* [PATCH 01/19] perf probe: Remove unused tracing_dir variable
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 02/19] perf tools: Fix error handling of lzma decompression Arnaldo Carvalho de Melo
` (12 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Masami Hiramatsu, Jiri Olsa, Peter Zijlstra,
Arnaldo Carvalho de Melo
From: Masami Hiramatsu <mhiramat@kernel.org>
Remove unused tracing_dir variable from open_probe_events().
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/147201827792.5713.4165387506020511920.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/probe-file.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/tools/perf/util/probe-file.c b/tools/perf/util/probe-file.c
index 697ef66bff91..6f931e442f14 100644
--- a/tools/perf/util/probe-file.c
+++ b/tools/perf/util/probe-file.c
@@ -73,11 +73,10 @@ static void print_both_open_warning(int kerr, int uerr)
static int open_probe_events(const char *trace_file, bool readwrite)
{
char buf[PATH_MAX];
- const char *tracing_dir = "";
int ret;
- ret = e_snprintf(buf, PATH_MAX, "%s/%s%s",
- tracing_path, tracing_dir, trace_file);
+ ret = e_snprintf(buf, PATH_MAX, "%s/%s",
+ tracing_path, trace_file);
if (ret >= 0) {
pr_debug("Opening %s write=%d\n", buf, readwrite);
if (readwrite && !probe_event_dry_run)
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 02/19] perf tools: Fix error handling of lzma decompression
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 01/19] perf probe: Remove unused tracing_dir variable Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 06/19] perf symbols: Demangle symbols for synthesized @plt entries Arnaldo Carvalho de Melo
` (11 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Shawn Lin, Peter Zijlstra, Arnaldo Carvalho de Melo
From: Shawn Lin <shawn.lin@rock-chips.com>
lzma_decompress_to_file() never actually closes the file pointer, let's
fix it.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1471766253-1964-1-git-send-email-shawn.lin@rock-chips.com
[ Make err = -1, the common case, set it to 0 before the error label ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/lzma.c | 15 +++++++++------
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/tools/perf/util/lzma.c b/tools/perf/util/lzma.c
index 95a1acb61245..9ddea5cecd94 100644
--- a/tools/perf/util/lzma.c
+++ b/tools/perf/util/lzma.c
@@ -29,6 +29,7 @@ int lzma_decompress_to_file(const char *input, int output_fd)
lzma_action action = LZMA_RUN;
lzma_stream strm = LZMA_STREAM_INIT;
lzma_ret ret;
+ int err = -1;
u8 buf_in[BUFSIZE];
u8 buf_out[BUFSIZE];
@@ -45,7 +46,7 @@ int lzma_decompress_to_file(const char *input, int output_fd)
if (ret != LZMA_OK) {
pr_err("lzma: lzma_stream_decoder failed %s (%d)\n",
lzma_strerror(ret), ret);
- return -1;
+ goto err_fclose;
}
strm.next_in = NULL;
@@ -60,7 +61,7 @@ int lzma_decompress_to_file(const char *input, int output_fd)
if (ferror(infile)) {
pr_err("lzma: read error: %s\n", strerror(errno));
- return -1;
+ goto err_fclose;
}
if (feof(infile))
@@ -74,7 +75,7 @@ int lzma_decompress_to_file(const char *input, int output_fd)
if (writen(output_fd, buf_out, write_size) != write_size) {
pr_err("lzma: write error: %s\n", strerror(errno));
- return -1;
+ goto err_fclose;
}
strm.next_out = buf_out;
@@ -83,13 +84,15 @@ int lzma_decompress_to_file(const char *input, int output_fd)
if (ret != LZMA_OK) {
if (ret == LZMA_STREAM_END)
- return 0;
+ break;
pr_err("lzma: failed %s\n", lzma_strerror(ret));
- return -1;
+ goto err_fclose;
}
}
+ err = 0;
+err_fclose:
fclose(infile);
- return 0;
+ return err;
}
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 06/19] perf symbols: Demangle symbols for synthesized @plt entries.
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 01/19] perf probe: Remove unused tracing_dir variable Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 02/19] perf tools: Fix error handling of lzma decompression Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 07/19] perf config: Show default report configuration in example and docs Arnaldo Carvalho de Melo
` (10 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar; +Cc: linux-kernel, Milian Wolff, Arnaldo Carvalho de Melo
From: Milian Wolff <milian.wolff@kdab.com>
The symbols in the synthesized @plt entries where not demangled before,
i.e. we could end up with entries such as:
$ perf report
Samples: 7K of event 'cycles:ppp', Event count (approx.): 6223833141
Children Self Command Shared Object Symbol
- 93.63% 28.89% lab_mandelbrot lab_mandelbrot [.] main
- 73.81% main
- 33.57% hypot
27.76% __hypot_finite
15.97% __muldc3
2.90% __muldc3@plt
2.40% _ZNK6QImage6heightEv@plt
+ 2.14% QColor::rgb
1.94% _ZNK6QImage5widthEv@plt
1.92% cabs@plt
This patch remedies this issue by also applying demangling to the
synthesized symbols. The output for the above is now:
$ perf report
Samples: 7K of event 'cycles:ppp', Event count (approx.): 6223833141
Children Self Command Shared Object Symbol
- 93.63% 28.89% lab_mandelbrot lab_mandelbrot [.] main
- 73.81% main
- 33.57% hypot
27.76% __hypot_finite
15.97% __muldc3
2.90% __muldc3@plt
2.40% QImage::height() const@plt
+ 2.14% QColor::rgb
1.94% QImage::width() const@plt
1.92% cabs@plt
Signed-off-by: Milian Wolff <milian.wolff@kdab.com>
LPU-Reference: 20160830114102.30863-1-milian.wolff@kdab.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/symbol-elf.c | 81 ++++++++++++++++++++++++++++----------------
1 file changed, 52 insertions(+), 29 deletions(-)
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index a811c13a74d6..fbe31eff1beb 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -206,6 +206,37 @@ Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
return NULL;
}
+static bool want_demangle(bool is_kernel_sym)
+{
+ return is_kernel_sym ? symbol_conf.demangle_kernel : symbol_conf.demangle;
+}
+
+static char *demangle_sym(struct dso *dso, int kmodule, const char *elf_name)
+{
+ int demangle_flags = verbose ? (DMGL_PARAMS | DMGL_ANSI) : DMGL_NO_OPTS;
+ char *demangled = NULL;
+
+ /*
+ * We need to figure out if the object was created from C++ sources
+ * DWARF DW_compile_unit has this, but we don't always have access
+ * to it...
+ */
+ if (!want_demangle(dso->kernel || kmodule))
+ return demangled;
+
+ demangled = bfd_demangle(NULL, elf_name, demangle_flags);
+ if (demangled == NULL)
+ demangled = java_demangle_sym(elf_name, JAVA_DEMANGLE_NORET);
+ else if (rust_is_mangled(demangled))
+ /*
+ * Input to Rust demangling is the BFD-demangled
+ * name which it Rust-demangles in place.
+ */
+ rust_demangle_sym(demangled);
+
+ return demangled;
+}
+
#define elf_section__for_each_rel(reldata, pos, pos_mem, idx, nr_entries) \
for (idx = 0, pos = gelf_getrel(reldata, 0, &pos_mem); \
idx < nr_entries; \
@@ -301,11 +332,19 @@ int dso__synthesize_plt_symbols(struct dso *dso, struct symsrc *ss, struct map *
elf_section__for_each_rela(reldata, pos, pos_mem, idx,
nr_rel_entries) {
+ const char *elf_name = NULL;
+ char *demangled = NULL;
symidx = GELF_R_SYM(pos->r_info);
plt_offset += shdr_plt.sh_entsize;
gelf_getsym(syms, symidx, &sym);
+
+ elf_name = elf_sym__name(&sym, symstrs);
+ demangled = demangle_sym(dso, 0, elf_name);
+ if (demangled != NULL)
+ elf_name = demangled;
snprintf(sympltname, sizeof(sympltname),
- "%s@plt", elf_sym__name(&sym, symstrs));
+ "%s@plt", elf_name);
+ free(demangled);
f = symbol__new(plt_offset, shdr_plt.sh_entsize,
STB_GLOBAL, sympltname);
@@ -323,11 +362,19 @@ int dso__synthesize_plt_symbols(struct dso *dso, struct symsrc *ss, struct map *
GElf_Rel pos_mem, *pos;
elf_section__for_each_rel(reldata, pos, pos_mem, idx,
nr_rel_entries) {
+ const char *elf_name = NULL;
+ char *demangled = NULL;
symidx = GELF_R_SYM(pos->r_info);
plt_offset += shdr_plt.sh_entsize;
gelf_getsym(syms, symidx, &sym);
+
+ elf_name = elf_sym__name(&sym, symstrs);
+ demangled = demangle_sym(dso, 0, elf_name);
+ if (demangled != NULL)
+ elf_name = demangled;
snprintf(sympltname, sizeof(sympltname),
- "%s@plt", elf_sym__name(&sym, symstrs));
+ "%s@plt", elf_name);
+ free(demangled);
f = symbol__new(plt_offset, shdr_plt.sh_entsize,
STB_GLOBAL, sympltname);
@@ -775,11 +822,6 @@ static u64 ref_reloc(struct kmap *kmap)
return 0;
}
-static bool want_demangle(bool is_kernel_sym)
-{
- return is_kernel_sym ? symbol_conf.demangle_kernel : symbol_conf.demangle;
-}
-
void __weak arch__sym_update(struct symbol *s __maybe_unused,
GElf_Sym *sym __maybe_unused) { }
@@ -1070,29 +1112,10 @@ int dso__load_sym(struct dso *dso, struct map *map,
sym.st_value -= shdr.sh_addr - shdr.sh_offset;
}
new_symbol:
- /*
- * We need to figure out if the object was created from C++ sources
- * DWARF DW_compile_unit has this, but we don't always have access
- * to it...
