Re: [PATCH v4 0/3] perf diff: Add new filter options

Re: [PATCH v4 0/3] perf diff: Add new filter options

[Jiri Olsa]

On Tue, Mar 05, 2019 at 09:05:40PM +0800, Jin Yao wrote:
> This patch series creates following new filter options for perf diff.
> 
> --time:
>     It supports time percent with multipe time ranges. Time string is
>     'a%/n,b%/m,...' or 'a%-b%,c%-%d,...'.
> 
>     For example:
> 
>     Select the second 10% time slice to diff:
>     perf diff --time 10%/2
> 
>     Select from 0% to 10% time slice to diff:
>     perf diff --time 0%-10%
> 
>     Select the first and the second 10% time slices to diff:
>     perf diff --time 10%/1,10%/2
> 
>     Select from 0% to 10% and 30% to 40% slices to diff:
>     perf diff --time 0%-10%,30%-40%
> 
>     It also supports to analyze samples within given time window as:
>     <start>,<stop>. Times have the format seconds.microseconds. If start
>     is not given (i.e., time string is ',x.y') then analysis starts at
>     the beginning of the file. If stop time is not given (i.e, time
>     string is 'x.y,') then analysis goes to end of file. Time string is
>     'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for different
>     perf.data files.
> 
>     For example, we get the timestamp information from perf script.
> 
>     perf script -i perf.data.old
>       mgen 13940 [000]  3946.361400: ...
> 
>     perf script -i perf.data
>       mgen 13940 [000]  3971.150589 ...
> 
>     perf diff --time 3946.361400,:3971.150589,
> 
>     It analyzes the perf.data.old from the timestamp 3946.361400 to
>     the end of perf.data.old and analyzes the perf.data from the
>     timestamp 3971.150589 to the end of perf.data.
> 
> --cpu:
>     Only diff samples for the list of CPUs provided. Multiple CPUs can
>     be provided as a comma-separated list with no space: 0,1. Ranges of
>     CPUs are specified with -: 0-2. Default is to report samples on all
>     CPUs.
> 
>     For example:
>     perf diff --cpu 0,1
> 
>     It only diff the samples for CPU0 and CPU1.
> 
> --pid:
>     Only diff samples for given process ID (comma separated list).
> 
> --tid:
>     Only diff samples for given thread ID (comma separated list).
> 
>     For example,
>     perf diff --tid 13965
> 
>     It only diff the samples for thread 13965.
> 
>  v4:
>  ---
>  Update abstime_str_dup(), let it return error if strdup
>  is failed, and update __cmd_diff() accordingly.
> 
>  Only the patch "perf diff: Support --time filter option" is impacted
>  Others are no logical changes.
> 
> Jin Yao (3):
>   perf diff: Support --time filter option
>   perf diff: Support --cpu filter option
>   perf diff: Support --pid/--tid filter options

Acked-by: Jiri Olsa <jolsa@kernel.org>

thanks,
jirka

[PATCH v4 0/3] perf diff: Add new filter options

[Jin Yao]

This patch series creates following new filter options for perf diff.

--time:
    It supports time percent with multipe time ranges. Time string is
    'a%/n,b%/m,...' or 'a%-b%,c%-%d,...'.

    For example:

    Select the second 10% time slice to diff:
    perf diff --time 10%/2

    Select from 0% to 10% time slice to diff:
    perf diff --time 0%-10%

    Select the first and the second 10% time slices to diff:
    perf diff --time 10%/1,10%/2

    Select from 0% to 10% and 30% to 40% slices to diff:
    perf diff --time 0%-10%,30%-40%

    It also supports to analyze samples within given time window as:
    <start>,<stop>. Times have the format seconds.microseconds. If start
    is not given (i.e., time string is ',x.y') then analysis starts at
    the beginning of the file. If stop time is not given (i.e, time
    string is 'x.y,') then analysis goes to end of file. Time string is
    'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for different
    perf.data files.

    For example, we get the timestamp information from perf script.

    perf script -i perf.data.old
      mgen 13940 [000]  3946.361400: ...

    perf script -i perf.data
      mgen 13940 [000]  3971.150589 ...

    perf diff --time 3946.361400,:3971.150589,

    It analyzes the perf.data.old from the timestamp 3946.361400 to
    the end of perf.data.old and analyzes the perf.data from the
    timestamp 3971.150589 to the end of perf.data.

--cpu:
    Only diff samples for the list of CPUs provided. Multiple CPUs can
    be provided as a comma-separated list with no space: 0,1. Ranges of
    CPUs are specified with -: 0-2. Default is to report samples on all
    CPUs.

    For example:
    perf diff --cpu 0,1

    It only diff the samples for CPU0 and CPU1.

--pid:
    Only diff samples for given process ID (comma separated list).

--tid:
    Only diff samples for given thread ID (comma separated list).

