[GIT PULL 0/3] perf/core improvements and fixes

[GIT PULL 0/3] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

        Please consider pulling from:

git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 perf/core

	This one has everything in acme/perf/urgent, i.e. starts the next
tools/ devel cycle :-)

Regards,

- Arnaldo

Arnaldo Carvalho de Melo (1):
  perf tools: Fixup exit path when not able to open events

David Ahern (1):
  perf tools: Emit clearer message for sys_perf_event_open ENOENT
    return

Lin Ming (1):
  perf probe: Add fastpath to do lookup by function name

 tools/perf/builtin-record.c    |   19 +++++++---
 tools/perf/builtin-top.c       |   44 ++++++++++++++++--------
 tools/perf/util/debug.c        |   10 +++++
 tools/perf/util/debug.h        |    1 +
 tools/perf/util/probe-finder.c |   72 ++++++++++++++++++++++++++++++++++++++++
 tools/perf/util/probe-finder.h |    2 +
 6 files changed, 128 insertions(+), 20 deletions(-)

[PATCH 1/3] perf tools: Fixup exit path when not able to open events

[Arnaldo Carvalho de Melo]

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

We have to deal with the TUI mode in perf top, so that we don't end up
with a garbled screen when, say, a non root user on a machine with a
paranoid setting (the default) tries to use 'perf top'.

Introduce a ui__warning_paranoid() routine shared by top and record that
tells the user the valid values for /proc/sys/kernel/perf_event_paranoid.

Cc: David Ahern <daahern@cisco.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-record.c |    9 ++++-----
 tools/perf/builtin-top.c    |   38 ++++++++++++++++++++++++--------------
 tools/perf/util/debug.c     |   10 ++++++++++
 tools/perf/util/debug.h     |    1 +
 4 files changed, 39 insertions(+), 19 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 6febcc1..623695e 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -275,11 +275,10 @@ try_again:
 				     !no_inherit) < 0) {
 			int err = errno;
 
-			if (err == EPERM || err == EACCES)
-				die("Permission error - are you root?\n"
-					"\t Consider tweaking"
-					" /proc/sys/kernel/perf_event_paranoid.\n");
-			else if (err ==  ENODEV && cpu_list) {
+			if (err == EPERM || err == EACCES) {
+				ui__warning_paranoid();
+				exit(EXIT_FAILURE);
+			} else if (err ==  ENODEV && cpu_list) {
 				die("No such device - did you specify"
 					" an out-of-range profile CPU?\n");
 			} else if (err == EINVAL && sample_id_all_avail) {
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 676b4fb..935fc4f 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -850,10 +850,10 @@ try_again:
 				     top.evlist->threads, group, inherit) < 0) {
 			int err = errno;
 
-			if (err == EPERM || err == EACCES)
-				die("Permission error - are you root?\n"
-					"\t Consider tweaking"
-					" /proc/sys/kernel/perf_event_paranoid.\n");
+			if (err == EPERM || err == EACCES) {
+				ui__warning_paranoid();
+				goto out_err;
+			}
 			/*
 			 * If it's cycles then fall back to hrtimer
 			 * based cpu-clock-tick sw counter, which
@@ -861,25 +861,35 @@ try_again:
 			 */
 			if (attr->type == PERF_TYPE_HARDWARE &&
 			    attr->config == PERF_COUNT_HW_CPU_CYCLES) {
-
 				if (verbose)
-					warning(" ... trying to fall back to cpu-clock-ticks\n");
+					ui__warning("Cycles event not supported,\n"
+						    "trying to fall back to cpu-clock-ticks\n");
 
 				attr->type = PERF_TYPE_SOFTWARE;
 				attr->config = PERF_COUNT_SW_CPU_CLOCK;
 				goto try_again;
 			}
-			printf("\n");
-			error("sys_perf_event_open() syscall returned with %d "
-			      "(%s).  /bin/dmesg may provide additional information.\n",
-			      err, strerror(err));
-			die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
-			exit(-1);
+
+			ui__warning("The sys_perf_event_open() syscall "
+				    "returned with %d (%s).  /bin/dmesg "
+				    "may provide additional information.\n"
+				    "No CONFIG_PERF_EVENTS=y kernel support "
+				    "configured?\n", err, strerror(err));
+			goto out_err;
 		}
 	}
 
