEGAL), \ + EXC(BREAKPOINT), EXC(LOAD_MISALIGNED), EXC(LOAD_ACCESS), \ + EXC(STORE_MISALIGNED), EXC(STORE_ACCESS), EXC(SYSCALL), \ + EXC(HYPERVISOR_SYSCALL), EXC(SUPERVISOR_SYSCALL), \ + EXC(INST_PAGE_FAULT), EXC(LOAD_PAGE_FAULT), EXC(STORE_PAGE_FAULT), \ + EXC(INST_GUEST_PAGE_FAULT), EXC(LOAD_GUEST_PAGE_FAULT), \ + EXC(VIRTUAL_INST_FAULT), EXC(STORE_GUEST_PAGE_FAULT)) + +#endif /* ARCH_PERF_RISCV_EXCEPTION_TYPES_H */ -- 2.37.2 _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH v2 2/2] perf kvm/riscv: Port perf kvm stat to RISC-V @ 2024-04-15 3:11 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-15 3:11 UTC (permalink / raw) To: anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users Cc: Shenlin Liang 'perf kvm stat report/record' generates a statistical analysis of KVM events and can be used to analyze guest exit reasons. "report" reports statistical analysis of guest exit events. To record kvm events on the host: # perf kvm stat record -a To report kvm VM EXIT events: # perf kvm stat report --event=vmexit Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> --- tools/perf/arch/riscv/Makefile | 1 + tools/perf/arch/riscv/util/Build | 1 + tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ 4 files changed, 121 insertions(+) create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h diff --git a/tools/perf/arch/riscv/Makefile b/tools/perf/arch/riscv/Makefile index a8d25d005207..e1e445615536 100644 --- a/tools/perf/arch/riscv/Makefile +++ b/tools/perf/arch/riscv/Makefile @@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1 endif PERF_HAVE_ARCH_REGS_QUERY_REGISTER_OFFSET := 1 PERF_HAVE_JITDUMP := 1 +HAVE_KVM_STAT_SUPPORT := 1 \ No newline at end of file diff --git a/tools/perf/arch/riscv/util/Build b/tools/perf/arch/riscv/util/Build index 603dbb5ae4dc..d72b04f8d32b 100644 --- a/tools/perf/arch/riscv/util/Build +++ b/tools/perf/arch/riscv/util/Build @@ -1,5 +1,6 @@ perf-y += perf_regs.o perf-y += header.o +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o perf-$(CONFIG_DWARF) += dwarf-regs.o perf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o diff --git a/tools/perf/arch/riscv/util/kvm-stat.c b/tools/perf/arch/riscv/util/kvm-stat.c new file mode 100644 index 000000000000..db7183e0f09f --- /dev/null +++ b/tools/perf/arch/riscv/util/kvm-stat.c @@ -0,0 +1,78 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Arch specific functions for perf kvm stat. + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#include <errno.h> +#include <memory.h> +#include "../../../util/evsel.h" +#include "../../../util/kvm-stat.h" +#include "riscv_exception_types.h" +#include "debug.h" + +define_exit_reasons_table(riscv_exit_reasons, kvm_riscv_exception_class); + +const char *kvm_exit_reason = "scause"; +const char *kvm_entry_trace = "kvm:kvm_entry"; +const char *kvm_exit_trace = "kvm:kvm_exit"; + +const char *kvm_events_tp[] = { + "kvm:kvm_entry", + "kvm:kvm_exit", + NULL, +}; + +static void event_get_key(struct evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + key->info = 0; + key->key = evsel__intval(evsel, sample, kvm_exit_reason); + key->key = (int)key->key; + key->exit_reasons = riscv_exit_reasons; +} + +static bool event_begin(struct evsel *evsel, + struct perf_sample *sample __maybe_unused, + struct event_key *key __maybe_unused) +{ + return evsel__name_is(evsel, kvm_entry_trace); +} + +static bool event_end(struct evsel *evsel, + struct perf_sample *sample, + struct event_key *key) +{ + if (evsel__name_is(evsel, kvm_exit_trace)) { + event_get_key(evsel, sample, key); + return true; + } + return false; +} + +static struct kvm_events_ops exit_events = { + .is_begin_event = event_begin, + .is_end_event = event_end, + .decode_key = exit_event_decode_key, + .name = "VM-EXIT" +}; + +struct kvm_reg_events_ops kvm_reg_events_ops[] = { + { + .name = "vmexit", + .ops = &exit_events, + }, + { NULL, NULL }, +}; + +const char * const kvm_skip_events[] = { + NULL, +}; + +int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid __maybe_unused) +{ + kvm->exit_reasons_isa = "riscv64"; + return 0; +} diff --git a/tools/perf/arch/riscv/util/riscv_exception_types.h b/tools/perf/arch/riscv/util/riscv_exception_types.h new file mode 100644 index 000000000000..2e42150f72b2 --- /dev/null +++ b/tools/perf/arch/riscv/util/riscv_exception_types.h @@ -0,0 +1,41 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * RISCV exception types + * + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. + * + */ +#ifndef ARCH_PERF_RISCV_EXCEPTION_TYPES_H +#define ARCH_PERF_RISCV_EXCEPTION_TYPES_H + +#define EXC_INST_MISALIGNED 0 +#define EXC_INST_ACCESS 1 +#define EXC_INST_ILLEGAL 2 +#define EXC_BREAKPOINT 3 +#define EXC_LOAD_MISALIGNED 4 +#define EXC_LOAD_ACCESS 5 +#define EXC_STORE_MISALIGNED 6 +#define EXC_STORE_ACCESS 7 +#define EXC_SYSCALL 8 +#define EXC_HYPERVISOR_SYSCALL 9 +#define EXC_SUPERVISOR_SYSCALL 10 +#define EXC_INST_PAGE_FAULT 12 +#define EXC_LOAD_PAGE_FAULT 13 +#define EXC_STORE_PAGE_FAULT 15 +#define EXC_INST_GUEST_PAGE_FAULT 20 +#define EXC_LOAD_GUEST_PAGE_FAULT 21 +#define EXC_VIRTUAL_INST_FAULT 22 +#define EXC_STORE_GUEST_PAGE_FAULT 23 + +#define EXC(x) {EXC_##x, #x } + +#define kvm_riscv_exception_class \ + (EXC(INST_MISALIGNED), EXC(INST_ACCESS), EXC(INST_ILLEGAL), \ + EXC(BREAKPOINT), EXC(LOAD_MISALIGNED), EXC(LOAD_ACCESS), \ + EXC(STORE_MISALIGNED), EXC(STORE_ACCESS), EXC(SYSCALL), \ + EXC(HYPERVISOR_SYSCALL), EXC(SUPERVISOR_SYSCALL), \ + EXC(INST_PAGE_FAULT), EXC(LOAD_PAGE_FAULT), EXC(STORE_PAGE_FAULT), \ + EXC(INST_GUEST_PAGE_FAULT), EXC(LOAD_GUEST_PAGE_FAULT), \ + EXC(VIRTUAL_INST_FAULT), EXC(STORE_GUEST_PAGE_FAULT)) + +#endif /* ARCH_PERF_RISCV_EXCEPTION_TYPES_H */ -- 2.37.2 ^ permalink raw reply related [flat|nested] 21+ messages in thread
