perf kwork: Implement BPF trace
'perf record' generates perf.data, which generates extra interrupts
for hard disk, amount of data to be collected increases with time.
Using eBPF trace can process the data in kernel, which solves the
preceding two problems.
Add -b/--use-bpf option for latency and report to support
tracing kwork events using eBPF:
1. Create bpf prog and attach to tracepoints,
2. Start tracing after command is entered,
3. After user hit "ctrl+c", stop tracing and report,
4. Support CPU and name filtering.
This commit implements the framework code and
does not add specific event support.
Test cases:
# perf kwork rep -h
Usage: perf kwork report [<options>]
-b, --use-bpf Use BPF to measure kwork runtime
-C, --cpu <cpu> list of cpus to profile
-i, --input <file> input file name
-n, --name <name> event name to profile
-s, --sort <key[,key2...]>
sort by key(s): runtime, max, count
-S, --with-summary Show summary with statistics
--time <str> Time span for analysis (start,stop)
# perf kwork lat -h
Usage: perf kwork latency [<options>]
-b, --use-bpf Use BPF to measure kwork latency
-C, --cpu <cpu> list of cpus to profile
-i, --input <file> input file name
-n, --name <name> event name to profile
-s, --sort <key[,key2...]>
sort by key(s): avg, max, count
--time <str> Time span for analysis (start,stop)
# perf kwork lat -b
Unsupported bpf trace class irq
# perf kwork rep -b
Unsupported bpf trace class irq
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220709015033.38326-15-yangjihong1@huawei.com
[ Simplify work_findnew() ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2022-07-09 09:50:30 +08:00
|
|
|
// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
|
|
|
|
// Copyright (c) 2022, Huawei
|
|
|
|
|
|
|
|
#include "vmlinux.h"
|
|
|
|
#include <bpf/bpf_helpers.h>
|
|
|
|
#include <bpf/bpf_tracing.h>
|
|
|
|
|
|
|
|
#define KWORK_COUNT 100
|
|
|
|
#define MAX_KWORKNAME 128
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This should be in sync with "util/kwork.h"
|
|
|
|
*/
|
|
|
|
enum kwork_class_type {
|
|
|
|
KWORK_CLASS_IRQ,
|
|
|
|
KWORK_CLASS_SOFTIRQ,
|
|
|
|
KWORK_CLASS_WORKQUEUE,
|
|
|
|
KWORK_CLASS_MAX,
|
|
|
|
};
|
|
|
|
|
|
|
|
struct work_key {
|
|
|
|
__u32 type;
|
|
|
|
__u32 cpu;
|
|
|
|
__u64 id;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct report_data {
|
|
|
|
__u64 nr;
|
|
|
|
__u64 total_time;
|
|
|
|
__u64 max_time;
|
|
|
|
__u64 max_time_start;
|
|
|
|
__u64 max_time_end;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct {
|
|
|
|
__uint(type, BPF_MAP_TYPE_HASH);
|
|
|
|
__uint(key_size, sizeof(struct work_key));
|
|
|
|
__uint(value_size, MAX_KWORKNAME);
|
|
|
|
__uint(max_entries, KWORK_COUNT);
|
|
|
|
} perf_kwork_names SEC(".maps");
|
|
|
|
|
|
|
|
struct {
|
|
|
|
__uint(type, BPF_MAP_TYPE_HASH);
|
|
|
|
__uint(key_size, sizeof(struct work_key));
|
|
|
|
__uint(value_size, sizeof(__u64));
|
|
|
|
__uint(max_entries, KWORK_COUNT);
|
|
|
|
} perf_kwork_time SEC(".maps");
|
|
|
|
|
|
|
|
struct {
|
|
|
|
__uint(type, BPF_MAP_TYPE_HASH);
|
|
|
|
__uint(key_size, sizeof(struct work_key));
|
|
|
|
__uint(value_size, sizeof(struct report_data));
|
|
|
|
__uint(max_entries, KWORK_COUNT);
|
|
|
|
} perf_kwork_report SEC(".maps");
|
|
|
|
|
|
|
|
struct {
|
|
|
|
__uint(type, BPF_MAP_TYPE_HASH);
|
|
|
|
__uint(key_size, sizeof(__u32));
|
|
|
|
__uint(value_size, sizeof(__u8));
|
|
|
|
__uint(max_entries, 1);
|
|
|
|
} perf_kwork_cpu_filter SEC(".maps");
|
|
|
|
|
|
|
|
struct {
|
|
|
|
__uint(type, BPF_MAP_TYPE_ARRAY);
|
|
|
|
__uint(key_size, sizeof(__u32));
|
|
|
|
__uint(value_size, MAX_KWORKNAME);
|
|
|
|
__uint(max_entries, 1);
|
|
|
|
} perf_kwork_name_filter SEC(".maps");
|
|
|
|
|
|
|
|
int enabled = 0;
|
perf kwork: Constify control data for BPF
The control knobs set before loading BPF programs should be declared as
'const volatile' so that it can be optimized by the BPF core.
