linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: Clark Williams <williams@redhat.com>,
	linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
	Arnaldo Carvalho de Melo <acme@redhat.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	David Ahern <dsahern@gmail.com>, Jiri Olsa <jolsa@kernel.org>,
	Milian Wolff <milian.wolff@kdab.com>,
	Namhyung Kim <namhyung@kernel.org>,
	Wang Nan <wangnan0@huawei.com>
Subject: [PATCH 08/37] perf evsel: Introduce per event max_events property
Date: Thu, 25 Oct 2018 08:10:02 -0300	[thread overview]
Message-ID: <20181025111031.3440-9-acme@kernel.org> (raw)
In-Reply-To: <20181025111031.3440-1-acme@kernel.org>

From: Arnaldo Carvalho de Melo <acme@redhat.com>

This simply adds the field to 'struct perf_evsel' and allows setting
it via the event parser, to test it lets trace trace:

First look at where in a function that receives an evsel we can put a probe
to read how evsel->max_events was setup:

  # perf probe -x ~/bin/perf -L trace__event_handler
  <trace__event_handler@/home/acme/git/perf/tools/perf/builtin-trace.c:0>
        0  static int trace__event_handler(struct trace *trace, struct perf_evsel *evsel,
                                          union perf_event *event __maybe_unused,
                                          struct perf_sample *sample)
        3  {
        4         struct thread *thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
        5         int callchain_ret = 0;

        7         if (sample->callchain) {
        8                 callchain_ret = trace__resolve_callchain(trace, evsel, sample, &callchain_cursor);
        9                 if (callchain_ret == 0) {
       10                         if (callchain_cursor.nr < trace->min_stack)
       11                                 goto out;
       12                         callchain_ret = 1;
                          }
                  }

See what variables we can probe at line 7:

  # perf probe -x ~/bin/perf -V trace__event_handler:7
  Available variables at trace__event_handler:7
          @<trace__event_handler+89>
                  int     callchain_ret
                  struct perf_evsel*      evsel
                  struct perf_sample*     sample
                  struct thread*  thread
                  struct trace*   trace
                  union perf_event*       event

Add a probe at that line asking for evsel->max_events to be collected and named
as "max_events":

  # perf probe -x ~/bin/perf trace__event_handler:7 'max_events=evsel->max_events'
  Added new event:
    probe_perf:trace__event_handler (on trace__event_handler:7 in /home/acme/bin/perf with max_events=evsel->max_events)

  You can now use it in all perf tools, such as:

  	perf record -e probe_perf:trace__event_handler -aR sleep 1

Now use 'perf trace', here aliased to just 'trace' and trace trace, i.e.
the first 'trace' is tracing just that 'probe_perf:trace__event_handler' event,
while the traced trace is tracing all scheduler tracepoints, will stop at two
events (--max-events 2) and will just set evsel->max_events for all the sched
tracepoints to 9, we will see the output of both traces intermixed:

  # trace -e *perf:*event_handler trace --max-events 2 -e sched:*/nr=9/
       0.000 :0/0 sched:sched_waking:comm=rcu_sched pid=10 prio=120 target_cpu=000
       0.009 :0/0 sched:sched_wakeup:comm=rcu_sched pid=10 prio=120 target_cpu=000
       0.000 trace/23949 probe_perf:trace__event_handler:(48c34a) max_events=0x9
       0.046 trace/23949 probe_perf:trace__event_handler:(48c34a) max_events=0x9
  #

Now, if the traced trace sends its output to /dev/null, we'll see just
what the first level trace outputs: that evsel->max_events is indeed
being set to 9:

  # trace -e *perf:*event_handler trace -o /dev/null --max-events 2 -e sched:*/nr=9/
       0.000 trace/23961 probe_perf:trace__event_handler:(48c34a) max_events=0x9
       0.030 trace/23961 probe_perf:trace__event_handler:(48c34a) max_events=0x9
  #

Now that we can set evsel->max_events, we can go to the next step, honour that
per-event property in 'perf trace'.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-og00yasj276joem6e14l1eas@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evsel.c        | 4 ++++
 tools/perf/util/evsel.h        | 3 +++
 tools/perf/util/parse-events.c | 8 ++++++++
 tools/perf/util/parse-events.h | 1 +
 tools/perf/util/parse-events.l | 1 +
 5 files changed, 17 insertions(+)

diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 29d7b97f66fb..7e95ec1c19a8 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -232,6 +232,7 @@ void perf_evsel__init(struct perf_evsel *evsel,
 	evsel->leader	   = evsel;
 	evsel->unit	   = "";
 	evsel->scale	   = 1.0;
+	evsel->max_events  = ULONG_MAX;
 	evsel->evlist	   = NULL;
 	evsel->bpf_fd	   = -1;
 	INIT_LIST_HEAD(&evsel->node);
@@ -793,6 +794,9 @@ static void apply_config_terms(struct perf_evsel *evsel,
 		case PERF_EVSEL__CONFIG_TERM_MAX_STACK:
 			max_stack = term->val.max_stack;
 			break;
+		case PERF_EVSEL__CONFIG_TERM_MAX_EVENTS:
+			evsel->max_events = term->val.max_events;
+			break;
 		case PERF_EVSEL__CONFIG_TERM_INHERIT:
 			/*
 			 * attr->inherit should has already been set by
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 4107c39f4a54..ad5d615c6db6 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -46,6 +46,7 @@ enum term_type {
 	PERF_EVSEL__CONFIG_TERM_STACK_USER,
 	PERF_EVSEL__CONFIG_TERM_INHERIT,
 	PERF_EVSEL__CONFIG_TERM_MAX_STACK,
+	PERF_EVSEL__CONFIG_TERM_MAX_EVENTS,
 	PERF_EVSEL__CONFIG_TERM_OVERWRITE,
 	PERF_EVSEL__CONFIG_TERM_DRV_CFG,
 	PERF_EVSEL__CONFIG_TERM_BRANCH,
@@ -65,6 +66,7 @@ struct perf_evsel_config_term {
 		bool	inherit;
 		bool	overwrite;
 		char	*branch;
+		unsigned long max_events;
 	} val;
 	bool weak;
 };
@@ -99,6 +101,7 @@ struct perf_evsel {
 	struct perf_counts	*prev_raw_counts;
 	int			idx;
 	u32			ids;
+	unsigned long		max_events;
 	char			*name;
 	double			scale;
 	const char		*unit;
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index f8cd3e7c9186..59be3466d64d 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -926,6 +926,7 @@ static const char *config_term_names[__PARSE_EVENTS__TERM_TYPE_NR] = {
 	[PARSE_EVENTS__TERM_TYPE_NOINHERIT]		= "no-inherit",
 	[PARSE_EVENTS__TERM_TYPE_INHERIT]		= "inherit",
 	[PARSE_EVENTS__TERM_TYPE_MAX_STACK]		= "max-stack",
+	[PARSE_EVENTS__TERM_TYPE_MAX_EVENTS]		= "nr",
 	[PARSE_EVENTS__TERM_TYPE_OVERWRITE]		= "overwrite",
 	[PARSE_EVENTS__TERM_TYPE_NOOVERWRITE]		= "no-overwrite",
 	[PARSE_EVENTS__TERM_TYPE_DRV_CFG]		= "driver-config",
@@ -1037,6 +1038,9 @@ do {									   \
 	case PARSE_EVENTS__TERM_TYPE_MAX_STACK:
 		CHECK_TYPE_VAL(NUM);
 		break;
+	case PARSE_EVENTS__TERM_TYPE_MAX_EVENTS:
+		CHECK_TYPE_VAL(NUM);
+		break;
 	default:
 		err->str = strdup("unknown term");
 		err->idx = term->err_term;
@@ -1084,6 +1088,7 @@ static int config_term_tracepoint(struct perf_event_attr *attr,
 	case PARSE_EVENTS__TERM_TYPE_INHERIT:
 	case PARSE_EVENTS__TERM_TYPE_NOINHERIT:
 	case PARSE_EVENTS__TERM_TYPE_MAX_STACK:
+	case PARSE_EVENTS__TERM_TYPE_MAX_EVENTS:
 	case PARSE_EVENTS__TERM_TYPE_OVERWRITE:
 	case PARSE_EVENTS__TERM_TYPE_NOOVERWRITE:
 		return config_term_common(attr, term, err);
@@ -1162,6 +1167,9 @@ do {								\
 		case PARSE_EVENTS__TERM_TYPE_MAX_STACK:
 			ADD_CONFIG_TERM(MAX_STACK, max_stack, term->val.num);
 			break;
+		case PARSE_EVENTS__TERM_TYPE_MAX_EVENTS:
+			ADD_CONFIG_TERM(MAX_EVENTS, max_events, term->val.num);
+			break;
 		case PARSE_EVENTS__TERM_TYPE_OVERWRITE:
 			ADD_CONFIG_TERM(OVERWRITE, overwrite, term->val.num ? 1 : 0);
 			break;
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
index 4473dac27aee..5ed035cbcbb7 100644
--- a/tools/perf/util/parse-events.h
+++ b/tools/perf/util/parse-events.h
@@ -71,6 +71,7 @@ enum {
 	PARSE_EVENTS__TERM_TYPE_NOINHERIT,
 	PARSE_EVENTS__TERM_TYPE_INHERIT,
 	PARSE_EVENTS__TERM_TYPE_MAX_STACK,
+	PARSE_EVENTS__TERM_TYPE_MAX_EVENTS,
 	PARSE_EVENTS__TERM_TYPE_NOOVERWRITE,
 	PARSE_EVENTS__TERM_TYPE_OVERWRITE,
 	PARSE_EVENTS__TERM_TYPE_DRV_CFG,
diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
index 5f761f3ed0f3..7805c71aaae2 100644
--- a/tools/perf/util/parse-events.l
+++ b/tools/perf/util/parse-events.l
@@ -269,6 +269,7 @@ time			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_TIME); }
 call-graph		{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CALLGRAPH); }
 stack-size		{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_STACKSIZE); }
 max-stack		{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_MAX_STACK); }
+nr			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_MAX_EVENTS); }
 inherit			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_INHERIT); }
 no-inherit		{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_NOINHERIT); }
 overwrite		{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_OVERWRITE); }
-- 
2.14.4