- */
- if (want_demangle(dso->kernel || kmodule)) {
- int demangle_flags = DMGL_NO_OPTS;
- if (verbose)
- demangle_flags = DMGL_PARAMS | DMGL_ANSI;
-
- demangled = bfd_demangle(NULL, elf_name, demangle_flags);
- if (demangled == NULL)
- demangled = java_demangle_sym(elf_name, JAVA_DEMANGLE_NORET);
- else if (rust_is_mangled(demangled))
- /*
- * Input to Rust demangling is the BFD-demangled
- * name which it Rust-demangles in place.
- */
- rust_demangle_sym(demangled);
+ demangled = demangle_sym(dso, kmodule, elf_name);
+ if (demangled != NULL)
+ elf_name = demangled;
- if (demangled != NULL)
- elf_name = demangled;
- }
f = symbol__new(sym.st_value, sym.st_size,
GELF_ST_BIND(sym.st_info), elf_name);
free(demangled);
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 07/19] perf config: Show default report configuration in example and docs
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (2 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 06/19] perf symbols: Demangle symbols for synthesized @plt entries Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 08/19] perf probe: Show trace event definition Arnaldo Carvalho de Melo
` (9 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar; +Cc: linux-kernel, Milian Wolff, Arnaldo Carvalho de Melo
From: Milian Wolff <milian.wolff@kdab.com>
Signed-off-by: Milian Wolff <milian.wolff@kdab.com>
LPU-Reference: 20160830134106.21240-2-milian.wolff@kdab.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/Documentation/perf-config.txt | 8 ++++++++
tools/perf/Documentation/perfconfig.example | 9 +++++++++
2 files changed, 17 insertions(+)
diff --git a/tools/perf/Documentation/perf-config.txt b/tools/perf/Documentation/perf-config.txt
index 68c89190e79c..cb081ac59fd1 100644
--- a/tools/perf/Documentation/perf-config.txt
+++ b/tools/perf/Documentation/perf-config.txt
@@ -110,6 +110,14 @@ Given a $HOME/.perfconfig like this:
order = caller
sort-key = function
+ [report]
+ # Defaults
+ sort-order = comm,dso,symbol
+ percent-limit = 0
+ queue-size = 0
+ children = true
+ group = true
+
Variables
~~~~~~~~~
diff --git a/tools/perf/Documentation/perfconfig.example b/tools/perf/Documentation/perfconfig.example
index 1d8d5bc4cd2d..2b477c1d1efe 100644
--- a/tools/perf/Documentation/perfconfig.example
+++ b/tools/perf/Documentation/perfconfig.example
@@ -27,3 +27,12 @@
use_offset = true
jump_arrows = true
show_nr_jumps = false
+
+[report]
+
+ # Defaults
+ sort-order = comm,dso,symbol
+ percent-limit = 0
+ queue-size = 0
+ children = true
+ group = true
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 08/19] perf probe: Show trace event definition
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (3 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 07/19] perf config: Show default report configuration in example and docs Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 12/19] perf test vmlinux: Clarify which -v lines are errors or warning Arnaldo Carvalho de Melo
` (8 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Masami Hiramatsu, Ingo Molnar, Jiri Olsa,
Peter Zijlstra, Arnaldo Carvalho de Melo
From: Masami Hiramatsu <mhiramat@kernel.org>
Add --definition/-D option for showing the trace-event definition in
stdout. This can be useful in debugging or combined with a shell script.
e.g.
----
# perf probe --definition 'do_sys_open $params'
p:probe/do_sys_open _text+2261728 dfd=%di:s32 filename=%si:u64 flags=%dx:s32 mode=%cx:u16
----
Suggested-and-Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/147214226712.23638.2240534040014013658.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/Documentation/perf-probe.txt | 7 +++++
tools/perf/builtin-probe.c | 26 ++++++++++++++-----
tools/perf/util/probe-event.c | 46 +++++++++++++++++++++++++++++++++
tools/perf/util/probe-event.h | 1 +
4 files changed, 73 insertions(+), 7 deletions(-)
diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt
index f37d123d5dac..56db4d40d27e 100644
--- a/tools/perf/Documentation/perf-probe.txt
+++ b/tools/perf/Documentation/perf-probe.txt
@@ -21,6 +21,8 @@ or
'perf probe' [options] --vars='PROBEPOINT'
or
'perf probe' [options] --funcs
+or
+'perf probe' [options] --definition='PROBE' [...]
DESCRIPTION
-----------
@@ -96,6 +98,11 @@ OPTIONS
can also list functions in a user space executable / shared library.
This also can accept a FILTER rule argument.
+-D::
+--definition=::
+ Show trace-event definition converted from given probe-event instead
+ of write it into tracing/[k,u]probe_events.
+
--filter=FILTER::
(Only for --vars and --funcs) Set filter. FILTER is a combination of glob
pattern, see FILTER PATTERN for detail.
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
index ee5b42173ba3..7a3d8c47ed29 100644
--- a/tools/perf/builtin-probe.c
+++ b/tools/perf/builtin-probe.c
@@ -326,6 +326,11 @@ static int perf_add_probe_events(struct perf_probe_event *pevs, int npevs)
if (ret < 0)
goto out_cleanup;
+ if (params.command == 'D') { /* it shows definition */
+ ret = show_probe_trace_events(pevs, npevs);
+ goto out_cleanup;
+ }
+
ret = apply_perf_probe_events(pevs, npevs);
if (ret < 0)
goto out_cleanup;
@@ -454,6 +459,14 @@ out:
return ret;
}
+#ifdef HAVE_DWARF_SUPPORT
+#define PROBEDEF_STR \
+ "[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT [[NAME=]ARG ...]"
+#else
+#define PROBEDEF_STR "[EVENT=]FUNC[+OFF|%return] [[NAME=]ARG ...]"