Committer testing:
root@x1:~# perf kwork report --use-bpf
Starting trace, Hit <Ctrl+C> to stop and report
^C
Kwork Name | Cpu | Total Runtime | Count | Max runtime | Max runtime start | Max runtime end |
--------------------------------------------------------------------------------------------------------------------------------
(w)intel_atomic_commit_work [ | 0009 | 18.680 ms | 2 | 18.553 ms | 362410.681580 s | 362410.700133 s |
(w)pm_runtime_work | 0007 | 13.300 ms | 1 | 13.300 ms | 362410.254996 s | 362410.268295 s |
(w)intel_atomic_commit_work [ | 0009 | 9.846 ms | 2 | 9.717 ms | 362410.172352 s | 362410.182069 s |
(w)acpi_ec_event_processor | 0002 | 8.106 ms | 1 | 8.106 ms | 362410.463187 s | 362410.471293 s |
(s)SCHED:7 | 0000 | 1.351 ms | 106 | 0.063 ms | 362410.658017 s | 362410.658080 s |
i915:157 | 0008 | 0.994 ms | 13 | 0.361 ms | 362411.222125 s | 362411.222486 s |
(s)SCHED:7 | 0001 | 0.703 ms | 98 | 0.047 ms | 362410.245004 s | 362410.245051 s |
(s)SCHED:7 | 0005 | 0.674 ms | 42 | 0.074 ms | 362411.483039 s | 362411.483113 s |
(s)NET_RX:3 | 0001 | 0.556 ms | 10 | 0.079 ms | 362411.066388 s | 362411.066467 s |
<SNIP>
root@x1:~# perf trace -e bpf --max-events 5 perf kwork report --use-bpf
0.000 ( 0.016 ms): perf/2948007 bpf(cmd: 36, uattr: 0x7ffededa6660, size: 8) = -1 EOPNOTSUPP (Operation not supported)
0.026 ( 0.106 ms): perf/2948007 bpf(cmd: PROG_LOAD, uattr: 0x7ffededa6390, size: 148) = 12
0.152 ( 0.032 ms): perf/2948007 bpf(cmd: PROG_LOAD, uattr: 0x7ffededa6450, size: 148) = 12
26.247 ( 0.138 ms): perf/2948007 bpf(cmd: PROG_LOAD, uattr: 0x7ffededa6300, size: 148) = 12
26.396 ( 0.012 ms): perf/2948007 bpf(uattr: 0x7ffededa64b0, size: 80) = 12
Starting trace, Hit <Ctrl+C> to stop and report
root@x1:~#
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <song@kernel.org>
Cc: Yang Jihong <yangjihong@bytedance.com>
Link: https://lore.kernel.org/r/20240902200515.2103769-4-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-02 13:05:13 -07:00
|
|
|
|
|
|
|
const volatile int has_cpu_filter = 0;
|
|
|
|
const volatile int has_name_filter = 0;
|
perf kwork: Implement BPF trace
'perf record' generates perf.data, which generates extra interrupts
for hard disk, amount of data to be collected increases with time.
Using eBPF trace can process the data in kernel, which solves the
preceding two problems.
Add -b/--use-bpf option for latency and report to support
tracing kwork events using eBPF:
1. Create bpf prog and attach to tracepoints,
2. Start tracing after command is entered,
3. After user hit "ctrl+c", stop tracing and report,
4. Support CPU and name filtering.
This commit implements the framework code and
does not add specific event support.