* [PATCH v2 2/2] perf kvm/riscv: Port perf kvm stat to RISC-V 2024-04-15 3:11 ` Shenlin Liang (?) @ 2024-04-16 4:17 ` Anup Patel -1 siblings, 0 replies; 21+ messages in thread From: Anup Patel @ 2024-04-16 4:17 UTC (permalink / raw) To: kvm-riscv On Mon, Apr 15, 2024 at 8:45?AM Shenlin Liang <liangshenlin@eswincomputing.com> wrote: > > 'perf kvm stat report/record' generates a statistical analysis of KVM > events and can be used to analyze guest exit reasons. > > "report" reports statistical analysis of guest exit events. > > To record kvm events on the host: > # perf kvm stat record -a > > To report kvm VM EXIT events: > # perf kvm stat report --event=vmexit > > Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> LGTM. Reviewed-by: Anup Patel <anup@brainfault.org> @Peter Zijlstra, is it okay to take this through the KVM RISC-V tree ? Regards, Anup > --- > tools/perf/arch/riscv/Makefile | 1 + > tools/perf/arch/riscv/util/Build | 1 + > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > 4 files changed, 121 insertions(+) > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > > diff --git a/tools/perf/arch/riscv/Makefile b/tools/perf/arch/riscv/Makefile > index a8d25d005207..e1e445615536 100644 > --- a/tools/perf/arch/riscv/Makefile > +++ b/tools/perf/arch/riscv/Makefile > @@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1 > endif > PERF_HAVE_ARCH_REGS_QUERY_REGISTER_OFFSET := 1 > PERF_HAVE_JITDUMP := 1 > +HAVE_KVM_STAT_SUPPORT := 1 > \ No newline at end of file > diff --git a/tools/perf/arch/riscv/util/Build b/tools/perf/arch/riscv/util/Build > index 603dbb5ae4dc..d72b04f8d32b 100644 > --- a/tools/perf/arch/riscv/util/Build > +++ b/tools/perf/arch/riscv/util/Build > @@ -1,5 +1,6 @@ > perf-y += perf_regs.o > perf-y += header.o > > +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o > perf-$(CONFIG_DWARF) += dwarf-regs.o > perf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o > diff --git a/tools/perf/arch/riscv/util/kvm-stat.c b/tools/perf/arch/riscv/util/kvm-stat.c > new file mode 100644 > index 000000000000..db7183e0f09f > --- /dev/null > +++ b/tools/perf/arch/riscv/util/kvm-stat.c > @@ -0,0 +1,78 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Arch specific functions for perf kvm stat. > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#include <errno.h> > +#include <memory.h> > +#include "../../../util/evsel.h" > +#include "../../../util/kvm-stat.h" > +#include "riscv_exception_types.h" > +#include "debug.h" > + > +define_exit_reasons_table(riscv_exit_reasons, kvm_riscv_exception_class); > + > +const char *kvm_exit_reason = "scause"; > +const char *kvm_entry_trace = "kvm:kvm_entry"; > +const char *kvm_exit_trace = "kvm:kvm_exit"; > + > +const char *kvm_events_tp[] = { > + "kvm:kvm_entry", > + "kvm:kvm_exit", > + NULL, > +}; > + > +static void event_get_key(struct evsel *evsel, > + struct perf_sample *sample, > + struct event_key *key) > +{ > + key->info = 0; > + key->key = evsel__intval(evsel, sample, kvm_exit_reason); > + key->key = (int)key->key; > + key->exit_reasons = riscv_exit_reasons; > +} > + > +static bool event_begin(struct evsel *evsel, > + struct perf_sample *sample __maybe_unused, > + struct event_key *key __maybe_unused) > +{ > + return evsel__name_is(evsel, kvm_entry_trace); > +} > + > +static bool event_end(struct evsel *evsel, > + struct perf_sample *sample, > + struct event_key *key) > +{ > + if (evsel__name_is(evsel, kvm_exit_trace)) { > + event_get_key(evsel, sample, key); > + return true; > + } > + return false; > +} > + > +static struct kvm_events_ops exit_events = { > + .is_begin_event = event_begin, > + .is_end_event = event_end, > + .decode_key = exit_event_decode_key, > + .name = "VM-EXIT" > +}; > + > +struct kvm_reg_events_ops kvm_reg_events_ops[] = { > + { > + .name = "vmexit", > + .ops = &exit_events, > + }, > + { NULL, NULL }, > +}; > + > +const char * const kvm_skip_events[] = { > + NULL, > +}; > + > +int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid __maybe_unused) > +{ > + kvm->exit_reasons_isa = "riscv64"; > + return 0; > +} > diff --git a/tools/perf/arch/riscv/util/riscv_exception_types.h b/tools/perf/arch/riscv/util/riscv_exception_types.h > new file mode 100644 > index 000000000000..2e42150f72b2 > --- /dev/null > +++ b/tools/perf/arch/riscv/util/riscv_exception_types.h > @@ -0,0 +1,41 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * RISCV exception types > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#ifndef ARCH_PERF_RISCV_EXCEPTION_TYPES_H > +#define ARCH_PERF_RISCV_EXCEPTION_TYPES_H > + > +#define EXC_INST_MISALIGNED 0 > +#define EXC_INST_ACCESS 1 > +#define EXC_INST_ILLEGAL 2 > +#define EXC_BREAKPOINT 3 > +#define EXC_LOAD_MISALIGNED 4 > +#define