perf sched stats: Add schedstat v17 support

The /proc/schedstat file output is standardized with version number.

Add support to record and raw dump v17 version layout.

Version 17 of schedstats removed 'lb_imbalance' field as it has no
significance anymore and instead added more relevant fields namely
'lb_imbalance_load', 'lb_imbalance_util', 'lb_imbalance_task' and
'lb_imbalance_misfit'.

The domain field prints the name of the corresponding sched domain from
this version onwards.

Co-developed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Swapnil Sapkal <swapnil.sapkal@amd.com>
Tested-by: Chen Yu <yu.c.chen@intel.com>
Acked-by: Ian Rogers <irogers@google.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Anubhav Shelat <ashelat@redhat.com>
Cc: Ben Gainey <ben.gainey@arm.com>
Cc: Blake Jones <blakejones@google.com>
Cc: Chun-Tse Shao <ctshao@google.com>
Cc: David Vernet <void@manifault.com>
Cc: Dmitriy Vyukov <dvyukov@google.com>
Cc: Dr. David Alan Gilbert <linux@treblig.org>
Cc: Gautham Shenoy <gautham.shenoy@amd.com>
Cc: Graham Woodward <graham.woodward@arm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Madadi Vineeth Reddy <vineethr@linux.ibm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Sandipan Das <sandipan.das@amd.com>
Cc: Santosh Shukla <santosh.shukla@amd.com>
Cc: Shrikanth Hegde <sshegde@linux.ibm.com>
Cc: Steven Rostedt (VMware) <rostedt@goodmis.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Falcon <thomas.falcon@intel.com>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Yang Jihong <yangjihong@bytedance.com>
Cc: Yujie Liu <yujie.liu@intel.com>
Cc: Zhongqiu Han <quic_zhonhan@quicinc.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

diff --git a/tools/lib/perf/Makefile b/tools/lib/perf/Makefile
index 965e066fd7806e26216aaac502d0ee89868bb1a1..27e6490f64dc4ff7c38959fb3d81bc25d21c4f59 100644 (file)
--- a/tools/lib/perf/Makefile
+++ b/tools/lib/perf/Makefile
@@ -179,7 +179,7 @@ install_lib: libs
                cp -fpR $(LIBPERF_ALL) $(DESTDIR)$(libdir_SQ)
 
 HDRS := bpf_perf.h core.h cpumap.h threadmap.h evlist.h evsel.h event.h mmap.h
-HDRS += schedstat-v15.h schedstat-v16.h
+HDRS += schedstat-v15.h schedstat-v16.h schedstat-v17.h
 INTERNAL_HDRS := cpumap.h evlist.h evsel.h lib.h mmap.h rc_check.h threadmap.h xyarray.h
 
 INSTALL_HDRS_PFX := $(DESTDIR)$(prefix)/include/perf

diff --git a/tools/lib/perf/include/perf/event.h b/tools/lib/perf/include/perf/event.h
index bd4d507ea8abba4c0b66645cb41f0889bb94db61..9043dc72b5d68d586a296bc921e8a09de99f7ec0 100644 (file)
--- a/tools/lib/perf/include/perf/event.h
+++ b/tools/lib/perf/include/perf/event.h
@@ -508,6 +508,12 @@ struct perf_record_schedstat_cpu_v16 {
 #undef CPU_FIELD
 };
 
+struct perf_record_schedstat_cpu_v17 {
+#define CPU_FIELD(_type, _name, _desc, _format, _is_pct, _pct_of, _ver)                _type _name
+#include "schedstat-v17.h"
+#undef CPU_FIELD
+};
+
 struct perf_record_schedstat_cpu {
        struct perf_event_header header;
        __u64                    timestamp;
@@ -518,6 +524,7 @@ struct perf_record_schedstat_cpu {
        union {
                struct perf_record_schedstat_cpu_v15 v15;
                struct perf_record_schedstat_cpu_v16 v16;
+               struct perf_record_schedstat_cpu_v17 v17;
        };
 };
 
@@ -533,6 +540,12 @@ struct perf_record_schedstat_domain_v16 {
 #undef DOMAIN_FIELD
 };
 