+#endif
+
+
static int
__cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
{
@@ -479,13 +492,7 @@ __cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
opt_set_filter_with_command, DEFAULT_LIST_FILTER),
OPT_CALLBACK('d', "del", NULL, "[GROUP:]EVENT", "delete a probe event.",
opt_set_filter_with_command),
- OPT_CALLBACK('a', "add", NULL,
-#ifdef HAVE_DWARF_SUPPORT
- "[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT"
- " [[NAME=]ARG ...]",
-#else
- "[EVENT=]FUNC[+OFF|%return] [[NAME=]ARG ...]",
-#endif
+ OPT_CALLBACK('a', "add", NULL, PROBEDEF_STR,
"probe point definition, where\n"
"\t\tGROUP:\tGroup name (optional)\n"
"\t\tEVENT:\tEvent name\n"
@@ -503,6 +510,9 @@ __cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
"\t\tARG:\tProbe argument (kprobe-tracer argument format.)\n",
#endif
opt_add_probe_event),
+ OPT_CALLBACK('D', "definition", NULL, PROBEDEF_STR,
+ "Show trace event definition of given traceevent for k/uprobe_events.",
+ opt_add_probe_event),
OPT_BOOLEAN('f', "force", &probe_conf.force_add, "forcibly add events"
" with existing name"),
OPT_CALLBACK('L', "line", NULL,
@@ -548,6 +558,7 @@ __cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
set_option_flag(options, 'a', "add", PARSE_OPT_EXCLUSIVE);
set_option_flag(options, 'd', "del", PARSE_OPT_EXCLUSIVE);
+ set_option_flag(options, 'D', "definition", PARSE_OPT_EXCLUSIVE);
set_option_flag(options, 'l', "list", PARSE_OPT_EXCLUSIVE);
#ifdef HAVE_DWARF_SUPPORT
set_option_flag(options, 'L', "line", PARSE_OPT_EXCLUSIVE);
@@ -644,6 +655,7 @@ __cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
}
break;
case 'a':
+ case 'D':
/* Ensure the last given target is used */
if (params.target && !params.target_used) {
pr_err(" Error: -x/-m must follow the probe definitions.\n");
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 0bed2ee16292..4a49cb8ce7ab 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -3207,6 +3207,52 @@ int convert_perf_probe_events(struct perf_probe_event *pevs, int npevs)
return 0;
}
+static int show_probe_trace_event(struct probe_trace_event *tev)
+{
+ char *buf = synthesize_probe_trace_command(tev);
+
+ if (!buf) {
+ pr_debug("Failed to synthesize probe trace event.\n");
+ return -EINVAL;
+ }
+
+ /* Showing definition always go stdout */
+ printf("%s\n", buf);
+ free(buf);
+
+ return 0;
+}
+
+int show_probe_trace_events(struct perf_probe_event *pevs, int npevs)
+{
+ struct strlist *namelist = strlist__new(NULL, NULL);
+ struct probe_trace_event *tev;
+ struct perf_probe_event *pev;
+ int i, j, ret = 0;
+
+ if (!namelist)
+ return -ENOMEM;
+
+ for (j = 0; j < npevs && !ret; j++) {
+ pev = &pevs[j];
+ for (i = 0; i < pev->ntevs && !ret; i++) {
+ tev = &pev->tevs[i];
+ /* Skip if the symbol is out of .text or blacklisted */
+ if (!tev->point.symbol && !pev->uprobes)
+ continue;
+
+ /* Set new name for tev (and update namelist) */
+ ret = probe_trace_event__set_name(tev, pev,
+ namelist, true);
+ if (!ret)
+ ret = show_probe_trace_event(tev);
+ }
+ }
+ strlist__delete(namelist);
+
+ return ret;
+}
+
int apply_perf_probe_events(struct perf_probe_event *pevs, int npevs)
{
int i, ret = 0;
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
index f4f45db77c1c..62094082d940 100644
--- a/tools/perf/util/probe-event.h
+++ b/tools/perf/util/probe-event.h
@@ -147,6 +147,7 @@ int line_range__init(struct line_range *lr);
int add_perf_probe_events(struct perf_probe_event *pevs, int npevs);
int convert_perf_probe_events(struct perf_probe_event *pevs, int npevs);
int apply_perf_probe_events(struct perf_probe_event *pevs, int npevs);
+int show_probe_trace_events(struct perf_probe_event *pevs, int npevs);
void cleanup_perf_probe_events(struct perf_probe_event *pevs, int npevs);
int del_perf_probe_events(struct strfilter *filter);
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 12/19] perf test vmlinux: Clarify which -v lines are errors or warning
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (4 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 08/19] perf probe: Show trace event definition Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 13/19] perf test vmlinux: Avoid printing headers for empty lists Arnaldo Carvalho de Melo
` (7 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Arnaldo Carvalho de Melo, Adrian Hunter,
David Ahern, Jiri Olsa, Masami Hiramatsu, Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
When the 'perf test -v vmlinux' test fails, it is not clear which of the
lines are errors or warnings, clarify that adding ERR/WARN prefixes:
# perf test -F -v 1
1: vmlinux symtab matches kallsyms :
--- start ---
Looking at the vmlinux_path (8 entries long)
Using /lib/modules/4.8.0-rc4+/build/vmlinux for symbols
ERR : 0xffffffffb7001000: diff name v: xen_hypercall_set_trap_table k: hypercall_page
WARN: 0xffffffffb7077970: diff end addr for aesni_gcm_dec v: 0xffffffffb707a2f2 k: 0xffffffffb7077a02
WARN: 0xffffffffb707a300: diff end addr for aesni_gcm_enc v: 0xffffffffb707cc03 k: 0xffffffffb707a392
WARN: 0xffffffffb707f950: diff end addr for aesni_gcm_enc_avx_gen2 v: 0xffffffffb7084ef6 k: 0xffffffffb707f9c3
WARN: 0xffffffffb7084f00: diff end addr for aesni_gcm_dec_avx_gen2 v: 0xffffffffb708a691 k: 0xffffffffb7084f73
WARN: 0xffffffffb708aa10: diff end addr for aesni_gcm_enc_avx_gen4 v: 0xffffffffb708f844 k: 0xffffffffb708aa83
WARN: 0xffffffffb708f850: diff end addr for aesni_gcm_dec_avx_gen4 v: 0xffffffffb709486f k: 0xffffffffb708f8c3
ERR : 0xffffffffb71a6e50: diff name v: perf_pmu_commit_txn.part.98 k: perf_pmu_cancel_txn.part.97
ERR : 0xffffffffb752e480: diff name v: wakeup_expire_count_show.part.5 k: wakeup_active_count_show.part.7
ERR : 0xffffffffb76e8d00: diff name v: phys_switch_id_show.part.11 k: phys_port_name_show.part.12
WARN: Maps only in vmlinux:
ffffffffb7d7d000-ffffffffb7eeaac8 117d000 [kernel].init.text
ffffffffb7eeaac8-ffffffffc03ad000 12eaac8 [kernel].exit.text
WARN: Maps in vmlinux with a different name in kallsyms:
WARN: Maps only in kallsyms:
---- end ----
vmlinux symtab matches kallsyms: FAILED!
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-n5ml8m7y9x8kzvxt09ipku88@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/tests/vmlinux-kallsyms.c | 18 +++++++++---------
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/tools/perf/tests/vmlinux-kallsyms.c b/tools/perf/tests/vmlinux-kallsyms.c
index e63abab7d5a1..6bd5bf980de2 100644
--- a/tools/perf/tests/vmlinux-kallsyms.c
+++ b/tools/perf/tests/vmlinux-kallsyms.c
@@ -143,7 +143,7 @@ next_pair:
*/
s64 skew = mem_end - UM(pair->end);
if (llabs(skew) >= page_size)
- pr_debug("%#" PRIx64 ": diff end addr for %s v: %#" PRIx64 " k: %#" PRIx64 "\n",
+ pr_debug("WARN: %#" PRIx64 ": diff end addr for %s v: %#" PRIx64 " k: %#" PRIx64 "\n",
mem_start, sym->name, mem_end,
UM(pair->end));
@@ -161,15 +161,15 @@ next_pair:
if (UM(pair->start) == mem_start)
goto next_pair;
- pr_debug("%#" PRIx64 ": diff name v: %s k: %s\n",
+ pr_debug("ERR : %#" PRIx64 ": diff name v: %s k: %s\n",
mem_start, sym->name, pair->name);
} else {
- pr_debug("%#" PRIx64 ": diff name v: %s k: %s\n",
+ pr_debug("ERR : %#" PRIx64 ": diff name v: %s k: %s\n",
mem_start, sym->name, first_pair->name);
}
}
} else
- pr_debug("%#" PRIx64 ": %s not on kallsyms\n",
+ pr_debug("ERR : %#" PRIx64 ": %s not on kallsyms\n",
mem_start, sym->name);
err = -1;
@@ -178,7 +178,7 @@ next_pair:
if (!verbose)
goto out;
- pr_info("Maps only in vmlinux:\n");
+ pr_info("WARN: Maps only in vmlinux:\n");
for (map = maps__first(maps); map; map = map__next(map)) {
struct map *
@@ -198,7 +198,7 @@ next_pair:
map__fprintf(map, stderr);
}
- pr_info("Maps in vmlinux with a different name in kallsyms:\n");
+ pr_info("WARN: Maps in vmlinux with a different name in kallsyms:\n");
for (map = maps__first(maps); map; map = map__next(map)) {
struct map *pair;
@@ -212,17 +212,17 @@ next_pair:
if (pair->start == mem_start) {
pair->priv = 1;
- pr_info(" %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s in kallsyms as",
+ pr_info("WARN: %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s in kallsyms as",
map->start, map->end, map->pgoff, map->dso->name);
if (mem_end != pair->end)
- pr_info(":\n*%" PRIx64 "-%" PRIx64 " %" PRIx64,
+ pr_info(":\nWARN: *%" PRIx64 "-%" PRIx64 " %" PRIx64,
pair->start, pair->end, pair->pgoff);
pr_info(" %s\n", pair->dso->name);
pair->priv = 1;
}
}
- pr_info("Maps only in kallsyms:\n");
+ pr_info("WARN: Maps only in kallsyms:\n");
maps = &kallsyms.kmaps.maps[type];
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 13/19] perf test vmlinux: Avoid printing headers for empty lists
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (5 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 12/19] perf test vmlinux: Clarify which -v lines are errors or warning Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 14/19] perf test vmlinux: Tolerate symbol aliases Arnaldo Carvalho de Melo
` (6 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Arnaldo Carvalho de Melo, Adrian Hunter,
David Ahern, Jiri Olsa, Masami Hiramatsu, Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Before:
# perf test -F -v 1
1: vmlinux symtab matches kallsyms:
--- start ---
<SNIP>
WARN: Maps only in vmlinux:
ffffffffb7d7d000-ffffffffb7eeaac8 117d000 [kernel].init.text
ffffffffb7eeaac8-ffffffffc03ad000 12eaac8 [kernel].exit.text
WARN: Maps in vmlinux with a different name in kallsyms:
WARN: Maps only in kallsyms:
---- end ----
vmlinux symtab matches kallsyms: Ok
#
The two last WARN lines are now suppressed, since there are no such
cases detected.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-9ww8uvzl682ykaw8ht1tozlr@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/tests/vmlinux-kallsyms.c | 29 ++++++++++++++++++++++-------
1 file changed, 22 insertions(+), 7 deletions(-)
diff --git a/tools/perf/tests/vmlinux-kallsyms.c b/tools/perf/tests/vmlinux-kallsyms.c
index 6bd5bf980de2..450f69960931 100644
--- a/tools/perf/tests/vmlinux-kallsyms.c
+++ b/tools/perf/tests/vmlinux-kallsyms.c
@@ -28,6 +28,7 @@ int test__vmlinux_matches_kallsyms(int subtest __maybe_unused)
enum map_type type = MAP__FUNCTION;
struct maps *maps = &vmlinux.kmaps.maps[type];
u64 mem_start, mem_end;
+ bool header_printed;
/*
* Step 1:
@@ -178,7 +179,7 @@ next_pair:
if (!verbose)
goto out;
- pr_info("WARN: Maps only in vmlinux:\n");
+ header_printed = false;
for (map = maps__first(maps); map; map = map__next(map)) {
struct map *
@@ -192,13 +193,18 @@ next_pair:
(map->dso->kernel ?