EXC_LOAD_ACCESS 5 > +#define EXC_STORE_MISALIGNED 6 > +#define EXC_STORE_ACCESS 7 > +#define EXC_SYSCALL 8 > +#define EXC_HYPERVISOR_SYSCALL 9 > +#define EXC_SUPERVISOR_SYSCALL 10 > +#define EXC_INST_PAGE_FAULT 12 > +#define EXC_LOAD_PAGE_FAULT 13 > +#define EXC_STORE_PAGE_FAULT 15 > +#define EXC_INST_GUEST_PAGE_FAULT 20 > +#define EXC_LOAD_GUEST_PAGE_FAULT 21 > +#define EXC_VIRTUAL_INST_FAULT 22 > +#define EXC_STORE_GUEST_PAGE_FAULT 23 > + > +#define EXC(x) {EXC_##x, #x } > + > +#define kvm_riscv_exception_class \ > + (EXC(INST_MISALIGNED), EXC(INST_ACCESS), EXC(INST_ILLEGAL), \ > + EXC(BREAKPOINT), EXC(LOAD_MISALIGNED), EXC(LOAD_ACCESS), \ > + EXC(STORE_MISALIGNED), EXC(STORE_ACCESS), EXC(SYSCALL), \ > + EXC(HYPERVISOR_SYSCALL), EXC(SUPERVISOR_SYSCALL), \ > + EXC(INST_PAGE_FAULT), EXC(LOAD_PAGE_FAULT), EXC(STORE_PAGE_FAULT), \ > + EXC(INST_GUEST_PAGE_FAULT), EXC(LOAD_GUEST_PAGE_FAULT), \ > + EXC(VIRTUAL_INST_FAULT), EXC(STORE_GUEST_PAGE_FAULT)) > + > +#endif /* ARCH_PERF_RISCV_EXCEPTION_TYPES_H */ > -- > 2.37.2 > ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 2/2] perf kvm/riscv: Port perf kvm stat to RISC-V @ 2024-04-16 4:17 ` Anup Patel 0 siblings, 0 replies; 21+ messages in thread From: Anup Patel @ 2024-04-16 4:17 UTC (permalink / raw) To: peterz Cc: atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users, Shenlin Liang On Mon, Apr 15, 2024 at 8:45 AM Shenlin Liang <liangshenlin@eswincomputing.com> wrote: > > 'perf kvm stat report/record' generates a statistical analysis of KVM > events and can be used to analyze guest exit reasons. > > "report" reports statistical analysis of guest exit events. > > To record kvm events on the host: > # perf kvm stat record -a > > To report kvm VM EXIT events: > # perf kvm stat report --event=vmexit > > Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> LGTM. Reviewed-by: Anup Patel <anup@brainfault.org> @Peter Zijlstra, is it okay to take this through the KVM RISC-V tree ? Regards, Anup > --- > tools/perf/arch/riscv/Makefile | 1 + > tools/perf/arch/riscv/util/Build | 1 + > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > 4 files changed, 121 insertions(+) > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > > diff --git a/tools/perf/arch/riscv/Makefile b/tools/perf/arch/riscv/Makefile > index a8d25d005207..e1e445615536 100644 > --- a/tools/perf/arch/riscv/Makefile > +++ b/tools/perf/arch/riscv/Makefile > @@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1 > endif > PERF_HAVE_ARCH_REGS_QUERY_REGISTER_OFFSET := 1 > PERF_HAVE_JITDUMP := 1 > +HAVE_KVM_STAT_SUPPORT := 1 > \ No newline at end of file > diff --git a/tools/perf/arch/riscv/util/Build b/tools/perf/arch/riscv/util/Build > index 603dbb5ae4dc..d72b04f8d32b 100644 > --- a/tools/perf/arch/riscv/util/Build > +++ b/tools/perf/arch/riscv/util/Build > @@ -1,5 +1,6 @@ > perf-y += perf_regs.o > perf-y += header.o > > +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o > perf-$(CONFIG_DWARF) += dwarf-regs.o > perf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o > diff --git a/tools/perf/arch/riscv/util/kvm-stat.c b/tools/perf/arch/riscv/util/kvm-stat.c > new file mode 100644 > index 000000000000..db7183e0f09f > --- /dev/null > +++ b/tools/perf/arch/riscv/util/kvm-stat.c > @@ -0,0 +1,78 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Arch specific functions for perf kvm stat. > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#include <errno.h> > +#include <memory.h> > +#include "../../../util/evsel.h" > +#include "../../../util/kvm-stat.h" > +#include "riscv_exception_types.h" > +#include "debug.h" > + > +define_exit_reasons_table(riscv_exit_reasons, kvm_riscv_exception_class); > + > +const char *kvm_exit_reason = "scause"; > +const char *kvm_entry_trace = "kvm:kvm_entry"; > +const char *kvm_exit_trace = "kvm:kvm_exit"; > + > +const char *kvm_events_tp[] = { > + "kvm:kvm_entry", > + "kvm:kvm_exit", > + NULL, > +}; > + > +static void event_get_key(struct evsel *evsel, > + struct perf_sample *sample, > + struct event_key *key) > +{ > + key->info = 0; > + key->key = evsel__intval(evsel, sample, kvm_exit_reason); > + key->key = (int)key->key; > + key->exit_reasons = riscv_exit_reasons; > +} > + > +static bool event_begin(struct evsel *evsel, > + struct perf_sample *sample __maybe_unused, > + struct event_key *key __maybe_unused) > +{ > + return evsel__name_is(evsel, kvm_entry_trace); > +} > + > +static bool event_end(struct evsel *evsel, > + struct perf_sample *sample, > + struct event_key *key) > +{ > + if (evsel__name_is(evsel, kvm_exit_trace)) { > + event_get_key(evsel, sample, key); > + return true; > + } > + return false; > +} > + > +static struct kvm_events_ops exit_events = { > + .is_begin_event = event_begin, > + .is_end_event = event_end, > + .decode_key = exit_event_decode_key, > + .name = "VM-EXIT" > +}; > + > +struct kvm_reg_events_ops kvm_reg_events_ops[] = { > + { > + .name = "vmexit", > + .ops = &exit_events, > + }, > + { NULL, NULL }, > +}; > + > +const char * const kvm_skip_events[] = { > + NULL, > +}; > + > +int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid __maybe_unused) > +{ > + kvm->exit_reasons_isa = "riscv64"; > + return 0; > +} > diff --git a/tools/perf/arch/riscv/util/riscv_exception_types.h