    For example,
    perf diff --tid 13965

    It only diff the samples for thread 13965.

 v4:
 ---
 Update abstime_str_dup(), let it return error if strdup
 is failed, and update __cmd_diff() accordingly.

 Only the patch "perf diff: Support --time filter option" is impacted
 Others are no logical changes.

Jin Yao (3):
  perf diff: Support --time filter option
  perf diff: Support --cpu filter option
  perf diff: Support --pid/--tid filter options

 tools/perf/Documentation/perf-diff.txt |  52 ++++++++++
 tools/perf/builtin-diff.c              | 168 ++++++++++++++++++++++++++++++---
 2 files changed, 206 insertions(+), 14 deletions(-)

-- 
2.7.4

[PATCH v4 1/3] perf diff: Support --time filter option

[Jin Yao]

For better support for perf diff, it would be useful to add --time filter
option to diff the samples within given time window.

It supports time percent with multipe time ranges. Time string is
'a%/n,b%/m,...' or 'a%-b%,c%-%d,...'.

For example:

Select the second 10% time slice to diff:
perf diff --time 10%/2

Select from 0% to 10% time slice to diff:
perf diff --time 0%-10%

Select the first and the second 10% time slices to diff:
perf diff --time 10%/1,10%/2

Select from 0% to 10% and 30% to 40% slices to diff:
perf diff --time 0%-10%,30%-40%

It also supports to analyze samples within given time window as:
<start>,<stop>. Times have the format seconds.microseconds. If start
is not given (i.e., time string is ',x.y') then analysis starts at
the beginning of the file. If stop time is not given (i.e, time
string is 'x.y,') then analysis goes to end of file. Time string is
'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for different
perf.data files.

For example, we get the timestamp information from perf script.

perf script -i perf.data.old
  mgen 13940 [000]  3946.361400: ...

perf script -i perf.data
  mgen 13940 [000]  3971.150589 ...

perf diff --time 3946.361400,:3971.150589,

It analyzes the perf.data.old from the timestamp 3946.361400 to
the end of perf.data.old and analyzes the perf.data from the
timestamp 3971.150589 to the end of perf.data.

 v4:
 ---
 Update abstime_str_dup(), let it return error if strdup
 is failed, and update __cmd_diff() accordingly.

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
---
 tools/perf/Documentation/perf-diff.txt |  41 +++++++++
 tools/perf/builtin-diff.c              | 148 +++++++++++++++++++++++++++++----
 2 files changed, 175 insertions(+), 14 deletions(-)

diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
index a79c84a..5bfe876 100644
--- a/tools/perf/Documentation/perf-diff.txt
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -118,6 +118,47 @@ OPTIONS
 	sum of shown entries will be always 100%.  "absolute" means it retains
 	the original value before and after the filter is applied.
 
+--time::
+	Analyze samples within given time window. It supports time
+	percent with multipe time ranges. Time string is 'a%/n,b%/m,...'
+	or 'a%-b%,c%-%d,...'.
+
+	For example:
+
+	Select the second 10% time slice to diff:
+	perf diff --time 10%/2
+
+	Select from 0% to 10% time slice to diff:
+	perf diff --time 0%-10%
+
+	Select the first and the second 10% time slices to diff:
+	perf diff --time 10%/1,10%/2
+
+	Select from 0% to 10% and 30% to 40% slices to diff:
+	perf diff --time 0%-10%,30%-40%
+
+	It also supports to analyze samples within given time window:
+	<start>,<stop>. Times have the format seconds.microseconds. If start
+	is not given (i.e., time string is ',x.y') then analysis starts at
+	the beginning of the file. If stop time is not given (i.e, time
+	string is 'x.y,') then analysis goes to end of file. Time string is
+	'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for different
+	perf.data files.
+
+	For example, we get the timestamp information from perf script.
+
+	perf script -i perf.data.old
+	  mgen 13940 [000]  3946.361400: ...
+
+	perf script -i perf.data
+	  mgen 13940 [000]  3971.150589 ...
+
+	perf diff --time 3946.361400,:3971.150589,
+
+	It analyzes the perf.data.old from the timestamp 3946.361400 to
+	the end of perf.data.old and analyzes the perf.data from the
+	timestamp 3971.150589 to the end of perf.data.
+
 COMPARISON
 ----------
 The comparison is governed by the baseline file. The baseline perf.data
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
index 58fe0e8..17cd898 100644
--- a/tools/perf/builtin-diff.c
+++ b/tools/perf/builtin-diff.c
@@ -19,12 +19,21 @@
 #include "util/util.h"
 #include "util/data.h"
 #include "util/config.h"
+#include "util/time-utils.h"
 
 #include <errno.h>
 #include <inttypes.h>
 #include <stdlib.h>
 #include <math.h>
 
