[GIT PULL 00/30] perf/core improvements and fixes

[GIT PULL 00/30] perf/core improvements and fixes

[Arnaldo Carvalho de Melo]

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

Hi Ingo,

	Please consider pulling,

- Arnaldo

The following changes since commit 67d61296ffcc850bffdd4466430cb91e5328f39a:

  Merge tag 'perf-core-for-mingo-20160419' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux (2016-04-23 14:50:39 +0200)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-20160427

for you to fetch changes up to 4cb93446c587d56e2a54f4f83113daba2c0b6dee:

  perf tools: Set the maximum allowed stack from /proc/sys/kernel/perf_event_max_stack (2016-04-27 10:29:07 -0300)

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

User visible:

- perf trace --pf maj/min/all works with --call-graph: (Arnaldo Carvalho de Melo)

  Tracing write syscalls and major page faults with callchains while starting
  firefox, limiting the stack to 5 frames:

 # perf trace -e write --pf maj --max-stack 5 firefox
   589.549 ( 0.014 ms): firefox/15377 write(fd: 4, buf: 0x7fff80acc898, count: 151) = 151
                                       [0xfaed] (/usr/lib64/libpthread-2.22.so)
                                       fire_glxtest_process+0x5c (/usr/lib64/firefox/libxul.so)
                                       InstallGdkErrorHandler+0x41 (/usr/lib64/firefox/libxul.so)
                                       XREMain::XRE_mainInit+0x12c (/usr/lib64/firefox/libxul.so)
                                       XREMain::XRE_main+0x1e4 (/usr/lib64/firefox/libxul.so)
   760.704 ( 0.000 ms): firefox/15332 majfault [gtk_tree_view_accessible_get_type+0x0] => /usr/lib64/libgtk-3.so.0.1800.9@0xa0850 (x.)
                                       gtk_tree_view_accessible_get_type+0x0 (/usr/lib64/libgtk-3.so.0.1800.9)
                                       gtk_tree_view_class_intern_init+0x1a54 (/usr/lib64/libgtk-3.so.0.1800.9)
                                       g_type_class_ref+0x6dd (/usr/lib64/libgobject-2.0.so.0.4600.2)
                                       [0x115378] (/usr/lib64/libgnutls.so.30.6.3)

  This automagically selects "--call-graph dwarf", use "--call-graph fp" on systems
  where -fno-omit-frame-pointer was used to built the components of interest, to
  incur in less overhead, or tune "--call-graph dwarf" appropriately, see 'perf record --help'.

- Allow /proc/sys/kernel/perf_event_max_stack, that defaults to the old hard coded value
  of PERF_MAX_STACK_DEPTH (127), useful for huge callstacks for things like Groovy, Ruby, etc,
  and also to reduce overhead by limiting it to a smaller value, upcoming work will allow
  this to be done per-event (Arnaldo Carvalho de Melo)

- Make 'perf trace --min-stack' be honoured by --pf and --event (Arnaldo Carvalho de Melo)

- Make 'perf evlist -v' decode perf_event_attr->branch_sample_type (Arnaldo Carvalho de Melo)

   # perf record --call lbr usleep 1
   # perf evlist -v
   cycles:ppp: ... sample_type: IP|TID|TIME|CALLCHAIN|PERIOD|BRANCH_STACK, ...
            branch_sample_type: USER|CALL_STACK|NO_FLAGS|NO_CYCLES
   #

- Clear dummy entry accumulated period, fixing such 'perf top/report' output
  as: (Kan Liang)

    4769.98%  0.01%  0.00%  0.01%  tchain_edit  [kernel] [k] update_fast_timekeeper

- System calls with pid_t arguments gets them augmented with the COMM event
  more thoroughly:

  # trace -e perf_event_open perf stat -e cycles -p 15608
   6.876 ( 0.014 ms): perf_event_open(attr_uptr: 0x2ae20d8, pid: 15608 (hexchat), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 3
   6.882 ( 0.005 ms): perf_event_open(attr_uptr: 0x2ae20d8, pid: 15639 (gmain), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 4
   6.889 ( 0.005 ms): perf_event_open(attr_uptr: 0x2ae20d8, pid: 15640 (gdbus), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 5
                                                            ^^^^^^^^^^^^^^^^^^
   ^C

- Fix offline module name mismatch issue in 'perf probe' (Ravi Bangoria)

- Fix module probe issue if no dwarf support in (Ravi Bangoria)

Assorted fixes:

- Fix off-by-one in write_buildid() (Andrey Ryabinin)

- Fix segfault when printing callchains in 'perf script' (Chris Phlipot)

- Replace assignment with comparison on assert check in 'perf test' entry (Colin Ian King)

- Fix off-by-one comparison in intel-pt code (Colin Ian King)

- Close target file on error path in 'perf probe' (Masami Hiramatsu)

- Set default kprobe group name if not given in 'perf probe' (Masami Hiramatsu)

- Avoid partial perf_event_header reads (Wang Nan)

Infrastructure:

- Update x86's syscall_64.tbl copy, adding preadv2 & pwritev2 (Arnaldo Carvalho de Melo)

- Make the x86 clean quiet wrt syscall table removal (Jiri Olsa)

Cleanups:

- Simplify wrapper for LOCK_PI in 'perf bench futex' (Davidlohr Bueso)

- Remove duplicate const qualifier (Eric Engestrom)

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

----------------------------------------------------------------
Andrey Ryabinin (1):
      perf buildid: Fix off-by-one in write_buildid()

Arnaldo Carvalho de Melo (14):
      perf trace: Extract evsel contructor from perf_evlist__add_pgfault
      perf trace: Make --pf maj/min/all use callchains too
      perf trace: Make --event honour --min-stack too
      perf trace: Make --pf honour --min-stack too
      perf evlist: Decode perf_event_attr->branch_sample_type
      perf trace: Move perf_flags beautifier to tools/perf/trace/beauty/
      perf trace: Do not beautify the 'pid' parameter as a simple integer
      tools lib api fs: Add helper to read string from procfs file
      perf thread: Introduce method to set comm from /proc/pid/self
      perf trace: Read thread's COMM from /proc when not set
      perf tools: Update x86's syscall_64.tbl, adding preadv2 & pwritev2
      perf bench: Remove one more die() call
      perf core: Allow setting up max frame stack depth via sysctl
      perf tools: Set the maximum allowed stack from /proc/sys/kernel/perf_event_max_stack

Chris Phlipot (1):
      perf script: Fix segfault when printing callchains

Colin Ian King (2):
      perf tests: Replace assignment with comparison on assert check
      perf intel-pt: Fix off-by-one comparison on maximum code

Davidlohr Bueso (1):
      perf bench futex: Simplify wrapper for LOCK_PI

Eric Engestrom (1):
      perf tools: Remove duplicate const qualifier

Jiri Olsa (2):
      perf tools: Make the x86 clean quiet
      tools build: Fix perf_clean target

Kan Liang (1):
      perf hists: Clear dummy entry accumulated period

Masami Hiramatsu (4):
      perf probe: Close target file on error path
      perf tools: Add lsdir() helper to read a directory
      perf probe: Let probe_file__add_event return 0 if succeeded
      perf probe: Set default kprobe group name if it is not given

Ravi Bangoria (2):
      perf probe: Fix offline module name missmatch issue
      perf probe: Fix module probe issue if no dwarf support

Wang Nan (1):
      perf evlist: Enforce ring buffer reading

 Documentation/sysctl/kernel.txt                    |  14 ++
 arch/arm/kernel/perf_callchain.c                   |   2 +-
 arch/arm64/kernel/perf_callchain.c                 |   4 +-
 arch/metag/kernel/perf_callchain.c                 |   2 +-
 arch/mips/kernel/perf_event.c                      |   4 +-
 arch/powerpc/perf/callchain.c                      |   4 +-
 arch/sparc/kernel/perf_event.c                     |   6 +-
 arch/x86/events/core.c                             |   4 +-
 arch/xtensa/kernel/perf_event.c                    |   4 +-
 include/linux/perf_event.h                         |   8 +-
 kernel/bpf/stackmap.c                              |   8 +-
 kernel/events/callchain.c                          |  35 ++++-
 kernel/sysctl.c                                    |  12 ++
 tools/Makefile                                     |   3 +-
 tools/lib/api/fs/fs.c                              |  13 ++
 tools/lib/api/fs/fs.h                              |   2 +
 tools/perf/Documentation/perf-report.txt           |   2 +-
 tools/perf/Documentation/perf-script.txt           |   2 +-
 tools/perf/Documentation/perf-top.txt              |   2 +-
 tools/perf/Documentation/perf-trace.txt            |   2 +-
 tools/perf/arch/x86/Makefile                       |   2 +-
 tools/perf/arch/x86/entry/syscalls/syscall_64.tbl  |   2 +
 tools/perf/bench/futex-lock-pi.c                   |   2 +-
 tools/perf/bench/futex.h                           |   6 +-
 tools/perf/bench/mem-functions.c                   |  22 ++-
 tools/perf/builtin-report.c                        |   4 +-
 tools/perf/builtin-script.c                        |  16 +-
 tools/perf/builtin-top.c                           |   4 +-
 tools/perf/builtin-trace.c                         | 174 +++++++++++----------
 tools/perf/perf.c                                  |   5 +
 tools/perf/tests/event_update.c                    |   2 +-
 tools/perf/tests/hists_cumulate.c                  |   2 +-
 tools/perf/tests/hists_filter.c                    |   2 +-
 tools/perf/tests/hists_output.c                    |   2 +-
 tools/perf/trace/beauty/perf_event_open.c          |  43 +++++
 tools/perf/trace/beauty/pid.c                      |   5 +-
 tools/perf/util/build-id.c                         |   6 +-
 tools/perf/util/evlist.c                           |  12 +-
 tools/perf/util/evsel.c                            |  18 ++-
 tools/perf/util/hist.c                             |   2 +
 .../perf/util/intel-pt-decoder/intel-pt-decoder.c  |   2 +-
 tools/perf/util/machine.c                          |   6 +-
 tools/perf/util/probe-event.c                      | 114 +++++++++++---
 tools/perf/util/probe-file.c                       |   3 +-
 .../perf/util/scripting-engines/trace-event-perl.c |   2 +-
 tools/perf/util/thread.c                           |  21 ++-
 tools/perf/util/thread.h                           |   2 +
 tools/perf/util/util.c                             |  36 +++++
 tools/perf/util/util.h                             |   4 +
 49 files changed, 475 insertions(+), 179 deletions(-)
 create mode 100644 tools/perf/trace/beauty/perf_event_open.c

[PATCH 01/30] perf buildid: Fix off-by-one in write_buildid()

[Arnaldo Carvalho de Melo]

From: Andrey Ryabinin <aryabinin@virtuozzo.com>

write_buildid() increments 'name_len' with intention to take into
account trailing zero byte. However, 'name_len' was already incremented
in machine__write_buildid_table() before.  So this leads to
out-of-bounds read in do_write():

  $ ./perf record sleep 0
  [ perf record: Woken up 1 times to write data ]
  =================================================================
  ==15899==ERROR: AddressSanitizer: global-buffer-overflow on address 0x00000099fc92 at pc 0x7f1aa9c7eab5 bp 0x7fff940f84d0 sp 0x7fff940f7c78
  READ of size 19 at 0x00000099fc92 thread T0
      #0 0x7f1aa9c7eab4  (/usr/lib/gcc/x86_64-pc-linux-gnu/5.3.0/libasan.so.2+0x44ab4)
      #1 0x649c5b in do_write util/header.c:67
      #2 0x649c5b in write_padded util/header.c:82
      #3 0x57e8bc in write_buildid util/build-id.c:239
      #4 0x57e8bc in machine__write_buildid_table util/build-id.c:278
  ...

  0x00000099fc92 is located 0 bytes to the right of global variable '*.LC99' defined in 'util/symbol.c' (0x99fc80) of size 18
    '*.LC99' is ascii string '[kernel.kallsyms]'
  ...

  Shadow bytes around the buggy address:
    0x00008012bf80: f9 f9 f9 f9 00 00 00 00 00 00 03 f9 f9 f9 f9 f9
  =>0x00008012bf90: 00 00[02]f9 f9 f9 f9 f9 00 00 00 00 00 05 f9 f9
    0x00008012bfa0: f9 f9 f9 f9 00 03 f9 f9 f9 f9 f9 f9 00 00 00 00

Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1461053847-5633-1-git-send-email-aryabinin@virtuozzo.com
[ Remove the off-by one at the origin, to keep len(s) == strlen(s) assumption ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/build-id.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c
index 0573c2ec861d..b6ecf87bc3e3 100644
--- a/tools/perf/util/build-id.c
+++ b/tools/perf/util/build-id.c
@@ -261,14 +261,14 @@ static int machine__write_buildid_table(struct machine *machine, int fd)
 
 		if (dso__is_vdso(pos)) {
 			name = pos->short_name;
-			name_len = pos->short_name_len + 1;
+			name_len = pos->short_name_len;
 		} else if (dso__is_kcore(pos)) {
 			machine__mmap_name(machine, nm, sizeof(nm));
 			name = nm;
-			name_len = strlen(nm) + 1;
+			name_len = strlen(nm);
 		} else {
 			name = pos->long_name;
-			name_len = pos->long_name_len + 1;
+			name_len = pos->long_name_len;
 		}
 
 		in_kernel = pos->kernel ||
-- 
2.5.5

[PATCH 02/30] perf trace: Extract evsel contructor from perf_evlist__add_pgfault

[Arnaldo Carvalho de Melo]

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

Prep work for next patches, where we'll need access to the created
evsels, to possibly configure callchains.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-2pcgsgnkgellhlcao4aub8tu@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 5e2614bbb48d..69b460354201 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2381,8 +2381,7 @@ static bool perf_evlist__add_vfs_getname(struct perf_evlist *evlist)
 	return true;
 }
 
-static int perf_evlist__add_pgfault(struct perf_evlist *evlist,
-				    u64 config)
+static struct perf_evsel *perf_evsel__new_pgfault(u64 config)
 {
 	struct perf_evsel *evsel;
 	struct perf_event_attr attr = {
@@ -2396,13 +2395,10 @@ static int perf_evlist__add_pgfault(struct perf_evlist *evlist,
 	event_attr_init(&attr);
 
 	evsel = perf_evsel__new(&attr);
-	if (!evsel)
-		return -ENOMEM;
-
-	evsel->handler = trace__pgfault;
-	perf_evlist__add(evlist, evsel);
+	if (evsel)
+		evsel->handler = trace__pgfault;
 