b/tools/perf/arch/riscv/util/riscv_exception_types.h > new file mode 100644 > index 000000000000..2e42150f72b2 > --- /dev/null > +++ b/tools/perf/arch/riscv/util/riscv_exception_types.h > @@ -0,0 +1,41 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * RISCV exception types > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#ifndef ARCH_PERF_RISCV_EXCEPTION_TYPES_H > +#define ARCH_PERF_RISCV_EXCEPTION_TYPES_H > + > +#define EXC_INST_MISALIGNED 0 > +#define EXC_INST_ACCESS 1 > +#define EXC_INST_ILLEGAL 2 > +#define EXC_BREAKPOINT 3 > +#define EXC_LOAD_MISALIGNED 4 > +#define EXC_LOAD_ACCESS 5 > +#define EXC_STORE_MISALIGNED 6 > +#define EXC_STORE_ACCESS 7 > +#define EXC_SYSCALL 8 > +#define EXC_HYPERVISOR_SYSCALL 9 > +#define EXC_SUPERVISOR_SYSCALL 10 > +#define EXC_INST_PAGE_FAULT 12 > +#define EXC_LOAD_PAGE_FAULT 13 > +#define EXC_STORE_PAGE_FAULT 15 > +#define EXC_INST_GUEST_PAGE_FAULT 20 > +#define EXC_LOAD_GUEST_PAGE_FAULT 21 > +#define EXC_VIRTUAL_INST_FAULT 22 > +#define EXC_STORE_GUEST_PAGE_FAULT 23 > + > +#define EXC(x) {EXC_##x, #x } > + > +#define kvm_riscv_exception_class \ > + (EXC(INST_MISALIGNED), EXC(INST_ACCESS), EXC(INST_ILLEGAL), \ > + EXC(BREAKPOINT), EXC(LOAD_MISALIGNED), EXC(LOAD_ACCESS), \ > + EXC(STORE_MISALIGNED), EXC(STORE_ACCESS), EXC(SYSCALL), \ > + EXC(HYPERVISOR_SYSCALL), EXC(SUPERVISOR_SYSCALL), \ > + EXC(INST_PAGE_FAULT), EXC(LOAD_PAGE_FAULT), EXC(STORE_PAGE_FAULT), \ > + EXC(INST_GUEST_PAGE_FAULT), EXC(LOAD_GUEST_PAGE_FAULT), \ > + EXC(VIRTUAL_INST_FAULT), EXC(STORE_GUEST_PAGE_FAULT)) > + > +#endif /* ARCH_PERF_RISCV_EXCEPTION_TYPES_H */ > -- > 2.37.2 > _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 2/2] perf kvm/riscv: Port perf kvm stat to RISC-V @ 2024-04-16 4:17 ` Anup Patel 0 siblings, 0 replies; 21+ messages in thread From: Anup Patel @ 2024-04-16 4:17 UTC (permalink / raw) To: peterz Cc: atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users, Shenlin Liang On Mon, Apr 15, 2024 at 8:45 AM Shenlin Liang <liangshenlin@eswincomputing.com> wrote: > > 'perf kvm stat report/record' generates a statistical analysis of KVM > events and can be used to analyze guest exit reasons. > > "report" reports statistical analysis of guest exit events. > > To record kvm events on the host: > # perf kvm stat record -a > > To report kvm VM EXIT events: > # perf kvm stat report --event=vmexit > > Signed-off-by: Shenlin Liang <liangshenlin@eswincomputing.com> LGTM. Reviewed-by: Anup Patel <anup@brainfault.org> @Peter Zijlstra, is it okay to take this through the KVM RISC-V tree ? Regards, Anup > --- > tools/perf/arch/riscv/Makefile | 1 + > tools/perf/arch/riscv/util/Build | 1 + > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > 4 files changed, 121 insertions(+) > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > > diff --git a/tools/perf/arch/riscv/Makefile b/tools/perf/arch/riscv/Makefile > index a8d25d005207..e1e445615536 100644 > --- a/tools/perf/arch/riscv/Makefile > +++ b/tools/perf/arch/riscv/Makefile > @@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1 > endif > PERF_HAVE_ARCH_REGS_QUERY_REGISTER_OFFSET := 1 > PERF_HAVE_JITDUMP := 1 > +HAVE_KVM_STAT_SUPPORT := 1 > \ No newline at end of file > diff --git a/tools/perf/arch/riscv/util/Build b/tools/perf/arch/riscv/util/Build > index 603dbb5ae4dc..d72b04f8d32b 100644 > --- a/tools/perf/arch/riscv/util/Build > +++ b/tools/perf/arch/riscv/util/Build > @@ -1,5 +1,6 @@ > perf-y += perf_regs.o > perf-y += header.o > > +perf-$(CONFIG_LIBTRACEEVENT) += kvm-stat.o > perf-$(CONFIG_DWARF) += dwarf-regs.o > perf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o > diff --git a/tools/perf/arch/riscv/util/kvm-stat.c b/tools/perf/arch/riscv/util/kvm-stat.c > new file mode 100644 > index 000000000000..db7183e0f09f > --- /dev/null > +++ b/tools/perf/arch/riscv/util/kvm-stat.c > @@ -0,0 +1,78 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Arch specific functions for perf kvm stat. > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#include <errno.h> > +#include <memory.h> > +#include "../../../util/evsel.h" > +#include "../../../util/kvm-stat.h" > +#include "riscv_exception_types.h" > +#include "debug.h" > + > +define_exit_reasons_table(riscv_exit_reasons, kvm_riscv_exception_class); > + > +const char *kvm_exit_reason = "scause"; > +const char *kvm_entry_trace = "kvm:kvm_entry"; > +const char *kvm_exit_trace = "kvm:kvm_exit"; > + > +const char *kvm_events_tp[] = { > + "kvm:kvm_entry", > + "kvm:kvm_exit", > + NULL, > +}; > + > +static void event_get_key(struct evsel *evsel, > + struct perf_sample *sample, > + struct event_key *key) > +{ > + key->info = 0; > + key->key = evsel__intval(evsel, sample, kvm_exit_reason); > + key->key = (int)key->key; > + key->exit_reasons = riscv_exit_reasons; > +} > + > +static bool event_begin(struct evsel *evsel, > + struct perf_sample *sample __maybe_unused, > + struct event_key *key __maybe_unused) > +{ > + return evsel__name_is(evsel, kvm_entry_trace); > +} > + > +static bool event_end(struct evsel *evsel, > + struct perf_sample *sample, > + struct event_key *key) > +{ > + if (evsel__name_is(evsel, kvm_exit_trace)) { > + event_get_key(evsel, sample, key); > + return true; > + } > + return false; > +} > + > +static struct kvm_events_ops