map->dso->short_name :
map->dso->name));
- if (pair)
+ if (pair) {
pair->priv = 1;
- else
+ } else {
+ if (!header_printed) {
+ pr_info("WARN: Maps only in vmlinux:\n");
+ header_printed = true;
+ }
map__fprintf(map, stderr);
+ }
}
- pr_info("WARN: Maps in vmlinux with a different name in kallsyms:\n");
+ header_printed = false;
for (map = maps__first(maps); map; map = map__next(map)) {
struct map *pair;
@@ -211,7 +217,11 @@ next_pair:
continue;
if (pair->start == mem_start) {
- pair->priv = 1;
+ if (!header_printed) {
+ pr_info("WARN: Maps in vmlinux with a different name in kallsyms:\n");
+ header_printed = true;
+ }
+
pr_info("WARN: %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s in kallsyms as",
map->start, map->end, map->pgoff, map->dso->name);
if (mem_end != pair->end)
@@ -222,13 +232,18 @@ next_pair:
}
}
- pr_info("WARN: Maps only in kallsyms:\n");
+ header_printed = false;
maps = &kallsyms.kmaps.maps[type];
for (map = maps__first(maps); map; map = map__next(map)) {
- if (!map->priv)
+ if (!map->priv) {
+ if (!header_printed) {
+ pr_info("WARN: Maps only in kallsyms:\n");
+ header_printed = true;
+ }
map__fprintf(map, stderr);
+ }
}
out:
machine__exit(&kallsyms);
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 14/19] perf test vmlinux: Tolerate symbol aliases
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (6 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 13/19] perf test vmlinux: Avoid printing headers for empty lists Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 15/19] perf symbols: Check symbol_conf.allow_aliases for kallsyms loading too Arnaldo Carvalho de Melo
` (5 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Arnaldo Carvalho de Melo, Adrian Hunter,
David Ahern, Jiri Olsa, Masami Hiramatsu, Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
The algorithms used to prune aliases in symbols__fixup_duplicate() uses
information available on ELF symtabs that are not present on
/proc/kallsyms, so it picks different aliases as "best" for vmlinux and
kallsyms.
We could probably improve a bit this by having a list of aliases for the
"best" symbols picked, instead of throwing this info, but that is left
for when we find a real need.
With this, 'perf test vmlinux' passes:
# perf test -F 1
1: vmlinux symtab matches kallsyms: Ok
#
When we ask for verbose mode, we can see those warning:
# perf test -F -v 1
1: vmlinux symtab matches kallsyms:
--- start ---
Looking at the vmlinux_path (8 entries long)
Using /lib/modules/4.8.0-rc4+/build/vmlinux for symbols
WARN: 0xffffffffb7001000: diff name v: xen_hypercall_set_trap_table k: hypercall_page
WARN: 0xffffffffb7077970: diff end addr for aesni_gcm_dec v: 0xffffffffb707a2f2 k: 0xffffffffb7077a02
WARN: 0xffffffffb707a300: diff end addr for aesni_gcm_enc v: 0xffffffffb707cc03 k: 0xffffffffb707a392
WARN: 0xffffffffb707f950: diff end addr for aesni_gcm_enc_avx_gen2 v: 0xffffffffb7084ef6 k: 0xffffffffb707f9c3
WARN: 0xffffffffb7084f00: diff end addr for aesni_gcm_dec_avx_gen2 v: 0xffffffffb708a691 k: 0xffffffffb7084f73
WARN: 0xffffffffb708aa10: diff end addr for aesni_gcm_enc_avx_gen4 v: 0xffffffffb708f844 k: 0xffffffffb708aa83
WARN: 0xffffffffb708f850: diff end addr for aesni_gcm_dec_avx_gen4 v: 0xffffffffb709486f k: 0xffffffffb708f8c3
WARN: 0xffffffffb71a6e50: diff name v: perf_pmu_commit_txn.part.98 k: perf_pmu_cancel_txn.part.97
WARN: 0xffffffffb752e480: diff name v: wakeup_expire_count_show.part.5 k: wakeup_active_count_show.part.7
WARN: 0xffffffffb76e8d00: diff name v: phys_switch_id_show.part.11 k: phys_port_name_show.part.12
WARN: Maps only in vmlinux:
ffffffffb7d7d000-ffffffffb7eeaac8 117d000 [kernel].init.text
ffffffffb7eeaac8-ffffffffc03ad000 12eaac8 [kernel].exit.text
---- end ----
vmlinux symtab matches kallsyms: Ok
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-6v5w1k8rpx4ggczlkw730vt0@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/tests/vmlinux-kallsyms.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/tools/perf/tests/vmlinux-kallsyms.c b/tools/perf/tests/vmlinux-kallsyms.c
index 450f69960931..77513bf99d1b 100644
--- a/tools/perf/tests/vmlinux-kallsyms.c
+++ b/tools/perf/tests/vmlinux-kallsyms.c
@@ -155,19 +155,20 @@ next_pair:
* kallsyms.
*/
continue;
-
} else {
pair = machine__find_kernel_symbol_by_name(&kallsyms, type, sym->name, NULL, NULL);
if (pair) {
if (UM(pair->start) == mem_start)
goto next_pair;
- pr_debug("ERR : %#" PRIx64 ": diff name v: %s k: %s\n",
+ pr_debug("WARN: %#" PRIx64 ": diff name v: %s k: %s\n",
mem_start, sym->name, pair->name);
} else {
- pr_debug("ERR : %#" PRIx64 ": diff name v: %s k: %s\n",
+ pr_debug("WARN: %#" PRIx64 ": diff name v: %s k: %s\n",
mem_start, sym->name, first_pair->name);
}
+
+ continue;
}
} else
pr_debug("ERR : %#" PRIx64 ": %s not on kallsyms\n",
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 15/19] perf symbols: Check symbol_conf.allow_aliases for kallsyms loading too
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (7 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 14/19] perf test vmlinux: Tolerate symbol aliases Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 16/19] perf symbols: Fixup symbol sizes before picking best ones Arnaldo Carvalho de Melo
` (4 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Arnaldo Carvalho de Melo, Adrian Hunter,
David Ahern, Jiri Olsa, Masami Hiramatsu, Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
We can allow aliases to be kept, but we were checking this just when
loading vmlinux files, be consistent, do it for any symbol table loading
code that calls symbol__fixup_duplicate() by making this function check
.allow_aliases instead.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: 680d926a8cb0 ("perf symbols: Allow symbol alias when loading map for symbol name")
Link: http://lkml.kernel.org/n/tip-z0avp0s6cfjckc4xj3pdfjdz@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/symbol-elf.c | 3 +--
tools/perf/util/symbol.c | 3 +++
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index e680371bd5c9..9ba6e49ead62 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -1136,8 +1136,7 @@ new_symbol:
* For misannotated, zeroed, ASM function sizes.
*/
if (nr > 0) {
- if (!symbol_conf.allow_aliases)
- symbols__fixup_duplicate(&dso->symbols[map->type]);
+ symbols__fixup_duplicate(&dso->symbols[map->type]);
symbols__fixup_end(&dso->symbols[map->type]);
if (kmap) {
/*
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 863d69c45b8a..ef2f913de662 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -153,6 +153,9 @@ void symbols__fixup_duplicate(struct rb_root *symbols)
struct rb_node *nd;
struct symbol *curr, *next;
+ if (symbol_conf.allow_aliases)
+ return;
+
nd = rb_first(symbols);
while (nd) {
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 16/19] perf symbols: Fixup symbol sizes before picking best ones
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (8 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 15/19] perf symbols: Check symbol_conf.allow_aliases for kallsyms loading too Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 17/19] perf probe: Add helper function to check if probe with variable Arnaldo Carvalho de Melo
` (3 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Arnaldo Carvalho de Melo, Anton Blanchard,
Adrian Hunter, David Ahern, Jiri Olsa, Masami Hiramatsu,
Namhyung Kim, Wang Nan
From: Arnaldo Carvalho de Melo <acme@redhat.com>
When we call symbol__fixup_duplicate() we use algorithms to pick the
"best" symbols for cases where there are various functions/aliases to an
address, and those check zero size symbols, which, before calling
symbol__fixup_end() are _all_ symbols in a just parsed kallsyms file.
So first fixup the end, then fixup the duplicates.
Found while trying to figure out why 'perf test vmlinux' failed, see the
output of 'perf test -v vmlinux' to see cases where the symbols picked
as best for vmlinux don't match the ones picked for kallsyms.