-	return 0;
+	return evsel;
 }
 
 static void trace__handle_event(struct trace *trace, union perf_event *event, struct perf_sample *sample)
@@ -2504,7 +2500,7 @@ out_enomem:
 static int trace__run(struct trace *trace, int argc, const char **argv)
 {
 	struct perf_evlist *evlist = trace->evlist;
-	struct perf_evsel *evsel;
+	struct perf_evsel *evsel, *pgfault_maj = NULL, *pgfault_min = NULL;
 	int err = -1, i;
 	unsigned long before;
 	const bool forks = argc > 0;
@@ -2518,14 +2514,19 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
 	if (trace->trace_syscalls)
 		trace->vfs_getname = perf_evlist__add_vfs_getname(evlist);
 
-	if ((trace->trace_pgfaults & TRACE_PFMAJ) &&
-	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MAJ)) {
-		goto out_error_mem;
+	if ((trace->trace_pgfaults & TRACE_PFMAJ)) {
+		pgfault_maj = perf_evsel__new_pgfault(PERF_COUNT_SW_PAGE_FAULTS_MAJ);
+		if (pgfault_maj == NULL)
+			goto out_error_mem;
+		perf_evlist__add(evlist, pgfault_maj);
 	}
 
-	if ((trace->trace_pgfaults & TRACE_PFMIN) &&
-	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MIN))
-		goto out_error_mem;
+	if ((trace->trace_pgfaults & TRACE_PFMIN)) {
+		pgfault_min = perf_evsel__new_pgfault(PERF_COUNT_SW_PAGE_FAULTS_MIN);
+		if (pgfault_min == NULL)
+			goto out_error_mem;
+		perf_evlist__add(evlist, pgfault_min);
+	}
 
 	if (trace->sched &&
 	    perf_evlist__add_newtp(evlist, "sched", "sched_stat_runtime",
-- 
2.5.5

[PATCH 03/30] perf trace: Make --pf maj/min/all use callchains too

[Arnaldo Carvalho de Melo]

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

Forgot about page faults, a software event, when adding support for callchains,
fix it:

  # trace --no-syscalls --pf maj --call dwarf
     0.000 ( 0.000 ms): Xorg/2068 majfault [sfbSegment1+0x0] => /usr/lib64/xorg/modules/drivers/intel_drv.so@0x11b490 (x.)
                                       sfbSegment1+0x0 (/usr/lib64/xorg/modules/drivers/intel_drv.so)
                                       fbPolySegment32+0x361 (/usr/lib64/xorg/modules/drivers/intel_drv.so)
                                       sna_poly_segment+0x743 (/usr/lib64/xorg/modules/drivers/intel_drv.so)
                                       damagePolySegment+0x77 (/usr/libexec/Xorg)
                                       ProcPolySegment+0xe7 (/usr/libexec/Xorg)
                                       Dispatch+0x25f (/usr/libexec/Xorg)
                                       dix_main+0x3c3 (/usr/libexec/Xorg)
                                       __libc_start_main+0xf0 (/usr/lib64/libc-2.22.so)
                                       _start+0x29 (/usr/libexec/Xorg)
     0.257 ( 0.000 ms): Xorg/2068 majfault [miZeroClipLine+0x0] => /usr/libexec/Xorg@0x18e830 (x.)
                                       miZeroClipLine+0x0 (/usr/libexec/Xorg)
                                       _fbSegment+0x2c0 (/usr/lib64/xorg/modules/drivers/intel_drv.so)
                                       sfbSegment1+0x67 (/usr/lib64/xorg/modules/drivers/intel_drv.so)
                                       fbPolySegment32+0x361 (/usr/lib64/xorg/modules/drivers/intel_drv.so)
                                       sna_poly_segment+0x743 (/usr/lib64/xorg/modules/drivers/intel_drv.so)
                                       damagePolySegment+0x77 (/usr/libexec/Xorg)
                                       ProcPolySegment+0xe7 (/usr/libexec/Xorg)
                                       Dispatch+0x25f (/usr/libexec/Xorg)
                                       dix_main+0x3c3 (/usr/libexec/Xorg)
                                       __libc_start_main+0xf0 (/usr/lib64/libc-2.22.so)
                                       _start+0x29 (/usr/libexec/Xorg)
^C#

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-8h6ssirw5z15qyhy2lwd6f89@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 59 ++++++++++++++++++++++++++++++++--------------
 1 file changed, 41 insertions(+), 18 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 69b460354201..d1bbcb9abca3 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2222,6 +2222,11 @@ static int trace__pgfault(struct trace *trace,
 	print_location(trace->output, sample, &al, true, false);
 
 	fprintf(trace->output, " (%c%c)\n", map_type, al.level);
+
+	if (sample->callchain) {
+		if (trace__resolve_callchain(trace, evsel, sample, &callchain_cursor) == 0)
+			trace__fprintf_callchain(trace, sample);
+	}
 out:
 	err = 0;
 out_put:
@@ -2547,24 +2552,42 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
 
 	perf_evlist__config(evlist, &trace->opts, NULL);
 
-	if (callchain_param.enabled && trace->syscalls.events.sys_exit) {
-		perf_evsel__config_callchain(trace->syscalls.events.sys_exit,
-					     &trace->opts, &callchain_param);
-               /*
-                * Now we have evsels with different sample_ids, use
-                * PERF_SAMPLE_IDENTIFIER to map from sample to evsel
-                * from a fixed position in each ring buffer record.
-                *
-                * As of this the changeset introducing this comment, this
-                * isn't strictly needed, as the fields that can come before
-                * PERF_SAMPLE_ID are all used, but we'll probably disable
-                * some of those for things like copying the payload of
-                * pointer syscall arguments, and for vfs_getname we don't
-                * need PERF_SAMPLE_ADDR and PERF_SAMPLE_IP, so do this
-                * here as a warning we need to use PERF_SAMPLE_IDENTIFIER.
-                */
-		perf_evlist__set_sample_bit(evlist, IDENTIFIER);
-		perf_evlist__reset_sample_bit(evlist, ID);
+	if (callchain_param.enabled) {
+		bool use_identifier = false;
+
+		if (trace->syscalls.events.sys_exit) {
+			perf_evsel__config_callchain(trace->syscalls.events.sys_exit,
+						     &trace->opts, &callchain_param);
+			use_identifier = true;
+		}
+
+		if (pgfault_maj) {
+			perf_evsel__config_callchain(pgfault_maj, &trace->opts, &callchain_param);
+			use_identifier = true;
+		}
+
+		if (pgfault_min) {
+			perf_evsel__config_callchain(pgfault_min, &trace->opts, &callchain_param);
+			use_identifier = true;
+		}
+
+		if (use_identifier) {
+		       /*
+			* Now we have evsels with different sample_ids, use
+			* PERF_SAMPLE_IDENTIFIER to map from sample to evsel
+			* from a fixed position in each ring buffer record.
+			*
+			* As of this the changeset introducing this comment, this
+			* isn't strictly needed, as the fields that can come before
+			* PERF_SAMPLE_ID are all used, but we'll probably disable
+			* some of those for things like copying the payload of
+			* pointer syscall arguments, and for vfs_getname we don't
+			* need PERF_SAMPLE_ADDR and PERF_SAMPLE_IP, so do this
+			* here as a warning we need to use PERF_SAMPLE_IDENTIFIER.
+			*/
+			perf_evlist__set_sample_bit(evlist, IDENTIFIER);
+			perf_evlist__reset_sample_bit(evlist, ID);
+		}
 	}
 
 	signal(SIGCHLD, sig_handler);
-- 
2.5.5

[PATCH 04/30] perf script: Fix segfault when printing callchains

[Arnaldo Carvalho de Melo]

From: Chris Phlipot <cphlipot0@gmail.com>

This fixes a bug caused by an unitialized callchain cursor. The crash
frist appeared in:

6f736735e30f ("perf evsel: Require that callchains be resolved before
calling fprintf_{sym,callchain}")

The callchain cursor is a struct that contains pointers, that when
uninitialized will cause unpredictable behavior (usually a crash)
when trying to append to the callchain.

The existing implementation has the following issues:

1. The callchain cursor used is not initialized, resulting in
	unpredictable behavior when used.
2. The cursor is declared on the stack. Even if it is properly initalized,
	the implmentation will leak memory when the function returns,
	since all the references to the callchain_nodes allocated by
	callchain_cursor_append will be lost when the cursor goes out of
	scope.
3. Storing the cursor on the stack is inefficient. Even if memory is
	properly freed when it goes out of scope, a performance penalty
	will be incurred due to reallocation of callchain nodes.
	callchain_cursor_append is designed to avoid these reallocations
	when an existing cursor is reused.

This patch fixes the crash by replacing cursor_callchain with a reference
to the global callchain_cursor which also resolves all 3 issues mentioned
above.

How to reproduce the crash:

  $ perf record --call-graph=dwarf stress -t 1 -c 1
  $ perf script > /dev/null
  Segfault

Signed-off-by: Chris Phlipot <cphlipot0@gmail.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Fixes: 6f736735e30f ("perf evsel: Require that callchains be resolved before calling fprintf_{sym,callchain}")
Link: http://lkml.kernel.org/r/1461119531-2529-1-git-send-email-cphlipot0@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-script.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 5099740aa50b..f43b0c6f88f4 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -570,12 +570,12 @@ static void print_sample_bts(struct perf_sample *sample,
 	/* print branch_from information */
 	if (PRINT_FIELD(IP)) {
 		unsigned int print_opts = output[attr->type].print_ip_opts;
-		struct callchain_cursor *cursor = NULL, cursor_callchain;
+		struct callchain_cursor *cursor = NULL;
 
 		if (symbol_conf.use_callchain && sample->callchain &&
-		    thread__resolve_callchain(al->thread, &cursor_callchain, evsel,
+		    thread__resolve_callchain(al->thread, &callchain_cursor, evsel,
 					      sample, NULL, NULL, scripting_max_stack) == 0)
-			cursor = &cursor_callchain;
+			cursor = &callchain_cursor;
 
 		if (cursor == NULL) {
 			putchar(' ');
@@ -789,12 +789,12 @@ static void process_event(struct perf_script *script,
 		printf("%16" PRIu64, sample->weight);
 
 	if (PRINT_FIELD(IP)) {
-		struct callchain_cursor *cursor = NULL, cursor_callchain;
+		struct callchain_cursor *cursor = NULL;
 
 		if (symbol_conf.use_callchain && sample->callchain &&
-		    thread__resolve_callchain(al->thread, &cursor_callchain, evsel,
+		    thread__resolve_callchain(al->thread, &callchain_cursor, evsel,
 					      sample, NULL, NULL, scripting_max_stack) == 0)
-			cursor = &cursor_callchain;
+			cursor = &callchain_cursor;
 
 		putchar(cursor ? '\n' : ' ');
 		sample__fprintf_sym(sample, al, 0, output[attr->type].print_ip_opts, cursor, stdout);
-- 
2.5.5

[PATCH 05/30] perf trace: Make --event honour --min-stack too

[Arnaldo Carvalho de Melo]

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

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-shj0fazntmskhjild5i6x73l@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index d1bbcb9abca3..fc276d718172 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2126,6 +2126,17 @@ static int trace__event_handler(struct trace *trace, struct perf_evsel *evsel,
 				union perf_event *event __maybe_unused,
 				struct perf_sample *sample)
 {
+	int callchain_ret = 0;
+
+	if (sample->callchain) {
+		callchain_ret = trace__resolve_callchain(trace, evsel, sample, &callchain_cursor);
+		if (callchain_ret == 0) {
+			if (callchain_cursor.nr < trace->min_stack)
+				goto out;
+			callchain_ret = 1;
+		}
+	}
+
 	trace__printf_interrupted_entry(trace, sample);
 	trace__fprintf_tstamp(trace, sample->time, trace->output);
 
@@ -2144,11 +2155,11 @@ static int trace__event_handler(struct trace *trace, struct perf_evsel *evsel,
 
 	fprintf(trace->output, ")\n");
 
-	if (sample->callchain) {
-		if (trace__resolve_callchain(trace, evsel, sample, &callchain_cursor) == 0)
-			trace__fprintf_callchain(trace, sample);
-	}
-
+	if (callchain_ret > 0)
+		trace__fprintf_callchain(trace, sample);
+	else if (callchain_ret < 0)
+		pr_err("Problem processing %s callchain, skipping...\n", perf_evsel__name(evsel));
+out:
 	return 0;
 }
 
-- 
2.5.5

[PATCH 06/30] perf trace: Make --pf honour --min-stack too

[Arnaldo Carvalho de Melo]

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

To check deeply nested page fault callchains.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-wuji34xx003kr88nmqt6jkgf@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index fc276d718172..a4b133fac82b 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -2190,8 +2190,19 @@ static int trace__pgfault(struct trace *trace,
 	char map_type = 'd';
 	struct thread_trace *ttrace;
 	int err = -1;
+	int callchain_ret = 0;
 
 	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
+
+	if (sample->callchain) {
+		callchain_ret = trace__resolve_callchain(trace, evsel, sample, &callchain_cursor);
+		if (callchain_ret == 0) {
+			if (callchain_cursor.nr < trace->min_stack)
+				goto out_put;
+			callchain_ret = 1;
+		}
+	}
+
 	ttrace = thread__trace(thread, trace->output);
 	if (ttrace == NULL)
 		goto out_put;
@@ -2234,10 +2245,10 @@ static int trace__pgfault(struct trace *trace,
 
 	fprintf(trace->output, " (%c%c)\n", map_type, al.level);
 
-	if (sample->callchain) {
-		if (trace__resolve_callchain(trace, evsel, sample, &callchain_cursor) == 0)
-			trace__fprintf_callchain(trace, sample);
-	}
+	if (callchain_ret > 0)
+		trace__fprintf_callchain(trace, sample);
+	else if (callchain_ret < 0)
+		pr_err("Problem processing %s callchain, skipping...\n", perf_evsel__name(evsel));
 out:
 	err = 0;
 out_put:
-- 
2.5.5

[PATCH 07/30] perf evlist: Decode perf_event_attr->branch_sample_type

[Arnaldo Carvalho de Melo]

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

While trying to use --call-graph lbr in 'perf trace', since we only are
interested in the callchain for userspace, up to the callchain, I found
that 'perf evlist' is not decoding the branch_sample_type field, fix it.