exit_events = { > + .is_begin_event = event_begin, > + .is_end_event = event_end, > + .decode_key = exit_event_decode_key, > + .name = "VM-EXIT" > +}; > + > +struct kvm_reg_events_ops kvm_reg_events_ops[] = { > + { > + .name = "vmexit", > + .ops = &exit_events, > + }, > + { NULL, NULL }, > +}; > + > +const char * const kvm_skip_events[] = { > + NULL, > +}; > + > +int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid __maybe_unused) > +{ > + kvm->exit_reasons_isa = "riscv64"; > + return 0; > +} > diff --git a/tools/perf/arch/riscv/util/riscv_exception_types.h b/tools/perf/arch/riscv/util/riscv_exception_types.h > new file mode 100644 > index 000000000000..2e42150f72b2 > --- /dev/null > +++ b/tools/perf/arch/riscv/util/riscv_exception_types.h > @@ -0,0 +1,41 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * RISCV exception types > + * > + * Copyright 2024 Beijing ESWIN Computing Technology Co., Ltd. > + * > + */ > +#ifndef ARCH_PERF_RISCV_EXCEPTION_TYPES_H > +#define ARCH_PERF_RISCV_EXCEPTION_TYPES_H > + > +#define EXC_INST_MISALIGNED 0 > +#define EXC_INST_ACCESS 1 > +#define EXC_INST_ILLEGAL 2 > +#define EXC_BREAKPOINT 3 > +#define EXC_LOAD_MISALIGNED 4 > +#define EXC_LOAD_ACCESS 5 > +#define EXC_STORE_MISALIGNED 6 > +#define EXC_STORE_ACCESS 7 > +#define EXC_SYSCALL 8 > +#define EXC_HYPERVISOR_SYSCALL 9 > +#define EXC_SUPERVISOR_SYSCALL 10 > +#define EXC_INST_PAGE_FAULT 12 > +#define EXC_LOAD_PAGE_FAULT 13 > +#define EXC_STORE_PAGE_FAULT 15 > +#define EXC_INST_GUEST_PAGE_FAULT 20 > +#define EXC_LOAD_GUEST_PAGE_FAULT 21 > +#define EXC_VIRTUAL_INST_FAULT 22 > +#define EXC_STORE_GUEST_PAGE_FAULT 23 > + > +#define EXC(x) {EXC_##x, #x } > + > +#define kvm_riscv_exception_class \ > + (EXC(INST_MISALIGNED), EXC(INST_ACCESS), EXC(INST_ILLEGAL), \ > + EXC(BREAKPOINT), EXC(LOAD_MISALIGNED), EXC(LOAD_ACCESS), \ > + EXC(STORE_MISALIGNED), EXC(STORE_ACCESS), EXC(SYSCALL), \ > + EXC(HYPERVISOR_SYSCALL), EXC(SUPERVISOR_SYSCALL), \ > + EXC(INST_PAGE_FAULT), EXC(LOAD_PAGE_FAULT), EXC(STORE_PAGE_FAULT), \ > + EXC(INST_GUEST_PAGE_FAULT), EXC(LOAD_GUEST_PAGE_FAULT), \ > + EXC(VIRTUAL_INST_FAULT), EXC(STORE_GUEST_PAGE_FAULT)) > + > +#endif /* ARCH_PERF_RISCV_EXCEPTION_TYPES_H */ > -- > 2.37.2 > ^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv 2024-04-15 3:11 ` Shenlin Liang (?) @ 2024-04-17 0:25 ` Atish Patra -1 siblings, 0 replies; 21+ messages in thread From: Atish Patra @ 2024-04-17 0:25 UTC (permalink / raw) To: kvm-riscv On 4/14/24 20:11, Shenlin Liang wrote: > Changes from v1->v2: > - Rebased on Linux 6.9-rc3. > > 'perf kvm stat report/record' generates a statistical analysis of KVM > events and can be used to analyze guest exit reasons. This patch tries > to add stat support on riscv. > > Map the return value of trace_kvm_exit() to the specific cause of the > exception, and export it to userspace. > > It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" > and "kvm:kvm_exit", and reports statistical data which includes events > handles time, samples, and so on. > > Simple tests go below: > > # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" > Lowering default frequency rate from 4000 to 2500. > Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. > [ perf record: Woken up 18 times to write data ] > [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) > I want to test these patches but couldn't build a perf for RISC-V with libtraceevent enabled. It fails with pkg-config dependencies when I tried to build it (both via buildroot and directly from kernel source). > # ./perf kvm report > 31K kvm:kvm_entry > 31K kvm:kvm_exit > > # ./perf kvm stat record -a > [ perf record: Woken up 3 times to write data ] > [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] > > # ./perf kvm stat report --event=vmexit > Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) > STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 > LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 > VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 > INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 > SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 > LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 > INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 > HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 > > Shenlin Liang (2): > RISCV: KVM: add tracepoints for entry and exit events > perf kvm/riscv: Port perf kvm stat to RISC-V > > arch/riscv/kvm/trace.h | 67 ++++++++++++++++ > arch/riscv/kvm/vcpu.c | 7 ++ > tools/perf/arch/riscv/Makefile | 1 + > tools/perf/arch/riscv/util/Build | 1 + > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > 6 files changed, 195 insertions(+) > create mode 100644 arch/riscv/kvm/trace.h > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv @ 2024-04-17 0:25 ` Atish Patra 0 siblings, 0 replies; 21+ messages in thread From: Atish Patra @ 2024-04-17 0:25 UTC (permalink / raw) To: Shenlin Liang, anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users On 4/14/24 20:11, Shenlin Liang wrote: > Changes from v1->v2: > - Rebased on Linux 6.9-rc3. > > 'perf kvm stat report/record' generates a statistical analysis of KVM > events and can be used to analyze guest exit reasons. This patch tries > to add stat