Test cases:
# perf kwork rep -h
Usage: perf kwork report [<options>]
-b, --use-bpf Use BPF to measure kwork runtime
-C, --cpu <cpu> list of cpus to profile
-i, --input <file> input file name
-n, --name <name> event name to profile
-s, --sort <key[,key2...]>
sort by key(s): runtime, max, count
-S, --with-summary Show summary with statistics
--time <str> Time span for analysis (start,stop)
# perf kwork lat -h
Usage: perf kwork latency [<options>]
-b, --use-bpf Use BPF to measure kwork latency
-C, --cpu <cpu> list of cpus to profile
-i, --input <file> input file name
-n, --name <name> event name to profile
-s, --sort <key[,key2...]>
sort by key(s): avg, max, count
--time <str> Time span for analysis (start,stop)
# perf kwork lat -b
Unsupported bpf trace class irq
# perf kwork rep -b
Unsupported bpf trace class irq
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220709015033.38326-15-yangjihong1@huawei.com
[ Simplify work_findnew() ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2022-07-09 09:50:30 +08:00
|
|
|
|
2022-07-09 09:50:31 +08:00
|
|
|
static __always_inline int local_strncmp(const char *s1,
|
|
|
|
unsigned int sz, const char *s2)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
unsigned int i;
|
|
|
|
|
|
|
|
for (i = 0; i < sz; i++) {
|
|
|
|
ret = (unsigned char)s1[i] - (unsigned char)s2[i];
|
2025-03-14 11:10:13 +08:00
|
|
|
if (ret || !s1[i])
|
2022-07-09 09:50:31 +08:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline int trace_event_match(struct work_key *key, char *name)
|
|
|
|
{
|
|
|
|
__u8 *cpu_val;
|
|
|
|
char *name_val;
|
|
|
|
__u32 zero = 0;
|
|
|
|
__u32 cpu = bpf_get_smp_processor_id();
|
|
|
|
|
|
|
|
if (!enabled)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (has_cpu_filter) {
|
|
|
|
cpu_val = bpf_map_lookup_elem(&perf_kwork_cpu_filter, &cpu);
|
|
|
|
if (!cpu_val)
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (has_name_filter && (name != NULL)) {
|
|
|
|
name_val = bpf_map_lookup_elem(&perf_kwork_name_filter, &zero);
|
|
|
|
if (name_val &&
|
|
|
|
(local_strncmp(name_val, MAX_KWORKNAME, name) != 0)) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline void do_update_time(void *map, struct work_key *key,
|
|
|
|
__u64 time_start, __u64 time_end)
|
|
|
|
{
|
|
|
|
struct report_data zero, *data;
|
|
|
|
__s64 delta = time_end - time_start;
|
|
|
|
|
|
|
|
if (delta < 0)
|
|
|
|
return;
|
|
|
|
|
|
|
|
data = bpf_map_lookup_elem(map, key);
|
|
|
|
if (!data) {
|
|
|
|
__builtin_memset(&zero, 0, sizeof(zero));
|
|
|
|
bpf_map_update_elem(map, key, &zero, BPF_NOEXIST);
|
|
|
|
data = bpf_map_lookup_elem(map, key);
|
|
|
|
if (!data)
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((delta > data->max_time) ||
|
|
|
|
(data->max_time == 0)) {
|
|
|
|
data->max_time = delta;
|
|
|
|
data->max_time_start = time_start;
|
|
|
|
data->max_time_end = time_end;
|
|
|
|
}
|
|
|
|
|
|
|
|
data->total_time += delta;
|
|
|
|
data->nr++;
|
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline void do_update_timestart(void *map, struct work_key *key)
|
|
|
|
{
|
|
|
|
__u64 ts = bpf_ktime_get_ns();
|
|
|
|
|
|
|
|
bpf_map_update_elem(map, key, &ts, BPF_ANY);
|
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline void do_update_timeend(void *report_map, void *time_map,
|
|
|
|
struct work_key *key)
|
|
|
|
{
|
|
|
|
__u64 *time = bpf_map_lookup_elem(time_map, key);
|
|
|
|
|
|
|
|
if (time) {
|
|
|
|
bpf_map_delete_elem(time_map, key);
|
|
|
|
do_update_time(report_map, key, *time, bpf_ktime_get_ns());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline void do_update_name(void *map,
|
|
|
|
struct work_key *key, char *name)
|
|
|
|
{
|
|
|
|
if (!bpf_map_lookup_elem(map, key))
|
|
|
|
bpf_map_update_elem(map, key, name, BPF_ANY);
|
|
|
|
}
|
|
|
|
|
2022-07-09 09:50:33 +08:00
|
|
|
static __always_inline int update_timestart(void *map, struct work_key *key)
|
|
|
|
{
|
|
|
|
if (!trace_event_match(key, NULL))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
do_update_timestart(map, key);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2022-07-09 09:50:31 +08:00