Before:

  # perf record --call-graph lbr usleep 1
  # perf evlist -v
  cycles:ppp: size: 112, { sample_period, sample_freq }: 4000,
  sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD|BRANCH_STACK,
  disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1,
  precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1,
  comm_exec: 1, branch_sample_type: 51201
                ^^^^^^^^^^^^^^^^^^^^^^^^^

After:

  # perf evlist -v
  cycles:ppp: size: 112, { sample_period, sample_freq }: 4000,
  sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD|BRANCH_STACK,
  disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, task: 1,
  precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1,
  comm_exec: 1, branch_sample_type: USER|CALL_STACK|NO_FLAGS|NO_CYCLES
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-hozai7974u0ulgx13k96fcaw@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evsel.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 545bb3f0b2b0..334364e25bbe 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1231,6 +1231,21 @@ static void __p_sample_type(char *buf, size_t size, u64 value)
 	__p_bits(buf, size, value, bits);
 }
 
+static void __p_branch_sample_type(char *buf, size_t size, u64 value)
+{
+#define bit_name(n) { PERF_SAMPLE_BRANCH_##n, #n }
+	struct bit_names bits[] = {
+		bit_name(USER), bit_name(KERNEL), bit_name(HV), bit_name(ANY),
+		bit_name(ANY_CALL), bit_name(ANY_RETURN), bit_name(IND_CALL),
+		bit_name(ABORT_TX), bit_name(IN_TX), bit_name(NO_TX),
+		bit_name(COND), bit_name(CALL_STACK), bit_name(IND_JUMP),
+		bit_name(CALL), bit_name(NO_FLAGS), bit_name(NO_CYCLES),
+		{ .name = NULL, }
+	};
+#undef bit_name
+	__p_bits(buf, size, value, bits);
+}
+
 static void __p_read_format(char *buf, size_t size, u64 value)
 {
 #define bit_name(n) { PERF_FORMAT_##n, #n }
@@ -1249,6 +1264,7 @@ static void __p_read_format(char *buf, size_t size, u64 value)
 #define p_unsigned(val)		snprintf(buf, BUF_SIZE, "%"PRIu64, (uint64_t)(val))
 #define p_signed(val)		snprintf(buf, BUF_SIZE, "%"PRId64, (int64_t)(val))
 #define p_sample_type(val)	__p_sample_type(buf, BUF_SIZE, val)
+#define p_branch_sample_type(val) __p_branch_sample_type(buf, BUF_SIZE, val)
 #define p_read_format(val)	__p_read_format(buf, BUF_SIZE, val)
 
 #define PRINT_ATTRn(_n, _f, _p)				\
@@ -1305,7 +1321,7 @@ int perf_event_attr__fprintf(FILE *fp, struct perf_event_attr *attr,
 	PRINT_ATTRf(bp_type, p_unsigned);
 	PRINT_ATTRn("{ bp_addr, config1 }", bp_addr, p_hex);
 	PRINT_ATTRn("{ bp_len, config2 }", bp_len, p_hex);
-	PRINT_ATTRf(branch_sample_type, p_unsigned);
+	PRINT_ATTRf(branch_sample_type, p_branch_sample_type);
 	PRINT_ATTRf(sample_regs_user, p_hex);
 	PRINT_ATTRf(sample_stack_user, p_unsigned);
 	PRINT_ATTRf(clockid, p_signed);
-- 
2.5.5

[PATCH 08/30] perf tools: Make the x86 clean quiet

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Turn current clean output:

  $ make clean
  rm -f arch/x86/include/generated/asm/syscalls_64.c
    CLEAN    libbpf
    CLEAN    libapi

into:

  $ make clean
    CLEAN    x86
    CLEAN    libapi
    CLEAN    libbpf

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: TJ <linux@iam.tj>
Link: http://lkml.kernel.org/r/1461615438-27894-1-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/arch/x86/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/arch/x86/Makefile b/tools/perf/arch/x86/Makefile
index a33729173b13..6c9211b18ec0 100644
--- a/tools/perf/arch/x86/Makefile
+++ b/tools/perf/arch/x86/Makefile
@@ -24,6 +24,6 @@ $(header): $(sys)/syscall_64.tbl $(systbl)
 	$(Q)$(SHELL) '$(systbl)' $(sys)/syscall_64.tbl 'x86_64' > $@
 
 clean::
-	rm -f $(header)
+	$(call QUIET_CLEAN, x86) $(RM) $(header)
 
 archheaders: $(header)
-- 
2.5.5

[PATCH 09/30] tools build: Fix perf_clean target

[Arnaldo Carvalho de Melo]

From: Jiri Olsa <jolsa@kernel.org>

Fix perf_clean target to follow the same logic as perf target.

Fixes the following make invokation:

  $ cd <kernelsrc> && make tools/perf_clean

Reported-by: TJ <linux@iam.tj>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=116411
Link: http://lkml.kernel.org/r/1461615438-27894-2-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/Makefile | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/Makefile b/tools/Makefile
index 60c7e6c8ff17..6bf68fe7dd29 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -137,7 +137,8 @@ libsubcmd_clean:
 	$(call descend,lib/subcmd,clean)
 
 perf_clean:
-	$(call descend,$(@:_clean=),clean)
+	$(Q)mkdir -p $(PERF_O) .
+	$(Q)$(MAKE) --no-print-directory -C perf O=$(PERF_O) subdir= clean
 
 selftests_clean:
 	$(call descend,testing/$(@:_clean=),clean)
-- 
2.5.5

[PATCH 10/30] perf tools: Remove duplicate const qualifier

[Arnaldo Carvalho de Melo]

From: Eric Engestrom <eric.engestrom@imgtec.com>

Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1461577678-29517-1-git-send-email-eric.engestrom@imgtec.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/thread.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
index dfd00c6dad6e..de2036d1251b 100644
--- a/tools/perf/util/thread.c
+++ b/tools/perf/util/thread.c
@@ -233,7 +233,7 @@ void thread__find_cpumode_addr_location(struct thread *thread,
 					struct addr_location *al)
 {
 	size_t i;
-	const u8 const cpumodes[] = {
+	const u8 cpumodes[] = {
 		PERF_RECORD_MISC_USER,
 		PERF_RECORD_MISC_KERNEL,
 		PERF_RECORD_MISC_GUEST_USER,
-- 
2.5.5

[PATCH 11/30] perf tests: Replace assignment with comparison on assert check

[Arnaldo Carvalho de Melo]

From: Colin Ian King <colin.king@canonical.com>

The current assert check is checking an assignment, which will always be
true.  Instead, the assert should be checking if scale is equal to 0.122

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1461419154-16918-1-git-send-email-colin.king@canonical.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/tests/event_update.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/tests/event_update.c b/tools/perf/tests/event_update.c
index 012eab5d1df1..63ecf21750eb 100644
--- a/tools/perf/tests/event_update.c
+++ b/tools/perf/tests/event_update.c
@@ -30,7 +30,7 @@ static int process_event_scale(struct perf_tool *tool __maybe_unused,
 
 	TEST_ASSERT_VAL("wrong id", ev->id == 123);
 	TEST_ASSERT_VAL("wrong id", ev->type == PERF_EVENT_UPDATE__SCALE);
-	TEST_ASSERT_VAL("wrong scale", ev_data->scale = 0.123);
+	TEST_ASSERT_VAL("wrong scale", ev_data->scale == 0.123);
 	return 0;
 }
 
-- 
2.5.5

[PATCH 12/30] perf bench futex: Simplify wrapper for LOCK_PI

[Arnaldo Carvalho de Melo]

From: Davidlohr Bueso <dave@stgolabs.net>

Given that the 'val' parameter is ignored for FUTEX_LOCK_PI, get rid of
the bogus deadlock detection flag in the wrapper code and avoid the
extra argument, making it resemble its unlock counterpart. And if
nothing else, we already only pass 0 anyway.

Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Cc: Davidlohr Bueso <dbueso@suse.de>
Link: http://lkml.kernel.org/r/1461208447-29328-1-git-send-email-dave@stgolabs.net
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/bench/futex-lock-pi.c | 2 +-
 tools/perf/bench/futex.h         | 6 ++----
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/tools/perf/bench/futex-lock-pi.c b/tools/perf/bench/futex-lock-pi.c
index 6a18ce21f865..6952db65508a 100644
--- a/tools/perf/bench/futex-lock-pi.c
+++ b/tools/perf/bench/futex-lock-pi.c
@@ -83,7 +83,7 @@ static void *workerfn(void *arg)
 	do {
 		int ret;
 	again:
-		ret = futex_lock_pi(w->futex, NULL, 0, futex_flag);
+		ret = futex_lock_pi(w->futex, NULL, futex_flag);
 
 		if (ret) { /* handle lock acquisition */
 			if (!silent)
diff --git a/tools/perf/bench/futex.h b/tools/perf/bench/futex.h
index d44de9f44281..b2e06d1190d0 100644
--- a/tools/perf/bench/futex.h
+++ b/tools/perf/bench/futex.h
@@ -57,13 +57,11 @@ futex_wake(u_int32_t *uaddr, int nr_wake, int opflags)
 
 /**
  * futex_lock_pi() - block on uaddr as a PI mutex
- * @detect:	whether (1) or not (0) to perform deadlock detection
  */
 static inline int
-futex_lock_pi(u_int32_t *uaddr, struct timespec *timeout, int detect,
-	      int opflags)
+futex_lock_pi(u_int32_t *uaddr, struct timespec *timeout, int opflags)
 {
-	return futex(uaddr, FUTEX_LOCK_PI, detect, timeout, NULL, 0, opflags);
+	return futex(uaddr, FUTEX_LOCK_PI, 0, timeout, NULL, 0, opflags);
 }
 
 /**
-- 
2.5.5

[PATCH 13/30] perf intel-pt: Fix off-by-one comparison on maximum code

[Arnaldo Carvalho de Melo]

From: Colin Ian King <colin.king@canonical.com>

The check for the maximum code is off-by-one; the current comparison of
a code that is INTEL_PT_ERR_MAX will cause the strlcpy to perform an out
of bounds array access on the intel_pt_err_msgs array.

Fix this with a >= comparison.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1461524203-10224-1-git-send-email-colin.king@canonical.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/intel-pt-decoder/intel-pt-decoder.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
index 9409d014b46c..9c8f15da86ce 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
@@ -356,7 +356,7 @@ static const char *intel_pt_err_msgs[] = {
 
 int intel_pt__strerror(int code, char *buf, size_t buflen)
 {
-	if (code < 1 || code > INTEL_PT_ERR_MAX)
+	if (code < 1 || code >= INTEL_PT_ERR_MAX)
 		code = INTEL_PT_ERR_UNK;
 	strlcpy(buf, intel_pt_err_msgs[code], buflen);
 	return 0;
-- 
2.5.5

[PATCH 14/30] perf hists: Clear dummy entry accumulated period

[Arnaldo Carvalho de Melo]

From: Kan Liang <kan.liang@intel.com>

The accumulated period for dummy entry should also be 0.  Otherwise, the
total overhead could be overcounted.

  $ perf record -e '{LLC-load-misses,cpu/instructions/}' --call-graph=lbr ./tchain
  $ perf report --stdio
  # To display the perf.data header info, please use --header/--header-only options.
  #
  # Total Lost Samples: 0
  #
  # Samples: 21K of event 'anon group { LLC-load-misses, cpu/instructions/ }'
  # Event count (approx.): 16313667937
  #
  #         Children              Self  Command      Shared Object     Symbol
  # ................  ................  ...........  ................  ............................
  #
    4769.98%   0.01%     0.00%   0.01%  tchain_edit  [kernel.vmlinux]  [k] update_fast_timekeeper
    4356.18%   0.01%     0.00%   0.01%  tchain_edit  [kernel.vmlinux]  [k] trigger_load_balance
    3181.12%   0.01%     0.00%   0.01%  tchain_edit  [kernel.vmlinux]  [k] irq_work_tick
    1592.37%   0.00%     0.00%   0.00%  tchain_edit  [kernel.vmlinux]  [k] cpu_needs_another_gp

Signed-off-by: Kan Liang <kan.liang@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1461565689-5862-1-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/hist.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index 991a351a8a41..0f33d7e698c4 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -2062,6 +2062,8 @@ static struct hist_entry *hists__add_dummy_entry(struct hists *hists,
 	if (he) {
 		memset(&he->stat, 0, sizeof(he->stat));
 		he->hists = hists;
+		if (symbol_conf.cumulate_callchain)
+			memset(he->stat_acc, 0, sizeof(he->stat));
 		rb_link_node(&he->rb_node_in, parent, p);
 		rb_insert_color(&he->rb_node_in, root);
 		hists__inc_stats(hists, he);
-- 
2.5.5

[PATCH 15/30] perf evlist: Enforce ring buffer reading

[Arnaldo Carvalho de Melo]

From: Wang Nan <wangnan0@huawei.com>

Don't read broken data after 'head' pointer.

Following commits will feed perf_evlist__mmap_read() with some 'head'
pointers not maintained by kernel. If 'head' pointer breaks an event, we
should avoid reading from the broken event. This can happen in backward
ring buffer.

For example:

                              old     head
                                |     |
                                V     V
     +---+------+----------+----+-----+--+
     |..E|D....D|C........C|B..B|A....|E.|
     +---+------+----------+----+-----+--+

'old' pointer points to the beginning of 'A' and trying read from it,
but 'A' has been overwritten. In this case, don't try to read from 'A',
simply return NULL.

