[GIT PULL 0/3] perf/urgent 'perf probe' fixes

[GIT PULL 0/3] perf/urgent 'perf probe' fixes

[Arnaldo Carvalho de Melo]

Hi Ingo,

	Please consider pulling,

- Arnaldo

Test results at the end of this message, as usual.

The following changes since commit 18e7a45af91acdde99d3aa1372cc40e1f8142f7b:

  perf/x86: Reject non sampling events with precise_ip (2017-01-14 11:06:50 +0100)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-urgent-for-mingo-4.10-20170117

for you to fetch changes up to 613f050d68a8ed3c0b18b9568698908ef7bbc1f7:

  perf probe: Fix to probe on gcc generated functions in modules (2017-01-16 15:43:04 -0300)

----------------------------------------------------------------
perf/urgent 'perf probe' fixes:

Fixes:

- Show correct locations for 'perf probe' on modules (Masami Hiramatsu)

- Correctly handle 'perf probe's on gcc generated functions in modules (Masami Hiramatsu)

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

----------------------------------------------------------------
Masami Hiramatsu (3):
      perf probe: Fix to show correct locations for events on modules
      perf probe: Add error checks to offline probe post-processing
      perf probe: Fix to probe on gcc generated functions in modules

 tools/perf/util/probe-event.c  | 95 +++++++++++++++++++++++++++---------------
 tools/perf/util/probe-finder.c | 15 ++++---
 tools/perf/util/probe-finder.h |  3 ++
 3 files changed, 72 insertions(+), 41 deletions(-)

Test results:

The first ones are container (docker) based builds of tools/perf with and
without libelf support, objtool where it is supported and samples/bpf/, ditto.

Several are cross builds, the ones with -x-ARCH, and the android one, and those
may not have all the features built, due to lack of multi-arch devel packages,
available and being used so far on just a few, like
debian:experimental-x-{arm64,mipsel}.

The 'perf test' one will perform a variety of tests exercising
tools/perf/util/, tools/lib/{bpf,traceevent,etc}, as well as run perf commands
with a variety of command line event specifications to then intercept the
sys_perf_event syscall to check that the perf_event_attr fields are set up as
expected, among a variety of other unit tests.

Then there is the 'make -C tools/perf build-test' ones, that build tools/perf/
with a variety of feature sets, exercising the build with an incomplete set of
features as well as with a complete one. It is planned to have it run on each
of the containers mentioned above, using some container orchestration
infrastructure. Get in contact if interested in helping having this in place.

  # dm
   1  alpine:3.4: Ok
   2  android-ndk:r12b-arm: Ok
   3  archlinux:latest: Ok
   4  centos:5: Ok
   5  centos:6: Ok
   6  centos:7: Ok
   7  debian:7: Ok
   8  debian:8: Ok
   9  debian:experimental: Ok
  10  debian:experimental-x-arm64: Ok
  11  debian:experimental-x-mips: Ok
  12  debian:experimental-x-mips64: Ok
  13  debian:experimental-x-mipsel: Ok
  14  fedora:20: Ok
  15  fedora:21: Ok
  16  fedora:22: Ok
  17  fedora:23: Ok
  18  fedora:24: Ok
  19  fedora:24-x-ARC-uClibc: Ok
  20  fedora:25: Ok
  21  fedora:rawhide: Ok
  22  mageia:5: Ok
  23  opensuse:13.2: Ok
  24  opensuse:42.1: Ok
  25  opensuse:tumbleweed: Ok
  26  ubuntu:12.04.5: Ok
  27  ubuntu:14.04.4-x-linaro-arm64: Ok
  28  ubuntu:15.10: Ok
  29  ubuntu:16.04: Ok
  30  ubuntu:16.04-x-arm: Ok
  31  ubuntu:16.04-x-arm64: Ok
  32  ubuntu:16.04-x-powerpc: Ok
  33  ubuntu:16.04-x-powerpc64: Ok
  34  ubuntu:16.04-x-powerpc64el: Ok
  35  ubuntu:16.04-x-s390: Ok
  36  ubuntu:16.10: Ok
  #