+struct perf_diff {
+	struct perf_tool		 tool;
+	const char			*time_str;
+	struct perf_time_interval	*ptime_range;
+	int				 range_size;
+	int				 range_num;
+};
+
 /* Diff command specific HPP columns. */
 enum {
 	PERF_HPP_DIFF__BASELINE,
@@ -323,16 +332,22 @@ static int formula_fprintf(struct hist_entry *he, struct hist_entry *pair,
 	return -1;
 }
 
-static int diff__process_sample_event(struct perf_tool *tool __maybe_unused,
+static int diff__process_sample_event(struct perf_tool *tool,
 				      union perf_event *event,
 				      struct perf_sample *sample,
 				      struct perf_evsel *evsel,
 				      struct machine *machine)
 {
+	struct perf_diff *pdiff = container_of(tool, struct perf_diff, tool);
 	struct addr_location al;
 	struct hists *hists = evsel__hists(evsel);
 	int ret = -1;
 
+	if (perf_time__ranges_skip_sample(pdiff->ptime_range, pdiff->range_num,
+					  sample->time)) {
+		return 0;
+	}
+
 	if (machine__resolve(machine, &al, sample) < 0) {
 		pr_warning("problem processing %d event, skipping it.\n",
 			   event->header.type);
@@ -359,17 +374,19 @@ static int diff__process_sample_event(struct perf_tool *tool __maybe_unused,
 	return ret;
 }
 
-static struct perf_tool tool = {
-	.sample	= diff__process_sample_event,
-	.mmap	= perf_event__process_mmap,
-	.mmap2	= perf_event__process_mmap2,
-	.comm	= perf_event__process_comm,
-	.exit	= perf_event__process_exit,
-	.fork	= perf_event__process_fork,
-	.lost	= perf_event__process_lost,
-	.namespaces = perf_event__process_namespaces,
-	.ordered_events = true,
-	.ordering_requires_timestamps = true,
+static struct perf_diff pdiff = {
+	.tool = {
+		.sample	= diff__process_sample_event,
+		.mmap	= perf_event__process_mmap,
+		.mmap2	= perf_event__process_mmap2,
+		.comm	= perf_event__process_comm,
+		.exit	= perf_event__process_exit,
+		.fork	= perf_event__process_fork,
+		.lost	= perf_event__process_lost,
+		.namespaces = perf_event__process_namespaces,
+		.ordered_events = true,
+		.ordering_requires_timestamps = true,
+	},
 };
 
 static struct perf_evsel *evsel_match(struct perf_evsel *evsel,
@@ -771,19 +788,110 @@ static void data__free(struct data__file *d)
 	}
 }
 
+static int abstime_str_dup(char **pstr)
+{
+	char *str = NULL;
+
+	if (pdiff.time_str && strchr(pdiff.time_str, ':')) {
+		str = strdup(pdiff.time_str);
+		if (!str)
+			return -ENOMEM;
+	}
+
+	*pstr = str;
+	return 0;
+}
+
+static int parse_absolute_time(struct data__file *d, char **pstr)
+{
+	char *p = *pstr;
+	int ret;
+
+	/*
+	 * Absolute timestamp for one file has the format: a.b,c.d
+	 * For multiple files, the format is: a.b,c.d:a.b,c.d
+	 */
+	p = strchr(*pstr, ':');
+	if (p) {
+		if (p == *pstr) {
+			pr_err("Invalid time string\n");
+			return -EINVAL;
+		}
+
+		*p = 0;
+		p++;
+		if (*p == 0) {
+			pr_err("Invalid time string\n");
+			return -EINVAL;
+		}
+	}
+
+	ret = perf_time__parse_for_ranges(*pstr, d->session,
+					  &pdiff.ptime_range,
+					  &pdiff.range_size,
+					  &pdiff.range_num);
+	if (ret < 0)
+		return ret;
+
+	if (!p || *p == 0)
+		*pstr = NULL;
+	else
+		*pstr = p;
+
+	return ret;
+}
+
+static int parse_percent_time(struct data__file *d)
+{
+	int ret;
+
+	ret = perf_time__parse_for_ranges(pdiff.time_str, d->session,
+					  &pdiff.ptime_range,
+					  &pdiff.range_size,
+					  &pdiff.range_num);
+	return ret;
+}
+
+static int parse_time_str(struct data__file *d, char *abstime_ostr,
+			   char **pabstime_tmp)
+{
+	int ret = 0;
+
+	if (abstime_ostr)
+		ret = parse_absolute_time(d, pabstime_tmp);
+	else if (pdiff.time_str)
+		ret = parse_percent_time(d);
+
+	return ret;
+}
+
 static int __cmd_diff(void)
 {
 	struct data__file *d;
-	int ret = -EINVAL, i;
+	int ret, i;
+	char *abstime_ostr, *abstime_tmp;
+
+	ret = abstime_str_dup(&abstime_ostr);
+	if (ret)
+		return ret;
+
+	abstime_tmp = abstime_ostr;
+	ret = -EINVAL;
 
 	data__for_each_file(i, d) {
-		d->session = perf_session__new(&d->data, false, &tool);
+		d->session = perf_session__new(&d->data, false, &pdiff.tool);
 		if (!d->session) {
 			pr_err("Failed to open %s\n", d->data.path);
 			ret = -1;
 			goto out_delete;
 		}
 