Signed-off-by: Wang Nan <wangnan0@huawei.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Link: http://lkml.kernel.org/r/1461637738-62722-2-git-send-email-wangnan0@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/evlist.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 6fb5725821de..85271e54a63b 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -684,6 +684,7 @@ union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
 	struct perf_mmap *md = &evlist->mmap[idx];
 	u64 head;
 	u64 old = md->prev;
+	int diff;
 	unsigned char *data = md->base + page_size;
 	union perf_event *event = NULL;
 
@@ -694,6 +695,7 @@ union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
 		return NULL;
 
 	head = perf_mmap__read_head(md);
+	diff = head - old;
 	if (evlist->overwrite) {
 		/*
 		 * If we're further behind than half the buffer, there's a chance
@@ -703,7 +705,6 @@ union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
 		 *
 		 * In either case, truncate and restart at head.
 		 */
-		int diff = head - old;
 		if (diff > md->mask / 2 || diff < 0) {
 			fprintf(stderr, "WARNING: failed to keep up with mmap data.\n");
 
@@ -711,15 +712,21 @@ union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
 			 * head points to a known good entry, start there.
 			 */
 			old = head;
+			diff = 0;
 		}
 	}
 
-	if (old != head) {
+	if (diff >= (int)sizeof(event->header)) {
 		size_t size;
 
 		event = (union perf_event *)&data[old & md->mask];
 		size = event->header.size;
 
+		if (size < sizeof(event->header) || diff < (int)size) {
+			event = NULL;
+			goto broken_event;
+		}
+
 		/*
 		 * Event straddles the mmap boundary -- header should always
 		 * be inside due to u64 alignment of output.
@@ -743,6 +750,7 @@ union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
 		old += size;
 	}
 
+broken_event:
 	md->prev = old;
 
 	return event;
-- 
2.5.5

[PATCH 16/30] perf probe: Close target file on error path

[Arnaldo Carvalho de Melo]

From: Masami Hiramatsu <mhiramat@kernel.org>

Fix a bug to close target elf file in get_text_start_address().

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20160426064737.1443.44093.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-event.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 8319fbb08636..97b7f8e5fe69 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -486,8 +486,10 @@ static int get_text_start_address(const char *exec, unsigned long *address)
 		return -errno;
 
 	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
-	if (elf == NULL)
-		return -EINVAL;
+	if (elf == NULL) {
+		ret = -EINVAL;
+		goto out_close;
+	}
 
 	if (gelf_getehdr(elf, &ehdr) == NULL)
 		goto out;
@@ -499,6 +501,9 @@ static int get_text_start_address(const char *exec, unsigned long *address)
 	ret = 0;
 out:
 	elf_end(elf);
+out_close:
+	close(fd);
+
 	return ret;
 }
 
-- 
2.5.5

[PATCH 17/30] perf tools: Add lsdir() helper to read a directory

[Arnaldo Carvalho de Melo]

From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>

As a utility function, add lsdir() which reads given directory and store
entry name into a strlist.  lsdir accepts a filter function so that user
can filter out unneeded entries.

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
Cc: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20160426090242.11891.79014.stgit@devbox
[ Do not use the 'dirname' it is used in some distros ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/util.c | 34 ++++++++++++++++++++++++++++++++++
 tools/perf/util/util.h |  3 +++
 2 files changed, 37 insertions(+)

diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index b7766c577b01..9473d46c00bb 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -117,6 +117,40 @@ int rm_rf(char *path)
 	return rmdir(path);
 }
 
+/* A filter which removes dot files */
+bool lsdir_no_dot_filter(const char *name __maybe_unused, struct dirent *d)
+{
+	return d->d_name[0] != '.';
+}
+
+/* lsdir reads a directory and store it in strlist */
+struct strlist *lsdir(const char *name,
+		      bool (*filter)(const char *, struct dirent *))
+{
+	struct strlist *list = NULL;
+	DIR *dir;
+	struct dirent *d;
+
+	dir = opendir(name);
+	if (!dir)
+		return NULL;
+
+	list = strlist__new(NULL, NULL);
+	if (!list) {
+		errno = -ENOMEM;
+		goto out;
+	}
+
+	while ((d = readdir(dir)) != NULL) {
+		if (!filter || filter(name, d))
+			strlist__add(list, d->d_name);
+	}
+
+out:
+	closedir(dir);
+	return list;
+}
+
 static int slow_copyfile(const char *from, const char *to)
 {
 	int err = -1;
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 3bf3de86d429..26a924651e7b 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -79,6 +79,7 @@
 #include <termios.h>
 #include <linux/bitops.h>
 #include <termios.h>
+#include "strlist.h"
 
 extern const char *graph_line;
 extern const char *graph_dotted_line;
@@ -222,6 +223,8 @@ static inline int sane_case(int x, int high)
 
 int mkdir_p(char *path, mode_t mode);
 int rm_rf(char *path);
+struct strlist *lsdir(const char *name, bool (*filter)(const char *, struct dirent *));
+bool lsdir_no_dot_filter(const char *name, struct dirent *d);
 int copyfile(const char *from, const char *to);
 int copyfile_mode(const char *from, const char *to, mode_t mode);
 int copyfile_offset(int fromfd, loff_t from_ofs, int tofd, loff_t to_ofs, u64 size);
-- 
2.5.5

[PATCH 18/30] perf probe: Let probe_file__add_event return 0 if succeeded

[Arnaldo Carvalho de Melo]

From: Masami Hiramatsu <mhiramat@kernel.org>

Since other methods return 0 if succeeded (or filedesc), let
probe_file__add_event() return 0 instead of the length of written bytes.

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
Cc: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20160426090303.11891.18232.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-file.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/tools/perf/util/probe-file.c b/tools/perf/util/probe-file.c
index e3b3b92e4458..3fe6214970e6 100644
--- a/tools/perf/util/probe-file.c
+++ b/tools/perf/util/probe-file.c
@@ -220,8 +220,7 @@ int probe_file__add_event(int fd, struct probe_trace_event *tev)
 
 	pr_debug("Writing event: %s\n", buf);
 	if (!probe_event_dry_run) {
-		ret = write(fd, buf, strlen(buf));
-		if (ret <= 0) {
+		if (write(fd, buf, strlen(buf)) < (int)strlen(buf)) {
 			ret = -errno;
 			pr_warning("Failed to write event: %s\n",
 				   strerror_r(errno, sbuf, sizeof(sbuf)));
-- 
2.5.5

[PATCH 19/30] perf probe: Set default kprobe group name if it is not given

[Arnaldo Carvalho de Melo]

From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>

Set kprobe group name as "probe" if it is not given.

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
Cc: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20160426090413.11891.95640.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-event.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 97b7f8e5fe69..0de5d10dda71 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -2748,9 +2748,13 @@ static int convert_to_probe_trace_events(struct perf_probe_event *pev,
 {
 	int ret;
 
-	if (pev->uprobes && !pev->group) {
-		/* Replace group name if not given */
-		ret = convert_exec_to_group(pev->target, &pev->group);
+	if (!pev->group) {
+		/* Set group name if not given */
+		if (!pev->uprobes) {
+			pev->group = strdup(PERFPROBE_GROUP);
+			ret = pev->group ? 0 : -ENOMEM;
+		} else
+			ret = convert_exec_to_group(pev->target, &pev->group);
 		if (ret != 0) {
 			pr_warning("Failed to make a group name.\n");
 			return ret;
-- 
2.5.5

[PATCH 20/30] perf trace: Move perf_flags beautifier to tools/perf/trace/beauty/

[Arnaldo Carvalho de Melo]

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

To reduce the size of builtin-trace.c.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-8r3gmymyn3r0ynt4yuzspp9g@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c                | 45 +------------------------------
 tools/perf/trace/beauty/perf_event_open.c | 43 +++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+), 44 deletions(-)
 create mode 100644 tools/perf/trace/beauty/perf_event_open.c

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index a4b133fac82b..903deda92d9e 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -56,22 +56,6 @@
 # define MSG_CMSG_CLOEXEC	0x40000000
 #endif
 
-#ifndef PERF_FLAG_FD_NO_GROUP
-# define PERF_FLAG_FD_NO_GROUP		(1UL << 0)
-#endif
-
-#ifndef PERF_FLAG_FD_OUTPUT
-# define PERF_FLAG_FD_OUTPUT		(1UL << 1)
-#endif
-
-#ifndef PERF_FLAG_PID_CGROUP
-# define PERF_FLAG_PID_CGROUP		(1UL << 2) /* pid=cgroup id, per-cpu mode only */
-#endif
-
-#ifndef PERF_FLAG_FD_CLOEXEC
-# define PERF_FLAG_FD_CLOEXEC		(1UL << 3) /* O_CLOEXEC */
-#endif
-
 struct trace {
 	struct perf_tool	tool;
 	struct syscalltbl	*sctbl;
@@ -674,34 +658,6 @@ static size_t syscall_arg__scnprintf_open_flags(char *bf, size_t size,
 
 #define SCA_OPEN_FLAGS syscall_arg__scnprintf_open_flags
 
-static size_t syscall_arg__scnprintf_perf_flags(char *bf, size_t size,
-						struct syscall_arg *arg)
-{
-	int printed = 0, flags = arg->val;
-
-	if (flags == 0)
-		return 0;
-
-#define	P_FLAG(n) \
-	if (flags & PERF_FLAG_##n) { \
-		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
-		flags &= ~PERF_FLAG_##n; \
-	}
-
-	P_FLAG(FD_NO_GROUP);
-	P_FLAG(FD_OUTPUT);
-	P_FLAG(PID_CGROUP);
-	P_FLAG(FD_CLOEXEC);
-#undef P_FLAG
-
-	if (flags)
-		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
-
-	return printed;
-}
-
-#define SCA_PERF_FLAGS syscall_arg__scnprintf_perf_flags
-
 static size_t syscall_arg__scnprintf_pipe_flags(char *bf, size_t size,
 						struct syscall_arg *arg)
 {
@@ -894,6 +850,7 @@ static size_t syscall_arg__scnprintf_getrandom_flags(char *bf, size_t size,
 #include "trace/beauty/pid.c"
 #include "trace/beauty/mmap.c"
 #include "trace/beauty/mode_t.c"
+#include "trace/beauty/perf_event_open.c"
 #include "trace/beauty/sched_policy.c"
 #include "trace/beauty/socket_type.c"
 #include "trace/beauty/waitid_options.c"
diff --git a/tools/perf/trace/beauty/perf_event_open.c b/tools/perf/trace/beauty/perf_event_open.c
new file mode 100644
index 000000000000..311f09dd718d
--- /dev/null
+++ b/tools/perf/trace/beauty/perf_event_open.c
@@ -0,0 +1,43 @@
+#ifndef PERF_FLAG_FD_NO_GROUP
+# define PERF_FLAG_FD_NO_GROUP		(1UL << 0)
+#endif
+
+#ifndef PERF_FLAG_FD_OUTPUT
+# define PERF_FLAG_FD_OUTPUT		(1UL << 1)
+#endif
+
+#ifndef PERF_FLAG_PID_CGROUP
+# define PERF_FLAG_PID_CGROUP		(1UL << 2) /* pid=cgroup id, per-cpu mode only */
+#endif
+
+#ifndef PERF_FLAG_FD_CLOEXEC
+# define PERF_FLAG_FD_CLOEXEC		(1UL << 3) /* O_CLOEXEC */
+#endif
+
+static size_t syscall_arg__scnprintf_perf_flags(char *bf, size_t size,
+						struct syscall_arg *arg)
+{
+	int printed = 0, flags = arg->val;
+
+	if (flags == 0)
+		return 0;
+
+#define	P_FLAG(n) \
+	if (flags & PERF_FLAG_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		flags &= ~PERF_FLAG_##n; \
+	}
+
+	P_FLAG(FD_NO_GROUP);
+	P_FLAG(FD_OUTPUT);
+	P_FLAG(PID_CGROUP);
+	P_FLAG(FD_CLOEXEC);
+#undef P_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
+
+	return printed;
+}
+
+#define SCA_PERF_FLAGS syscall_arg__scnprintf_perf_flags
-- 
2.5.5

[PATCH 21/30] perf trace: Do not beautify the 'pid' parameter as a simple integer

[Arnaldo Carvalho de Melo]

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

Leave it alone so that it ends up assigned to SCA_PID via its type,
'pid_t', that will look up the pid on the machine thread rb_tree and
possibly find its COMM.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-r7dujgmhtxxfajuunpt1bkuo@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/builtin-trace.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 903deda92d9e..48b00f042599 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -1043,8 +1043,7 @@ static struct syscall_fmt {
 			     [1] = SCA_FILENAME, /* filename */
 			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
 	{ .name	    = "perf_event_open", .errmsg = true,
-	  .arg_scnprintf = { [1] = SCA_INT, /* pid */
-			     [2] = SCA_INT, /* cpu */
+	  .arg_scnprintf = { [2] = SCA_INT, /* cpu */
 			     [3] = SCA_FD,  /* group_fd */
 			     [4] = SCA_PERF_FLAGS,  /* flags */ }, },
 	{ .name	    = "pipe2",	    .errmsg = true,
-- 
2.5.5

[PATCH 22/30] tools lib api fs: Add helper to read string from procfs file

[Arnaldo Carvalho de Melo]

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

To read things like /proc/self/comm.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-ztpkbmseidt0hq2psr46o0h9@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/lib/api/fs/fs.c | 13 +++++++++++++
 tools/lib/api/fs/fs.h |  2 ++
 2 files changed, 15 insertions(+)

diff --git a/tools/lib/api/fs/fs.c b/tools/lib/api/fs/fs.c
index ef78c22ff44d..08556cf2c70d 100644
--- a/tools/lib/api/fs/fs.c
+++ b/tools/lib/api/fs/fs.c
@@ -351,6 +351,19 @@ int filename__read_str(const char *filename, char **buf, size_t *sizep)
 	return err;
 }
 
+int procfs__read_str(const char *entry, char **buf, size_t *sizep)
+{
+	char path[PATH_MAX];
+	const char *procfs = procfs__mountpoint();
+
+	if (!procfs)
+		return -1;
+
+	snprintf(path, sizeof(path), "%s/%s", procfs, entry);
+
+	return filename__read_str(path, buf, sizep);
+}
+
 int sysfs__read_ull(const char *entry, unsigned long long *value)
 {
 	char path[PATH_MAX];
diff --git a/tools/lib/api/fs/fs.h b/tools/lib/api/fs/fs.h
index 9f6598098dc5..16c9c2ed7c5b 100644
--- a/tools/lib/api/fs/fs.h
+++ b/tools/lib/api/fs/fs.h
@@ -29,6 +29,8 @@ int filename__read_int(const char *filename, int *value);
 int filename__read_ull(const char *filename, unsigned long long *value);
 int filename__read_str(const char *filename, char **buf, size_t *sizep);
 
+int procfs__read_str(const char *entry, char **buf, size_t *sizep);
+
 int sysctl__read_int(const char *sysctl, int *value);
 int sysfs__read_int(const char *entry, int *value);
 int sysfs__read_ull(const char *entry, unsigned long long *value);
-- 
2.5.5

