[GIT PULL 0/6] perf/core cleanups, fixes and improvements

[GIT PULL 0/6] perf/core cleanups, fixes and improvements

[Arnaldo Carvalho de Melo]

Hi Ingo,

        Please consider pulling from:

git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux perf/core

Regards,

- Arnaldo

The following changes since commit 9e183426bfb52bb44bf3c443d6587e4d02478603:

  perf kvm: Fix copy & paste error in description (2012-01-03 15:17:14 -0200)

are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux perf/core

Arnaldo Carvalho de Melo (3):
      perf hists: Rename total_session to total_period
      perf hists: Stop using 'self' for struct hist_entry
      perf top: Don't update total_period on process_sample

Joerg Roedel (2):
      perf kvm: Do guest-only counting by default
      perf tools: Add support for guest/host-only profiling

Sebastian Andrzej Siewior (1):
      perf tools: Add const.h to MANIFEST to make perf-tar-src-pkg work again

 tools/perf/Documentation/perf-list.txt |    2 +
 tools/perf/MANIFEST                    |    1 +
 tools/perf/builtin-kvm.c               |    6 +-
 tools/perf/builtin-top.c               |    1 -
 tools/perf/util/evlist.c               |    5 +-
 tools/perf/util/hist.c                 |   97 ++++++++++++++++---------------
 tools/perf/util/hist.h                 |    7 +--
 tools/perf/util/parse-events.c         |   15 ++++-
 tools/perf/util/util.c                 |   15 +++++
 tools/perf/util/util.h                 |    4 +
 10 files changed, 93 insertions(+), 60 deletions(-)

[PATCH 1/6] perf hists: Rename total_session to total_period

[Arnaldo Carvalho de Melo]

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

Nowadays we do it per evsel, not per session (that may have multiple
evsels), so rename it to avoid confusion.

Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-azsgomr5h4dmaudoogw48w49@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/hist.c |   16 ++++++++--------
 1 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index abef270..20059d1 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -733,7 +733,7 @@ void hists__output_recalc_col_len(struct hists *hists, int max_rows)
 static int hist_entry__pcnt_snprintf(struct hist_entry *self, char *s,
 				     size_t size, struct hists *pair_hists,
 				     bool show_displacement, long displacement,
-				     bool color, u64 session_total)
+				     bool color, u64 total_period)
 {
 	u64 period, total, period_sys, period_us, period_guest_sys, period_guest_us;
 	u64 nr_events;
@@ -754,7 +754,7 @@ static int hist_entry__pcnt_snprintf(struct hist_entry *self, char *s,
 	} else {
 		period = self->period;
 		nr_events = self->nr_events;
-		total = session_total;
+		total = total_period;
 		period_sys = self->period_sys;
 		period_us = self->period_us;
 		period_guest_sys = self->period_guest_sys;
@@ -812,8 +812,8 @@ static int hist_entry__pcnt_snprintf(struct hist_entry *self, char *s,
 
 		if (total > 0)
 			old_percent = (period * 100.0) / total;
-		if (session_total > 0)
-			new_percent = (self->period * 100.0) / session_total;
+		if (total_period > 0)
+			new_percent = (self->period * 100.0) / total_period;
 
 		diff = new_percent - old_percent;
 
@@ -864,7 +864,7 @@ int hist_entry__snprintf(struct hist_entry *he, char *s, size_t size,
 
 int hist_entry__fprintf(struct hist_entry *he, size_t size, struct hists *hists,
 			struct hists *pair_hists, bool show_displacement,
-			long displacement, FILE *fp, u64 session_total)
+			long displacement, FILE *fp, u64 total_period)
 {
 	char bf[512];
 	int ret;
@@ -874,14 +874,14 @@ int hist_entry__fprintf(struct hist_entry *he, size_t size, struct hists *hists,
 
 	ret = hist_entry__pcnt_snprintf(he, bf, size, pair_hists,
 					show_displacement, displacement,
-					true, session_total);
+					true, total_period);
 	hist_entry__snprintf(he, bf + ret, size - ret, hists);
 	return fprintf(fp, "%s\n", bf);
 }
 
 static size_t hist_entry__fprintf_callchain(struct hist_entry *self,
 					    struct hists *hists, FILE *fp,
-					    u64 session_total)
+					    u64 total_period)
 {
 	int left_margin = 0;
 