+		if (pdiff.time_str) {
+			ret = parse_time_str(d, abstime_ostr, &abstime_tmp);
+			if (ret < 0)
+				goto out_delete;
+		}
+
 		ret = perf_session__process_events(d->session);
 		if (ret) {
 			pr_err("Failed to process %s\n", d->data.path);
@@ -791,6 +899,9 @@ static int __cmd_diff(void)
 		}
 
 		perf_evlist__collapse_resort(d->session->evlist);
+
+		if (pdiff.ptime_range)
+			zfree(&pdiff.ptime_range);
 	}
 
 	data_process();
@@ -802,6 +913,13 @@ static int __cmd_diff(void)
 	}
 
 	free(data__files);
+
+	if (pdiff.ptime_range)
+		zfree(&pdiff.ptime_range);
+
+	if (abstime_ostr)
+		free(abstime_ostr);
+
 	return ret;
 }
 
@@ -849,6 +967,8 @@ static const struct option options[] = {
 	OPT_UINTEGER('o', "order", &sort_compute, "Specify compute sorting."),
 	OPT_CALLBACK(0, "percentage", NULL, "relative|absolute",
 		     "How to display percentage of filtered entries", parse_filter_percentage),
+	OPT_STRING(0, "time", &pdiff.time_str, "str",
+		   "Time span (time percent or absolute timestamp)"),
 	OPT_END()
 };
 
-- 
2.7.4

[PATCH v4 2/3] perf diff: Support --cpu filter option

[Jin Yao]

For better support for perf diff, it would be useful to add --cpu filter
option.

Multiple CPUs can be provided as a comma-separated list with no space: 0,1.
Ranges of CPUs are specified with -: 0-2. Default is to report samples on
all CPUs.

For example,

perf diff --cpu 0,1

It only diff the samples for CPU0 and CPU1.

 v4:
 ---
 No logical changes

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
---
 tools/perf/Documentation/perf-diff.txt |  5 +++++
 tools/perf/builtin-diff.c              | 16 ++++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
index 5bfe876..ed13b3f 100644
--- a/tools/perf/Documentation/perf-diff.txt
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -159,6 +159,11 @@ OPTIONS
 	the end of perf.data.old and analyzes the perf.data from the
 	timestamp 3971.150589 to the end of perf.data.
 
+--cpu:: Only diff samples for the list of CPUs provided. Multiple CPUs can
+	be provided as a comma-separated list with no space: 0,1. Ranges of
+	CPUs are specified with -: 0-2. Default is to report samples on all
+	CPUs.
+
 COMPARISON
 ----------
 The comparison is governed by the baseline file. The baseline perf.data
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
index 17cd898..dfe6c76 100644
--- a/tools/perf/builtin-diff.c
+++ b/tools/perf/builtin-diff.c
@@ -83,6 +83,9 @@ static unsigned int sort_compute = 1;
 static s64 compute_wdiff_w1;
 static s64 compute_wdiff_w2;
 
+static const char		*cpu_list;
+static DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
+
 enum {
 	COMPUTE_DELTA,
 	COMPUTE_RATIO,
@@ -354,6 +357,11 @@ static int diff__process_sample_event(struct perf_tool *tool,
 		return -1;
 	}
 
+	if (cpu_list && !test_bit(sample->cpu, cpu_bitmap)) {
+		ret = 0;
+		goto out_put;
+	}
+
 	if (!hists__add_entry(hists, &al, NULL, NULL, NULL, sample, true)) {
 		pr_warning("problem incrementing symbol period, skipping event\n");
 		goto out_put;
@@ -892,6 +900,13 @@ static int __cmd_diff(void)
 				goto out_delete;
 		}
 
+		if (cpu_list) {
+			ret = perf_session__cpu_bitmap(d->session, cpu_list,
+						       cpu_bitmap);
+			if (ret < 0)
+				goto out_delete;
+		}
+
 		ret = perf_session__process_events(d->session);
 		if (ret) {
 			pr_err("Failed to process %s\n", d->data.path);
@@ -969,6 +984,7 @@ static const struct option options[] = {
 		     "How to display percentage of filtered entries", parse_filter_percentage),
 	OPT_STRING(0, "time", &pdiff.time_str, "str",
 		   "Time span (time percent or absolute timestamp)"),
+	OPT_STRING(0, "cpu", &cpu_list, "cpu", "list of cpus to profile"),
 	OPT_END()
 };
 
-- 
2.7.4

[PATCH v4 3/3] perf diff: Support --pid/--tid filter options

[Jin Yao]

For better filtering support for perf diff, it would be useful to
add --pid and --tid filter options.