-	if (perf_evlist__mmap(evlist, mmap_pages, false) < 0)
-		die("failed to mmap with %d (%s)\n", errno, strerror(errno));
+	if (perf_evlist__mmap(evlist, mmap_pages, false) < 0) {
+		ui__warning("Failed to mmap with %d (%s)\n",
+			    errno, strerror(errno));
+		goto out_err;
+	}
+
+	return;
+
+out_err:
+	exit_browser(0);
+	exit(0);
 }
 
 static int __cmd_top(void)
diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c
index d4536a9..155749d 100644
--- a/tools/perf/util/debug.c
+++ b/tools/perf/util/debug.c
@@ -57,6 +57,16 @@ void ui__warning(const char *format, ...)
 }
 #endif
 
+void ui__warning_paranoid(void)
+{
+	ui__warning("Permission error - are you root?\n"
+		    "Consider tweaking /proc/sys/kernel/perf_event_paranoid:\n"
+		    " -1 - Not paranoid at all\n"
+		    "  0 - Disallow raw tracepoint access for unpriv\n"
+		    "  1 - Disallow cpu events for unpriv\n"
+		    "  2 - Disallow kernel profiling for unpriv\n");
+}
+
 void trace_event(union perf_event *event)
 {
 	unsigned char *raw_event = (void *)event;
diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h
index 93516cf..fd53db4 100644
--- a/tools/perf/util/debug.h
+++ b/tools/perf/util/debug.h
@@ -36,5 +36,6 @@ int ui_helpline__show_help(const char *format, va_list ap);
 #endif
 
 void ui__warning(const char *format, ...) __attribute__((format(printf, 1, 2)));
+void ui__warning_paranoid(void);
 
 #endif	/* __PERF_DEBUG_H */
-- 
1.6.2.5

[PATCH 2/3] perf tools: Emit clearer message for sys_perf_event_open ENOENT return

[Arnaldo Carvalho de Melo]

From: David Ahern <daahern@cisco.com>

Resend of patch sent back in January 2011 in light of recent confusion around
unsupported events for a given platform.

Improve sys_perf_event_open ENOENT return handling in top and record, just
like 5a3446b does for stat.

Retry of Arnaldo's patch using ui_warning instead of die which allows the
fallback from hardware cycles to software clock.

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: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
LKML-Reference: <1301080271-20945-1-git-send-email-daahern@cisco.com>
Signed-off-by: David Ahern <daahern@cisco.com>
[ committer note: Some adjustments to make it apply to newer codebase ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-record.c |   10 +++++++++-
 tools/perf/builtin-top.c    |    6 ++++++
 2 files changed, 15 insertions(+), 1 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 623695e..db6adec 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -301,11 +301,19 @@ try_again:
 					&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
 
 				if (verbose)
-					warning(" ... trying to fall back to cpu-clock-ticks\n");
+					ui__warning("The cycles event is not supported, "
+						    "trying to fall back to cpu-clock-ticks\n");
 				attr->type = PERF_TYPE_SOFTWARE;
 				attr->config = PERF_COUNT_SW_CPU_CLOCK;
 				goto try_again;
 			}
+
+			if (err == ENOENT) {
+				ui__warning("The %s event is not supported.\n",
+					    event_name(pos));
+				exit(EXIT_FAILURE);
+			}
+
 			printf("\n");
 			error("sys_perf_event_open() syscall returned with %d (%s).  /bin/dmesg may provide additional information.\n",
 			      err, strerror(err));
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 935fc4f..fc1273e 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -870,6 +870,12 @@ try_again:
 				goto try_again;
 			}
 
+			if (err == ENOENT) {
+				ui__warning("The %s event is not supported.\n",
+					    event_name(counter));
+				goto out_err;
+			}
+
 			ui__warning("The sys_perf_event_open() syscall "
 				    "returned with %d (%s).  /bin/dmesg "
 				    "may provide additional information.\n"
-- 
1.6.2.5

[PATCH 3/3] perf probe: Add fastpath to do lookup by function name

[Arnaldo Carvalho de Melo]

From: Lin Ming <ming.m.lin@intel.com>

v3 -> v2:
- Make pubname_search_cb more generic
- Add fastpath to find_probes also

v2 -> v1:
- Don't compare file names with cu_find_realpath(...), instead, compare
  them with the name returned by dwarf_decl_file(sp_die)

The vmlinux file may have thousands of CUs.
We can lookup function name from .debug_pubnames section
to avoid the slow loop on CUs.