  # uname -a
  Linux jouet 4.9.0+ #2 SMP Wed Dec 21 11:54:44 BRT 2016 x86_64 x86_64 x86_64 GNU/Linux
  # perf test
   1: vmlinux symtab matches kallsyms            : Ok
   2: Detect openat syscall event                : Ok
   3: Detect openat syscall event on all cpus    : Ok
   4: Read samples using the mmap interface      : Ok
   5: Parse event definition strings             : Ok
   6: PERF_RECORD_* events & perf_sample fields  : Ok
   7: Parse perf pmu format                      : Ok
   8: DSO data read                              : Ok
   9: DSO data cache                             : Ok
  10: DSO data reopen                            : Ok
  11: Roundtrip evsel->name                      : Ok
  12: Parse sched tracepoints fields             : Ok
  13: syscalls:sys_enter_openat event fields     : Ok
  14: Setup struct perf_event_attr               : Ok
  15: Match and link multiple hists              : Ok
  16: 'import perf' in python                    : Ok
  17: Breakpoint overflow signal handler         : Ok
  18: Breakpoint overflow sampling               : Ok
  19: Number of exit events of a simple workload : Ok
  20: Software clock events period values        : Ok
  21: Object code reading                        : Ok
  22: Sample parsing                             : Ok
  23: Use a dummy software event to keep tracking: Ok
  24: Parse with no sample_id_all bit set        : Ok
  25: Filter hist entries                        : Ok
  26: Lookup mmap thread                         : Ok
  27: Share thread mg                            : Ok
  28: Sort output of hist entries                : Ok
  29: Cumulate child hist entries                : Ok
  30: Track with sched_switch                    : Ok
  31: Filter fds with revents mask in a fdarray  : Ok
  32: Add fd to a fdarray, making it autogrow    : Ok
  33: kmod_path__parse                           : Ok
  34: Thread map                                 : Ok
  35: LLVM search and compile                    :
  35.1: Basic BPF llvm compile                    : Ok
  35.2: kbuild searching                          : Ok
  35.3: Compile source for BPF prologue generation: Ok
  35.4: Compile source for BPF relocation         : Ok
  36: Session topology                           : Ok
  37: BPF filter                                 :
  37.1: Basic BPF filtering                      : Ok
  37.2: BPF prologue generation                  : Ok
  37.3: BPF relocation checker                   : Ok
  38: Synthesize thread map                      : Ok
  39: Remove thread map                          : Ok
  40: Synthesize cpu map                         : Ok
  41: Synthesize stat config                     : Ok
  42: Synthesize stat                            : Ok
  43: Synthesize stat round                      : Ok
  44: Synthesize attr update                     : Ok
  45: Event times                                : Ok
  46: Read backward ring buffer                  : Ok
  47: Print cpu map                              : Ok
  48: Probe SDT events                           : Ok
  49: is_printable_array                         : Ok
  50: Print bitmap                               : Ok
  51: perf hooks                                 : Ok
  52: builtin clang support                      : Skip (not compiled in)
  53: x86 rdpmc                                  : Ok
  54: Convert perf time to TSC                   : Ok
  55: DWARF unwind                               : Ok
  56: x86 instruction decoder - new instructions : Ok
  57: Intel cqm nmi context read                 : Skip