@@ -892,7 +892,7 @@ static size_t hist_entry__fprintf_callchain(struct hist_entry *self,
 		left_margin -= thread__comm_len(self->thread);
 	}
 
-	return hist_entry_callchain__fprintf(fp, self, session_total,
+	return hist_entry_callchain__fprintf(fp, self, total_period,
 					     left_margin);
 }
 
-- 
1.7.4.4

[PATCH 2/6] perf hists: Stop using 'self' for struct hist_entry

[Arnaldo Carvalho de Melo]

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

Stop using this python/OOP convention, doesn't really helps. Will do
more from time to time till we get it cleaned up in all of /perf.

Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-me4dyj6s5snh7jr8wb9gzt82@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/hist.c |   89 +++++++++++++++++++++++++-----------------------
 tools/perf/util/hist.h |    7 +---
 2 files changed, 48 insertions(+), 48 deletions(-)

diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index 20059d1..4df4495 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -76,21 +76,21 @@ static void hists__calc_col_len(struct hists *hists, struct hist_entry *h)
 	}
 }
 
-static void hist_entry__add_cpumode_period(struct hist_entry *self,
+static void hist_entry__add_cpumode_period(struct hist_entry *he,
 					   unsigned int cpumode, u64 period)
 {
 	switch (cpumode) {
 	case PERF_RECORD_MISC_KERNEL:
-		self->period_sys += period;
+		he->period_sys += period;
 		break;
 	case PERF_RECORD_MISC_USER:
-		self->period_us += period;
+		he->period_us += period;
 		break;
 	case PERF_RECORD_MISC_GUEST_KERNEL:
-		self->period_guest_sys += period;
+		he->period_guest_sys += period;
 		break;
 	case PERF_RECORD_MISC_GUEST_USER:
-		self->period_guest_us += period;
+		he->period_guest_us += period;
 		break;
 	default:
 		break;
@@ -165,18 +165,18 @@ void hists__decay_entries_threaded(struct hists *hists,
 static struct hist_entry *hist_entry__new(struct hist_entry *template)
 {
 	size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_root) : 0;
-	struct hist_entry *self = malloc(sizeof(*self) + callchain_size);
+	struct hist_entry *he = malloc(sizeof(*he) + callchain_size);
 
-	if (self != NULL) {
-		*self = *template;
-		self->nr_events = 1;
-		if (self->ms.map)
-			self->ms.map->referenced = true;
+	if (he != NULL) {
+		*he = *template;
+		he->nr_events = 1;
+		if (he->ms.map)
+			he->ms.map->referenced = true;
 		if (symbol_conf.use_callchain)
-			callchain_init(self->callchain);
+			callchain_init(he->callchain);
 	}
 
-	return self;
+	return he;
 }
 
 static void hists__inc_nr_entries(struct hists *hists, struct hist_entry *h)
@@ -677,15 +677,16 @@ static size_t callchain__fprintf_flat(FILE *fp, struct callchain_node *self,
 	return ret;
 }
 
-static size_t hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
-					    u64 total_samples, int left_margin)
+static size_t hist_entry_callchain__fprintf(struct hist_entry *he,
+					    u64 total_samples, int left_margin,
+					    FILE *fp)
 {
 	struct rb_node *rb_node;
 	struct callchain_node *chain;
 	size_t ret = 0;
 	u32 entries_printed = 0;
 
-	rb_node = rb_first(&self->sorted_chain);
+	rb_node = rb_first(&he->sorted_chain);
 	while (rb_node) {
 		double percent;
 
@@ -730,7 +731,7 @@ void hists__output_recalc_col_len(struct hists *hists, int max_rows)
 	}
 }
 