  parent reply	other threads:[~2018-10-25 11:11 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-25 11:09 [PATCH 00/37] perf/core improvements and fixes Arnaldo Carvalho de Melo
2018-10-25 11:09 ` [PATCH 01/37] perf record: Encode -k clockid frequency into Perf trace Arnaldo Carvalho de Melo
2018-10-25 11:09 ` [PATCH 02/37] perf annotate: Add Sparc support Arnaldo Carvalho de Melo
2018-10-25 11:09 ` [PATCH 03/37] perf jitdump: " Arnaldo Carvalho de Melo
2018-10-25 11:09 ` [PATCH 04/37] perf symbols: Set PLT entry/header sizes properly on Sparc Arnaldo Carvalho de Melo
2018-10-25 11:09 ` [PATCH 05/37] perf arm64: Fix generate system call table failed with /tmp mounted with noexec Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 06/37] tools lib subcmd: Introduce OPTION_ULONG Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 07/37] perf trace: Introduce --max-events Arnaldo Carvalho de Melo
2018-10-25 11:10 ` Arnaldo Carvalho de Melo [this message]
2018-10-25 11:10 ` [PATCH 09/37] perf evsel: Mark a evsel as disabled when asking the kernel do disable it Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 10/37] perf trace: Drop addr_location refcounts Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 11/37] perf trace: Drop thread refcount in trace__event_handler() Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 12/37] perf stat: Poll for monitored tasks being alive Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 13/37] perf script: Allow extended console debug output Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 14/37] perf script: Flush output stream after events in verbose mode Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 15/37] perf trace: Introduce per-event maximum number of events property Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 16/37] perf scripts python: call-graph-from-sql.py: Use SPDX license identifier Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 17/37] perf scripts python: call-graph-from-sql.py: Provide better default column sizes Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 18/37] perf scripts python: call-graph-from-sql.py: Set a minimum window size Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 19/37] perf scripts python: call-graph-from-sql.py: Change icon Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 20/37] perf scripts python: call-graph-from-sql.py: Make a "Main" function Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 21/37] perf scripts python: call-graph-from-sql.py: Separate the database details into a class Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 22/37] perf scripts python: call-graph-from-sql.py: Add a class for global data Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 23/37] perf scripts python: call-graph-from-sql.py: Remove use of setObjectName() Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 24/37] perf scripts python: call-graph-from-sql.py: Factor out CallGraphModel from TreeModel Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 25/37] perf scripts python: call-graph-from-sql.py: Add data helper functions Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 26/37] perf scripts python: call-graph-from-sql.py: Refactor TreeItem class Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 27/37] perf scripts python: call-graph-from-sql.py: Rename to exported-sql-viewer.py Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 28/37] perf scripts python: exported-sql-viewer.py: Add support for multiple sub-windows Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 29/37] perf scripts python: exported-sql-viewer.py: Add ability to find symbols in the call-graph Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 30/37] perf scripts python: exported-sql-viewer.py: Add ability to shrink / enlarge font Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 31/37] perf scripts python: exported-sql-viewer.py: Add ability to display all the database tables Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 32/37] perf scripts python: exported-sql-viewer.py: Add All branches report Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 33/37] perf script: Add --insn-trace for instruction decoding Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 34/37] perf script: Make itrace script default to all calls Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 35/37] tools script: Add --call-trace and --call-ret-trace Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 36/37] perf script: Implement --graph-function Arnaldo Carvalho de Melo
2018-10-25 11:10 ` [PATCH 37/37] perf script: Support total cycles count Arnaldo Carvalho de Melo
2018-10-26  7:25 ` [PATCH 00/37] perf/core improvements and fixes Ingo Molnar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20181025111031.3440-9-acme@kernel.org \
    --to=acme@kernel.org \
    --cc=acme@redhat.com \
    --cc=adrian.hunter@intel.com \
    --cc=dsahern@gmail.com \
    --cc=jolsa@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=milian.wolff@kdab.com \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --cc=wangnan0@huawei.com \
    --cc=williams@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).