1. Improvement data for find_line_range

./perf stat -e cycles -r 10 -- ./perf probe -k /home/mlin/vmlinux \
        -s /home/mlin/linux-2.6 \
        --line csum_partial_copy_to_user > tmp.log

before patch applied
=====================
       847,988,276 cycles

        0.355075856  seconds time elapsed

after patch applied
=====================
       206,102,622 cycles

        0.086883555  seconds time elapsed

2. Improvement data for find_probes

./perf stat -e cycles -r 10 -- ./perf probe -k /home/mlin/vmlinux \
        -s /home/mlin/linux-2.6 \
        --vars csum_partial_copy_to_user > tmp.log

before patch applied
=====================
       848,490,844 cycles

        0.355307901  seconds time elapsed

after patch applied
=====================
       205,684,469 cycles

        0.086694010  seconds time elapsed

Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1301041668.14111.52.camel@minggr.sh.intel.com>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-finder.c |   72 ++++++++++++++++++++++++++++++++++++++++
 tools/perf/util/probe-finder.h |    2 +
 2 files changed, 74 insertions(+), 0 deletions(-)

diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 194f9e2..ff416b8 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -1435,6 +1435,38 @@ static int find_probe_point_by_func(struct probe_finder *pf)
 	return _param.retval;
 }
 
+struct pubname_callback_param {
+	char *function;
+	char *file;
+	Dwarf_Die *cu_die;
+	Dwarf_Die *sp_die;
+	int found;
+};
+
+static int pubname_search_cb(Dwarf *dbg, Dwarf_Global *gl, void *data)
+{
+	struct pubname_callback_param *param = data;
+
+	if (dwarf_offdie(dbg, gl->die_offset, param->sp_die)) {
+		if (dwarf_tag(param->sp_die) != DW_TAG_subprogram)
+			return DWARF_CB_OK;
+
+		if (die_compare_name(param->sp_die, param->function)) {
+			if (!dwarf_offdie(dbg, gl->cu_offset, param->cu_die))
+				return DWARF_CB_OK;
+
+			if (param->file &&
+			    strtailcmp(param->file, dwarf_decl_file(param->sp_die)))
+				return DWARF_CB_OK;
+
+			param->found = 1;
+			return DWARF_CB_ABORT;
+		}
+	}
+
+	return DWARF_CB_OK;
+}
+
 /* Find probe points from debuginfo */
 static int find_probes(int fd, struct probe_finder *pf)
 {
@@ -1461,6 +1493,27 @@ static int find_probes(int fd, struct probe_finder *pf)
 
 	off = 0;
 	line_list__init(&pf->lcache);
+
+	/* Fastpath: lookup by function name from .debug_pubnames section */
+	if (pp->function) {
+		struct pubname_callback_param pubname_param = {
+			.function = pp->function,
+			.file	  = pp->file,
+			.cu_die	  = &pf->cu_die,
+			.sp_die	  = &pf->sp_die,
+		};
+		struct dwarf_callback_param probe_param = {
+			.data = pf,
+		};
+
+		dwarf_getpubnames(dbg, pubname_search_cb, &pubname_param, 0);
+		if (pubname_param.found) {
+			ret = probe_point_search_cb(&pf->sp_die, &probe_param);
+			if (ret)
+				goto found;
+		}
+	}
+
 	/* Loop on CUs (Compilation Unit) */
 	while (!dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL)) {
 		/* Get the DIE(Debugging Information Entry) of this CU */
@@ -1488,6 +1541,8 @@ static int find_probes(int fd, struct probe_finder *pf)
 		}
 		off = noff;
 	}
+
+found:
 	line_list__free(&pf->lcache);
 	if (dwfl)
 		dwfl_end(dwfl);
@@ -1895,6 +1950,22 @@ int find_line_range(int fd, struct line_range *lr)
 		return -EBADF;
 	}
 