-static int hist_entry__pcnt_snprintf(struct hist_entry *self, char *s,
+static int hist_entry__pcnt_snprintf(struct hist_entry *he, char *s,
 				     size_t size, struct hists *pair_hists,
 				     bool show_displacement, long displacement,
 				     bool color, u64 total_period)
@@ -740,25 +741,25 @@ static int hist_entry__pcnt_snprintf(struct hist_entry *self, char *s,
 	const char *sep = symbol_conf.field_sep;
 	int ret;
 
-	if (symbol_conf.exclude_other && !self->parent)
+	if (symbol_conf.exclude_other && !he->parent)
 		return 0;
 
 	if (pair_hists) {
-		period = self->pair ? self->pair->period : 0;
-		nr_events = self->pair ? self->pair->nr_events : 0;
+		period = he->pair ? he->pair->period : 0;
+		nr_events = he->pair ? he->pair->nr_events : 0;
 		total = pair_hists->stats.total_period;
-		period_sys = self->pair ? self->pair->period_sys : 0;
-		period_us = self->pair ? self->pair->period_us : 0;
-		period_guest_sys = self->pair ? self->pair->period_guest_sys : 0;
-		period_guest_us = self->pair ? self->pair->period_guest_us : 0;
+		period_sys = he->pair ? he->pair->period_sys : 0;
+		period_us = he->pair ? he->pair->period_us : 0;
+		period_guest_sys = he->pair ? he->pair->period_guest_sys : 0;
+		period_guest_us = he->pair ? he->pair->period_guest_us : 0;
 	} else {
-		period = self->period;
-		nr_events = self->nr_events;
+		period = he->period;
+		nr_events = he->nr_events;
 		total = total_period;
-		period_sys = self->period_sys;
-		period_us = self->period_us;
-		period_guest_sys = self->period_guest_sys;
-		period_guest_us = self->period_guest_us;
+		period_sys = he->period_sys;
+		period_us = he->period_us;
+		period_guest_sys = he->period_guest_sys;
+		period_guest_us = he->period_guest_us;
 	}
 
 	if (total) {
@@ -813,7 +814,7 @@ static int hist_entry__pcnt_snprintf(struct hist_entry *self, char *s,
 		if (total > 0)
 			old_percent = (period * 100.0) / total;
 		if (total_period > 0)
-			new_percent = (self->period * 100.0) / total_period;
+			new_percent = (he->period * 100.0) / total_period;
 
 		diff = new_percent - old_percent;
 
@@ -862,9 +863,10 @@ int hist_entry__snprintf(struct hist_entry *he, char *s, size_t size,
 	return ret;
 }
 
-int hist_entry__fprintf(struct hist_entry *he, size_t size, struct hists *hists,
-			struct hists *pair_hists, bool show_displacement,
-			long displacement, FILE *fp, u64 total_period)
+static int hist_entry__fprintf(struct hist_entry *he, size_t size,
+			       struct hists *hists, struct hists *pair_hists,
+			       bool show_displacement, long displacement,
+			       u64 total_period, FILE *fp)
 {
 	char bf[512];
 	int ret;
@@ -879,9 +881,9 @@ int hist_entry__fprintf(struct hist_entry *he, size_t size, struct hists *hists,
 	return fprintf(fp, "%s\n", bf);
 }
 
-static size_t hist_entry__fprintf_callchain(struct hist_entry *self,
-					    struct hists *hists, FILE *fp,
-					    u64 total_period)
+static size_t hist_entry__fprintf_callchain(struct hist_entry *he,
+					    struct hists *hists,
+					    u64 total_period, FILE *fp)
 {
 	int left_margin = 0;
 
@@ -889,11 +891,10 @@ static size_t hist_entry__fprintf_callchain(struct hist_entry *self,
 		struct sort_entry *se = list_first_entry(&hist_entry__sort_list,
 							 typeof(*se), list);
 		left_margin = hists__col_len(hists, se->se_width_idx);
-		left_margin -= thread__comm_len(self->thread);
+		left_margin -= thread__comm_len(he->thread);
 	}
 