Cc: Anton Blanchard <anton@samba.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: 694bf407b061 ("perf symbols: Add some heuristics for choosing the best duplicate symbol")
Link: http://lkml.kernel.org/n/tip-rxqvdgr0mqjdxee0kf8i2ufn@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/symbol-elf.c | 2 +-
tools/perf/util/symbol.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index 9ba6e49ead62..295d3147a803 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -1136,8 +1136,8 @@ new_symbol:
* For misannotated, zeroed, ASM function sizes.
*/
if (nr > 0) {
- symbols__fixup_duplicate(&dso->symbols[map->type]);
symbols__fixup_end(&dso->symbols[map->type]);
+ symbols__fixup_duplicate(&dso->symbols[map->type]);
if (kmap) {
/*
* We need to fixup this here too because we create new
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index ef2f913de662..98cd50384c32 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1243,8 +1243,8 @@ int __dso__load_kallsyms(struct dso *dso, const char *filename,
if (kallsyms__delta(map, filename, &delta))
return -1;
- symbols__fixup_duplicate(&dso->symbols[map->type]);
symbols__fixup_end(&dso->symbols[map->type]);
+ symbols__fixup_duplicate(&dso->symbols[map->type]);
if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
dso->symtab_type = DSO_BINARY_TYPE__GUEST_KALLSYMS;
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 17/19] perf probe: Add helper function to check if probe with variable
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (9 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 16/19] perf symbols: Fixup symbol sizes before picking best ones Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 18/19] perf uprobe: Skip prologue if program compiled without optimization Arnaldo Carvalho de Melo
` (2 subsequent siblings)
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Ravi Bangoria, Alexander Shishkin, Hemant Kumar,
Naveen N . Rao, Peter Zijlstra, Wang Nan,
Arnaldo Carvalho de Melo
From: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Introduce helper function instead of inline code and replace hardcoded
strings "$vars" and "$params" with their corresponding macros.
perf_probe_with_var() is not declared as static since it will be called
from different file in subsequent patch.
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: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1470214725-5023-1-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 | 22 +++++++++++++++-------
tools/perf/util/probe-event.h | 2 ++
2 files changed, 17 insertions(+), 7 deletions(-)
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 8a1e9e67614f..a543e9ca581c 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -1618,19 +1618,27 @@ out:
return ret;
}
+/* Returns true if *any* ARG is either C variable, $params or $vars. */
+bool perf_probe_with_var(struct perf_probe_event *pev)
+{
+ int i = 0;
+
+ for (i = 0; i < pev->nargs; i++)
+ if (is_c_varname(pev->args[i].var) ||
+ !strcmp(pev->args[i].var, PROBE_ARG_PARAMS) ||
+ !strcmp(pev->args[i].var, PROBE_ARG_VARS))
+ return true;
+ return false;
+}
+
/* Return true if this perf_probe_event requires debuginfo */
bool perf_probe_event_need_dwarf(struct perf_probe_event *pev)
{
- int i;
-
if (pev->point.file || pev->point.line || pev->point.lazy_line)
return true;
- for (i = 0; i < pev->nargs; i++)
- if (is_c_varname(pev->args[i].var) ||
- !strcmp(pev->args[i].var, "$params") ||
- !strcmp(pev->args[i].var, "$vars"))
- return true;
+ if (perf_probe_with_var(pev))
+ return true;
return false;
}
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
index 62094082d940..8091d15113f7 100644
--- a/tools/perf/util/probe-event.h
+++ b/tools/perf/util/probe-event.h
@@ -128,6 +128,8 @@ char *synthesize_perf_probe_point(struct perf_probe_point *pp);
int perf_probe_event__copy(struct perf_probe_event *dst,
struct perf_probe_event *src);
+bool perf_probe_with_var(struct perf_probe_event *pev);
+
/* Check the perf_probe_event needs debuginfo */
bool perf_probe_event_need_dwarf(struct perf_probe_event *pev);
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 18/19] perf uprobe: Skip prologue if program compiled without optimization
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (10 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 17/19] perf probe: Add helper function to check if probe with variable Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-01 16:45 ` [PATCH 19/19] perf probe: Move dwarf specific functions to dwarf-aux.c Arnaldo Carvalho de Melo
2016-09-05 13:16 ` [GIT PULL 00/19] perf/core improvements and fixes Ingo Molnar
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Ravi Bangoria, Alexander Shishkin, Hemant Kumar,
Peter Zijlstra, Wang Nan, Yauheni Kaliuta,
Arnaldo Carvalho de Melo
From: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
The function prologue prepares stack and registers before executing
function logic.
When target program is compiled without optimization, function parameter
information is only valid after the prologue.
When we probe entrypc of the function, and try to record a function
parameter, it contains a garbage value.
For example:
$ vim test.c
#include <stdio.h>
void foo(int i)
{
printf("i: %d\n", i);
}
int main()
{
foo(42);
return 0;
}
$ gcc -g test.c -o test
$ objdump -dl test | less
foo():
/home/ravi/test.c:4
400536: 55 push %rbp
400537: 48 89 e5 mov %rsp,%rbp
40053a: 48 83 ec 10 sub -bashx10,%rsp
40053e: 89 7d fc mov %edi,-0x4(%rbp)
/home/ravi/test.c:5
400541: 8b 45 fc mov -0x4(%rbp),%eax
...
...
main():
/home/ravi/test.c:9
400558: 55 push %rbp
400559: 48 89 e5 mov %rsp,%rbp
/home/ravi/test.c:10
40055c: bf 2a 00 00 00 mov -bashx2a,%edi
400561: e8 d0 ff ff ff callq 400536 <foo>
$ perf probe -x ./test 'foo i'
$ cat /sys/kernel/debug/tracing/uprobe_events
p:probe_test/foo /home/ravi/test:0x0000000000000536 i=-12(%sp):s32
$ perf record -e probe_test:foo ./test
$ perf script
test 5778 [001] 4918.562027: probe_test:foo: (400536) i=0
Here variable 'i' is passed via stack which is pushed on stack at
0x40053e. But we are probing at 0x400536.
To resolve this issues, we need to probe on next instruction after
prologue. gdb and systemtap also does same thing. I've implemented this
patch based on approach systemtap has used.
After applying patch:
$ perf probe -x ./test 'foo i'
$ cat /sys/kernel/debug/tracing/uprobe_events
p:probe_test/foo /home/ravi/test:0x0000000000000541 i=-4(%bp):s32
$ perf record -e probe_test:foo ./test
$ perf script
test 6300 [001] 5877.879327: probe_test:foo: (400541) i=42
No need to skip prologue for optimized case since debug info is correct
for each instructions for -O2 -g. For more details please visit:
https://bugzilla.redhat.com/show_bug.cgi?id=612253#c6
Changes in v2:
- Skipping prologue only when any ARG is either C variable, $params or
$vars.
- Probe on line(:1) may not be always possible. Recommend only address
to force probe on function entry.
Committer notes:
Testing it with 'perf trace':
# perf probe -x ./test foo i
Added new event:
probe_test:foo (on foo in /home/acme/c/test with i)
You can now use it in all perf tools, such as:
perf record -e probe_test:foo -aR sleep 1
# cat /sys/kernel/debug/tracing/uprobe_events
p:probe_test/foo /home/acme/c/test:0x0000000000000526 i=-12(%sp):s32
# trace --no-sys --event probe_*:* ./test
i: 42
0.000 probe_test:foo:(400526) i=0)
#
After the patch:
# perf probe -d *:*
Removed event: probe_test:foo
# perf probe -x ./test foo i
Target program is compiled without optimization. Skipping prologue.
Probe on address 0x400526 to force probing at the function entry.
Added new event:
probe_test:foo (on foo in /home/acme/c/test with i)
You can now use it in all perf tools, such as:
perf record -e probe_test:foo -aR sleep 1
# cat /sys/kernel/debug/tracing/uprobe_events
p:probe_test/foo /home/acme/c/test:0x0000000000000531 i=-4(%bp):s32
# trace --no-sys --event probe_*:* ./test
i: 42
0.000 probe_test:foo:(400531) i=42)
#
Reported-by: Michael Petlan <mpetlan@redhat.com>
Report-Link: https://www.mail-archive.com/linux-perf-users@vger.kernel.org/msg02348.html
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: Yauheni Kaliuta <yauheni.kaliuta@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1299021
Link: http://lkml.kernel.org/r/1470214725-5023-2-git-send-email-ravi.bangoria@linux.vnet.ibm.com
[ Rename 'die' to 'cu_die' to avoid shadowing a die() definition on at least centos 5, Debian 7 and ubuntu:12.04.5]
[ Use PRIx64 instead of lx to format a Dwarf_Addr, aka long long unsigned int, fixing the build on 32-bit systems ]
[ dwarf_getsrclines() expects a size_t * argument ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/probe-finder.c | 165 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 165 insertions(+)
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 508b61c6e5fb..003ecadae35d 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -907,6 +907,170 @@ static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf)
return die_walk_lines(sp_die, probe_point_lazy_walker, pf);
}
+static bool var_has_loclist(Dwarf_Die *cu_die)
+{
+ Dwarf_Attribute loc;
+ int tag = dwarf_tag(cu_die);
+
+ if (tag != DW_TAG_formal_parameter &&
+ tag != DW_TAG_variable)
+ return false;
+
+ return (dwarf_attr_integrate(cu_die, DW_AT_location, &loc) &&
+ dwarf_whatform(&loc) == DW_FORM_sec_offset);
+}
+
+/*
+ * For any object in given CU whose DW_AT_location is a location list,
+ * target program is compiled with optimization.