support on riscv. > > Map the return value of trace_kvm_exit() to the specific cause of the > exception, and export it to userspace. > > It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" > and "kvm:kvm_exit", and reports statistical data which includes events > handles time, samples, and so on. > > Simple tests go below: > > # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" > Lowering default frequency rate from 4000 to 2500. > Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. > [ perf record: Woken up 18 times to write data ] > [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) > I want to test these patches but couldn't build a perf for RISC-V with libtraceevent enabled. It fails with pkg-config dependencies when I tried to build it (both via buildroot and directly from kernel source). > # ./perf kvm report > 31K kvm:kvm_entry > 31K kvm:kvm_exit > > # ./perf kvm stat record -a > [ perf record: Woken up 3 times to write data ] > [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] > > # ./perf kvm stat report --event=vmexit > Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) > STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 > LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 > VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 > INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 > SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 > LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 > INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 > HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 > > Shenlin Liang (2): > RISCV: KVM: add tracepoints for entry and exit events > perf kvm/riscv: Port perf kvm stat to RISC-V > > arch/riscv/kvm/trace.h | 67 ++++++++++++++++ > arch/riscv/kvm/vcpu.c | 7 ++ > tools/perf/arch/riscv/Makefile | 1 + > tools/perf/arch/riscv/util/Build | 1 + > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > 6 files changed, 195 insertions(+) > create mode 100644 arch/riscv/kvm/trace.h > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv @ 2024-04-17 0:25 ` Atish Patra 0 siblings, 0 replies; 21+ messages in thread From: Atish Patra @ 2024-04-17 0:25 UTC (permalink / raw) To: Shenlin Liang, anup, atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users On 4/14/24 20:11, Shenlin Liang wrote: > Changes from v1->v2: > - Rebased on Linux 6.9-rc3. > > 'perf kvm stat report/record' generates a statistical analysis of KVM > events and can be used to analyze guest exit reasons. This patch tries > to add stat support on riscv. > > Map the return value of trace_kvm_exit() to the specific cause of the > exception, and export it to userspace. > > It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" > and "kvm:kvm_exit", and reports statistical data which includes events > handles time, samples, and so on. > > Simple tests go below: > > # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" > Lowering default frequency rate from 4000 to 2500. > Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. > [ perf record: Woken up 18 times to write data ] > [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) > I want to test these patches but couldn't build a perf for RISC-V with libtraceevent enabled. It fails with pkg-config dependencies when I tried to build it (both via buildroot and directly from kernel source). > # ./perf kvm report > 31K kvm:kvm_entry > 31K kvm:kvm_exit > > # ./perf kvm stat record -a > [ perf record: Woken up 3 times to write data ] > [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] > > # ./perf kvm stat report --event=vmexit > Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) > STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 > LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 > VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 > INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 > SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 > LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 > INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 > HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 > > Shenlin Liang (2): > RISCV: KVM: add tracepoints for entry and exit events > perf kvm/riscv: Port perf kvm stat to RISC-V > > arch/riscv/kvm/trace.h | 67 ++++++++++++++++ > arch/riscv/kvm/vcpu.c | 7 ++ > tools/perf/arch/riscv/Makefile | 1 + > tools/perf/arch/riscv/util/Build | 1 + > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > 6 files changed, 195 insertions(+) > create mode 100644 arch/riscv/kvm/trace.h > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > ^ permalink raw reply [flat|nested] 21+ messages in thread