-	return hist_entry_callchain__fprintf(fp, self, total_period,
-					     left_margin);
+	return hist_entry_callchain__fprintf(he, total_period, left_margin, fp);
 }
 
 size_t hists__fprintf(struct hists *hists, struct hists *pair,
@@ -903,6 +904,7 @@ size_t hists__fprintf(struct hists *hists, struct hists *pair,
 	struct sort_entry *se;
 	struct rb_node *nd;
 	size_t ret = 0;
+	u64 total_period;
 	unsigned long position = 1;
 	long displacement = 0;
 	unsigned int width;
@@ -1025,6 +1027,8 @@ size_t hists__fprintf(struct hists *hists, struct hists *pair,
 		goto out;
 
 print_entries:
+	total_period = hists->stats.total_period;
+
 	for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) {
 		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
 
@@ -1040,11 +1044,10 @@ print_entries:
 			++position;
 		}
 		ret += hist_entry__fprintf(h, max_cols, hists, pair, show_displacement,
-					   displacement, fp, hists->stats.total_period);
+					   displacement, total_period, fp);
 
 		if (symbol_conf.use_callchain)
-			ret += hist_entry__fprintf_callchain(h, hists, fp,
-							     hists->stats.total_period);
+			ret += hist_entry__fprintf_callchain(h, hists, total_period, fp);
 		if (max_rows && ++nr_rows >= max_rows)
 			goto out;
 
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index ff6f9d5..f55f0a8d 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -66,11 +66,8 @@ struct hists {
 struct hist_entry *__hists__add_entry(struct hists *self,
 				      struct addr_location *al,
 				      struct symbol *parent, u64 period);
-extern int64_t hist_entry__cmp(struct hist_entry *, struct hist_entry *);
-extern int64_t hist_entry__collapse(struct hist_entry *, struct hist_entry *);
-int hist_entry__fprintf(struct hist_entry *he, size_t size, struct hists *hists,
-			struct hists *pair_hists, bool show_displacement,
-			long displacement, FILE *fp, u64 session_total);
+int64_t hist_entry__cmp(struct hist_entry *left, struct hist_entry *right);
+int64_t hist_entry__collapse(struct hist_entry *left, struct hist_entry *right);
 int hist_entry__snprintf(struct hist_entry *self, char *bf, size_t size,
 			 struct hists *hists);
 void hist_entry__free(struct hist_entry *);
-- 
1.7.4.4

[PATCH 3/6] perf top: Don't update total_period on process_sample

[Arnaldo Carvalho de Melo]

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

It will be recalculated at __hists__output_resort, to take into account
filters possibly applied by the TUI, etc.

Since we do the percent math only for those entries that will appear on
the TUI instead of for _all_ the entries at decay time, updating it for
each sample makes the entries seem to decay faster when using the
navigation keys (since the screen will be refreshed), as we're not
coalescing the entries that are being batched to be merged at next
resort/decay time, but considering their periods.

Bug introduced in 743eb86.

Reported-by: Ingo Molnar <mingo@elte.hu>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-k0d0rq9a8nqtkqohov8cir72@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-top.c |    1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 4f81eeb..d89dec9 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -235,7 +235,6 @@ static struct hist_entry *perf_evsel__add_hist_entry(struct perf_evsel *evsel,
 	if (he == NULL)
 		return NULL;
 
-	evsel->hists.stats.total_period += sample->period;
 	hists__inc_nr_events(&evsel->hists, PERF_RECORD_SAMPLE);
 	return he;
 }
-- 
1.7.4.4

[PATCH 4/6] perf kvm: Do guest-only counting by default

[Arnaldo Carvalho de Melo]

From: Joerg Roedel <joerg.roedel@amd.com>

Make use of exclude_guest and exlude_host in perf-kvm to do only
guest-only counting by default.