+ */
+static bool optimized_target(Dwarf_Die *cu_die)
+{
+ Dwarf_Die tmp_die;
+
+ if (var_has_loclist(cu_die))
+ return true;
+
+ if (!dwarf_child(cu_die, &tmp_die) && optimized_target(&tmp_die))
+ return true;
+
+ if (!dwarf_siblingof(cu_die, &tmp_die) && optimized_target(&tmp_die))
+ return true;
+
+ return false;
+}
+
+static bool get_entrypc_idx(Dwarf_Lines *lines, unsigned long nr_lines,
+ Dwarf_Addr pf_addr, unsigned long *entrypc_idx)
+{
+ unsigned long i;
+ Dwarf_Addr addr;
+
+ for (i = 0; i < nr_lines; i++) {
+ if (dwarf_lineaddr(dwarf_onesrcline(lines, i), &addr))
+ return false;
+
+ if (addr == pf_addr) {
+ *entrypc_idx = i;
+ return true;
+ }
+ }
+ return false;
+}
+
+static bool get_postprologue_addr(unsigned long entrypc_idx,
+ Dwarf_Lines *lines,
+ unsigned long nr_lines,
+ Dwarf_Addr highpc,
+ Dwarf_Addr *postprologue_addr)
+{
+ unsigned long i;
+ int entrypc_lno, lno;
+ Dwarf_Line *line;
+ Dwarf_Addr addr;
+ bool p_end;
+
+ /* entrypc_lno is actual source line number */
+ line = dwarf_onesrcline(lines, entrypc_idx);
+ if (dwarf_lineno(line, &entrypc_lno))
+ return false;
+
+ for (i = entrypc_idx; i < nr_lines; i++) {
+ line = dwarf_onesrcline(lines, i);
+
+ if (dwarf_lineaddr(line, &addr) ||
+ dwarf_lineno(line, &lno) ||
+ dwarf_lineprologueend(line, &p_end))
+ return false;
+
+ /* highpc is exclusive. [entrypc,highpc) */
+ if (addr >= highpc)
+ break;
+
+ /* clang supports prologue-end marker */
+ if (p_end)
+ break;
+
+ /* Actual next line in source */
+ if (lno != entrypc_lno)
+ break;
+
+ /*
+ * Single source line can have multiple line records.
+ * For Example,
+ * void foo() { printf("hello\n"); }
+ * contains two line records. One points to declaration and
+ * other points to printf() line. Variable 'lno' won't get
+ * incremented in this case but 'i' will.
+ */
+ if (i != entrypc_idx)
+ break;
+ }
+
+ dwarf_lineaddr(line, postprologue_addr);
+ if (*postprologue_addr >= highpc)
+ dwarf_lineaddr(dwarf_onesrcline(lines, i - 1),
+ postprologue_addr);
+
+ return true;
+}
+
+static void __skip_prologue(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+ size_t nr_lines = 0;
+ unsigned long entrypc_idx = 0;
+ Dwarf_Lines *lines = NULL;
+ Dwarf_Addr postprologue_addr;
+ Dwarf_Addr highpc;
+
+ if (dwarf_highpc(sp_die, &highpc))
+ return;
+
+ if (dwarf_getsrclines(&pf->cu_die, &lines, &nr_lines))
+ return;
+
+ if (!get_entrypc_idx(lines, nr_lines, pf->addr, &entrypc_idx))
+ return;
+
+ if (!get_postprologue_addr(entrypc_idx, lines, nr_lines,
+ highpc, &postprologue_addr))
+ return;
+
+ pf->addr = postprologue_addr;
+}
+
+static void skip_prologue(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+ struct perf_probe_point *pp = &pf->pev->point;
+
+ /* Not uprobe? */
+ if (!pf->pev->uprobes)
+ return;
+
+ /* Compiled with optimization? */
+ if (optimized_target(&pf->cu_die))
+ return;
+
+ /* Don't know entrypc? */
+ if (!pf->addr)
+ return;
+
+ /* Only FUNC and FUNC@SRC are eligible. */
+ if (!pp->function || pp->line || pp->retprobe || pp->lazy_line ||
+ pp->offset || pp->abs_address)
+ return;
+
+ /* Not interested in func parameter? */
+ if (!perf_probe_with_var(pf->pev))
+ return;
+
+ pr_info("Target program is compiled without optimization. Skipping prologue.\n"
+ "Probe on address 0x%" PRIx64 " to force probing at the function entry.\n\n",
+ pf->addr);
+
+ __skip_prologue(sp_die, pf);
+}
+
static int probe_point_inline_cb(Dwarf_Die *in_die, void *data)
{
struct probe_finder *pf = data;
@@ -969,6 +1133,7 @@ static int probe_point_search_cb(Dwarf_Die *sp_die, void *data)
if (pp->lazy_line)
param->retval = find_probe_point_lazy(sp_die, pf);
else {
+ skip_prologue(sp_die, pf);
pf->addr += pp->offset;
/* TODO: Check the address in this function */
param->retval = call_probe_finder(sp_die, pf);
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* [PATCH 19/19] perf probe: Move dwarf specific functions to dwarf-aux.c
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (11 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 18/19] perf uprobe: Skip prologue if program compiled without optimization Arnaldo Carvalho de Melo
@ 2016-09-01 16:45 ` Arnaldo Carvalho de Melo
2016-09-05 13:16 ` [GIT PULL 00/19] perf/core improvements and fixes Ingo Molnar
13 siblings, 0 replies; 15+ messages in thread
From: Arnaldo Carvalho de Melo @ 2016-09-01 16:45 UTC (permalink / raw)
To: Ingo Molnar
Cc: linux-kernel, Ravi Bangoria, Alexander Shishkin, Hemant Kumar,
Naveen N . Rao, Peter Zijlstra, Wang Nan,
Arnaldo Carvalho de Melo
From: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Move generic dwarf related functions from util/probe-finder.c to
util/dwarf-aux.c. Functions name and their prototype are also changed
accordingly. No functionality changes.
Suggested-and-Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1472546377-25612-1-git-send-email-ravi.bangoria@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/dwarf-aux.c | 179 +++++++++++++++++++++++++++++++++++++++++
tools/perf/util/dwarf-aux.h | 8 ++
tools/perf/util/probe-finder.c | 136 +------------------------------
3 files changed, 189 insertions(+), 134 deletions(-)
diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c
index a347b19c961a..faec899435f2 100644
--- a/tools/perf/util/dwarf-aux.c
+++ b/tools/perf/util/dwarf-aux.c
@@ -1085,3 +1085,182 @@ int die_get_var_range(Dwarf_Die *sp_die __maybe_unused,
return -ENOTSUP;
}
#endif
+
+/*
+ * die_has_loclist - Check if DW_AT_location of @vr_die is a location list
+ * @vr_die: a variable DIE
+ */
+static bool die_has_loclist(Dwarf_Die *vr_die)
+{
+ Dwarf_Attribute loc;
+ int tag = dwarf_tag(vr_die);
+
+ if (tag != DW_TAG_formal_parameter &&
+ tag != DW_TAG_variable)
+ return false;
+
+ return (dwarf_attr_integrate(vr_die, DW_AT_location, &loc) &&
+ dwarf_whatform(&loc) == DW_FORM_sec_offset);
+}
+
+/*
+ * die_is_optimized_target - Check if target program is compiled with
+ * optimization
+ * @cu_die: a CU DIE
+ *
+ * For any object in given CU whose DW_AT_location is a location list,
+ * target program is compiled with optimization. This is applicable to
+ * clang as well.
+ */
+bool die_is_optimized_target(Dwarf_Die *cu_die)
+{
+ Dwarf_Die tmp_die;
+
+ if (die_has_loclist(cu_die))
+ return true;
+
+ if (!dwarf_child(cu_die, &tmp_die) &&
+ die_is_optimized_target(&tmp_die))
+ return true;
+
+ if (!dwarf_siblingof(cu_die, &tmp_die) &&
+ die_is_optimized_target(&tmp_die))
+ return true;
+
+ return false;
+}
+
+/*
+ * die_search_idx - Search index of given line address
+ * @lines: Line records of single CU
+ * @nr_lines: Number of @lines
+ * @addr: address we are looking for
+ * @idx: index to be set by this function (return value)
+ *
+ * Search for @addr by looping over every lines of CU. If address
+ * matches, set index of that line in @idx. Note that single source
+ * line can have multiple line records. i.e. single source line can
+ * have multiple index.
+ */
+static bool die_search_idx(Dwarf_Lines *lines, unsigned long nr_lines,
+ Dwarf_Addr addr, unsigned long *idx)
+{
+ unsigned long i;
+ Dwarf_Addr tmp;
+
+ for (i = 0; i < nr_lines; i++) {
+ if (dwarf_lineaddr(dwarf_onesrcline(lines, i), &tmp))
+ return false;
+
+ if (tmp == addr) {
+ *idx = i;
+ return true;
+ }
+ }
+ return false;
+}
+
+/*
+ * die_get_postprologue_addr - Search next address after function prologue
+ * @entrypc_idx: entrypc index
+ * @lines: Line records of single CU
+ * @nr_lines: Number of @lines
+ * @hignpc: high PC address of function
+ * @postprologue_addr: Next address after function prologue (return value)
+ *
+ * Look for prologue-end marker. If there is no explicit marker, return
+ * address of next line record or next source line.
