[GIT PULL 0/3] perf/urgent fixes

[GIT PULL 0/3] perf/urgent fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

	Please consider pulling,

- Arnaldo

The following changes since commit 1018faa6cf23b256bf25919ef203cd7c129f06f2:

  perf/x86/kvm: Fix Host-Only/Guest-Only counting with SVM disabled (2012-03-02 12:16:39 +0100)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux tags/perf-urgent-for-mingo

for you to fetch changes up to bc76efe64533305b55d1f0834fd03414da8a12b2:

  perf tools: Handle kernels that don't support attr.exclude_{guest,host} (2012-03-03 12:19:56 -0300)

----------------------------------------------------------------
Cherry picked fixes from perf/core, together with the kernel fix (1018faa),
the sampling tools (top, record) are back working on AMD systems.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

----------------------------------------------------------------
Arnaldo Carvalho de Melo (1):
      perf tools: Handle kernels that don't support attr.exclude_{guest,host}

David Ahern (1):
      perf record: No build id option fails

Joerg Roedel (1):
      perf tools: Change perf_guest default back to false

 tools/perf/builtin-record.c |   33 ++++++++++++++++++++++-----------
 tools/perf/builtin-top.c    |   23 +++++++++++++++++------
 tools/perf/perf.h           |    1 +
 tools/perf/util/top.h       |    1 +
 tools/perf/util/util.c      |    2 +-
 5 files changed, 42 insertions(+), 18 deletions(-)

[PATCH 1/3] perf record: No build id option fails

[Arnaldo Carvalho de Melo]

From: David Ahern <dsahern@gmail.com>

A recent refactoring of perf-record introduced the following:

perf record -a -B
Couldn't generating buildids. Use --no-buildid to profile anyway.
sleep: Terminated

I believe the triple negative was meant to be only a double negative.
:-) While I'm there, fixed the grammar on the error message.

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1328567272-13190-1-git-send-email-dsahern@gmail.com
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-record.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 0abfb18..39b6a40 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -503,9 +503,9 @@ static int __cmd_record(struct perf_record *rec, int argc, const char **argv)
 			return err;
 	}
 
-	if (!!rec->no_buildid
+	if (!rec->no_buildid
 	    && !perf_header__has_feat(&session->header, HEADER_BUILD_ID)) {
-		pr_err("Couldn't generating buildids. "
+		pr_err("Couldn't generate buildids. "
 		       "Use --no-buildid to profile anyway.\n");
 		return -1;
 	}
-- 
1.7.9.2.358.g22243

[PATCH 2/3] perf tools: Change perf_guest default back to false

[Arnaldo Carvalho de Melo]

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

Setting perf_guest to true by default makes no sense because the perf
subcommands can not setup guest symbol information and thus not process
and guest samples. The only exception is perf-kvm which changes the
perf_guest value on its own.  So change the default for perf_guest back
to false.

Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1328893505-4115-3-git-send-email-joerg.roedel@amd.com
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/util.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index 8131410..fb25d13 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -6,7 +6,7 @@
  * XXX We need to find a better place for these things...
  */
 bool perf_host  = true;
-bool perf_guest = true;
+bool perf_guest = false;
 
 void event_attr_init(struct perf_event_attr *attr)
 {
-- 
1.7.9.2.358.g22243

[PATCH 3/3] perf tools: Handle kernels that don't support attr.exclude_{guest,host}

[Arnaldo Carvalho de Melo]

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

Just fall back to resetting those fields, if set, warning the user that
that feature is not available.

If guest samples appear they will just be discarded because no struct
machine will be found and thus the event will be accounted as not
handled and dropped, see 0c09571.