+	/* Fastpath: lookup by function name from .debug_pubnames section */
+	if (lr->function) {
+		struct pubname_callback_param pubname_param = {
+			.function = lr->function, .file = lr->file,
+			.cu_die = &lf.cu_die, .sp_die = &lf.sp_die, .found = 0};
+		struct dwarf_callback_param line_range_param = {
+			.data = (void *)&lf, .retval = 0};
+
+		dwarf_getpubnames(dbg, pubname_search_cb, &pubname_param, 0);
+		if (pubname_param.found) {
+			line_range_search_cb(&lf.sp_die, &line_range_param);
+			if (lf.found)
+				goto found;
+		}
+	}
+
 	/* Loop on CUs (Compilation Unit) */
 	while (!lf.found && ret >= 0) {
 		if (dwarf_nextcu(dbg, off, &noff, &cuhl, NULL, NULL, NULL) != 0)
@@ -1923,6 +1994,7 @@ int find_line_range(int fd, struct line_range *lr)
 		off = noff;
 	}
 
+found:
 	/* Store comp_dir */
 	if (lf.found) {
 		comp_dir = cu_get_comp_dir(&lf.cu_die);
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
index beaefc3..605730a 100644
--- a/tools/perf/util/probe-finder.h
+++ b/tools/perf/util/probe-finder.h
@@ -49,6 +49,7 @@ struct probe_finder {
 	Dwarf_Addr		addr;		/* Address */
 	const char		*fname;		/* Real file name */
 	Dwarf_Die		cu_die;		/* Current CU */
+	Dwarf_Die		sp_die;
 	struct list_head	lcache;		/* Line cache for lazy match */
 
 	/* For variable searching */
@@ -83,6 +84,7 @@ struct line_finder {
 	int			lno_s;		/* Start line number */
 	int			lno_e;		/* End line number */
 	Dwarf_Die		cu_die;		/* Current CU */
+	Dwarf_Die		sp_die;
 	int			found;
 };
 
-- 
1.6.2.5

Re: [GIT PULL 0/3] perf/core improvements and fixes

[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-2.6 perf/core
> 
> 	This one has everything in acme/perf/urgent, i.e. starts the next
> tools/ devel cycle :-)
> 
> Regards,
> 
> - Arnaldo
> 
> Arnaldo Carvalho de Melo (1):
>   perf tools: Fixup exit path when not able to open events
> 
> David Ahern (1):
>   perf tools: Emit clearer message for sys_perf_event_open ENOENT
>     return
> 
> Lin Ming (1):
>   perf probe: Add fastpath to do lookup by function name
> 
>  tools/perf/builtin-record.c    |   19 +++++++---
>  tools/perf/builtin-top.c       |   44 ++++++++++++++++--------
>  tools/perf/util/debug.c        |   10 +++++
>  tools/perf/util/debug.h        |    1 +
>  tools/perf/util/probe-finder.c |   72 ++++++++++++++++++++++++++++++++++++++++
>  tools/perf/util/probe-finder.h |    2 +
>  6 files changed, 128 insertions(+), 20 deletions(-)

Pulled, thanks Arnaldo!

Note, i merged these two also into perf/urgent:

  ca6a42586fae: perf tools: Emit clearer message for sys_perf_event_open ENOENT return
  c286c419c784: perf tools: Fixup exit path when not able to open events

as both are simple and fix real bugs.