For example,
perf diff --tid 13965

It only diff the samples for thread 13965.

 v4:
 ---
 No logical changes.

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
---
 tools/perf/Documentation/perf-diff.txt | 6 ++++++
 tools/perf/builtin-diff.c              | 4 ++++
 2 files changed, 10 insertions(+)

diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
index ed13b3f..83c9496 100644
--- a/tools/perf/Documentation/perf-diff.txt
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -164,6 +164,12 @@ OPTIONS
 	CPUs are specified with -: 0-2. Default is to report samples on all
 	CPUs.
 
+--pid=::
+	Only diff samples for given process ID (comma separated list).
+
+--tid=::
+	Only diff samples for given thread ID (comma separated list).
+
 COMPARISON
 ----------
 The comparison is governed by the baseline file. The baseline perf.data
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
index dfe6c76..6e79207 100644
--- a/tools/perf/builtin-diff.c
+++ b/tools/perf/builtin-diff.c
@@ -985,6 +985,10 @@ static const struct option options[] = {
 	OPT_STRING(0, "time", &pdiff.time_str, "str",
 		   "Time span (time percent or absolute timestamp)"),
 	OPT_STRING(0, "cpu", &cpu_list, "cpu", "list of cpus to profile"),
+	OPT_STRING(0, "pid", &symbol_conf.pid_list_str, "pid[,pid...]",
+		   "only consider symbols in these pids"),
+	OPT_STRING(0, "tid", &symbol_conf.tid_list_str, "tid[,tid...]",
+		   "only consider symbols in these tids"),
 	OPT_END()
 };
 
-- 
2.7.4

Re: [PATCH v4 0/3] perf diff: Add new filter options

[Arnaldo Carvalho de Melo]

Em Tue, Mar 05, 2019 at 01:34:23PM +0100, Jiri Olsa escreveu:
> On Tue, Mar 05, 2019 at 09:05:40PM +0800, Jin Yao wrote:
> > Jin Yao (3):
> >   perf diff: Support --time filter option
> >   perf diff: Support --cpu filter option
> >   perf diff: Support --pid/--tid filter options
> 
> Acked-by: Jiri Olsa <jolsa@kernel.org>

Thanks, applied.

- Arnaldo

[tip:perf/urgent] perf diff: Support --time filter option

[tip-bot for Jin Yao]

Commit-ID:  4802138d78caed36cee2a859f77fb2035f230018
Gitweb:     https://git.kernel.org/tip/4802138d78caed36cee2a859f77fb2035f230018
Author:     Jin Yao <yao.jin@linux.intel.com>
AuthorDate: Tue, 5 Mar 2019 21:05:41 +0800
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Wed, 6 Mar 2019 18:03:23 -0300

perf diff: Support --time filter option

To improve 'perf diff', implement a --time filter option to diff the
samples within given time window.

It supports time percent with multiple time ranges. The time string
format is 'a%/n,b%/m,...' or 'a%-b%,c%-%d,...'.

For example:

Select the second 10% time slice to diff:

  perf diff --time 10%/2

Select from 0% to 10% time slice to diff:

  perf diff --time 0%-10%

Select the first and the second 10% time slices to diff:

  perf diff --time 10%/1,10%/2

Select from 0% to 10% and 30% to 40% slices to diff:

  perf diff --time 0%-10%,30%-40%

It also supports analysing samples within a given time window
<start>,<stop>.

Times have the format seconds.microseconds.

If 'start' is not given (i.e., time string is ',x.y') then analysis starts at
the beginning of the file.

If the stop time is not given (i.e, time string is 'x.y,') then analysis
goes to end of file.

Time string is 'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for
different perf.data files.

For example, we get the timestamp information from perf script.

  perf script -i perf.data.old

    mgen 13940 [000]  3946.361400: ...

  perf script -i perf.data

    mgen 13940 [000]  3971.150589 ...

  perf diff --time 3946.361400,:3971.150589,

It analyzes the perf.data.old from the timestamp 3946.361400 to the end of
perf.data.old and analyzes the perf.data from the timestamp 3971.150589 to the
end of perf.data.

 v4:
 ---
 Update abstime_str_dup(), let it return error if strdup
 is failed, and update __cmd_diff() accordingly.

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1551791143-10334-2-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-diff.txt |  45 ++++++++++
 tools/perf/builtin-diff.c              | 148 +++++++++++++++++++++++++++++----
 2 files changed, 179 insertions(+), 14 deletions(-)

diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
index a79c84ae61aa..036d65bded51 100644
--- a/tools/perf/Documentation/perf-diff.txt
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -118,6 +118,51 @@ OPTIONS
 	sum of shown entries will be always 100%.  "absolute" means it retains
 	the original value before and after the filter is applied.
 