* [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv 2024-04-17 0:25 ` Atish Patra (?) @ 2024-04-22 8:27 ` Shenlin Liang -1 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-22 8:27 UTC (permalink / raw) To: kvm-riscv Hi Atish, I assume that you are cross building it on X86. You need to build a pkg-config-riscv64 first. Or get a deb file from [1] if you are buiding on Ubuntu. Instead of cross building, it is recommended to build it natively. BTW, please test with V3 which I sent today. [1] https://answers.launchpad.net/~ci-train-ppa-service/+archive/ubuntu/3719-deletedppa/+build/16823862/+files/pkg-config-riscv64-linux-gnu_7.4.0-1ubuntu1.3_amd64.deb Thanks. Shenlin > > On 4/14/24 20:11, Shenlin Liang wrote: > > Changes from v1->v2: > > - Rebased on Linux 6.9-rc3. > > > > 'perf kvm stat report/record' generates a statistical analysis of KVM > > events and can be used to analyze guest exit reasons. This patch tries > > to add stat support on riscv. > > > > Map the return value of trace_kvm_exit() to the specific cause of the > > exception, and export it to userspace. > > > > It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" > > and "kvm:kvm_exit", and reports statistical data which includes events > > handles time, samples, and so on. > > > > Simple tests go below: > > > > # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" > > Lowering default frequency rate from 4000 to 2500. > > Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. > > [ perf record: Woken up 18 times to write data ] > > [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) > > > > I want to test these patches but couldn't build a perf for RISC-V with > libtraceevent enabled. It fails with pkg-config dependencies when I > tried to build it (both via buildroot and directly from kernel source). > > > # ./perf kvm report > > 31K kvm:kvm_entry > > 31K kvm:kvm_exit > > > > # ./perf kvm stat record -a > > [ perf record: Woken up 3 times to write data ] > > [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] > > > > # ./perf kvm stat report --event=vmexit > > Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) > > STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 > > LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 > > VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 > > INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 > > SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 > > LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 > > INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 > > HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 > > > > Shenlin Liang (2): > > RISCV: KVM: add tracepoints for entry and exit events > > perf kvm/riscv: Port perf kvm stat to RISC-V > > > > arch/riscv/kvm/trace.h | 67 ++++++++++++++++ > > arch/riscv/kvm/vcpu.c | 7 ++ > > tools/perf/arch/riscv/Makefile | 1 + > > tools/perf/arch/riscv/util/Build | 1 + > > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > > 6 files changed, 195 insertions(+) > > create mode 100644 arch/riscv/kvm/trace.h > > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > > ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv @ 2024-04-22 8:27 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-22 8:27 UTC (permalink / raw) To: atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users, atishp Hi Atish, I assume that you are cross building it on X86. You need to build a pkg-config-riscv64 first. Or get a deb file from [1] if you are buiding on Ubuntu. Instead of cross building, it is recommended to build it natively. BTW, please test with V3 which I sent today. [1] https://answers.launchpad.net/~ci-train-ppa-service/+archive/ubuntu/3719-deletedppa/+build/16823862/+files/pkg-config-riscv64-linux-gnu_7.4.0-1ubuntu1.3_amd64.deb Thanks. Shenlin > > On 4/14/24 20:11, Shenlin Liang wrote: > > Changes from v1->v2: > > - Rebased on Linux 6.9-rc3. > > > > 'perf kvm stat report/record' generates a statistical analysis of KVM > > events and can be used to analyze guest exit reasons. This patch tries > > to add stat support on riscv. > > > > Map the return value of trace_kvm_exit() to the specific cause of the > > exception, and export it to userspace. > > > > It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" > > and "kvm:kvm_exit", and reports statistical data which includes events > > handles time, samples, and so on. > > > > Simple tests go below: > > > > # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" > > Lowering default frequency rate from 4000 to 2500. > > Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. > > [ perf record: Woken up 18 times to write data ] > > [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) > > > > I want to test these patches but couldn't build a perf for RISC-V with > libtraceevent enabled. It fails with pkg-config dependencies when I > tried to build it (both via buildroot and directly from kernel source). > > > # ./perf kvm report > > 31K kvm:kvm_entry > > 31K kvm:kvm_exit > > > > # ./perf kvm stat record -a > > [ perf record: Woken up 3 times to write data ] > > [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] > > > > # ./perf kvm stat report --event=vmexit > > Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) > > STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 > > LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 > > VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 > > INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 > > SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 > > LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 > > INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 > > HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 > > > > Shenlin Liang (2): > > RISCV: KVM: add tracepoints for entry and exit events > > perf kvm/riscv: Port perf kvm stat to RISC-V > > > > arch/riscv/kvm/trace.h | 67 ++++++++++++++++ > > arch/riscv/kvm/vcpu.c | 7 ++ > > tools/perf/arch/riscv/Makefile | 1 + > > tools/perf/arch/riscv/util/Build | 1 + > > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > > 6 files changed, 195 insertions(+) > > create mode 100644 arch/riscv/kvm/trace.h > > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > > _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply [flat|nested] 21+ messages in thread