  $ make -C tools/perf build-test
  make: Entering directory '/home/acme/git/linux/tools/perf'
  - tarpkg: ./tests/perf-targz-src-pkg .
               make_no_slang_O: make NO_SLANG=1g
           make_no_backtrace_O: make NO_BACKTRACE=1g
        make_with_babeltrace_O: make LIBBABELTRACE=1g
            make_no_libaudit_O: make NO_LIBAUDIT=1g
           make_no_libunwind_O: make NO_LIBUNWIND=1g
                make_install_O: make installg
             make_no_libnuma_O: make NO_LIBNUMA=1g
                  make_debug_O: make DEBUG=1g
            make_no_auxtrace_O: make NO_AUXTRACE=1g
                make_minimal_O: make NO_LIBPERL=1 NO_LIBPYTHON=1 NO_NEWT=1 NO_GTK2=1 NO_DEMANGLE=1 NO_LIBELF=1 NO_LIBUNWIND=1 NO_BACKTRACE=1 NO_LIBNUMA=1 NO_LIBAUDIT=1 NO_LIBBIONIC=1 NO_LIBDW_DWARF_UNWIND=1 NO_AUXTRACE=1 NO_LIBBPF=1 NO_LIBCRYPTO=1 NO_SDT=1 NO_JVMTI=1g
         make_install_prefix_O: make install prefix=/tmp/kravag
              make_clean_all_O: make clean allg
       make_util_pmu_bison_o_O: make util/pmu-bison.og
           make_no_libbionic_O: make NO_LIBBIONIC=1g
            make_no_demangle_O: make NO_DEMANGLE=1g
                    make_doc_O: make docg
                   make_pure_O: makeg
                   make_help_O: make helpg
             make_util_map_o_O: make util/map.og
         make_with_clangllvm_O: make LIBCLANGLLVM=1g
              make_no_libbpf_O: make NO_LIBBPF=1g
                   make_tags_O: make tagsg
                make_no_gtk2_O: make NO_GTK2=1g
                make_no_newt_O: make NO_NEWT=1g
             make_no_scripts_O: make NO_LIBPYTHON=1 NO_LIBPERL=1g
           make_no_libpython_O: make NO_LIBPYTHON=1g
              make_no_libelf_O: make NO_LIBELF=1g
   make_install_prefix_slash_O: make install prefix=/tmp/krava/g
  make_no_libdw_dwarf_unwind_O: make NO_LIBDW_DWARF_UNWIND=1g
            make_install_bin_O: make install-bing
                 make_static_O: make LDFLAGS=-staticg
             make_no_libperl_O: make NO_LIBPERL=1g
                  make_no_ui_O: make NO_NEWT=1 NO_SLANG=1 NO_GTK2=1g
                 make_perf_o_O: make perf.og
  OK
  make: Leaving directory '/home/acme/git/linux/tools/perf'
  $ 
  
  

[PATCH 1/3] perf probe: Fix to show correct locations for events on modules

[Arnaldo Carvalho de Melo]

From: Masami Hiramatsu <mhiramat@kernel.org>

Fix to show correct locations for events on modules by relocating given
address instead of retrying after failure.

This happens when the module text size is big enough, bigger than
sh_addr, because the original code retries with given address + sh_addr
if it failed to find CU DIE at the given address.

Any address smaller than sh_addr always fails and it retries with the
correct address, but addresses bigger than sh_addr will get a CU DIE
which is on the given address (not adjusted by sh_addr).

In my environment(x86-64), the sh_addr of ".text" section is 0x10030.
Since i915 is a huge kernel module, we can see this issue as below.

  $ grep "[Tt] .*\[i915\]" /proc/kallsyms | sort | head -n1
  ffffffffc0270000 t i915_switcheroo_can_switch	[i915]

ffffffffc0270000 + 0x10030 = ffffffffc0280030, so we'll check
symbols cross this boundary.

  $ grep "[Tt] .*\[i915\]" /proc/kallsyms | grep -B1 ^ffffffffc028\
  | head -n 2
  ffffffffc027ff80 t haswell_init_clock_gating	[i915]
  ffffffffc0280110 t valleyview_init_clock_gating	[i915]

So setup probes on both function and see what happen.

  $ sudo ./perf probe -m i915 -a haswell_init_clock_gating \
        -a valleyview_init_clock_gating
  Added new events:
    probe:haswell_init_clock_gating (on haswell_init_clock_gating in i915)
    probe:valleyview_init_clock_gating (on valleyview_init_clock_gating in i915)

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

  	perf record -e probe:valleyview_init_clock_gating -aR sleep 1

  $ sudo ./perf probe -l
    probe:haswell_init_clock_gating (on haswell_init_clock_gating@gpu/drm/i915/intel_pm.c in i915)
    probe:valleyview_init_clock_gating (on i915_vga_set_decode:4@gpu/drm/i915/i915_drv.c in i915)

As you can see, haswell_init_clock_gating is correctly shown,
but valleyview_init_clock_gating is not.