+--time::
+	Analyze samples within given time window. It supports time
+	percent with multiple time ranges. Time string is 'a%/n,b%/m,...'
+	or 'a%-b%,c%-%d,...'.
+
+	For example:
+
+	Select the second 10% time slice to diff:
+
+	  perf diff --time 10%/2
+
+	Select from 0% to 10% time slice to diff:
+
+	  perf diff --time 0%-10%
+
+	Select the first and the second 10% time slices to diff:
+
+	  perf diff --time 10%/1,10%/2
+
+	Select from 0% to 10% and 30% to 40% slices to diff:
+
+	  perf diff --time 0%-10%,30%-40%
+
+	It also supports analyzing samples within a given time window
+	<start>,<stop>. Times have the format seconds.microseconds. If 'start'
+	is not given (i.e., time string is ',x.y') then analysis starts at
+	the beginning of the file. If stop time is not given (i.e, time
+	string is 'x.y,') then analysis goes to the end of the file. Time string is
+	'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for different
+	perf.data files.
+
+	For example, we get the timestamp information from 'perf script'.
+
+	  perf script -i perf.data.old
+	    mgen 13940 [000]  3946.361400: ...
+
+	  perf script -i perf.data
+	    mgen 13940 [000]  3971.150589 ...
+
+	  perf diff --time 3946.361400,:3971.150589,
+
+	It analyzes the perf.data.old from the timestamp 3946.361400 to
+	the end of perf.data.old and analyzes the perf.data from the
+	timestamp 3971.150589 to the end of perf.data.
+
 COMPARISON
 ----------
 The comparison is governed by the baseline file. The baseline perf.data
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
index 58fe0e88215c..17cd898074c8 100644
--- a/tools/perf/builtin-diff.c
+++ b/tools/perf/builtin-diff.c
@@ -19,12 +19,21 @@
 #include "util/util.h"
 #include "util/data.h"
 #include "util/config.h"
+#include "util/time-utils.h"
 
 #include <errno.h>
 #include <inttypes.h>
 #include <stdlib.h>
 #include <math.h>
 
+struct perf_diff {
+	struct perf_tool		 tool;
+	const char			*time_str;
+	struct perf_time_interval	*ptime_range;
+	int				 range_size;
+	int				 range_num;
+};
+
 /* Diff command specific HPP columns. */
 enum {
 	PERF_HPP_DIFF__BASELINE,
@@ -323,16 +332,22 @@ static int formula_fprintf(struct hist_entry *he, struct hist_entry *pair,
 	return -1;
 }
 
-static int diff__process_sample_event(struct perf_tool *tool __maybe_unused,
+static int diff__process_sample_event(struct perf_tool *tool,
 				      union perf_event *event,
 				      struct perf_sample *sample,
 				      struct perf_evsel *evsel,
 				      struct machine *machine)
 {
+	struct perf_diff *pdiff = container_of(tool, struct perf_diff, tool);
 	struct addr_location al;
 	struct hists *hists = evsel__hists(evsel);
 	int ret = -1;
 
+	if (perf_time__ranges_skip_sample(pdiff->ptime_range, pdiff->range_num,
+					  sample->time)) {
+		return 0;
+	}
+
 	if (machine__resolve(machine, &al, sample) < 0) {
 		pr_warning("problem processing %d event, skipping it.\n",
 			   event->header.type);
@@ -359,17 +374,19 @@ out_put:
 	return ret;
 }
 
-static struct perf_tool tool = {
-	.sample	= diff__process_sample_event,
-	.mmap	= perf_event__process_mmap,
-	.mmap2	= perf_event__process_mmap2,
-	.comm	= perf_event__process_comm,
-	.exit	= perf_event__process_exit,
-	.fork	= perf_event__process_fork,
-	.lost	= perf_event__process_lost,
-	.namespaces = perf_event__process_namespaces,
-	.ordered_events = true,
-	.ordering_requires_timestamps = true,
+static struct perf_diff pdiff = {
+	.tool = {
+		.sample	= diff__process_sample_event,
+		.mmap	= perf_event__process_mmap,
+		.mmap2	= perf_event__process_mmap2,
+		.comm	= perf_event__process_comm,
+		.exit	= perf_event__process_exit,
+		.fork	= perf_event__process_fork,
+		.lost	= perf_event__process_lost,
+		.namespaces = perf_event__process_namespaces,
+		.ordered_events = true,
+		.ordering_requires_timestamps = true,
+	},
 };
 
 static struct perf_evsel *evsel_match(struct perf_evsel *evsel,
@@ -771,19 +788,110 @@ static void data__free(struct data__file *d)
 	}
 }
 