* Re: [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv @ 2024-04-22 8:27 ` Shenlin Liang 0 siblings, 0 replies; 21+ messages in thread From: Shenlin Liang @ 2024-04-22 8:27 UTC (permalink / raw) To: atishp, paul.walmsley, palmer, aou, kvm, kvm-riscv, linux-riscv, linux-kernel, peterz, mingo, acme, namhyung, mark.rutland, alexander.shishkin, jolsa, irogers, adrian.hunter, linux-perf-users, atishp Hi Atish, I assume that you are cross building it on X86. You need to build a pkg-config-riscv64 first. Or get a deb file from [1] if you are buiding on Ubuntu. Instead of cross building, it is recommended to build it natively. BTW, please test with V3 which I sent today. [1] https://answers.launchpad.net/~ci-train-ppa-service/+archive/ubuntu/3719-deletedppa/+build/16823862/+files/pkg-config-riscv64-linux-gnu_7.4.0-1ubuntu1.3_amd64.deb Thanks. Shenlin > > On 4/14/24 20:11, Shenlin Liang wrote: > > Changes from v1->v2: > > - Rebased on Linux 6.9-rc3. > > > > 'perf kvm stat report/record' generates a statistical analysis of KVM > > events and can be used to analyze guest exit reasons. This patch tries > > to add stat support on riscv. > > > > Map the return value of trace_kvm_exit() to the specific cause of the > > exception, and export it to userspace. > > > > It records on two available KVM tracepoints for riscv: "kvm:kvm_entry" > > and "kvm:kvm_exit", and reports statistical data which includes events > > handles time, samples, and so on. > > > > Simple tests go below: > > > > # ./perf kvm record -e "kvm:kvm_entry" -e "kvm:kvm_exit" > > Lowering default frequency rate from 4000 to 2500. > > Please consider tweaking /proc/sys/kernel/perf_event_max_sample_rate. > > [ perf record: Woken up 18 times to write data ] > > [ perf record: Captured and wrote 5.433 MB perf.data.guest (62519 samples) > > > > I want to test these patches but couldn't build a perf for RISC-V with > libtraceevent enabled. It fails with pkg-config dependencies when I > tried to build it (both via buildroot and directly from kernel source). > > > # ./perf kvm report > > 31K kvm:kvm_entry > > 31K kvm:kvm_exit > > > > # ./perf kvm stat record -a > > [ perf record: Woken up 3 times to write data ] > > [ perf record: Captured and wrote 8.502 MB perf.data.guest (99338 samples) ] > > > > # ./perf kvm stat report --event=vmexit > > Event name Samples Sample% Time (ns) Time% Max Time (ns) Min Time (ns) Mean Time (ns) > > STORE_GUEST_PAGE_FAULT 26968 54.00% 2003031800 40.00% 3361400 27600 74274 > > LOAD_GUEST_PAGE_FAULT 17645 35.00% 1153338100 23.00% 2513400 30800 65363 > > VIRTUAL_INST_FAULT 1247 2.00% 340820800 6.00% 1190800 43300 273312 > > INST_GUEST_PAGE_FAULT 1128 2.00% 340645800 6.00% 2123200 30200 301990 > > SUPERVISOR_SYSCALL 1019 2.00% 245989900 4.00% 1851500 29300 241403 > > LOAD_ACCESS 986 1.00% 671556200 13.00% 4180200 100700 681091 > > INST_ACCESS 655 1.00% 170054800 3.00% 1808300 54600 259625 > > HYPERVISOR_SYSCALL 21 0.00% 4276400 0.00% 716500 116000 203638 > > > > Shenlin Liang (2): > > RISCV: KVM: add tracepoints for entry and exit events > > perf kvm/riscv: Port perf kvm stat to RISC-V > > > > arch/riscv/kvm/trace.h | 67 ++++++++++++++++ > > arch/riscv/kvm/vcpu.c | 7 ++ > > tools/perf/arch/riscv/Makefile | 1 + > > tools/perf/arch/riscv/util/Build | 1 + > > tools/perf/arch/riscv/util/kvm-stat.c | 78 +++++++++++++++++++ > > .../arch/riscv/util/riscv_exception_types.h | 41 ++++++++++ > > 6 files changed, 195 insertions(+) > > create mode 100644 arch/riscv/kvm/trace.h > > create mode 100644 tools/perf/arch/riscv/util/kvm-stat.c > > create mode 100644 tools/perf/arch/riscv/util/riscv_exception_types.h > > ^ permalink raw reply [flat|nested] 21+ messages in thread
end of thread, other threads:[~2024-04-22 8:29 UTC | newest] Thread overview: 21+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2024-04-15 3:11 [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv Shenlin Liang 2024-04-15 3:11 ` Shenlin Liang 2024-04-15 3:11 ` Shenlin Liang 2024-04-15 3:11 ` [PATCH v2 1/2] RISCV: KVM: add tracepoints for entry and exit events Shenlin Liang 2024-04-15 3:11 ` Shenlin Liang 2024-04-15 3:11 ` Shenlin Liang 2024-04-16 3:44 ` Anup Patel 2024-04-16 3:44 ` Anup Patel 2024-04-16 3:44 ` Anup Patel 2024-04-15 3:11 ` [PATCH v2 2/2] perf kvm/riscv: Port perf kvm stat to RISC-V Shenlin Liang 2024-04-15 3:11 ` Shenlin Liang 2024-04-15 3:11 ` Shenlin Liang 2024-04-16 4:17 ` Anup Patel 2024-04-16 4:17 ` Anup Patel 2024-04-16 4:17 ` Anup Patel 2024-04-17 0:25 ` [PATCH v2 0/2] perf kvm: Add kvm stat support on riscv Atish Patra 2024-04-17 0:25 ` Atish Patra 2024-04-17 0:25 ` Atish Patra 2024-04-22 8:27 ` Shenlin Liang 2024-04-22 8:27 ` Shenlin Liang 2024-04-22 8:27 ` Shenlin Liang
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.