Thanks,

	Ingo

[GIT PULL 0/3] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

From: Arnaldo Carvalho de Melo <acmeinfradead.org>

Hi Ingo,

	Please consider pulling,

- Arnaldo

The following changes since commit 9ec60972a38011ad8a5676f4cd5e51ac508c36b6:

  perf hists: Add missing period_* fields when collapsing a hist entry (2012-09-26 20:44:11 -0300)

are available in the git repository at:

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

for you to fetch changes up to aec1930b0f6f281a0ca038cd03aceace3fe0bb61:

  perf trace: Add aliases for some syscalls (2012-09-27 13:18:49 -0300)

----------------------------------------------------------------
perf/core improvements and fixes

. Improve listing of accessible enum perf probe variables, from Hyeoncheol Lee.

. Don't stop the build if the audit libraries are not installed, fix from Namhyung Kim.

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

----------------------------------------------------------------
Arnaldo Carvalho de Melo (1):
      perf trace: Add aliases for some syscalls

Hyeoncheol Lee (1):
      perf probe: Print an enum type variable in "enum variable-name" format when showing accessible variables

Namhyung Kim (1):
      perf tools: Check libaudit availability for perf-trace builtin

 tools/perf/Makefile                 |   16 ++++++++++++++--
 tools/perf/builtin-trace.c          |   26 ++++++++++++++++++--------
 tools/perf/config/feature-tests.mak |   11 +++++++++++
 tools/perf/perf.c                   |    2 ++
 tools/perf/util/dwarf-aux.c         |    2 ++
 5 files changed, 47 insertions(+), 10 deletions(-)

Re: [GIT PULL 0/3] perf/core improvements and fixes

[Ingo Molnar]

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

> From: Arnaldo Carvalho de Melo <acmeinfradead.org>
> 
> Hi Ingo,
> 
> 	Please consider pulling,
> 
> - Arnaldo
> 
> The following changes since commit 9ec60972a38011ad8a5676f4cd5e51ac508c36b6:
> 
>   perf hists: Add missing period_* fields when collapsing a hist entry (2012-09-26 20:44:11 -0300)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux tags/perf-core-for-mingo
> 
> for you to fetch changes up to aec1930b0f6f281a0ca038cd03aceace3fe0bb61:
> 
>   perf trace: Add aliases for some syscalls (2012-09-27 13:18:49 -0300)
> 
> ----------------------------------------------------------------
> perf/core improvements and fixes
> 
> . Improve listing of accessible enum perf probe variables, from Hyeoncheol Lee.
> 
> . Don't stop the build if the audit libraries are not installed, fix from Namhyung Kim.
> 
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> ----------------------------------------------------------------
> Arnaldo Carvalho de Melo (1):
>       perf trace: Add aliases for some syscalls
> 
> Hyeoncheol Lee (1):
>       perf probe: Print an enum type variable in "enum variable-name" format when showing accessible variables
> 
> Namhyung Kim (1):
>       perf tools: Check libaudit availability for perf-trace builtin
> 
>  tools/perf/Makefile                 |   16 ++++++++++++++--
>  tools/perf/builtin-trace.c          |   26 ++++++++++++++++++--------
>  tools/perf/config/feature-tests.mak |   11 +++++++++++
>  tools/perf/perf.c                   |    2 ++
>  tools/perf/util/dwarf-aux.c         |    2 ++
>  5 files changed, 47 insertions(+), 10 deletions(-)

Pulled, thanks Arnaldo!

	Ingo

[GIT PULL 0/3] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

        Please consider pulling from:

git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 perf/core

Regards,

- Arnaldo

Arnaldo Carvalho de Melo (3):
  perf annotate: Fix --stdio rendering
  perf annotate: Move locking to struct annotation
  perf annotate: Fix annotate context lines regression

 tools/perf/builtin-annotate.c          |   14 ++--
 tools/perf/builtin-report.c            |    5 +-
 tools/perf/builtin-top.c               |   67 ++++++++--------
 tools/perf/util/annotate.c             |  134 +++++++++++++++++++++-----------
 tools/perf/util/annotate.h             |   35 +++++----
 tools/perf/util/hist.c                 |    5 +-
 tools/perf/util/hist.h                 |    3 +-
 tools/perf/util/top.h                  |    6 --
 tools/perf/util/ui/browsers/annotate.c |   18 ++--
 9 files changed, 168 insertions(+), 119 deletions(-)