+ */
+static bool die_get_postprologue_addr(unsigned long entrypc_idx,
+ Dwarf_Lines *lines,
+ unsigned long nr_lines,
+ Dwarf_Addr highpc,
+ Dwarf_Addr *postprologue_addr)
+{
+ unsigned long i;
+ int entrypc_lno, lno;
+ Dwarf_Line *line;
+ Dwarf_Addr addr;
+ bool p_end;
+
+ /* entrypc_lno is actual source line number */
+ line = dwarf_onesrcline(lines, entrypc_idx);
+ if (dwarf_lineno(line, &entrypc_lno))
+ return false;
+
+ for (i = entrypc_idx; i < nr_lines; i++) {
+ line = dwarf_onesrcline(lines, i);
+
+ if (dwarf_lineaddr(line, &addr) ||
+ dwarf_lineno(line, &lno) ||
+ dwarf_lineprologueend(line, &p_end))
+ return false;
+
+ /* highpc is exclusive. [entrypc,highpc) */
+ if (addr >= highpc)
+ break;
+
+ /* clang supports prologue-end marker */
+ if (p_end)
+ break;
+
+ /* Actual next line in source */
+ if (lno != entrypc_lno)
+ break;
+
+ /*
+ * Single source line can have multiple line records.
+ * For Example,
+ * void foo() { printf("hello\n"); }
+ * contains two line records. One points to declaration and
+ * other points to printf() line. Variable 'lno' won't get
+ * incremented in this case but 'i' will.
+ */
+ if (i != entrypc_idx)
+ break;
+ }
+
+ dwarf_lineaddr(line, postprologue_addr);
+ if (*postprologue_addr >= highpc)
+ dwarf_lineaddr(dwarf_onesrcline(lines, i - 1),
+ postprologue_addr);
+
+ return true;
+}
+
+/*
+ * die_skip_prologue - Use next address after prologue as probe location
+ * @sp_die: a subprogram DIE
+ * @cu_die: a CU DIE
+ * @entrypc: entrypc of the function
+ *
+ * Function prologue prepares stack and registers before executing function
+ * logic. When target program is compiled without optimization, function
+ * parameter information is only valid after prologue. When we probe entrypc
+ * of the function, and try to record function parameter, it contains
+ * garbage value.
+ */
+void die_skip_prologue(Dwarf_Die *sp_die, Dwarf_Die *cu_die,
+ Dwarf_Addr *entrypc)
+{
+ size_t nr_lines = 0;
+ unsigned long entrypc_idx = 0;
+ Dwarf_Lines *lines = NULL;
+ Dwarf_Addr postprologue_addr;
+ Dwarf_Addr highpc;
+
+ if (dwarf_highpc(sp_die, &highpc))
+ return;
+
+ if (dwarf_getsrclines(cu_die, &lines, &nr_lines))
+ return;
+
+ if (!die_search_idx(lines, nr_lines, *entrypc, &entrypc_idx))
+ return;
+
+ if (!die_get_postprologue_addr(entrypc_idx, lines, nr_lines,
+ highpc, &postprologue_addr))
+ return;
+
+ *entrypc = postprologue_addr;
+}
diff --git a/tools/perf/util/dwarf-aux.h b/tools/perf/util/dwarf-aux.h
index dc0ce1adb075..8b6d2f83af02 100644
--- a/tools/perf/util/dwarf-aux.h
+++ b/tools/perf/util/dwarf-aux.h
@@ -125,4 +125,12 @@ int die_get_typename(Dwarf_Die *vr_die, struct strbuf *buf);
/* Get the name and type of given variable DIE, stored as "type\tname" */
int die_get_varname(Dwarf_Die *vr_die, struct strbuf *buf);
int die_get_var_range(Dwarf_Die *sp_die, Dwarf_Die *vr_die, struct strbuf *buf);
+
+/* Check if target program is compiled with optimization */
+bool die_is_optimized_target(Dwarf_Die *cu_die);
+
+/* Use next address after prologue as probe location */
+void die_skip_prologue(Dwarf_Die *sp_die, Dwarf_Die *cu_die,
+ Dwarf_Addr *entrypc);
+
#endif
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 003ecadae35d..8daca4fc1f8d 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -907,138 +907,6 @@ static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf)
return die_walk_lines(sp_die, probe_point_lazy_walker, pf);
}
-static bool var_has_loclist(Dwarf_Die *cu_die)
-{
- Dwarf_Attribute loc;
- int tag = dwarf_tag(cu_die);
-
- if (tag != DW_TAG_formal_parameter &&
- tag != DW_TAG_variable)
- return false;
-
- return (dwarf_attr_integrate(cu_die, DW_AT_location, &loc) &&
- dwarf_whatform(&loc) == DW_FORM_sec_offset);
-}
-
-/*
- * For any object in given CU whose DW_AT_location is a location list,
- * target program is compiled with optimization.
- */
-static bool optimized_target(Dwarf_Die *cu_die)
-{
- Dwarf_Die tmp_die;
-
- if (var_has_loclist(cu_die))
- return true;
-
- if (!dwarf_child(cu_die, &tmp_die) && optimized_target(&tmp_die))
- return true;
-
- if (!dwarf_siblingof(cu_die, &tmp_die) && optimized_target(&tmp_die))
- return true;
-
- return false;
-}
-
-static bool get_entrypc_idx(Dwarf_Lines *lines, unsigned long nr_lines,
- Dwarf_Addr pf_addr, unsigned long *entrypc_idx)
-{
- unsigned long i;
- Dwarf_Addr addr;
-
- for (i = 0; i < nr_lines; i++) {
- if (dwarf_lineaddr(dwarf_onesrcline(lines, i), &addr))
- return false;
-
- if (addr == pf_addr) {
- *entrypc_idx = i;
- return true;
- }
- }
- return false;
-}
-
-static bool get_postprologue_addr(unsigned long entrypc_idx,
- Dwarf_Lines *lines,
- unsigned long nr_lines,
- Dwarf_Addr highpc,
- Dwarf_Addr *postprologue_addr)
-{
- unsigned long i;
- int entrypc_lno, lno;
- Dwarf_Line *line;
- Dwarf_Addr addr;
- bool p_end;
-
- /* entrypc_lno is actual source line number */
- line = dwarf_onesrcline(lines, entrypc_idx);
- if (dwarf_lineno(line, &entrypc_lno))
- return false;
-
- for (i = entrypc_idx; i < nr_lines; i++) {
- line = dwarf_onesrcline(lines, i);
-
- if (dwarf_lineaddr(line, &addr) ||
- dwarf_lineno(line, &lno) ||
- dwarf_lineprologueend(line, &p_end))
- return false;
-
- /* highpc is exclusive. [entrypc,highpc) */
- if (addr >= highpc)
- break;
-
- /* clang supports prologue-end marker */
- if (p_end)
- break;
-
- /* Actual next line in source */
- if (lno != entrypc_lno)
- break;
-
- /*
- * Single source line can have multiple line records.
- * For Example,
- * void foo() { printf("hello\n"); }
- * contains two line records. One points to declaration and
- * other points to printf() line. Variable 'lno' won't get
- * incremented in this case but 'i' will.
- */
- if (i != entrypc_idx)
- break;
- }
-
- dwarf_lineaddr(line, postprologue_addr);
- if (*postprologue_addr >= highpc)
- dwarf_lineaddr(dwarf_onesrcline(lines, i - 1),
- postprologue_addr);
-
- return true;
-}
-
-static void __skip_prologue(Dwarf_Die *sp_die, struct probe_finder *pf)
-{
- size_t nr_lines = 0;
- unsigned long entrypc_idx = 0;
- Dwarf_Lines *lines = NULL;
- Dwarf_Addr postprologue_addr;
- Dwarf_Addr highpc;
-
- if (dwarf_highpc(sp_die, &highpc))
- return;
-
- if (dwarf_getsrclines(&pf->cu_die, &lines, &nr_lines))
- return;
-
- if (!get_entrypc_idx(lines, nr_lines, pf->addr, &entrypc_idx))
- return;
-
- if (!get_postprologue_addr(entrypc_idx, lines, nr_lines,
- highpc, &postprologue_addr))
- return;
-
- pf->addr = postprologue_addr;
-}
-
static void skip_prologue(Dwarf_Die *sp_die, struct probe_finder *pf)
{
struct perf_probe_point *pp = &pf->pev->point;
@@ -1048,7 +916,7 @@ static void skip_prologue(Dwarf_Die *sp_die, struct probe_finder *pf)
return;
/* Compiled with optimization? */
- if (optimized_target(&pf->cu_die))
+ if (die_is_optimized_target(&pf->cu_die))
return;
/* Don't know entrypc? */
@@ -1068,7 +936,7 @@ static void skip_prologue(Dwarf_Die *sp_die, struct probe_finder *pf)
"Probe on address 0x%" PRIx64 " to force probing at the function entry.\n\n",
pf->addr);
- __skip_prologue(sp_die, pf);
+ die_skip_prologue(sp_die, &pf->cu_die, &pf->addr);
}
static int probe_point_inline_cb(Dwarf_Die *in_die, void *data)
--
2.7.4
^ permalink raw reply related [flat|nested] 15+ messages in thread* Re: [GIT PULL 00/19] perf/core improvements and fixes
2016-09-01 16:45 [GIT PULL 00/19] perf/core improvements and fixes Arnaldo Carvalho de Melo
` (12 preceding siblings ...)