|
|
|
static __always_inline int update_timestart_and_name(void *time_map,
|
|
|
|
void *names_map,
|
|
|
|
struct work_key *key,
|
|
|
|
char *name)
|
|
|
|
{
|
|
|
|
if (!trace_event_match(key, name))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
do_update_timestart(time_map, key);
|
|
|
|
do_update_name(names_map, key, name);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static __always_inline int update_timeend(void *report_map,
|
|
|
|
void *time_map, struct work_key *key)
|
|
|
|
{
|
|
|
|
if (!trace_event_match(key, NULL))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
do_update_timeend(report_map, time_map, key);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2022-07-09 09:50:33 +08:00
|
|
|
static __always_inline int update_timeend_and_name(void *report_map,
|
|
|
|
void *time_map,
|
|
|
|
void *names_map,
|
|
|
|
struct work_key *key,
|
|
|
|
char *name)
|
|
|
|
{
|
|
|
|
if (!trace_event_match(key, name))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
do_update_timeend(report_map, time_map, key);
|
|
|
|
do_update_name(names_map, key, name);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2022-07-09 09:50:31 +08:00
|
|
|
SEC("tracepoint/irq/irq_handler_entry")
|
|
|
|
int report_irq_handler_entry(struct trace_event_raw_irq_handler_entry *ctx)
|
|
|
|
{
|
|
|
|
char name[MAX_KWORKNAME];
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_IRQ,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->irq,
|
|
|
|
};
|
|
|
|
void *name_addr = (void *)ctx + (ctx->__data_loc_name & 0xffff);
|
|
|
|
|
|
|
|
bpf_probe_read_kernel_str(name, sizeof(name), name_addr);
|
|
|
|
|
|
|
|
return update_timestart_and_name(&perf_kwork_time,
|
|
|
|
&perf_kwork_names, &key, name);
|
|
|
|
}
|
|
|
|
|
|
|
|
SEC("tracepoint/irq/irq_handler_exit")
|
|
|
|
int report_irq_handler_exit(struct trace_event_raw_irq_handler_exit *ctx)
|
|
|
|
{
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_IRQ,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->irq,
|
|
|
|
};
|
|
|
|
|
|
|
|
return update_timeend(&perf_kwork_report, &perf_kwork_time, &key);
|
|
|
|
}
|
|
|
|
|
2022-07-09 09:50:32 +08:00
|
|
|
static char softirq_name_list[NR_SOFTIRQS][MAX_KWORKNAME] = {
|
|
|
|
{ "HI" },
|
|
|
|
{ "TIMER" },
|
|
|
|
{ "NET_TX" },
|
|
|
|
{ "NET_RX" },
|
|
|
|
{ "BLOCK" },
|
|
|
|
{ "IRQ_POLL" },
|
|
|
|
{ "TASKLET" },
|
|
|
|
{ "SCHED" },
|
|
|
|
{ "HRTIMER" },
|
|
|
|
{ "RCU" },
|
|
|
|
};
|
|
|
|
|
|
|
|
SEC("tracepoint/irq/softirq_entry")
|
|
|
|
int report_softirq_entry(struct trace_event_raw_softirq *ctx)
|
|
|
|
{
|
|
|
|
unsigned int vec = ctx->vec;
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_SOFTIRQ,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)vec,
|
|
|
|
};
|
|
|
|
|
|
|
|
if (vec < NR_SOFTIRQS) {
|
|
|
|
return update_timestart_and_name(&perf_kwork_time,
|
|
|
|
&perf_kwork_names, &key,
|
|
|
|
softirq_name_list[vec]);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
SEC("tracepoint/irq/softirq_exit")
|
|
|
|
int report_softirq_exit(struct trace_event_raw_softirq *ctx)
|
|
|
|
{
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_SOFTIRQ,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->vec,
|
|
|
|
};
|
|
|
|
|
|
|
|
return update_timeend(&perf_kwork_report, &perf_kwork_time, &key);
|
|
|
|
}
|
|
|
|
|
|
|
|
SEC("tracepoint/irq/softirq_raise")
|
|
|
|
int latency_softirq_raise(struct trace_event_raw_softirq *ctx)
|
|
|
|
{
|
|
|
|
unsigned int vec = ctx->vec;
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_SOFTIRQ,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)vec,
|
|
|
|
};
|
|
|
|
|
|
|
|
if (vec < NR_SOFTIRQS) {
|
|
|
|
return update_timestart_and_name(&perf_kwork_time,
|
|
|
|
&perf_kwork_names, &key,
|
|
|
|
softirq_name_list[vec]);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
SEC("tracepoint/irq/softirq_entry")
|
|
|
|
int latency_softirq_entry(struct trace_event_raw_softirq *ctx)
|
|
|
|
{
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_SOFTIRQ,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->vec,