Cc: Gleb Natapov <gleb@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
[ committer note: Moved perf_{guest,host} & event_attr_init to util.c ]
[                 so as not to drag more stuff to the python binding]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-kvm.c       |    6 ++----
 tools/perf/util/evlist.c       |    5 ++++-
 tools/perf/util/parse-events.c |    1 +
 tools/perf/util/util.c         |   15 +++++++++++++++
 tools/perf/util/util.h         |    4 ++++
 5 files changed, 26 insertions(+), 5 deletions(-)

diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
index 032324a..9fc6e0f 100644
--- a/tools/perf/builtin-kvm.c
+++ b/tools/perf/builtin-kvm.c
@@ -22,9 +22,6 @@
 static const char		*file_name;
 static char			name_buffer[256];
 
-bool				perf_host = 1;
-bool				perf_guest;
-
 static const char * const kvm_usage[] = {
 	"perf kvm [<options>] {top|record|report|diff|buildid-list}",
 	NULL
@@ -107,7 +104,8 @@ static int __cmd_buildid_list(int argc, const char **argv)
 
 int cmd_kvm(int argc, const char **argv, const char *prefix __used)
 {
-	perf_host = perf_guest = 0;
+	perf_host  = 0;
+	perf_guest = 1;
 
 	argc = parse_options(argc, argv, kvm_options, kvm_usage,
 			PARSE_OPT_STOP_AT_NON_OPTION);
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index fa18370..3f16e08 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -111,8 +111,11 @@ int perf_evlist__add_default(struct perf_evlist *evlist)
 		.type = PERF_TYPE_HARDWARE,
 		.config = PERF_COUNT_HW_CPU_CYCLES,
 	};
-	struct perf_evsel *evsel = perf_evsel__new(&attr, 0);
+	struct perf_evsel *evsel;
+
+	event_attr_init(&attr);
 
+	evsel = perf_evsel__new(&attr, 0);
 	if (evsel == NULL)
 		goto error;
 
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 531c283..dcf999c 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -838,6 +838,7 @@ int parse_events(struct perf_evlist *evlist , const char *str, int unset __used)
 	for (;;) {
 		ostr = str;
 		memset(&attr, 0, sizeof(attr));
+		event_attr_init(&attr);
 		ret = parse_event_symbols(evlist, &str, &attr);
 		if (ret == EVT_FAILED)
 			return -1;
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index 5b3ea49..8131410 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -1,6 +1,21 @@
+#include "../perf.h"
 #include "util.h"
 #include <sys/mman.h>
 
+/*
+ * XXX We need to find a better place for these things...
+ */
+bool perf_host  = true;
+bool perf_guest = true;
+
+void event_attr_init(struct perf_event_attr *attr)
+{
+	if (!perf_host)
+		attr->exclude_host  = 1;
+	if (!perf_guest)
+		attr->exclude_guest = 1;
+}
+
 int mkdir_p(char *path, mode_t mode)
 {
 	struct stat st;
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 37be34d..b9c530c 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -242,6 +242,10 @@ int strtailcmp(const char *s1, const char *s2);
 unsigned long convert_unit(unsigned long value, char *unit);
 int readn(int fd, void *buf, size_t size);
 
+struct perf_event_attr;
+
+void event_attr_init(struct perf_event_attr *attr);
+
 #define _STR(x) #x
 #define STR(x) _STR(x)
 
-- 
1.7.4.4

[PATCH 5/6] perf tools: Add support for guest/host-only profiling

[Arnaldo Carvalho de Melo]

From: Joerg Roedel <joerg.roedel@amd.com>

To restrict a counter to either host or guest mode this patch introduces
two new event modifiers: G and H.

With G the counter is configured in guest-only mode and with H in
host-only mode.

Cc: Gleb Natapov <gleb@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Link: http://lkml.kernel.org/n/tip-or5aj3rghy9ngyg882z6kln9@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-list.txt |    2 ++
 tools/perf/util/parse-events.c         |   14 ++++++++++++--
 2 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt
index 7a527f7..ddc2252 100644
--- a/tools/perf/Documentation/perf-list.txt
+++ b/tools/perf/Documentation/perf-list.txt
@@ -21,6 +21,8 @@ EVENT MODIFIERS
 Events can optionally have a modifer by appending a colon and one or
 more modifiers.  Modifiers allow the user to restrict when events are
 counted with 'u' for user-space, 'k' for kernel, 'h' for hypervisor.
+Additional modifiers are 'G' for guest counting (in KVM guests) and 'H'
+for host counting (not in KVM guests).
 