With this patch, both events are shown correctly.

  $ sudo ./perf probe -l
    probe:haswell_init_clock_gating (on haswell_init_clock_gating@gpu/drm/i915/intel_pm.c in i915)
    probe:valleyview_init_clock_gating (on valleyview_init_clock_gating@gpu/drm/i915/intel_pm.c in i915)

Committer notes:

In my case:

  # perf probe -m i915 -a haswell_init_clock_gating -a valleyview_init_clock_gating
  Added new events:
    probe:haswell_init_clock_gating (on haswell_init_clock_gating in i915)
    probe:valleyview_init_clock_gating (on valleyview_init_clock_gating in i915)

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

	  perf record -e probe:valleyview_init_clock_gating -aR sleep 1

  # perf probe -l
    probe:haswell_init_clock_gating (on i915_getparam+432@gpu/drm/i915/i915_drv.c in i915)
    probe:valleyview_init_clock_gating (on __i915_printk+240@gpu/drm/i915/i915_drv.c in i915)
  #

  # readelf -SW /lib/modules/4.9.0+/build/vmlinux | egrep -w '.text|Name'
   [Nr] Name   Type      Address          Off    Size   ES Flg Lk Inf Al
   [ 1] .text  PROGBITS  ffffffff81000000 200000 822fd3 00  AX  0   0 4096
  #

  So both are b0rked, now with the fix:

  # perf probe -m i915 -a haswell_init_clock_gating -a valleyview_init_clock_gating
  Added new events:
    probe:haswell_init_clock_gating (on haswell_init_clock_gating in i915)
    probe:valleyview_init_clock_gating (on valleyview_init_clock_gating in i915)

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

	perf record -e probe:valleyview_init_clock_gating -aR sleep 1

  # perf probe -l
    probe:haswell_init_clock_gating (on haswell_init_clock_gating@gpu/drm/i915/intel_pm.c in i915)
    probe:valleyview_init_clock_gating (on valleyview_init_clock_gating@gpu/drm/i915/intel_pm.c in i915)
  #

Both looks correct.

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/148411436777.9978.1440275861947194930.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-finder.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index df4debe564da..0278fe1a4cc6 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -1543,16 +1543,12 @@ int debuginfo__find_probe_point(struct debuginfo *dbg, unsigned long addr,
 	Dwarf_Addr _addr = 0, baseaddr = 0;
 	const char *fname = NULL, *func = NULL, *basefunc = NULL, *tmp;
 	int baseline = 0, lineno = 0, ret = 0;
-	bool reloc = false;
 
-retry:
+	/* We always need to relocate the address for aranges */
+	if (debuginfo__get_text_offset(dbg, &baseaddr) == 0)
+		addr += baseaddr;
 	/* Find cu die */
 	if (!dwarf_addrdie(dbg->dbg, (Dwarf_Addr)addr, &cudie)) {
-		if (!reloc && debuginfo__get_text_offset(dbg, &baseaddr) == 0) {
-			addr += baseaddr;
-			reloc = true;
-			goto retry;
-		}
 		pr_warning("Failed to find debug information for address %lx\n",
 			   addr);
 		ret = -EINVAL;
-- 
2.9.3

[PATCH 2/3] perf probe: Add error checks to offline probe post-processing

[Arnaldo Carvalho de Melo]

From: Masami Hiramatsu <mhiramat@kernel.org>

Add error check codes on post processing and improve it for offline
probe events as:

 - post processing fails if no matched symbol found in map(-ENOENT)
   or strdup() failed(-ENOMEM).

 - Even if the symbol name is the same, it updates symbol address
   and offset.