|
|
|
|
};
|
|
|
|
|
|
|
|
return update_timeend(&perf_kwork_report, &perf_kwork_time, &key);
|
|
|
|
}
|
|
|
|
|
2022-07-09 09:50:33 +08:00
|
|
|
SEC("tracepoint/workqueue/workqueue_execute_start")
|
|
|
|
int report_workqueue_execute_start(struct trace_event_raw_workqueue_execute_start *ctx)
|
|
|
|
{
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_WORKQUEUE,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->work,
|
|
|
|
};
|
|
|
|
|
|
|
|
return update_timestart(&perf_kwork_time, &key);
|
|
|
|
}
|
|
|
|
|
|
|
|
SEC("tracepoint/workqueue/workqueue_execute_end")
|
|
|
|
int report_workqueue_execute_end(struct trace_event_raw_workqueue_execute_end *ctx)
|
|
|
|
{
|
|
|
|
char name[MAX_KWORKNAME];
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_WORKQUEUE,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->work,
|
|
|
|
};
|
|
|
|
unsigned long long func_addr = (unsigned long long)ctx->function;
|
|
|
|
|
|
|
|
__builtin_memset(name, 0, sizeof(name));
|
|
|
|
bpf_snprintf(name, sizeof(name), "%ps", &func_addr, sizeof(func_addr));
|
|
|
|
|
|
|
|
return update_timeend_and_name(&perf_kwork_report, &perf_kwork_time,
|
|
|
|
&perf_kwork_names, &key, name);
|
|
|
|
}
|
|
|
|
|
|
|
|
SEC("tracepoint/workqueue/workqueue_activate_work")
|
|
|
|
int latency_workqueue_activate_work(struct trace_event_raw_workqueue_activate_work *ctx)
|
|
|
|
{
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_WORKQUEUE,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->work,
|
|
|
|
};
|
|
|
|
|
|
|
|
return update_timestart(&perf_kwork_time, &key);
|
|
|
|
}
|
|
|
|
|
|
|
|
SEC("tracepoint/workqueue/workqueue_execute_start")
|
|
|
|
int latency_workqueue_execute_start(struct trace_event_raw_workqueue_execute_start *ctx)
|
|
|
|
{
|
|
|
|
char name[MAX_KWORKNAME];
|
|
|
|
struct work_key key = {
|
|
|
|
.type = KWORK_CLASS_WORKQUEUE,
|
|
|
|
.cpu = bpf_get_smp_processor_id(),
|
|
|
|
.id = (__u64)ctx->work,
|
|
|
|
};
|
|
|
|
unsigned long long func_addr = (unsigned long long)ctx->function;
|
|
|
|
|
|
|
|
__builtin_memset(name, 0, sizeof(name));
|
|
|
|
bpf_snprintf(name, sizeof(name), "%ps", &func_addr, sizeof(func_addr));
|
|
|
|
|
|
|
|
return update_timeend_and_name(&perf_kwork_report, &perf_kwork_time,
|
|
|
|
&perf_kwork_names, &key, name);
|
|
|
|
}
|
|
|
|
|
perf kwork: Implement BPF trace
'perf record' generates perf.data, which generates extra interrupts
for hard disk, amount of data to be collected increases with time.
Using eBPF trace can process the data in kernel, which solves the
preceding two problems.
Add -b/--use-bpf option for latency and report to support
tracing kwork events using eBPF:
1. Create bpf prog and attach to tracepoints,
2. Start tracing after command is entered,
3. After user hit "ctrl+c", stop tracing and report,
4. Support CPU and name filtering.
This commit implements the framework code and
does not add specific event support.
Test cases:
# perf kwork rep -h
Usage: perf kwork report [<options>]
-b, --use-bpf Use BPF to measure kwork runtime
-C, --cpu <cpu> list of cpus to profile
-i, --input <file> input file name
-n, --name <name> event name to profile
-s, --sort <key[,key2...]>
sort by key(s): runtime, max, count
-S, --with-summary Show summary with statistics
--time <str> Time span for analysis (start,stop)
# perf kwork lat -h
Usage: perf kwork latency [<options>]
-b, --use-bpf Use BPF to measure kwork latency
-C, --cpu <cpu> list of cpus to profile
-i, --input <file> input file name
-n, --name <name> event name to profile
-s, --sort <key[,key2...]>
sort by key(s): avg, max, count
--time <str> Time span for analysis (start,stop)
# perf kwork lat -b
Unsupported bpf trace class irq
# perf kwork rep -b
Unsupported bpf trace class irq
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220709015033.38326-15-yangjihong1@huawei.com
[ Simplify work_findnew() ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2022-07-09 09:50:30 +08:00
|
|
|
char LICENSE[] SEC("license") = "Dual BSD/GPL";
|