2016-09-01 16:45 ` [PATCH 19/19] perf probe: Move dwarf specific functions to dwarf-aux.c Arnaldo Carvalho de Melo
@ 2016-09-05 13:16 ` Ingo Molnar
13 siblings, 0 replies; 15+ messages in thread
From: Ingo Molnar @ 2016-09-05 13:16 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: linux-kernel, Adrian Hunter, Alexander Shishkin, Anton Blanchard,
David Ahern, Hemant Kumar, Jiri Olsa, Masami Hiramatsu,
Michael Petlan, Milian Wolff, Namhyung Kim, Naveen N . Rao,
Peter Zijlstra, Ravi Bangoria, Shawn Lin, Wang Nan,
Yauheni Kaliuta, Arnaldo Carvalho de Melo
* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> Hi Ingo,
>
> Please consider pulling,
>
> - Arnaldo
>
> The following changes since commit 36e674a05164cdbb9d4a5b1b0b279fabae6c13bd:
>
> Merge tag 'perf-core-for-mingo-20160823' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2016-08-24 11:08:10 +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-20160901
>
> for you to fetch changes up to 6243b9dc4c991fe8bdc53a0e029908aef3ddb101:
>
> perf probe: Move dwarf specific functions to dwarf-aux.c (2016-09-01 12:42:26 -0300)
>
> ----------------------------------------------------------------
> perf/core improvements and fixes:
>
> User visible:
>
> - Support generating cross arch probes, i.e. if you specify a vmlinux
> file for different arch than the one in the host machine,
>
> $ perf probe --definition function_name args
>
> will generate the probe definition string needed to append to the
> target machine /sys/kernel/debug/tracing/kprobes_events file, using
> scripting (Masami Hiramatsu).
>
> - Make 'perf probe' skip the function prologue in uprobes if program
> compiled without optimization, using the same strategy as gdb and
> systemtap uses, fixing a bug where:
>
> $ perf probe -x ./test 'foo i'
>
> When 'foo(42)' was used on the "./test" executable would produce i=0
> instead of the expected i=42 (Ravi Bangoria)
>
> - Demangle symbols for synthesized @plt entries too (Millian Wolff)
>
> Documentation:
>
> - Show default report configuration in 'perf config' example
> and docs (Millian Wolff)
>
> Infrastructure:
>
> - Make 'perf test vmlinux' tolerate the symbol aliasing pruning done when
> loading kallsyms and vmlinux (Arnaldo Carvalho de Melo)
>
> - Improve output of 'perf test vmlinux' test, to help identify on the verbose
> output which lines are warning and which are errors (Arnaldo Carvalho de Melo)
>
> - Prep work to stop having to pass symbol_filter_t to lots of functions,
> simplifying symtab loading routines (Arnaldo Carvalho de Melo)
>
> - Honor symbol_conf.allow_aliases when loading kallsyms as well, it was using
> it only when loading vmlinux files (Arnaldo Carvalho de Melo)
>
> - Fixup symbol->end before doing alias pruning when loading symbol tables
> (Arnaldo Carvalho de Melo)
>
> - Fix error handling of lzma kernel module decompression (Shawn Lin)
>
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
>
> ----------------------------------------------------------------
> Arnaldo Carvalho de Melo (8):
> perf annotate: Initialize the priv are in symbol__new()
> perf symbols: Rename ->ignore to ->idle
> perf probe: Do not use map_load filters for function
> perf test vmlinux: Clarify which -v lines are errors or warning
> perf test vmlinux: Avoid printing headers for empty lists
> perf test vmlinux: Tolerate symbol aliases
> perf symbols: Check symbol_conf.allow_aliases for kallsyms loading too
> perf symbols: Fixup symbol sizes before picking best ones
>
> Masami Hiramatsu (5):
> perf probe: Remove unused tracing_dir variable
> perf probe: Show trace event definition
> perf probe: Ignore vmlinux buildid if offline kernel is given
> perf probe: Support probing on offline cross-arch binary
> perf probe: Ignore vmlinux Build-id when offline vmlinux given
>
> Milian Wolff (2):
> perf symbols: Demangle symbols for synthesized @plt entries.
> perf config: Show default report configuration in example and docs
>
> Ravi Bangoria (3):
> perf probe: Add helper function to check if probe with variable
> perf uprobe: Skip prologue if program compiled without optimization
> perf probe: Move dwarf specific functions to dwarf-aux.c
>
> Shawn Lin (1):
> perf tools: Fix error handling of lzma decompression
>
> tools/perf/Documentation/perf-config.txt | 8 +
> tools/perf/Documentation/perf-probe.txt | 9 ++
> tools/perf/Documentation/perfconfig.example | 9 ++
> tools/perf/arch/arm/include/dwarf-regs-table.h | 9 ++
> tools/perf/arch/arm64/include/dwarf-regs-table.h | 13 ++
> tools/perf/arch/powerpc/include/dwarf-regs-table.h | 27 ++++
> tools/perf/arch/s390/include/dwarf-regs-table.h | 8 +
> tools/perf/arch/sh/include/dwarf-regs-table.h | 25 +++
> tools/perf/arch/sparc/include/dwarf-regs-table.h | 18 +++
> tools/perf/arch/x86/include/dwarf-regs-table.h | 14 ++
> tools/perf/arch/xtensa/include/dwarf-regs-table.h | 8 +
> tools/perf/builtin-annotate.c | 7 +-
> tools/perf/builtin-probe.c | 35 +++-
> tools/perf/builtin-report.c | 6 +-
> tools/perf/builtin-top.c | 8 +-
> tools/perf/tests/vmlinux-kallsyms.c | 44 +++--
> tools/perf/util/Build | 1 +
> tools/perf/util/annotate.c | 7 -
> tools/perf/util/annotate.h | 1 -
> tools/perf/util/dwarf-aux.c | 179 +++++++++++++++++++++
> tools/perf/util/dwarf-aux.h | 8 +
> tools/perf/util/dwarf-regs.c | 59 +++++++
> tools/perf/util/evsel_fprintf.c | 4 +-
> tools/perf/util/include/dwarf-regs.h | 6 +
> tools/perf/util/lzma.c | 15 +-
> tools/perf/util/probe-event.c | 101 +++++++++---
> tools/perf/util/probe-event.h | 3 +
> tools/perf/util/probe-file.c | 5 +-
> tools/perf/util/probe-finder.c | 60 +++++--
> tools/perf/util/probe-finder.h | 1 +
> tools/perf/util/symbol-elf.c | 86 ++++++----
> tools/perf/util/symbol.c | 30 +++-
> tools/perf/util/symbol.h | 5 +-
> 33 files changed, 698 insertions(+), 121 deletions(-)
> create mode 100644 tools/perf/arch/arm/include/dwarf-regs-table.h
> create mode 100644 tools/perf/arch/arm64/include/dwarf-regs-table.h
> create mode 100644 tools/perf/arch/powerpc/include/dwarf-regs-table.h
> create mode 100644 tools/perf/arch/s390/include/dwarf-regs-table.h
> create mode 100644 tools/perf/arch/sh/include/dwarf-regs-table.h
> create mode 100644 tools/perf/arch/sparc/include/dwarf-regs-table.h
> create mode 100644 tools/perf/arch/x86/include/dwarf-regs-table.h
> create mode 100644 tools/perf/arch/xtensa/include/dwarf-regs-table.h
> create mode 100644 tools/perf/util/dwarf-regs.c
>
> Build stats:
>
> 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 fedora:20: Ok
> 10 fedora:21: Ok
> 11 fedora:22: Ok
> 12 fedora:23: Ok
> 13 fedora:24: Ok
> 14 fedora:24-x-ARC-uClibc: Ok
> 15 fedora:rawhide: Ok
> 16 mageia:5: Ok
> 17 opensuse:13.2: Ok
> 18 opensuse:42.1: Ok
> 19 opensuse:tumbleweed: Ok
> 20 ubuntu:12.04.5: Ok
> 21 ubuntu:14.04.4: Ok
> 22 ubuntu:15.10: Ok
> 23 ubuntu:16.04: Ok
> 24 ubuntu:16.04-x-arm: Ok
> 25 ubuntu:16.04-x-arm64: Ok
> 26 ubuntu:16.04-x-powerpc64: Ok
> 27 ubuntu:16.04-x-powerpc64el: Ok
> 28 ubuntu:16.10: Ok
> 29 ubuntu:16.10-x-s390: Ok
Pulled, thanks a lot Arnaldo!
Ingo
^ permalink raw reply [flat|nested] 15+ messages in thread