[PATCH 23/30] perf thread: Introduce method to set comm from /proc/pid/self

[Arnaldo Carvalho de Melo]

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

Will be used for lazy comm loading in 'perf trace'.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-7ogbkuoka1y2qsmcckqxvl5m@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/thread.c | 19 +++++++++++++++++++
 tools/perf/util/thread.h |  2 ++
 2 files changed, 21 insertions(+)

diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
index de2036d1251b..45fcb715a36b 100644
--- a/tools/perf/util/thread.c
+++ b/tools/perf/util/thread.c
@@ -10,6 +10,8 @@
 #include "comm.h"
 #include "unwind.h"
 
+#include <api/fs/fs.h>
+
 int thread__init_map_groups(struct thread *thread, struct machine *machine)
 {
 	struct thread *leader;
@@ -153,6 +155,23 @@ int __thread__set_comm(struct thread *thread, const char *str, u64 timestamp,
 	return 0;
 }
 
+int thread__set_comm_from_proc(struct thread *thread)
+{
+	char path[64];
+	char *comm = NULL;
+	size_t sz;
+	int err = -1;
+
+	if (!(snprintf(path, sizeof(path), "%d/task/%d/comm",
+		       thread->pid_, thread->tid) >= (int)sizeof(path)) &&
+	    procfs__read_str(path, &comm, &sz) == 0) {
+		comm[sz - 1] = '\0';
+		err = thread__set_comm(thread, comm, 0);
+	}
+
+	return err;
+}
+
 const char *thread__comm_str(const struct thread *thread)
 {
 	const struct comm *comm = thread__comm(thread);
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index e214207bb13a..45fba13c800b 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -71,6 +71,8 @@ static inline int thread__set_comm(struct thread *thread, const char *comm,
 	return __thread__set_comm(thread, comm, timestamp, false);
 }
 
+int thread__set_comm_from_proc(struct thread *thread);
+
 int thread__comm_len(struct thread *thread);
 struct comm *thread__comm(const struct thread *thread);
 struct comm *thread__exec_comm(const struct thread *thread);
-- 
2.5.5

[PATCH 24/30] perf trace: Read thread's COMM from /proc when not set

[Arnaldo Carvalho de Melo]

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

We get notifications for threads that gets created while we're tracing,
but for preexisting threads we may end not having synthesized them, like
when tracing a 'perf trace' session that will use '--pid' to trace some
other thread.

And besides we should probably stop synthesizing those records and
instead read thread information in a lazy way, i.e. just when we need,
like done in this patch:

Now the 'pid_t' argument in 'perf_event_open' gets translated to a COMM:

  # perf trace -e perf_event_open perf stat -e cycles -p 31601
     0.027 ( 0.027 ms): perf/23393 perf_event_open(attr_uptr: 0x2fdd0d8, pid: 31601 (abrt-dump-journ), cpu: -1, group_fd: -1, flags: FD_CLOEXEC)
                                                                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
= 3
^C

And in other syscalls containing pid_t without thread->comm_set at the
time of the formatting.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-ioeps6dlwst17d6oozc9shtk@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/trace/beauty/pid.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/perf/trace/beauty/pid.c b/tools/perf/trace/beauty/pid.c
index 111ae08d38f1..07486ea65ae3 100644
--- a/tools/perf/trace/beauty/pid.c
+++ b/tools/perf/trace/beauty/pid.c
@@ -3,9 +3,12 @@ static size_t syscall_arg__scnprintf_pid(char *bf, size_t size, struct syscall_a
 	int pid = arg->val;
 	struct trace *trace = arg->trace;
 	size_t printed = scnprintf(bf, size, "%d", pid);
-	struct thread *thread = machine__find_thread(trace->host, pid, pid);
+	struct thread *thread = machine__findnew_thread(trace->host, pid, pid);
 
 	if (thread != NULL) {
+		if (!thread->comm_set)
+			thread__set_comm_from_proc(thread);
+
 		if (thread->comm_set)
 			printed += scnprintf(bf + printed, size - printed,
 					     " (%s)", thread__comm_str(thread));
-- 
2.5.5

[PATCH 25/30] perf probe: Fix offline module name missmatch issue

[Arnaldo Carvalho de Melo]

From: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>

Perf can add a probe on kernel module which has not been loaded yet.

The current implementation finds the module name from path. But if the
filename is different from the actual module name then perf fails to
register a probe while loading module because of mismatch in the names.

For example, samples/kobject/kobject-example.ko is loaded as
kobject_example.

Before applying patch:

  $ sudo ./perf probe -m /linux/samples/kobject/kobject-example.ko foo_show
    Added new event:
      probe:foo_show       (on foo_show in kobject-example)

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

    perf record -e probe:foo_show -aR sleep 1

  $ cat /sys/kernel/debug/tracing/kprobe_events
    p:probe/foo_show kobject-example:foo_show

  $ insmod kobject-example.ko

  $ lsmod
    Module                  Size  Used by
    kobject_example        16384  0

  Generate read to /sys/kernel/kobject_example/foo while recording data
  with below command
  $ sudo ./perf record -e probe:foo_show -a
    [ perf record: Woken up 1 times to write data ]
    [ perf record: Captured and wrote 0.093 MB perf.data ]

  $./perf report --stdio -F overhead,comm,dso,sym
    Error:
    The perf.data.old file has no samples!

After applying patch:

  $ sudo ./perf probe -m /linux/samples/kobject/kobject-example.ko foo_show
    Added new event:
      probe:foo_show       (on foo_show in kobject_example)

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

    perf record -e probe:foo_show -aR sleep 1

  $ sudo cat /sys/kernel/debug/tracing/kprobe_events
    p:probe/foo_show kobject_example:foo_show

  $ insmod kobject-example.ko

  $ lsmod
    Module                  Size  Used by
    kobject_example        16384  0

  Generate read to /sys/kernel/kobject_example/foo while recording data
  with below command
  $ sudo ./perf record -e probe:foo_show -a
    [ perf record: Woken up 1 times to write data ]
    [ perf record: Captured and wrote 0.097 MB perf.data (8 samples) ]

  $ sudo ./perf report  --stdio -F overhead,comm,dso,sym
    ...
    # Samples: 8  of event 'probe:foo_show'
    # Event count (approx.): 8
    #
    # Overhead  Command  Shared Object      Symbol
    # ........  .......  .................  ............
    #
       100.00%  cat      [kobject_example]  [k] foo_show

Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1461680741-12517-2-git-send-email-ravi.bangoria@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-event.c | 19 +++++--------------
 1 file changed, 5 insertions(+), 14 deletions(-)

diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 0de5d10dda71..bc2eb7cda2d1 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -588,32 +588,23 @@ static int add_module_to_probe_trace_events(struct probe_trace_event *tevs,
 					    int ntevs, const char *module)
 {
 	int i, ret = 0;
-	char *tmp;
+	char *mod_name = NULL;
 
 	if (!module)
 		return 0;
 
-	tmp = strrchr(module, '/');
-	if (tmp) {
-		/* This is a module path -- get the module name */
-		module = strdup(tmp + 1);
-		if (!module)
-			return -ENOMEM;
-		tmp = strchr(module, '.');
-		if (tmp)
-			*tmp = '\0';
-		tmp = (char *)module;	/* For free() */
-	}
+	mod_name = find_module_name(module);
 
 	for (i = 0; i < ntevs; i++) {
-		tevs[i].point.module = strdup(module);
+		tevs[i].point.module =
+			strdup(mod_name ? mod_name : module);
 		if (!tevs[i].point.module) {
 			ret = -ENOMEM;
 			break;
 		}
 	}
 
-	free(tmp);
+	free(mod_name);
 	return ret;
 }
 
-- 
2.5.5

[PATCH 26/30] perf probe: Fix module probe issue if no dwarf support

[Arnaldo Carvalho de Melo]

From: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>

Perf is not able to register probe in kernel module when dwarf supprt
is not there(and so it goes for symtab). Perf passes full path of
module where only module name is required which is causing the problem.
This patch fixes this issue.

Before applying patch:

  $ dpkg -s libdw-dev
  dpkg-query: package 'libdw-dev' is not installed and no information is...

  $ sudo ./perf probe -m /linux/samples/kprobes/kprobe_example.ko kprobe_init
  Added new event:
    probe:kprobe_init (on kprobe_init in /linux/samples/kprobes/kprobe_example.ko)

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

  perf record -e probe:kprobe_init -aR sleep 1

  $ sudo cat /sys/kernel/debug/tracing/kprobe_events
  p:probe/kprobe_init /linux/samples/kprobes/kprobe_example.ko:kprobe_init

  $ sudo ./perf record -a -e probe:kprobe_init
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.105 MB perf.data ]

  $ sudo ./perf script 	# No output here

After applying patch:

  $ sudo ./perf probe -m /linux/samples/kprobes/kprobe_example.ko kprobe_init
  Added new event:
    probe:kprobe_init    (on kprobe_init in kprobe_example)

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

  perf record -e probe:kprobe_init -aR sleep 1

  $ sudo cat /sys/kernel/debug/tracing/kprobe_events
  p:probe/kprobe_init kprobe_example:kprobe_init

  $ sudo ./perf record -a -e probe:kprobe_init
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.105 MB perf.data (2 samples) ]

  $ sudo ./perf script
  insmod 13990 [002]  5961.216833: probe:kprobe_init: ...
  insmod 13995 [002]  5962.889384: probe:kprobe_init: ...

Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1461680741-12517-1-git-send-email-ravi.bangoria@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-event.c | 76 +++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 73 insertions(+), 3 deletions(-)

diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index bc2eb7cda2d1..a9774628c6f6 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -265,6 +265,65 @@ static bool kprobe_warn_out_range(const char *symbol, unsigned long address)
 	return true;
 }
 
+/*
+ * NOTE:
+ * '.gnu.linkonce.this_module' section of kernel module elf directly
+ * maps to 'struct module' from linux/module.h. This section contains
+ * actual module name which will be used by kernel after loading it.
+ * But, we cannot use 'struct module' here since linux/module.h is not
+ * exposed to user-space. Offset of 'name' has remained same from long
+ * time, so hardcoding it here.
+ */
+#ifdef __LP64__
+#define MOD_NAME_OFFSET 24
+#else
+#define MOD_NAME_OFFSET 12
+#endif
+
+/*
+ * @module can be module name of module file path. In case of path,
+ * inspect elf and find out what is actual module name.
+ * Caller has to free mod_name after using it.
+ */
+static char *find_module_name(const char *module)
+{
+	int fd;
+	Elf *elf;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	Elf_Data *data;
+	Elf_Scn *sec;
+	char *mod_name = NULL;
+
+	fd = open(module, O_RDONLY);
+	if (fd < 0)
+		return NULL;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL)
+		goto elf_err;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL)
+		goto ret_err;
+
+	sec = elf_section_by_name(elf, &ehdr, &shdr,
+			".gnu.linkonce.this_module", NULL);
+	if (!sec)
+		goto ret_err;
+
+	data = elf_getdata(sec, NULL);
+	if (!data || !data->d_buf)
+		goto ret_err;
+
+	mod_name = strdup((char *)data->d_buf + MOD_NAME_OFFSET);
+
+ret_err:
+	elf_end(elf);
+elf_err:
+	close(fd);
+	return mod_name;
+}
+
 #ifdef HAVE_DWARF_SUPPORT
 
 static int kernel_get_module_dso(const char *module, struct dso **pdso)
@@ -2512,6 +2571,7 @@ static int find_probe_trace_events_from_map(struct perf_probe_event *pev,
 	struct probe_trace_point *tp;
 	int num_matched_functions;
 	int ret, i, j, skipped = 0;
+	char *mod_name;
 
 	map = get_target_map(pev->target, pev->uprobes);
 	if (!map) {
@@ -2596,9 +2656,19 @@ static int find_probe_trace_events_from_map(struct perf_probe_event *pev,
 		tp->realname = strdup_or_goto(sym->name, nomem_out);
 
 		tp->retprobe = pp->retprobe;
-		if (pev->target)
-			tev->point.module = strdup_or_goto(pev->target,
-							   nomem_out);
+		if (pev->target) {
+			if (pev->uprobes) {
+				tev->point.module = strdup_or_goto(pev->target,
+								   nomem_out);
+			} else {
+				mod_name = find_module_name(pev->target);
+				tev->point.module =
+					strdup(mod_name ? mod_name : pev->target);
+				free(mod_name);
+				if (!tev->point.module)
+					goto nomem_out;
+			}
+		}
 		tev->uprobes = pev->uprobes;
 		tev->nargs = pev->nargs;
 		if (tev->nargs) {
-- 
2.5.5

[PATCH 27/30] perf tools: Update x86's syscall_64.tbl, adding preadv2 & pwritev2

[Arnaldo Carvalho de Melo]

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

Introduced in commit 4babf2c5efb7 ("x86: wire up preadv2 and pwritev2").

This will make 'perf trace' aware of them.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-vojoylgce2cetsy36446s5ny@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/arch/x86/entry/syscalls/syscall_64.tbl | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl b/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
index 2e5b565adacc..cac6d17ce5db 100644
--- a/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
+++ b/tools/perf/arch/x86/entry/syscalls/syscall_64.tbl
@@ -333,6 +333,8 @@
 324	common	membarrier		sys_membarrier
 325	common	mlock2			sys_mlock2
 326	common	copy_file_range		sys_copy_file_range
+327	64	preadv2			sys_preadv2
+328	64	pwritev2		sys_pwritev2
 
 #
 # x32-specific system call numbers start at 512 to avoid cache impact
-- 
2.5.5

[PATCH 28/30] perf bench: Remove one more die() call

[Arnaldo Carvalho de Melo]

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

Propagate the error instead.