[GIT PULL 0/3] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

        Please consider pulling from:

git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 perf/core

Regards,

- Arnaldo

Arnaldo Carvalho de Melo (3):
  perf annotate: Fix --stdio rendering
  perf annotate: Move locking to struct annotation
  perf annotate: Fix annotate context lines regression

 tools/perf/builtin-annotate.c          |   14 ++--
 tools/perf/builtin-report.c            |    5 +-
 tools/perf/builtin-top.c               |   67 ++++++++--------
 tools/perf/util/annotate.c             |  134 +++++++++++++++++++++-----------
 tools/perf/util/annotate.h             |   35 +++++----
 tools/perf/util/hist.c                 |    5 +-
 tools/perf/util/hist.h                 |    3 +-
 tools/perf/util/top.h                  |    6 --
 tools/perf/util/ui/browsers/annotate.c |   18 ++--
 9 files changed, 168 insertions(+), 119 deletions(-)

[GIT PULL 0/3] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

        Please consider pulling from:

git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux-2.6 perf/core

Regards,

- Arnaldo

Arnaldo Carvalho de Melo (2):
  perf test: Clarify some error reports in the open syscall test
  perf evsel: Support perf_evsel__open(cpus > 1 && threads > 1)

Stephane Eranian (1):
  perf tools: Fix perf_event.h header usage

 tools/perf/builtin-test.c      |  124 +++++++++++++++++++++++++++++++++++++--
 tools/perf/util/evsel.c        |   75 +++++++++++++-----------
 tools/perf/util/evsel.h        |    2 +-
 tools/perf/util/parse-events.h |    2 +-
 4 files changed, 159 insertions(+), 44 deletions(-)

Re: [GIT PULL 0/3] perf/core improvements and fixes

[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-2.6 perf/core
> 
> Regards,
> 
> - Arnaldo
> 
> Arnaldo Carvalho de Melo (2):
>   perf test: Clarify some error reports in the open syscall test
>   perf evsel: Support perf_evsel__open(cpus > 1 && threads > 1)
> 
> Stephane Eranian (1):
>   perf tools: Fix perf_event.h header usage
> 
>  tools/perf/builtin-test.c      |  124 +++++++++++++++++++++++++++++++++++++--
>  tools/perf/util/evsel.c        |   75 +++++++++++++-----------
>  tools/perf/util/evsel.h        |    2 +-
>  tools/perf/util/parse-events.h |    2 +-
>  4 files changed, 159 insertions(+), 44 deletions(-)

Pulled, thanks a lot Arnaldo!

	Ingo

Re: [GIT PULL 0/3] perf/core improvements and fixes

[Ingo Molnar]

sorry, had to unpull it:

util/evsel.c: In function ‘perf_evsel__open_per_cpu’:
util/evsel.c:137: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
util/evsel.c:158: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
util/evsel.c:141: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
util/evsel.c:142: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
util/evsel.c:185: note: initialized from here
cc1: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
util/evsel.c:185: note: initialized from here
make: *** [util/evsel.o] Error 1
make: *** Waiting for unfinished jobs....

gcc version 4.4.4 20100630 (Red Hat 4.4.4-10) (GCC) 

on a 32-bit Fedora 12 box.

Thanks,

	Ingo

Re: [GIT PULL 0/3] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

Em Wed, Jan 05, 2011 at 02:14:11PM +0100, Ingo Molnar escreveu:
> 
> sorry, had to unpull it:
> 
> util/evsel.c: In function ‘perf_evsel__open_per_cpu’:
> util/evsel.c:137: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
> util/evsel.c:158: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
> util/evsel.c:141: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
> util/evsel.c:142: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
> util/evsel.c:185: note: initialized from here
> cc1: error: dereferencing pointer ‘map__empty.57’ does break strict-aliasing rules
> util/evsel.c:185: note: initialized from here
> make: *** [util/evsel.o] Error 1
> make: *** Waiting for unfinished jobs....
> 
> gcc version 4.4.4 20100630 (Red Hat 4.4.4-10) (GCC) 
> 
> on a 32-bit Fedora 12 box.

Checking it,

- Arnaldo