Reported-by: Namhyung Kim <namhyung@gmail.com>
Tested-by: Joerg Roedel <joerg.roedel@amd.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Joerg Roedel <joerg.roedel@amd.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-vuwxig36mzprl5n7nzvnxxsh@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-record.c |   29 ++++++++++++++++++++---------
 tools/perf/builtin-top.c    |   23 +++++++++++++++++------
 tools/perf/perf.h           |    1 +
 tools/perf/util/top.h       |    1 +
 4 files changed, 39 insertions(+), 15 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 39b6a40..227b6ae 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -204,6 +204,9 @@ static void perf_record__open(struct perf_record *rec)
 
 		if (opts->group && pos != first)
 			group_fd = first->fd;
+fallback_missing_features:
+		if (opts->exclude_guest_missing)
+			attr->exclude_guest = attr->exclude_host = 0;
 retry_sample_id:
 		attr->sample_id_all = opts->sample_id_all_avail ? 1 : 0;
 try_again:
@@ -217,15 +220,23 @@ try_again:
 			} else if (err ==  ENODEV && opts->cpu_list) {
 				die("No such device - did you specify"
 					" an out-of-range profile CPU?\n");
-			} else if (err == EINVAL && opts->sample_id_all_avail) {
-				/*
-				 * Old kernel, no attr->sample_id_type_all field
-				 */
-				opts->sample_id_all_avail = false;
-				if (!opts->sample_time && !opts->raw_samples && !time_needed)
-					attr->sample_type &= ~PERF_SAMPLE_TIME;
-
-				goto retry_sample_id;
+			} else if (err == EINVAL) {
+				if (!opts->exclude_guest_missing &&
+				    (attr->exclude_guest || attr->exclude_host)) {
+					pr_debug("Old kernel, cannot exclude "
+						 "guest or host samples.\n");
+					opts->exclude_guest_missing = true;
+					goto fallback_missing_features;
+				} else if (opts->sample_id_all_avail) {
+					/*
+					 * Old kernel, no attr->sample_id_type_all field
+					 */
+					opts->sample_id_all_avail = false;
+					if (!opts->sample_time && !opts->raw_samples && !time_needed)
+						attr->sample_type &= ~PERF_SAMPLE_TIME;
+
+					goto retry_sample_id;
+				}
 			}
 
 			/*
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index dd162aa..ecff312 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -857,6 +857,9 @@ static void perf_top__start_counters(struct perf_top *top)
 		attr->mmap = 1;
 		attr->comm = 1;
 		attr->inherit = top->inherit;
+fallback_missing_features:
+		if (top->exclude_guest_missing)
+			attr->exclude_guest = attr->exclude_host = 0;
 retry_sample_id:
 		attr->sample_id_all = top->sample_id_all_avail ? 1 : 0;
 try_again:
@@ -868,12 +871,20 @@ try_again:
 			if (err == EPERM || err == EACCES) {
 				ui__error_paranoid();
 				goto out_err;
-			} else if (err == EINVAL && top->sample_id_all_avail) {
-				/*
-				 * Old kernel, no attr->sample_id_type_all field
-				 */
-				top->sample_id_all_avail = false;
-				goto retry_sample_id;
+			} else if (err == EINVAL) {
+				if (!top->exclude_guest_missing &&
+				    (attr->exclude_guest || attr->exclude_host)) {
+					pr_debug("Old kernel, cannot exclude "
+						 "guest or host samples.\n");
+					top->exclude_guest_missing = true;
+					goto fallback_missing_features;
+				} else if (top->sample_id_all_avail) {
+					/*
+					 * Old kernel, no attr->sample_id_type_all field
+					 */
+					top->sample_id_all_avail = false;
+					goto retry_sample_id;
+				}
 			}
 			/*
 			 * If it's cycles then fall back to hrtimer
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 64f8bee..16e7d20 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -199,6 +199,7 @@ struct perf_record_opts {
 	bool	     sample_address;
 	bool	     sample_time;
 	bool	     sample_id_all_avail;
+	bool	     exclude_guest_missing;
 	bool	     system_wide;
 	bool	     period;
 	unsigned int freq;
diff --git a/tools/perf/util/top.h b/tools/perf/util/top.h
index a248f3c..f2eab81 100644
--- a/tools/perf/util/top.h
+++ b/tools/perf/util/top.h
@@ -34,6 +34,7 @@ struct perf_top {
 	bool		   inherit;
 	bool		   group;
 	bool		   sample_id_all_avail;
+	bool		   exclude_guest_missing;
 	bool		   dump_symtab;
 	const char	   *cpu_list;
 	struct hist_entry  *sym_filter_entry;
-- 
1.7.9.2.358.g22243