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/148411443738.9978.4617979132625405545.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-event.c | 50 ++++++++++++++++++++++++++++---------------
 1 file changed, 33 insertions(+), 17 deletions(-)

diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 4a57c8a60bd9..aa8a9227080a 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -610,6 +610,33 @@ static int find_perf_probe_point_from_dwarf(struct probe_trace_point *tp,
 	return ret ? : -ENOENT;
 }
 
+/* Adjust symbol name and address */
+static int post_process_probe_trace_point(struct probe_trace_point *tp,
+					   struct map *map, unsigned long offs)
+{
+	struct symbol *sym;
+	u64 addr = tp->address + tp->offset - offs;
+
+	sym = map__find_symbol(map, addr);
+	if (!sym)
+		return -ENOENT;
+
+	if (strcmp(sym->name, tp->symbol)) {
+		/* If we have no realname, use symbol for it */
+		if (!tp->realname)
+			tp->realname = tp->symbol;
+		else
+			free(tp->symbol);
+		tp->symbol = strdup(sym->name);
+		if (!tp->symbol)
+			return -ENOMEM;
+	}
+	tp->offset = addr - sym->start;
+	tp->address -= offs;
+
+	return 0;
+}
+
 /*
  * Rename DWARF symbols to ELF symbols -- gcc sometimes optimizes functions
  * and generate new symbols with suffixes such as .constprop.N or .isra.N
@@ -622,11 +649,9 @@ static int
 post_process_offline_probe_trace_events(struct probe_trace_event *tevs,
 					int ntevs, const char *pathname)
 {
-	struct symbol *sym;
 	struct map *map;
 	unsigned long stext = 0;
-	u64 addr;
-	int i;
+	int i, ret = 0;
 
 	/* Prepare a map for offline binary */
 	map = dso__new_map(pathname);
@@ -636,23 +661,14 @@ post_process_offline_probe_trace_events(struct probe_trace_event *tevs,
 	}
 
 	for (i = 0; i < ntevs; i++) {
-		addr = tevs[i].point.address + tevs[i].point.offset - stext;
-		sym = map__find_symbol(map, addr);
-		if (!sym)
-			continue;
-		if (!strcmp(sym->name, tevs[i].point.symbol))
-			continue;
-		/* If we have no realname, use symbol for it */
-		if (!tevs[i].point.realname)
-			tevs[i].point.realname = tevs[i].point.symbol;
-		else
-			free(tevs[i].point.symbol);
-		tevs[i].point.symbol = strdup(sym->name);
-		tevs[i].point.offset = addr - sym->start;
+		ret = post_process_probe_trace_point(&tevs[i].point,
+						     map, stext);
+		if (ret < 0)
+			break;
 	}
 	map__put(map);
 
-	return 0;
+	return ret;
 }
 
 static int add_exec_to_probe_trace_events(struct probe_trace_event *tevs,
-- 
2.9.3

[PATCH 3/3] perf probe: Fix to probe on gcc generated functions in modules

[Arnaldo Carvalho de Melo]

From: Masami Hiramatsu <mhiramat@kernel.org>

Fix to probe on gcc generated functions on modules. Since
probing on a module is based on its symbol name, it should
be adjusted on actual symbols.

E.g. without this fix, perf probe shows probe definition
on non-exist symbol as below.

  $ perf probe -m build-x86_64/net/netfilter/nf_nat.ko -F in_range*
  in_range.isra.12
  $ perf probe -m build-x86_64/net/netfilter/nf_nat.ko -D in_range
  p:probe/in_range nf_nat:in_range+0

With this fix, perf probe correctly shows a probe on
gcc-generated symbol.

  $ perf probe -m build-x86_64/net/netfilter/nf_nat.ko -D in_range
  p:probe/in_range nf_nat:in_range.isra.12+0

This also fixes same problem on online module as below.