+static int abstime_str_dup(char **pstr)
+{
+	char *str = NULL;
+
+	if (pdiff.time_str && strchr(pdiff.time_str, ':')) {
+		str = strdup(pdiff.time_str);
+		if (!str)
+			return -ENOMEM;
+	}
+
+	*pstr = str;
+	return 0;
+}
+
+static int parse_absolute_time(struct data__file *d, char **pstr)
+{
+	char *p = *pstr;
+	int ret;
+
+	/*
+	 * Absolute timestamp for one file has the format: a.b,c.d
+	 * For multiple files, the format is: a.b,c.d:a.b,c.d
+	 */
+	p = strchr(*pstr, ':');
+	if (p) {
+		if (p == *pstr) {
+			pr_err("Invalid time string\n");
+			return -EINVAL;
+		}
+
+		*p = 0;
+		p++;
+		if (*p == 0) {
+			pr_err("Invalid time string\n");
+			return -EINVAL;
+		}
+	}
+
+	ret = perf_time__parse_for_ranges(*pstr, d->session,
+					  &pdiff.ptime_range,
+					  &pdiff.range_size,
+					  &pdiff.range_num);
+	if (ret < 0)
+		return ret;
+
+	if (!p || *p == 0)
+		*pstr = NULL;
+	else
+		*pstr = p;
+
+	return ret;
+}
+
+static int parse_percent_time(struct data__file *d)
+{
+	int ret;
+
+	ret = perf_time__parse_for_ranges(pdiff.time_str, d->session,
+					  &pdiff.ptime_range,
+					  &pdiff.range_size,
+					  &pdiff.range_num);
+	return ret;
+}
+
+static int parse_time_str(struct data__file *d, char *abstime_ostr,
+			   char **pabstime_tmp)
+{
+	int ret = 0;
+
+	if (abstime_ostr)
+		ret = parse_absolute_time(d, pabstime_tmp);
+	else if (pdiff.time_str)
+		ret = parse_percent_time(d);
+
+	return ret;
+}
+
 static int __cmd_diff(void)
 {
 	struct data__file *d;
-	int ret = -EINVAL, i;
+	int ret, i;
+	char *abstime_ostr, *abstime_tmp;
+
+	ret = abstime_str_dup(&abstime_ostr);
+	if (ret)
+		return ret;
+
+	abstime_tmp = abstime_ostr;
+	ret = -EINVAL;
 
 	data__for_each_file(i, d) {
-		d->session = perf_session__new(&d->data, false, &tool);
+		d->session = perf_session__new(&d->data, false, &pdiff.tool);
 		if (!d->session) {
 			pr_err("Failed to open %s\n", d->data.path);
 			ret = -1;
 			goto out_delete;
 		}
 
+		if (pdiff.time_str) {
+			ret = parse_time_str(d, abstime_ostr, &abstime_tmp);
+			if (ret < 0)
+				goto out_delete;
+		}
+
 		ret = perf_session__process_events(d->session);
 		if (ret) {
 			pr_err("Failed to process %s\n", d->data.path);
@@ -791,6 +899,9 @@ static int __cmd_diff(void)
 		}
 
 		perf_evlist__collapse_resort(d->session->evlist);
+
+		if (pdiff.ptime_range)
+			zfree(&pdiff.ptime_range);
 	}
 
 	data_process();
@@ -802,6 +913,13 @@ static int __cmd_diff(void)
 	}
 
 	free(data__files);
+
+	if (pdiff.ptime_range)
+		zfree(&pdiff.ptime_range);
+
+	if (abstime_ostr)
+		free(abstime_ostr);
+
 	return ret;
 }
 
@@ -849,6 +967,8 @@ static const struct option options[] = {
 	OPT_UINTEGER('o', "order", &sort_compute, "Specify compute sorting."),
 	OPT_CALLBACK(0, "percentage", NULL, "relative|absolute",
 		     "How to display percentage of filtered entries", parse_filter_percentage),
+	OPT_STRING(0, "time", &pdiff.time_str, "str",
+		   "Time span (time percent or absolute timestamp)"),
 	OPT_END()
 };
 

[tip:perf/urgent] perf diff: Support --cpu filter option

[tip-bot for Jin Yao]

Commit-ID:  daca23b2007595b6a48255ca08c763f56050d1c5
Gitweb:     https://git.kernel.org/tip/daca23b2007595b6a48255ca08c763f56050d1c5
Author:     Jin Yao <yao.jin@linux.intel.com>
AuthorDate: Tue, 5 Mar 2019 21:05:42 +0800
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Wed, 6 Mar 2019 18:05:21 -0300

perf diff: Support --cpu filter option

To improve 'perf diff', implement a --cpu filter option.

Multiple CPUs can be provided as a comma-separated list with no space:
0,1.  Ranges of CPUs are specified with -: 0-2. Default is to report
samples on all CPUs.