+struct perf_record_schedstat_domain_v17 {
+#define DOMAIN_FIELD(_type, _name, _desc, _format, _is_jiffies, _ver)          _type _name
+#include "schedstat-v17.h"
+#undef DOMAIN_FIELD
+};
+
 #define DOMAIN_NAME_LEN                16
 
 struct perf_record_schedstat_domain {
@@ -544,6 +557,7 @@ struct perf_record_schedstat_domain {
        union {
                struct perf_record_schedstat_domain_v15 v15;
                struct perf_record_schedstat_domain_v16 v16;
+               struct perf_record_schedstat_domain_v17 v17;
        };
 };
 

diff --git a/tools/lib/perf/include/perf/schedstat-v17.h b/tools/lib/perf/include/perf/schedstat-v17.h
new file mode 100644 (file)
index 0000000..865dc7c
--- /dev/null
+++ b/tools/lib/perf/include/perf/schedstat-v17.h
@@ -0,0 +1,164 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifdef CPU_FIELD
+CPU_FIELD(__u32, yld_count, "sched_yield() count",
+         "%11u", false, yld_count, v17);
+CPU_FIELD(__u32, array_exp, "Legacy counter can be ignored",
+         "%11u", false, array_exp, v17);
+CPU_FIELD(__u32, sched_count, "schedule() called",
+         "%11u", false, sched_count, v17);
+CPU_FIELD(__u32, sched_goidle, "schedule() left the processor idle",
+         "%11u", true, sched_count, v17);
+CPU_FIELD(__u32, ttwu_count, "try_to_wake_up() was called",
+         "%11u", false, ttwu_count, v17);
+CPU_FIELD(__u32, ttwu_local, "try_to_wake_up() was called to wake up the local cpu",
+         "%11u", true, ttwu_count, v17);
+CPU_FIELD(__u64, rq_cpu_time, "total runtime by tasks on this processor (in jiffies)",
+         "%11llu", false, rq_cpu_time, v17);
+CPU_FIELD(__u64, run_delay, "total waittime by tasks on this processor (in jiffies)",
+         "%11llu", true, rq_cpu_time, v17);
+CPU_FIELD(__u64, pcount, "total timeslices run on this cpu",
+         "%11llu", false, pcount, v17);
+#endif /* CPU_FIELD */
+
+#ifdef DOMAIN_FIELD
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category busy> ");
+#endif
+DOMAIN_FIELD(__u32, busy_lb_count,
+            "load_balance() count on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_balanced,
+            "load_balance() found balanced on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_failed,
+            "load_balance() move task failed on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_load,
+            "imbalance in load on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_util,
+            "imbalance in utilization on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_task,
+            "imbalance in number of tasks on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_misfit,
+            "imbalance in misfit tasks on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_gained,
+            "pull_task() count on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_hot_gained,
+            "pull_task() when target task was cache-hot on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_nobusyq,
+            "load_balance() failed to find busier queue on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_nobusyg,
+            "load_balance() failed to find busier group on cpu busy", "%11u", true, v17);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(busy_lb_success_count, "load_balance() success count on cpu busy", "%11u",
+                 busy_lb_count, busy_lb_balanced, busy_lb_failed, v17);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(busy_lb_avg_pulled,
+                 "avg task pulled per successful lb attempt (cpu busy)", "%11.2Lf",
+                 busy_lb_count, busy_lb_balanced, busy_lb_failed, busy_lb_gained, v17);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category idle> ");
+#endif
+DOMAIN_FIELD(__u32, idle_lb_count,
+            "load_balance() count on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_balanced,
+            "load_balance() found balanced on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_failed,
+            "load_balance() move task failed on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_load,
+            "imbalance in load on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_util,
+            "imbalance in utilization on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_task,
+            "imbalance in number of tasks on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_misfit,
+            "imbalance in misfit tasks on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_gained,
+            "pull_task() count on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_hot_gained,
+            "pull_task() when target task was cache-hot on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_nobusyq,
+            "load_balance() failed to find busier queue on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_nobusyg,
+            "load_balance() failed to find busier group on cpu idle", "%11u", true, v17);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(idle_lb_success_count, "load_balance() success count on cpu idle", "%11u",
+                 idle_lb_count, idle_lb_balanced, idle_lb_failed, v17);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(idle_lb_avg_pulled,
+                 "avg task pulled per successful lb attempt (cpu idle)", "%11.2Lf",
+                 idle_lb_count, idle_lb_balanced, idle_lb_failed, idle_lb_gained, v17);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category newidle> ");
+#endif
+DOMAIN_FIELD(__u32, newidle_lb_count,
+            "load_balance() count on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_balanced,
+            "load_balance() found balanced on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_failed,
+            "load_balance() move task failed on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_load,
+            "imbalance in load on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_util,
+            "imbalance in utilization on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_task,
+            "imbalance in number of tasks on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_misfit,
+            "imbalance in misfit tasks on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_gained,
+            "pull_task() count on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_hot_gained,
+            "pull_task() when target task was cache-hot on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyq,
+            "load_balance() failed to find busier queue on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyg,
+            "load_balance() failed to find busier group on cpu newly idle", "%11u", true, v17);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(newidle_lb_success_count,
+                 "load_balance() success count on cpu newly idle", "%11u",
+                 newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, v17);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(newidle_lb_avg_pulled,
+                 "avg task pulled per successful lb attempt (cpu newly idle)", "%11.2Lf",
+                 newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, newidle_lb_gained, v17);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category active_load_balance()> ");
+#endif
+DOMAIN_FIELD(__u32, alb_count,
+            "active_load_balance() count", "%11u", false, v17);
+DOMAIN_FIELD(__u32, alb_failed,
+            "active_load_balance() move task failed", "%11u", false, v17);
+DOMAIN_FIELD(__u32, alb_pushed,
+            "active_load_balance() successfully moved a task", "%11u", false, v17);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_exec()> ");
+#endif
+DOMAIN_FIELD(__u32, sbe_count,
+            "sbe_count is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbe_balanced,
+            "sbe_balanced is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbe_pushed,
+            "sbe_pushed is not used", "%11u", false, v17);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_fork()> ");
+#endif
+DOMAIN_FIELD(__u32, sbf_count,
+            "sbf_count is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbf_balanced,
+            "sbf_balanced is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbf_pushed,
+            "sbf_pushed is not used", "%11u", false, v17);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Wakeup Info> ");
+#endif
+DOMAIN_FIELD(__u32, ttwu_wake_remote,
+            "try_to_wake_up() awoke a task that last ran on a diff cpu", "%11u", false, v17);
+DOMAIN_FIELD(__u32, ttwu_move_affine,
+            "try_to_wake_up() moved task because cache-cold on own cpu", "%11u", false, v17);
+DOMAIN_FIELD(__u32, ttwu_move_balance,
+            "try_to_wake_up() started passive balancing", "%11u", false, v17);
+#endif /* DOMAIN_FIELD */

diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index 730021cec1617ddaabd6e53c2431ca7aad040386..2dde1044b5a783a96a2b6f33ff7004def9992b58 100644 (file)
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -588,6 +588,9 @@ size_t perf_event__fprintf_schedstat_cpu(union perf_event *event, FILE *fp)
        } else if (version == 16) {
 #include <perf/schedstat-v16.h>
                return size;
+       } else if (version == 17) {
+#include <perf/schedstat-v17.h>
+               return size;
        }
 #undef CPU_FIELD
 
@@ -610,6 +613,9 @@ size_t perf_event__fprintf_schedstat_domain(union perf_event *event, FILE *fp)
        } else if (version == 16) {
 #include <perf/schedstat-v16.h>
                return size;
+       } else if (version == 17) {
+#include <perf/schedstat-v17.h>
+               return size;
        }
 #undef DOMAIN_FIELD
 

diff --git a/tools/perf/util/synthetic-events.c b/tools/perf/util/synthetic-events.c
index 4ce37357db05b1101f2ee009b39d6118e95d04aa..ef79433ebc3a663cf6561b110784bab1094e2862 100644 (file)
--- a/tools/perf/util/synthetic-events.c
+++ b/tools/perf/util/synthetic-events.c
@@ -2569,6 +2569,8 @@ static union perf_event *__synthesize_schedstat_cpu(struct io *io, __u16 version
 #include <perf/schedstat-v15.h>
        } else if (version == 16) {
 #include <perf/schedstat-v16.h>
+       } else if (version == 17) {
+#include <perf/schedstat-v17.h>
        }
 #undef CPU_FIELD
 
@@ -2595,6 +2597,11 @@ static union perf_event *__synthesize_schedstat_domain(struct io *io, __u16 vers
                return NULL;
 
        ch = io__get_dec(io, &d_num);
+       if (version >= 17) {
+               /* Skip domain name as it can be extracted from perf header */
+               while (io__get_char(io) != ' ')
+                       continue;
+       }
 
        /* Skip cpumask as it can be extracted from perf header */
        while (io__get_char(io) != ' ')
@@ -2624,6 +2631,8 @@ static union perf_event *__synthesize_schedstat_domain(struct io *io, __u16 vers
 #include <perf/schedstat-v15.h>
        } else if (version == 16) {
 #include <perf/schedstat-v16.h>
+       } else if (version == 17) {
+#include <perf/schedstat-v17.h>
        }
 #undef DOMAIN_FIELD
 
@@ -2667,6 +2676,8 @@ int perf_event__synthesize_schedstat(const struct perf_tool *tool,
                version = 15;
        } else if (!strcmp(line, "version 16\n")) {
                version = 16;
+       } else if (!strcmp(line, "version 17\n")) {
+               version = 17;
        } else {
                pr_err("Unsupported %s version: %s", path, line + 8);
                goto out_free_line;