Cc: David Ahern <dsahern@gmail.com>
Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-z6erjg35d1gekevwujoa0223@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/bench/mem-functions.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)

diff --git a/tools/perf/bench/mem-functions.c b/tools/perf/bench/mem-functions.c
index a91aa85d80ff..2b54d0f2672a 100644
--- a/tools/perf/bench/mem-functions.c
+++ b/tools/perf/bench/mem-functions.c
@@ -6,6 +6,7 @@
  * Written by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
  */
 
+#include "debug.h"
 #include "../perf.h"
 #include "../util/util.h"
 #include <subcmd/parse-options.h>
@@ -63,14 +64,16 @@ static struct perf_event_attr cycle_attr = {
 	.config		= PERF_COUNT_HW_CPU_CYCLES
 };
 
-static void init_cycles(void)
+static int init_cycles(void)
 {
 	cycles_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1, perf_event_open_cloexec_flag());
 
-	if (cycles_fd < 0 && errno == ENOSYS)
-		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
-	else
-		BUG_ON(cycles_fd < 0);
+	if (cycles_fd < 0 && errno == ENOSYS) {
+		pr_debug("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
+		return -1;
+	}
+
+	return cycles_fd;
 }
 
 static u64 get_cycles(void)
@@ -155,8 +158,13 @@ static int bench_mem_common(int argc, const char **argv, struct bench_mem_info *
 
 	argc = parse_options(argc, argv, options, info->usage, 0);
 
-	if (use_cycles)
-		init_cycles();
+	if (use_cycles) {
+		i = init_cycles();
+		if (i < 0) {
+			fprintf(stderr, "Failed to open cycles counter\n");
+			return i;
+		}
+	}
 
 	size = (size_t)perf_atoll((char *)size_str);
 	size_total = (double)size * nr_loops;
-- 
2.5.5

[PATCH 29/30] perf core: Allow setting up max frame stack depth via sysctl

[Arnaldo Carvalho de Melo]

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

The default remains 127, which is good for most cases, and not even hit
most of the time, but then for some cases, as reported by Brendan, 1024+
deep frames are appearing on the radar for things like groovy, ruby.

And in some workloads putting a _lower_ cap on this may make sense. One
that is per event still needs to be put in place tho.

The new file is:

  # cat /proc/sys/kernel/perf_event_max_stack
  127

Chaging it:

  # echo 256 > /proc/sys/kernel/perf_event_max_stack
  # cat /proc/sys/kernel/perf_event_max_stack
  256

But as soon as there is some event using callchains we get:

  # echo 512 > /proc/sys/kernel/perf_event_max_stack
  -bash: echo: write error: Device or resource busy
  #

Because we only allocate the callchain percpu data structures when there
is a user, which allows for changing the max easily, its just a matter
of having no callchain users at that point.

Reported-and-Tested-by: Brendan Gregg <brendan.d.gregg@gmail.com>
Reviewed-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: David Ahern <dsahern@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: He Kuang <hekuang@huawei.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: Zefan Li <lizefan@huawei.com>
Link: http://lkml.kernel.org/r/20160426002928.GB16708@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 Documentation/sysctl/kernel.txt    | 14 ++++++++++++++
 arch/arm/kernel/perf_callchain.c   |  2 +-
 arch/arm64/kernel/perf_callchain.c |  4 ++--
 arch/metag/kernel/perf_callchain.c |  2 +-
 arch/mips/kernel/perf_event.c      |  4 ++--
 arch/powerpc/perf/callchain.c      |  4 ++--
 arch/sparc/kernel/perf_event.c     |  6 +++---
 arch/x86/events/core.c             |  4 ++--
 arch/xtensa/kernel/perf_event.c    |  4 ++--
 include/linux/perf_event.h         |  8 ++++++--
 kernel/bpf/stackmap.c              |  8 ++++----
 kernel/events/callchain.c          | 35 +++++++++++++++++++++++++++++++++--
 kernel/sysctl.c                    | 12 ++++++++++++
 13 files changed, 84 insertions(+), 23 deletions(-)

diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
index 57653a44b128..260cde08e92e 100644
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -60,6 +60,7 @@ show up in /proc/sys/kernel:
 - panic_on_warn
 - perf_cpu_time_max_percent
 - perf_event_paranoid
+- perf_event_max_stack
 - pid_max
 - powersave-nap               [ PPC only ]
 - printk
@@ -654,6 +655,19 @@ users (without CAP_SYS_ADMIN).  The default value is 1.
 
 ==============================================================
 
+perf_event_max_stack:
+
+Controls maximum number of stack frames to copy for (attr.sample_type &
+PERF_SAMPLE_CALLCHAIN) configured events, for instance, when using
+'perf record -g' or 'perf trace --call-graph fp'.
+
+This can only be done when no events are in use that have callchains
+enabled, otherwise writing to this file will return -EBUSY.
+
+The default value is 127.
+
+==============================================================
+
 pid_max:
 
 PID allocation wrap value.  When the kernel's next PID value
diff --git a/arch/arm/kernel/perf_callchain.c b/arch/arm/kernel/perf_callchain.c
index 4e02ae5950ff..27563befa8a2 100644
--- a/arch/arm/kernel/perf_callchain.c
+++ b/arch/arm/kernel/perf_callchain.c
@@ -75,7 +75,7 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 
 	tail = (struct frame_tail __user *)regs->ARM_fp - 1;
 
-	while ((entry->nr < PERF_MAX_STACK_DEPTH) &&
+	while ((entry->nr < sysctl_perf_event_max_stack) &&
 	       tail && !((unsigned long)tail & 0x3))
 		tail = user_backtrace(tail, entry);
 }
diff --git a/arch/arm64/kernel/perf_callchain.c b/arch/arm64/kernel/perf_callchain.c
index ff4665462a02..32c3c6e70119 100644
--- a/arch/arm64/kernel/perf_callchain.c
+++ b/arch/arm64/kernel/perf_callchain.c
@@ -122,7 +122,7 @@ void perf_callchain_user(struct perf_callchain_entry *entry,
 
 		tail = (struct frame_tail __user *)regs->regs[29];
 
-		while (entry->nr < PERF_MAX_STACK_DEPTH &&
+		while (entry->nr < sysctl_perf_event_max_stack &&
 		       tail && !((unsigned long)tail & 0xf))
 			tail = user_backtrace(tail, entry);
 	} else {
@@ -132,7 +132,7 @@ void perf_callchain_user(struct perf_callchain_entry *entry,
 
 		tail = (struct compat_frame_tail __user *)regs->compat_fp - 1;
 
-		while ((entry->nr < PERF_MAX_STACK_DEPTH) &&
+		while ((entry->nr < sysctl_perf_event_max_stack) &&
 			tail && !((unsigned long)tail & 0x3))
 			tail = compat_user_backtrace(tail, entry);
 #endif
diff --git a/arch/metag/kernel/perf_callchain.c b/arch/metag/kernel/perf_callchain.c
index 315633461a94..252abc12a5a3 100644
--- a/arch/metag/kernel/perf_callchain.c
+++ b/arch/metag/kernel/perf_callchain.c
@@ -65,7 +65,7 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 
 	--frame;
 
-	while ((entry->nr < PERF_MAX_STACK_DEPTH) && frame)
+	while ((entry->nr < sysctl_perf_event_max_stack) && frame)
 		frame = user_backtrace(frame, entry);
 }
 
diff --git a/arch/mips/kernel/perf_event.c b/arch/mips/kernel/perf_event.c
index c1cf9c6c3f77..5021c546ad07 100644
--- a/arch/mips/kernel/perf_event.c
+++ b/arch/mips/kernel/perf_event.c
@@ -35,7 +35,7 @@ static void save_raw_perf_callchain(struct perf_callchain_entry *entry,
 		addr = *sp++;
 		if (__kernel_text_address(addr)) {
 			perf_callchain_store(entry, addr);
-			if (entry->nr >= PERF_MAX_STACK_DEPTH)
+			if (entry->nr >= sysctl_perf_event_max_stack)
 				break;
 		}
 	}
@@ -59,7 +59,7 @@ void perf_callchain_kernel(struct perf_callchain_entry *entry,
 	}
 	do {
 		perf_callchain_store(entry, pc);
-		if (entry->nr >= PERF_MAX_STACK_DEPTH)
+		if (entry->nr >= sysctl_perf_event_max_stack)
 			break;
 		pc = unwind_stack(current, &sp, pc, &ra);
 	} while (pc);
diff --git a/arch/powerpc/perf/callchain.c b/arch/powerpc/perf/callchain.c
index e04a6752b399..22d9015c1acc 100644
--- a/arch/powerpc/perf/callchain.c
+++ b/arch/powerpc/perf/callchain.c
@@ -247,7 +247,7 @@ static void perf_callchain_user_64(struct perf_callchain_entry *entry,
 	sp = regs->gpr[1];
 	perf_callchain_store(entry, next_ip);
 
-	while (entry->nr < PERF_MAX_STACK_DEPTH) {
+	while (entry->nr < sysctl_perf_event_max_stack) {
 		fp = (unsigned long __user *) sp;
 		if (!valid_user_sp(sp, 1) || read_user_stack_64(fp, &next_sp))
 			return;
@@ -453,7 +453,7 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry,
 	sp = regs->gpr[1];
 	perf_callchain_store(entry, next_ip);
 
-	while (entry->nr < PERF_MAX_STACK_DEPTH) {
+	while (entry->nr < sysctl_perf_event_max_stack) {
 		fp = (unsigned int __user *) (unsigned long) sp;
 		if (!valid_user_sp(sp, 0) || read_user_stack_32(fp, &next_sp))
 			return;
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 6596f66ce112..a4b8b5aed21c 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -1756,7 +1756,7 @@ void perf_callchain_kernel(struct perf_callchain_entry *entry,
 			}
 		}
 #endif
-	} while (entry->nr < PERF_MAX_STACK_DEPTH);
+	} while (entry->nr < sysctl_perf_event_max_stack);
 }
 
 static inline int
@@ -1790,7 +1790,7 @@ static void perf_callchain_user_64(struct perf_callchain_entry *entry,
 		pc = sf.callers_pc;
 		ufp = (unsigned long)sf.fp + STACK_BIAS;
 		perf_callchain_store(entry, pc);
-	} while (entry->nr < PERF_MAX_STACK_DEPTH);
+	} while (entry->nr < sysctl_perf_event_max_stack);
 }
 
 static void perf_callchain_user_32(struct perf_callchain_entry *entry,
@@ -1822,7 +1822,7 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry,
 			ufp = (unsigned long)sf.fp;
 		}
 		perf_callchain_store(entry, pc);
-	} while (entry->nr < PERF_MAX_STACK_DEPTH);
+	} while (entry->nr < sysctl_perf_event_max_stack);
 }
 
 void
diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index 041e442a3e28..41d93d0e972b 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -2277,7 +2277,7 @@ perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry)
 
 	fp = compat_ptr(ss_base + regs->bp);
 	pagefault_disable();
-	while (entry->nr < PERF_MAX_STACK_DEPTH) {
+	while (entry->nr < sysctl_perf_event_max_stack) {
 		unsigned long bytes;
 		frame.next_frame     = 0;
 		frame.return_address = 0;
@@ -2337,7 +2337,7 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 		return;
 
 	pagefault_disable();
-	while (entry->nr < PERF_MAX_STACK_DEPTH) {
+	while (entry->nr < sysctl_perf_event_max_stack) {
 		unsigned long bytes;
 		frame.next_frame	     = NULL;
 		frame.return_address = 0;
diff --git a/arch/xtensa/kernel/perf_event.c b/arch/xtensa/kernel/perf_event.c
index 54f01188c29c..a6b00b3af429 100644
--- a/arch/xtensa/kernel/perf_event.c
+++ b/arch/xtensa/kernel/perf_event.c
@@ -332,14 +332,14 @@ static int callchain_trace(struct stackframe *frame, void *data)
 void perf_callchain_kernel(struct perf_callchain_entry *entry,
 			   struct pt_regs *regs)
 {
-	xtensa_backtrace_kernel(regs, PERF_MAX_STACK_DEPTH,
+	xtensa_backtrace_kernel(regs, sysctl_perf_event_max_stack,
 				callchain_trace, NULL, entry);
 }
 
 void perf_callchain_user(struct perf_callchain_entry *entry,
 			 struct pt_regs *regs)
 {
-	xtensa_backtrace_user(regs, PERF_MAX_STACK_DEPTH,
+	xtensa_backtrace_user(regs, sysctl_perf_event_max_stack,
 			      callchain_trace, entry);
 }
 
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 85749ae8cb5f..a090700cccca 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -58,7 +58,7 @@ struct perf_guest_info_callbacks {
 
 struct perf_callchain_entry {
 	__u64				nr;
-	__u64				ip[PERF_MAX_STACK_DEPTH];
+	__u64				ip[0]; /* /proc/sys/kernel/perf_event_max_stack */
 };
 
 struct perf_raw_record {
@@ -993,9 +993,11 @@ get_perf_callchain(struct pt_regs *regs, u32 init_nr, bool kernel, bool user,
 extern int get_callchain_buffers(void);
 extern void put_callchain_buffers(void);
 
+extern int sysctl_perf_event_max_stack;
+
 static inline int perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
 {
-	if (entry->nr < PERF_MAX_STACK_DEPTH) {
+	if (entry->nr < sysctl_perf_event_max_stack) {
 		entry->ip[entry->nr++] = ip;
 		return 0;
 	} else {
@@ -1017,6 +1019,8 @@ extern int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write,
 		void __user *buffer, size_t *lenp,
 		loff_t *ppos);
 
+int perf_event_max_stack_handler(struct ctl_table *table, int write,
+				 void __user *buffer, size_t *lenp, loff_t *ppos);
 
 static inline bool perf_paranoid_tracepoint_raw(void)
 {
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 499d9e933f8e..f5a19548be12 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -66,7 +66,7 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
 	/* check sanity of attributes */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
 	    value_size < 8 || value_size % 8 ||
-	    value_size / 8 > PERF_MAX_STACK_DEPTH)
+	    value_size / 8 > sysctl_perf_event_max_stack)
 		return ERR_PTR(-EINVAL);
 
 	/* hash table size must be power of 2 */
@@ -124,8 +124,8 @@ static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5)
 	struct perf_callchain_entry *trace;
 	struct stack_map_bucket *bucket, *new_bucket, *old_bucket;
 	u32 max_depth = map->value_size / 8;
-	/* stack_map_alloc() checks that max_depth <= PERF_MAX_STACK_DEPTH */
-	u32 init_nr = PERF_MAX_STACK_DEPTH - max_depth;
+	/* stack_map_alloc() checks that max_depth <= sysctl_perf_event_max_stack */
+	u32 init_nr = sysctl_perf_event_max_stack - max_depth;
 	u32 skip = flags & BPF_F_SKIP_FIELD_MASK;
 	u32 hash, id, trace_nr, trace_len;
 	bool user = flags & BPF_F_USER_STACK;
@@ -143,7 +143,7 @@ static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5)
 		return -EFAULT;
 