For example,

  perf diff --cpu 0,1

It only diff the samples for CPU0 and CPU1.

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1551791143-10334-3-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-diff.txt |  5 +++++
 tools/perf/builtin-diff.c              | 16 ++++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
index 036d65bded51..8c2c229faf50 100644
--- a/tools/perf/Documentation/perf-diff.txt
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -163,6 +163,11 @@ OPTIONS
 	the end of perf.data.old and analyzes the perf.data from the
 	timestamp 3971.150589 to the end of perf.data.
 
+--cpu:: Only diff samples for the list of CPUs provided. Multiple CPUs can
+	be provided as a comma-separated list with no space: 0,1. Ranges of
+	CPUs are specified with -: 0-2. Default is to report samples on all
+	CPUs.
+
 COMPARISON
 ----------
 The comparison is governed by the baseline file. The baseline perf.data
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
index 17cd898074c8..dfe6c7606f5a 100644
--- a/tools/perf/builtin-diff.c
+++ b/tools/perf/builtin-diff.c
@@ -83,6 +83,9 @@ static unsigned int sort_compute = 1;
 static s64 compute_wdiff_w1;
 static s64 compute_wdiff_w2;
 
+static const char		*cpu_list;
+static DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
+
 enum {
 	COMPUTE_DELTA,
 	COMPUTE_RATIO,
@@ -354,6 +357,11 @@ static int diff__process_sample_event(struct perf_tool *tool,
 		return -1;
 	}
 
+	if (cpu_list && !test_bit(sample->cpu, cpu_bitmap)) {
+		ret = 0;
+		goto out_put;
+	}
+
 	if (!hists__add_entry(hists, &al, NULL, NULL, NULL, sample, true)) {
 		pr_warning("problem incrementing symbol period, skipping event\n");
 		goto out_put;
@@ -892,6 +900,13 @@ static int __cmd_diff(void)
 				goto out_delete;
 		}
 
+		if (cpu_list) {
+			ret = perf_session__cpu_bitmap(d->session, cpu_list,
+						       cpu_bitmap);
+			if (ret < 0)
+				goto out_delete;
+		}
+
 		ret = perf_session__process_events(d->session);
 		if (ret) {
 			pr_err("Failed to process %s\n", d->data.path);
@@ -969,6 +984,7 @@ static const struct option options[] = {
 		     "How to display percentage of filtered entries", parse_filter_percentage),
 	OPT_STRING(0, "time", &pdiff.time_str, "str",
 		   "Time span (time percent or absolute timestamp)"),
+	OPT_STRING(0, "cpu", &cpu_list, "cpu", "list of cpus to profile"),
 	OPT_END()
 };
 

[tip:perf/urgent] perf diff: Support --pid/--tid filter options

[tip-bot for Jin Yao]

Commit-ID:  c1d3e633e16db3eb64f519c7099171bfcef94b20
Gitweb:     https://git.kernel.org/tip/c1d3e633e16db3eb64f519c7099171bfcef94b20
Author:     Jin Yao <yao.jin@linux.intel.com>
AuthorDate: Tue, 5 Mar 2019 21:05:43 +0800
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Wed, 6 Mar 2019 18:06:16 -0300

perf diff: Support --pid/--tid filter options

Using the existing symbol_conf.pid_list_str and symbol_conf.tid_list_str
logic.

For example:

  perf diff --tid 13965

It'll only diff the samples for thread 13965.

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1551791143-10334-4-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-diff.txt | 6 ++++++
 tools/perf/builtin-diff.c              | 4 ++++
 2 files changed, 10 insertions(+)

diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
index 8c2c229faf50..da7809b15cc9 100644
--- a/tools/perf/Documentation/perf-diff.txt
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -168,6 +168,12 @@ OPTIONS
 	CPUs are specified with -: 0-2. Default is to report samples on all
 	CPUs.
 
+--pid=::
+	Only diff samples for given process ID (comma separated list).
+
+--tid=::
+	Only diff samples for given thread ID (comma separated list).
+
 COMPARISON
 ----------
 The comparison is governed by the baseline file. The baseline perf.data
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
index dfe6c7606f5a..6e7920793729 100644
--- a/tools/perf/builtin-diff.c
+++ b/tools/perf/builtin-diff.c
@@ -985,6 +985,10 @@ static const struct option options[] = {
 	OPT_STRING(0, "time", &pdiff.time_str, "str",
 		   "Time span (time percent or absolute timestamp)"),
 	OPT_STRING(0, "cpu", &cpu_list, "cpu", "list of cpus to profile"),
+	OPT_STRING(0, "pid", &symbol_conf.pid_list_str, "pid[,pid...]",
+		   "only consider symbols in these pids"),
+	OPT_STRING(0, "tid", &symbol_conf.tid_list_str, "tid[,tid...]",
+		   "only consider symbols in these tids"),
 	OPT_END()
 };