 The 'p' modifier can be used for specifying how precise the instruction
 address should be. The 'p' modifier is currently only implemented for
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index dcf999c..b029296 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -735,8 +735,8 @@ static int
 parse_event_modifier(const char **strp, struct perf_event_attr *attr)
 {
 	const char *str = *strp;
-	int exclude = 0;
-	int eu = 0, ek = 0, eh = 0, precise = 0;
+	int exclude = 0, exclude_GH = 0;
+	int eu = 0, ek = 0, eh = 0, eH = 0, eG = 0, precise = 0;
 
 	if (!*str)
 		return 0;
@@ -760,6 +760,14 @@ parse_event_modifier(const char **strp, struct perf_event_attr *attr)
 			if (!exclude)
 				exclude = eu = ek = eh = 1;
 			eh = 0;
+		} else if (*str == 'G') {
+			if (!exclude_GH)
+				exclude_GH = eG = eH = 1;
+			eG = 0;
+		} else if (*str == 'H') {
+			if (!exclude_GH)
+				exclude_GH = eG = eH = 1;
+			eH = 0;
 		} else if (*str == 'p') {
 			precise++;
 		} else
@@ -776,6 +784,8 @@ parse_event_modifier(const char **strp, struct perf_event_attr *attr)
 	attr->exclude_kernel = ek;
 	attr->exclude_hv     = eh;
 	attr->precise_ip     = precise;
+	attr->exclude_host   = eH;
+	attr->exclude_guest  = eG;
 
 	return 0;
 }
-- 
1.7.4.4

[PATCH 6/6] perf tools: Add const.h to MANIFEST to make perf-tar-src-pkg work again

[Arnaldo Carvalho de Melo]

From: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>

Fixes:
|make: *** No rule to make target `../../include/linux/const.h', needed by `builtin-annotate.o'.  Stop.

Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1324128938-17553-1-git-send-email-sebastian@breakpoint.cc
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/MANIFEST |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index c12659d..1078c5f 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -1,4 +1,5 @@
 tools/perf
+include/linux/const.h
 include/linux/perf_event.h
 include/linux/rbtree.h
 include/linux/list.h
-- 
1.7.4.4

Re: [GIT PULL 0/6] perf/core cleanups, fixes and improvements

[Ingo Molnar]

* Arnaldo Carvalho de Melo <acme@infradead.org> wrote:

> Hi Ingo,
> 
>         Please consider pulling from:
> 
> git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux perf/core
> 
> Regards,
> 
> - Arnaldo
> 
> The following changes since commit 9e183426bfb52bb44bf3c443d6587e4d02478603:
> 
>   perf kvm: Fix copy & paste error in description (2012-01-03 15:17:14 -0200)
> 
> are available in the git repository at:
>   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux perf/core
> 
> Arnaldo Carvalho de Melo (3):
>       perf hists: Rename total_session to total_period
>       perf hists: Stop using 'self' for struct hist_entry
>       perf top: Don't update total_period on process_sample
> 
> Joerg Roedel (2):
>       perf kvm: Do guest-only counting by default
>       perf tools: Add support for guest/host-only profiling
> 
> Sebastian Andrzej Siewior (1):
>       perf tools: Add const.h to MANIFEST to make perf-tar-src-pkg work again
> 
>  tools/perf/Documentation/perf-list.txt |    2 +
>  tools/perf/MANIFEST                    |    1 +
>  tools/perf/builtin-kvm.c               |    6 +-
>  tools/perf/builtin-top.c               |    1 -
>  tools/perf/util/evlist.c               |    5 +-
>  tools/perf/util/hist.c                 |   97 ++++++++++++++++---------------
>  tools/perf/util/hist.h                 |    7 +--
>  tools/perf/util/parse-events.c         |   15 ++++-
>  tools/perf/util/util.c                 |   15 +++++
>  tools/perf/util/util.h                 |    4 +
>  10 files changed, 93 insertions(+), 60 deletions(-)

Pulled, thanks Arnaldo!

	Ingo