  $ perf probe -m i915 -D assert_plane
  p:probe/assert_plane i915:assert_plane.constprop.134+0

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/148411450673.9978.14905987549651656075.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-event.c  | 45 +++++++++++++++++++++++++++---------------
 tools/perf/util/probe-finder.c |  7 +++++--
 tools/perf/util/probe-finder.h |  3 +++
 3 files changed, 37 insertions(+), 18 deletions(-)

diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index aa8a9227080a..6a6f44dd594b 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -698,18 +698,31 @@ static int add_exec_to_probe_trace_events(struct probe_trace_event *tevs,
 	return ret;
 }
 
-static int add_module_to_probe_trace_events(struct probe_trace_event *tevs,
-					    int ntevs, const char *module)
+static int
+post_process_module_probe_trace_events(struct probe_trace_event *tevs,
+				       int ntevs, const char *module,
+				       struct debuginfo *dinfo)
 {
+	Dwarf_Addr text_offs = 0;
 	int i, ret = 0;
 	char *mod_name = NULL;
+	struct map *map;
 
 	if (!module)
 		return 0;
 
-	mod_name = find_module_name(module);
+	map = get_target_map(module, false);
+	if (!map || debuginfo__get_text_offset(dinfo, &text_offs, true) < 0) {
+		pr_warning("Failed to get ELF symbols for %s\n", module);
+		return -EINVAL;
+	}
 
+	mod_name = find_module_name(module);
 	for (i = 0; i < ntevs; i++) {
+		ret = post_process_probe_trace_point(&tevs[i].point,
+						map, (unsigned long)text_offs);
+		if (ret < 0)
+			break;
 		tevs[i].point.module =
 			strdup(mod_name ? mod_name : module);
 		if (!tevs[i].point.module) {
@@ -719,6 +732,8 @@ static int add_module_to_probe_trace_events(struct probe_trace_event *tevs,
 	}
 
 	free(mod_name);
+	map__put(map);
+
 	return ret;
 }
 
@@ -776,7 +791,7 @@ arch__post_process_probe_trace_events(struct perf_probe_event *pev __maybe_unuse
 static int post_process_probe_trace_events(struct perf_probe_event *pev,
 					   struct probe_trace_event *tevs,
 					   int ntevs, const char *module,
-					   bool uprobe)
+					   bool uprobe, struct debuginfo *dinfo)
 {
 	int ret;
 
@@ -784,7 +799,8 @@ static int post_process_probe_trace_events(struct perf_probe_event *pev,
 		ret = add_exec_to_probe_trace_events(tevs, ntevs, module);
 	else if (module)
 		/* Currently ref_reloc_sym based probe is not for drivers */
-		ret = add_module_to_probe_trace_events(tevs, ntevs, module);
+		ret = post_process_module_probe_trace_events(tevs, ntevs,
+							     module, dinfo);
 	else
 		ret = post_process_kernel_probe_trace_events(tevs, ntevs);
 
@@ -828,30 +844,27 @@ static int try_to_find_probe_trace_events(struct perf_probe_event *pev,
 		}
 	}
 
-	debuginfo__delete(dinfo);
-
 	if (ntevs > 0) {	/* Succeeded to find trace events */
 		pr_debug("Found %d probe_trace_events.\n", ntevs);
 		ret = post_process_probe_trace_events(pev, *tevs, ntevs,
-						pev->target, pev->uprobes);
+					pev->target, pev->uprobes, dinfo);
 		if (ret < 0 || ret == ntevs) {
+			pr_debug("Post processing failed or all events are skipped. (%d)\n", ret);
 			clear_probe_trace_events(*tevs, ntevs);
 			zfree(tevs);
+			ntevs = 0;
 		}
-		if (ret != ntevs)
-			return ret < 0 ? ret : ntevs;
-		ntevs = 0;
-		/* Fall through */
 	}
 
+	debuginfo__delete(dinfo);
+
 	if (ntevs == 0)	{	/* No error but failed to find probe point. */
 		pr_warning("Probe point '%s' not found.\n",
 			   synthesize_perf_probe_point(&pev->point));
 		return -ENOENT;
-	}
-	/* Error path : ntevs < 0 */
-	pr_debug("An error occurred in debuginfo analysis (%d).\n", ntevs);
-	if (ntevs < 0) {
+	} else if (ntevs < 0) {
+		/* Error path : ntevs < 0 */
+		pr_debug("An error occurred in debuginfo analysis (%d).\n", ntevs);
 		if (ntevs == -EBADF)
 			pr_warning("Warning: No dwarf info found in the vmlinux - "
 				"please rebuild kernel with CONFIG_DEBUG_INFO=y.\n");
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 0278fe1a4cc6..0d9d6e0803b8 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -1501,7 +1501,8 @@ int debuginfo__find_available_vars_at(struct debuginfo *dbg,
 }
 