 	/* get_perf_callchain() guarantees that trace->nr >= init_nr
-	 * and trace-nr <= PERF_MAX_STACK_DEPTH, so trace_nr <= max_depth
+	 * and trace-nr <= sysctl_perf_event_max_stack, so trace_nr <= max_depth
 	 */
 	trace_nr = trace->nr - init_nr;
 
diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c
index 343c22f5e867..b9325e7dcba1 100644
--- a/kernel/events/callchain.c
+++ b/kernel/events/callchain.c
@@ -18,6 +18,14 @@ struct callchain_cpus_entries {
 	struct perf_callchain_entry	*cpu_entries[0];
 };
 
+int sysctl_perf_event_max_stack __read_mostly = PERF_MAX_STACK_DEPTH;
+
+static inline size_t perf_callchain_entry__sizeof(void)
+{
+	return (sizeof(struct perf_callchain_entry) +
+		sizeof(__u64) * sysctl_perf_event_max_stack);
+}
+
 static DEFINE_PER_CPU(int, callchain_recursion[PERF_NR_CONTEXTS]);
 static atomic_t nr_callchain_events;
 static DEFINE_MUTEX(callchain_mutex);
@@ -73,7 +81,7 @@ static int alloc_callchain_buffers(void)
 	if (!entries)
 		return -ENOMEM;
 
-	size = sizeof(struct perf_callchain_entry) * PERF_NR_CONTEXTS;
+	size = perf_callchain_entry__sizeof() * PERF_NR_CONTEXTS;
 
 	for_each_possible_cpu(cpu) {
 		entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL,
@@ -147,7 +155,8 @@ static struct perf_callchain_entry *get_callchain_entry(int *rctx)
 
 	cpu = smp_processor_id();
 
-	return &entries->cpu_entries[cpu][*rctx];
+	return (((void *)entries->cpu_entries[cpu]) +
+		(*rctx * perf_callchain_entry__sizeof()));
 }
 
 static void
@@ -215,3 +224,25 @@ exit_put:
 
 	return entry;
 }
+
+int perf_event_max_stack_handler(struct ctl_table *table, int write,
+				 void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	int new_value = sysctl_perf_event_max_stack, ret;
+	struct ctl_table new_table = *table;
+
+	new_table.data = &new_value;
+	ret = proc_dointvec_minmax(&new_table, write, buffer, lenp, ppos);
+	if (ret || !write)
+		return ret;
+
+	mutex_lock(&callchain_mutex);
+	if (atomic_read(&nr_callchain_events))
+		ret = -EBUSY;
+	else
+		sysctl_perf_event_max_stack = new_value;
+
+	mutex_unlock(&callchain_mutex);
+
+	return ret;
+}
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 725587f10667..c8b318663525 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -130,6 +130,9 @@ static int one_thousand = 1000;
 #ifdef CONFIG_PRINTK
 static int ten_thousand = 10000;
 #endif
+#ifdef CONFIG_PERF_EVENTS
+static int six_hundred_forty_kb = 640 * 1024;
+#endif
 
 /* this is needed for the proc_doulongvec_minmax of vm_dirty_bytes */
 static unsigned long dirty_bytes_min = 2 * PAGE_SIZE;
@@ -1144,6 +1147,15 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &zero,
 		.extra2		= &one_hundred,
 	},
+	{
+		.procname	= "perf_event_max_stack",
+		.data		= NULL, /* filled in by handler */
+		.maxlen		= sizeof(sysctl_perf_event_max_stack),
+		.mode		= 0644,
+		.proc_handler	= perf_event_max_stack_handler,
+		.extra1		= &zero,
+		.extra2		= &six_hundred_forty_kb,
+	},
 #endif
 #ifdef CONFIG_KMEMCHECK
 	{
-- 
2.5.5

[PATCH 30/30] perf tools: Set the maximum allowed stack from /proc/sys/kernel/perf_event_max_stack

[Arnaldo Carvalho de Melo]

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

There is an upper limit to what tooling considers a valid callchain,
and it was tied to the hardcoded value in the kernel,
PERF_MAX_STACK_DEPTH (127), now that this can be tuned via a sysctl,
make it read it and use that as the upper limit, falling back to
PERF_MAX_STACK_DEPTH for kernels where this sysctl isn't present.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-yjqsd30nnkogvj5oyx9ghir9@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/Documentation/perf-report.txt             | 2 +-
 tools/perf/Documentation/perf-script.txt             | 2 +-
 tools/perf/Documentation/perf-top.txt                | 2 +-
 tools/perf/Documentation/perf-trace.txt              | 2 +-
 tools/perf/builtin-report.c                          | 4 ++--
 tools/perf/builtin-script.c                          | 4 +++-
 tools/perf/builtin-top.c                             | 4 ++--
 tools/perf/builtin-trace.c                           | 4 ++--
 tools/perf/perf.c                                    | 5 +++++
 tools/perf/tests/hists_cumulate.c                    | 2 +-
 tools/perf/tests/hists_filter.c                      | 2 +-
 tools/perf/tests/hists_output.c                      | 2 +-
 tools/perf/util/machine.c                            | 6 +++---
 tools/perf/util/scripting-engines/trace-event-perl.c | 2 +-
 tools/perf/util/util.c                               | 2 ++
 tools/perf/util/util.h                               | 1 +
 16 files changed, 28 insertions(+), 18 deletions(-)

diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 496d42cdf02b..ebaf849e30ef 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -248,7 +248,7 @@ OPTIONS
 	Note that when using the --itrace option the synthesized callchain size
 	will override this value if the synthesized callchain size is bigger.
 
-	Default: 127
+	Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise.
 
 -G::
 --inverted::
diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
index 4fc44c75263f..a856a1095893 100644
--- a/tools/perf/Documentation/perf-script.txt
+++ b/tools/perf/Documentation/perf-script.txt
@@ -267,7 +267,7 @@ include::itrace.txt[]
         Note that when using the --itrace option the synthesized callchain size
         will override this value if the synthesized callchain size is bigger.
 
-        Default: 127
+        Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise.
 
 --ns::
 	Use 9 decimal places when displaying time (i.e. show the nanoseconds)
diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
index 19f046f027cd..91d638df3a6b 100644
--- a/tools/perf/Documentation/perf-top.txt
+++ b/tools/perf/Documentation/perf-top.txt
@@ -177,7 +177,7 @@ Default is to monitor all CPUS.
 	between information loss and faster processing especially for
 	workloads that can have a very long callchain stack.
 
-	Default: 127
+	Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise.
 
 --ignore-callees=<regex>::
         Ignore callees of the function(s) matching the given regex.
diff --git a/tools/perf/Documentation/perf-trace.txt b/tools/perf/Documentation/perf-trace.txt
index c075c002eaa4..6afe20121bc0 100644
--- a/tools/perf/Documentation/perf-trace.txt
+++ b/tools/perf/Documentation/perf-trace.txt
@@ -143,7 +143,7 @@ the thread executes on the designated CPUs. Default is to monitor all CPUs.
         Implies '--call-graph dwarf' when --call-graph not present on the
         command line, on systems where DWARF unwinding was built in.
 
-        Default: 127
+        Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise.
 
 --min-stack::
         Set the stack depth limit when parsing the callchain, anything
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 1d5be0bd426f..8d9b88af901d 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -691,7 +691,7 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 			.ordered_events	 = true,
 			.ordering_requires_timestamps = true,
 		},
-		.max_stack		 = PERF_MAX_STACK_DEPTH,
+		.max_stack		 = sysctl_perf_event_max_stack,
 		.pretty_printing_style	 = "normal",
 		.socket_filter		 = -1,
 	};
@@ -744,7 +744,7 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 	OPT_INTEGER(0, "max-stack", &report.max_stack,
 		    "Set the maximum stack depth when parsing the callchain, "
 		    "anything beyond the specified depth will be ignored. "
-		    "Default: " __stringify(PERF_MAX_STACK_DEPTH)),
+		    "Default: kernel.perf_event_max_stack or " __stringify(PERF_MAX_STACK_DEPTH)),
 	OPT_BOOLEAN('G', "inverted", &report.inverted_callchain,
 		    "alias for inverted call graph"),
 	OPT_CALLBACK(0, "ignore-callees", NULL, "regex",
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index f43b0c6f88f4..efca81679bb3 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -2031,7 +2031,7 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 	OPT_UINTEGER(0, "max-stack", &scripting_max_stack,
 		     "Set the maximum stack depth when parsing the callchain, "
 		     "anything beyond the specified depth will be ignored. "
-		     "Default: " __stringify(PERF_MAX_STACK_DEPTH)),
+		     "Default: kernel.perf_event_max_stack or " __stringify(PERF_MAX_STACK_DEPTH)),
 	OPT_BOOLEAN('I', "show-info", &show_full_info,
 		    "display extended information from perf.data file"),
 	OPT_BOOLEAN('\0', "show-kernel-path", &symbol_conf.show_kernel_path,
@@ -2067,6 +2067,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 		NULL
 	};
 
+	scripting_max_stack = sysctl_perf_event_max_stack;
+
 	setup_scripting();
 
 	argc = parse_options_subcommand(argc, argv, options, script_subcommands, script_usage,
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index c130a11d3a0d..da18517b1d40 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -1103,7 +1103,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 			},
 			.proc_map_timeout    = 500,
 		},
-		.max_stack	     = PERF_MAX_STACK_DEPTH,
+		.max_stack	     = sysctl_perf_event_max_stack,
 		.sym_pcnt_filter     = 5,
 	};
 	struct record_opts *opts = &top.record_opts;
@@ -1171,7 +1171,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 		    "Accumulate callchains of children and show total overhead as well"),
 	OPT_INTEGER(0, "max-stack", &top.max_stack,
 		    "Set the maximum stack depth when parsing the callchain. "
-		    "Default: " __stringify(PERF_MAX_STACK_DEPTH)),
+		    "Default: kernel.perf_event_max_stack or " __stringify(PERF_MAX_STACK_DEPTH)),
 	OPT_CALLBACK(0, "ignore-callees", NULL, "regex",
 		   "ignore callees of these functions in call graphs",
 		   report_parse_ignore_callees_opt),
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index 48b00f042599..f4f3389c92c7 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -3106,7 +3106,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 	OPT_UINTEGER(0, "max-stack", &trace.max_stack,
 		     "Set the maximum stack depth when parsing the callchain, "
 		     "anything beyond the specified depth will be ignored. "
-		     "Default: " __stringify(PERF_MAX_STACK_DEPTH)),
+		     "Default: kernel.perf_event_max_stack or " __stringify(PERF_MAX_STACK_DEPTH)),
 	OPT_UINTEGER(0, "proc-map-timeout", &trace.opts.proc_map_timeout,
 			"per thread proc mmap processing timeout in ms"),
 	OPT_END()
@@ -3150,7 +3150,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 		mmap_pages_user_set = false;
 
 	if (trace.max_stack == UINT_MAX) {
-		trace.max_stack = PERF_MAX_STACK_DEPTH;
+		trace.max_stack = sysctl_perf_event_max_stack;
 		max_stack_user_set = false;
 	}
 
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index 7b2df2b46525..83ffe7cd7330 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -17,6 +17,7 @@
 #include <subcmd/parse-options.h>
 #include "util/bpf-loader.h"
 #include "util/debug.h"
+#include <api/fs/fs.h>
 #include <api/fs/tracing_path.h>
 #include <pthread.h>
 #include <stdlib.h>
@@ -533,6 +534,7 @@ int main(int argc, const char **argv)
 {
 	const char *cmd;
 	char sbuf[STRERR_BUFSIZE];
+	int value;
 
 	/* libsubcmd init */
 	exec_cmd_init("perf", PREFIX, PERF_EXEC_PATH, EXEC_PATH_ENVIRONMENT);
@@ -542,6 +544,9 @@ int main(int argc, const char **argv)
 	page_size = sysconf(_SC_PAGE_SIZE);
 	cacheline_size = sysconf(_SC_LEVEL1_DCACHE_LINESIZE);
 
+	if (sysctl__read_int("kernel/perf_event_max_stack", &value) == 0)
+		sysctl_perf_event_max_stack = value;
+
 	cmd = extract_argv0_path(argv[0]);
 	if (!cmd)
 		cmd = "perf-help";
diff --git a/tools/perf/tests/hists_cumulate.c b/tools/perf/tests/hists_cumulate.c
index ed5aa9eaeb6c..4a2bbff9b1ee 100644
--- a/tools/perf/tests/hists_cumulate.c
+++ b/tools/perf/tests/hists_cumulate.c
@@ -101,7 +101,7 @@ static int add_hist_entries(struct hists *hists, struct machine *machine)
 		if (machine__resolve(machine, &al, &sample) < 0)
 			goto out;
 
-		if (hist_entry_iter__add(&iter, &al, PERF_MAX_STACK_DEPTH,
+		if (hist_entry_iter__add(&iter, &al, sysctl_perf_event_max_stack,
 					 NULL) < 0) {
 			addr_location__put(&al);
 			goto out;
diff --git a/tools/perf/tests/hists_filter.c b/tools/perf/tests/hists_filter.c
index b825d24f8186..e846f8c42013 100644
--- a/tools/perf/tests/hists_filter.c
+++ b/tools/perf/tests/hists_filter.c
@@ -81,7 +81,7 @@ static int add_hist_entries(struct perf_evlist *evlist,
 
 			al.socket = fake_samples[i].socket;
 			if (hist_entry_iter__add(&iter, &al,
-						 PERF_MAX_STACK_DEPTH, NULL) < 0) {
+						 sysctl_perf_event_max_stack, NULL) < 0) {
 				addr_location__put(&al);
 				goto out;
 			}
diff --git a/tools/perf/tests/hists_output.c b/tools/perf/tests/hists_output.c
index d3556fbe8c5c..7cd8738e842f 100644
--- a/tools/perf/tests/hists_output.c
+++ b/tools/perf/tests/hists_output.c
@@ -67,7 +67,7 @@ static int add_hist_entries(struct hists *hists, struct machine *machine)
 		if (machine__resolve(machine, &al, &sample) < 0)
 			goto out;
 