 /* For the kernel module, we need a special code to get a DIE */
-static int debuginfo__get_text_offset(struct debuginfo *dbg, Dwarf_Addr *offs)
+int debuginfo__get_text_offset(struct debuginfo *dbg, Dwarf_Addr *offs,
+				bool adjust_offset)
 {
 	int n, i;
 	Elf32_Word shndx;
@@ -1530,6 +1531,8 @@ static int debuginfo__get_text_offset(struct debuginfo *dbg, Dwarf_Addr *offs)
 			if (!shdr)
 				return -ENOENT;
 			*offs = shdr->sh_addr;
+			if (adjust_offset)
+				*offs -= shdr->sh_offset;
 		}
 	}
 	return 0;
@@ -1545,7 +1548,7 @@ int debuginfo__find_probe_point(struct debuginfo *dbg, unsigned long addr,
 	int baseline = 0, lineno = 0, ret = 0;
 
 	/* We always need to relocate the address for aranges */
-	if (debuginfo__get_text_offset(dbg, &baseaddr) == 0)
+	if (debuginfo__get_text_offset(dbg, &baseaddr, false) == 0)
 		addr += baseaddr;
 	/* Find cu die */
 	if (!dwarf_addrdie(dbg->dbg, (Dwarf_Addr)addr, &cudie)) {
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
index f1d8558f498e..2956c5198652 100644
--- a/tools/perf/util/probe-finder.h
+++ b/tools/perf/util/probe-finder.h
@@ -46,6 +46,9 @@ int debuginfo__find_trace_events(struct debuginfo *dbg,
 int debuginfo__find_probe_point(struct debuginfo *dbg, unsigned long addr,
 				struct perf_probe_point *ppt);
 
+int debuginfo__get_text_offset(struct debuginfo *dbg, Dwarf_Addr *offs,
+			       bool adjust_offset);
+
 /* Find a line range */
 int debuginfo__find_line_range(struct debuginfo *dbg, struct line_range *lr);
 
-- 
2.9.3

Re: [GIT PULL 0/3] perf/urgent 'perf probe' fixes

[Ingo Molnar]

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

> Hi Ingo,
> 
> 	Please consider pulling,
> 
> - Arnaldo
> 
> Test results at the end of this message, as usual.
> 
> The following changes since commit 18e7a45af91acdde99d3aa1372cc40e1f8142f7b:
> 
>   perf/x86: Reject non sampling events with precise_ip (2017-01-14 11:06:50 +0100)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-urgent-for-mingo-4.10-20170117
> 
> for you to fetch changes up to 613f050d68a8ed3c0b18b9568698908ef7bbc1f7:
> 
>   perf probe: Fix to probe on gcc generated functions in modules (2017-01-16 15:43:04 -0300)
> 
> ----------------------------------------------------------------
> perf/urgent 'perf probe' fixes:
> 
> Fixes:
> 
> - Show correct locations for 'perf probe' on modules (Masami Hiramatsu)
> 
> - Correctly handle 'perf probe's on gcc generated functions in modules (Masami Hiramatsu)
> 
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> ----------------------------------------------------------------
> Masami Hiramatsu (3):
>       perf probe: Fix to show correct locations for events on modules
>       perf probe: Add error checks to offline probe post-processing
>       perf probe: Fix to probe on gcc generated functions in modules
> 
>  tools/perf/util/probe-event.c  | 95 +++++++++++++++++++++++++++---------------
>  tools/perf/util/probe-finder.c | 15 ++++---
>  tools/perf/util/probe-finder.h |  3 ++
>  3 files changed, 72 insertions(+), 41 deletions(-)

Pulled, thanks a lot Arnaldo!

	Ingo