-		if (hist_entry_iter__add(&iter, &al, PERF_MAX_STACK_DEPTH,
+		if (hist_entry_iter__add(&iter, &al, sysctl_perf_event_max_stack,
 					 NULL) < 0) {
 			addr_location__put(&al);
 			goto out;
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 656c1d7ee7d4..2cb95bbf9ea6 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1764,7 +1764,7 @@ static int resolve_lbr_callchain_sample(struct thread *thread,
 		 */
 		int mix_chain_nr = i + 1 + lbr_nr + 1;
 
-		if (mix_chain_nr > PERF_MAX_STACK_DEPTH + PERF_MAX_BRANCH_DEPTH) {
+		if (mix_chain_nr > (int)sysctl_perf_event_max_stack + PERF_MAX_BRANCH_DEPTH) {
 			pr_warning("corrupted callchain. skipping...\n");
 			return 0;
 		}
@@ -1825,7 +1825,7 @@ static int thread__resolve_callchain_sample(struct thread *thread,
 	 * Based on DWARF debug information, some architectures skip
 	 * a callchain entry saved by the kernel.
 	 */
-	if (chain->nr < PERF_MAX_STACK_DEPTH)
+	if (chain->nr < sysctl_perf_event_max_stack)
 		skip_idx = arch_skip_callchain_idx(thread, chain);
 
 	/*
@@ -1886,7 +1886,7 @@ static int thread__resolve_callchain_sample(struct thread *thread,
 	}
 
 check_calls:
-	if (chain->nr > PERF_MAX_STACK_DEPTH && (int)chain->nr > max_stack) {
+	if (chain->nr > sysctl_perf_event_max_stack && (int)chain->nr > max_stack) {
 		pr_warning("corrupted callchain. skipping...\n");
 		return 0;
 	}
diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c
index ae1cebc307c5..62c7f6988e0e 100644
--- a/tools/perf/util/scripting-engines/trace-event-perl.c
+++ b/tools/perf/util/scripting-engines/trace-event-perl.c
@@ -265,7 +265,7 @@ static SV *perl_process_callchain(struct perf_sample *sample,
 
 	if (thread__resolve_callchain(al->thread, &callchain_cursor, evsel,
 				      sample, NULL, NULL,
-				      PERF_MAX_STACK_DEPTH) != 0) {
+				      sysctl_perf_event_max_stack) != 0) {
 		pr_err("Failed to resolve callchain. Skipping\n");
 		goto exit;
 	}
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index 9473d46c00bb..619ba2061b62 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -33,6 +33,8 @@ struct callchain_param	callchain_param = {
 unsigned int page_size;
 int cacheline_size;
 
+unsigned int sysctl_perf_event_max_stack = PERF_MAX_STACK_DEPTH;
+
 bool test_attr__enabled;
 
 bool perf_host  = true;
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 26a924651e7b..88f607af1f47 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -267,6 +267,7 @@ void sighandler_dump_stack(int sig);
 
 extern unsigned int page_size;
 extern int cacheline_size;
+extern unsigned int sysctl_perf_event_max_stack;
 
 struct parse_tag {
 	char tag;
-- 
2.5.5

Re: [GIT PULL 00/30] perf/core improvements and fixes

[Ingo Molnar]

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

> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> Hi Ingo,
> 
> 	Please consider pulling,
> 
> - Arnaldo
> 
> The following changes since commit 67d61296ffcc850bffdd4466430cb91e5328f39a:
> 
>   Merge tag 'perf-core-for-mingo-20160419' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux (2016-04-23 14:50:39 +0200)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-20160427
> 
> for you to fetch changes up to 4cb93446c587d56e2a54f4f83113daba2c0b6dee:
> 
>   perf tools: Set the maximum allowed stack from /proc/sys/kernel/perf_event_max_stack (2016-04-27 10:29:07 -0300)
> 
> ----------------------------------------------------------------
> perf/core improvements and fixes:
> 
> User visible:
> 
> - perf trace --pf maj/min/all works with --call-graph: (Arnaldo Carvalho de Melo)
> 
>   Tracing write syscalls and major page faults with callchains while starting
>   firefox, limiting the stack to 5 frames:
> 
>  # perf trace -e write --pf maj --max-stack 5 firefox
>    589.549 ( 0.014 ms): firefox/15377 write(fd: 4, buf: 0x7fff80acc898, count: 151) = 151
>                                        [0xfaed] (/usr/lib64/libpthread-2.22.so)
>                                        fire_glxtest_process+0x5c (/usr/lib64/firefox/libxul.so)
>                                        InstallGdkErrorHandler+0x41 (/usr/lib64/firefox/libxul.so)
>                                        XREMain::XRE_mainInit+0x12c (/usr/lib64/firefox/libxul.so)
>                                        XREMain::XRE_main+0x1e4 (/usr/lib64/firefox/libxul.so)
>    760.704 ( 0.000 ms): firefox/15332 majfault [gtk_tree_view_accessible_get_type+0x0] => /usr/lib64/libgtk-3.so.0.1800.9@0xa0850 (x.)
>                                        gtk_tree_view_accessible_get_type+0x0 (/usr/lib64/libgtk-3.so.0.1800.9)
>                                        gtk_tree_view_class_intern_init+0x1a54 (/usr/lib64/libgtk-3.so.0.1800.9)
>                                        g_type_class_ref+0x6dd (/usr/lib64/libgobject-2.0.so.0.4600.2)
>                                        [0x115378] (/usr/lib64/libgnutls.so.30.6.3)
> 
>   This automagically selects "--call-graph dwarf", use "--call-graph fp" on systems
>   where -fno-omit-frame-pointer was used to built the components of interest, to
>   incur in less overhead, or tune "--call-graph dwarf" appropriately, see 'perf record --help'.
> 
> - Allow /proc/sys/kernel/perf_event_max_stack, that defaults to the old hard coded value
>   of PERF_MAX_STACK_DEPTH (127), useful for huge callstacks for things like Groovy, Ruby, etc,
>   and also to reduce overhead by limiting it to a smaller value, upcoming work will allow
>   this to be done per-event (Arnaldo Carvalho de Melo)
> 
> - Make 'perf trace --min-stack' be honoured by --pf and --event (Arnaldo Carvalho de Melo)
> 
> - Make 'perf evlist -v' decode perf_event_attr->branch_sample_type (Arnaldo Carvalho de Melo)
> 
>    # perf record --call lbr usleep 1
>    # perf evlist -v
>    cycles:ppp: ... sample_type: IP|TID|TIME|CALLCHAIN|PERIOD|BRANCH_STACK, ...
>             branch_sample_type: USER|CALL_STACK|NO_FLAGS|NO_CYCLES
>    #
> 
> - Clear dummy entry accumulated period, fixing such 'perf top/report' output
>   as: (Kan Liang)
> 
>     4769.98%  0.01%  0.00%  0.01%  tchain_edit  [kernel] [k] update_fast_timekeeper
> 
> - System calls with pid_t arguments gets them augmented with the COMM event
>   more thoroughly:
> 
>   # trace -e perf_event_open perf stat -e cycles -p 15608
>    6.876 ( 0.014 ms): perf_event_open(attr_uptr: 0x2ae20d8, pid: 15608 (hexchat), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 3
>    6.882 ( 0.005 ms): perf_event_open(attr_uptr: 0x2ae20d8, pid: 15639 (gmain), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 4
>    6.889 ( 0.005 ms): perf_event_open(attr_uptr: 0x2ae20d8, pid: 15640 (gdbus), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 5
>                                                             ^^^^^^^^^^^^^^^^^^
>    ^C
> 
> - Fix offline module name mismatch issue in 'perf probe' (Ravi Bangoria)
> 
> - Fix module probe issue if no dwarf support in (Ravi Bangoria)
> 
> Assorted fixes:
> 
> - Fix off-by-one in write_buildid() (Andrey Ryabinin)
> 
> - Fix segfault when printing callchains in 'perf script' (Chris Phlipot)
> 
> - Replace assignment with comparison on assert check in 'perf test' entry (Colin Ian King)
> 
> - Fix off-by-one comparison in intel-pt code (Colin Ian King)
> 
> - Close target file on error path in 'perf probe' (Masami Hiramatsu)
> 
> - Set default kprobe group name if not given in 'perf probe' (Masami Hiramatsu)
> 
> - Avoid partial perf_event_header reads (Wang Nan)
> 
> Infrastructure:
> 
> - Update x86's syscall_64.tbl copy, adding preadv2 & pwritev2 (Arnaldo Carvalho de Melo)
> 
> - Make the x86 clean quiet wrt syscall table removal (Jiri Olsa)
> 
> Cleanups:
> 
> - Simplify wrapper for LOCK_PI in 'perf bench futex' (Davidlohr Bueso)
> 
> - Remove duplicate const qualifier (Eric Engestrom)
> 
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> ----------------------------------------------------------------
> Andrey Ryabinin (1):
>       perf buildid: Fix off-by-one in write_buildid()
> 
> Arnaldo Carvalho de Melo (14):
>       perf trace: Extract evsel contructor from perf_evlist__add_pgfault
>       perf trace: Make --pf maj/min/all use callchains too
>       perf trace: Make --event honour --min-stack too
>       perf trace: Make --pf honour --min-stack too
>       perf evlist: Decode perf_event_attr->branch_sample_type
>       perf trace: Move perf_flags beautifier to tools/perf/trace/beauty/
>       perf trace: Do not beautify the 'pid' parameter as a simple integer
>       tools lib api fs: Add helper to read string from procfs file
>       perf thread: Introduce method to set comm from /proc/pid/self
>       perf trace: Read thread's COMM from /proc when not set
>       perf tools: Update x86's syscall_64.tbl, adding preadv2 & pwritev2
>       perf bench: Remove one more die() call
>       perf core: Allow setting up max frame stack depth via sysctl
>       perf tools: Set the maximum allowed stack from /proc/sys/kernel/perf_event_max_stack
> 
> Chris Phlipot (1):
>       perf script: Fix segfault when printing callchains
> 
> Colin Ian King (2):
>       perf tests: Replace assignment with comparison on assert check
>       perf intel-pt: Fix off-by-one comparison on maximum code
> 
> Davidlohr Bueso (1):
>       perf bench futex: Simplify wrapper for LOCK_PI
> 
> Eric Engestrom (1):
>       perf tools: Remove duplicate const qualifier
> 
> Jiri Olsa (2):
>       perf tools: Make the x86 clean quiet
>       tools build: Fix perf_clean target
> 
> Kan Liang (1):
>       perf hists: Clear dummy entry accumulated period
> 
> Masami Hiramatsu (4):
>       perf probe: Close target file on error path
>       perf tools: Add lsdir() helper to read a directory
>       perf probe: Let probe_file__add_event return 0 if succeeded
>       perf probe: Set default kprobe group name if it is not given
> 
> Ravi Bangoria (2):
>       perf probe: Fix offline module name missmatch issue
>       perf probe: Fix module probe issue if no dwarf support
> 
> Wang Nan (1):
>       perf evlist: Enforce ring buffer reading
> 
>  Documentation/sysctl/kernel.txt                    |  14 ++
>  arch/arm/kernel/perf_callchain.c                   |   2 +-
>  arch/arm64/kernel/perf_callchain.c                 |   4 +-
>  arch/metag/kernel/perf_callchain.c                 |   2 +-
>  arch/mips/kernel/perf_event.c                      |   4 +-
>  arch/powerpc/perf/callchain.c                      |   4 +-
>  arch/sparc/kernel/perf_event.c                     |   6 +-
>  arch/x86/events/core.c                             |   4 +-
>  arch/xtensa/kernel/perf_event.c                    |   4 +-
>  include/linux/perf_event.h                         |   8 +-
>  kernel/bpf/stackmap.c                              |   8 +-
>  kernel/events/callchain.c                          |  35 ++++-
>  kernel/sysctl.c                                    |  12 ++
>  tools/Makefile                                     |   3 +-
>  tools/lib/api/fs/fs.c                              |  13 ++
>  tools/lib/api/fs/fs.h                              |   2 +
>  tools/perf/Documentation/perf-report.txt           |   2 +-
>  tools/perf/Documentation/perf-script.txt           |   2 +-
>  tools/perf/Documentation/perf-top.txt              |   2 +-
>  tools/perf/Documentation/perf-trace.txt            |   2 +-
>  tools/perf/arch/x86/Makefile                       |   2 +-
>  tools/perf/arch/x86/entry/syscalls/syscall_64.tbl  |   2 +
>  tools/perf/bench/futex-lock-pi.c                   |   2 +-
>  tools/perf/bench/futex.h                           |   6 +-
>  tools/perf/bench/mem-functions.c                   |  22 ++-
>  tools/perf/builtin-report.c                        |   4 +-
>  tools/perf/builtin-script.c                        |  16 +-
>  tools/perf/builtin-top.c                           |   4 +-
>  tools/perf/builtin-trace.c                         | 174 +++++++++++----------
>  tools/perf/perf.c                                  |   5 +
>  tools/perf/tests/event_update.c                    |   2 +-
>  tools/perf/tests/hists_cumulate.c                  |   2 +-
>  tools/perf/tests/hists_filter.c                    |   2 +-
>  tools/perf/tests/hists_output.c                    |   2 +-
>  tools/perf/trace/beauty/perf_event_open.c          |  43 +++++
>  tools/perf/trace/beauty/pid.c                      |   5 +-
>  tools/perf/util/build-id.c                         |   6 +-
>  tools/perf/util/evlist.c                           |  12 +-
>  tools/perf/util/evsel.c                            |  18 ++-
>  tools/perf/util/hist.c                             |   2 +
>  .../perf/util/intel-pt-decoder/intel-pt-decoder.c  |   2 +-
>  tools/perf/util/machine.c                          |   6 +-
>  tools/perf/util/probe-event.c                      | 114 +++++++++++---
>  tools/perf/util/probe-file.c                       |   3 +-
>  .../perf/util/scripting-engines/trace-event-perl.c |   2 +-
>  tools/perf/util/thread.c                           |  21 ++-
>  tools/perf/util/thread.h                           |   2 +
>  tools/perf/util/util.c                             |  36 +++++
>  tools/perf/util/util.h                             |   4 +
>  49 files changed, 475 insertions(+), 179 deletions(-)
>  create mode 100644 tools/perf/trace/beauty/perf_event_open.c

Pulled, thanks a lot Arnaldo!

	Ingo