* [PATCH] perf script: Fix --call-trace for Intel PT
@ 2020-05-27 18:02 Adrian Hunter
2020-05-28 14:32 ` Arnaldo Carvalho de Melo
0 siblings, 1 reply; 2+ messages in thread
From: Adrian Hunter @ 2020-05-27 18:02 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo; +Cc: Jiri Olsa, linux-kernel
Make process_attr() respect -F-ip, noting also that the condition in
process_attr() (callchain_param.record_mode != CALLCHAIN_NONE) is always
true so test the sample type directly.
Example:
Before:
$ perf record -e intel_pt//u uname
Linux
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.033 MB perf.data ]
$ perf script --call-trace | head -5
uname 30992 [006] 41758.313696574: cbr: 42 freq: 4219 MHz (156%) 0 [unknown] ([unknown] )
uname 30992 [006] 41758.313696907: _start 7f71792c4100 _start+0x0 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
uname 30992 [006] 41758.313699574: _dl_start 7f71792c4103 _start+0x3 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
uname 30992 [006] 41758.313699907: _dl_start 7f71792c4e18 _dl_start+0x28 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
uname 30992 [006] 41758.313701574: _dl_start 7f71792c5128 _dl_start+0x338 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
After:
$ perf script --call-trace | head -5
uname 30992 [006] 41758.313696574: cbr: 42 freq: 4219 MHz (156%)
uname 30992 [006] 41758.313696907: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _start
uname 30992 [006] 41758.313699574: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _dl_start
uname 30992 [006] 41758.313699907: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _dl_start
uname 30992 [006] 41758.313701574: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _dl_start
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Fixes: f288e8e1aa4f ("perf script: Enable IP fields for callchains")
---
tools/perf/builtin-script.c | 19 +++++++++++++++----
1 file changed, 15 insertions(+), 4 deletions(-)
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index 5bdd1a393399..3249ead2deef 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -167,6 +167,7 @@ static struct {
u64 fields;
u64 invalid_fields;
u64 user_set_fields;
+ u64 user_unset_fields;
} output[OUTPUT_TYPE_MAX] = {
[PERF_TYPE_HARDWARE] = {
@@ -2132,10 +2133,18 @@ static int process_attr(struct perf_tool *tool, union perf_event *event,
sample_type = perf_evlist__combined_sample_type(evlist);
callchain_param_setup(sample_type);
- /* Enable fields for callchain entries, if it got enabled. */
- if (callchain_param.record_mode != CALLCHAIN_NONE) {
- output[output_type(evsel->core.attr.type)].fields |= PERF_OUTPUT_IP |
- PERF_OUTPUT_SYM;
+ /* Enable fields for callchain entries */
+ if (symbol_conf.use_callchain &&
+ (sample_type & PERF_SAMPLE_CALLCHAIN ||
+ sample_type & PERF_SAMPLE_BRANCH_STACK ||
+ (sample_type & PERF_SAMPLE_REGS_USER &&
+ sample_type & PERF_SAMPLE_STACK_USER))) {
+ int type = output_type(evsel->core.attr.type);
+
+ if (!(output[type].user_unset_fields & PERF_OUTPUT_IP))
+ output[type].fields |= PERF_OUTPUT_IP;
+ if (!(output[type].user_unset_fields & PERF_OUTPUT_SYM))
+ output[type].fields |= PERF_OUTPUT_SYM;
}
set_print_ip_opts(&evsel->core.attr);
return 0;
@@ -2704,9 +2713,11 @@ static int parse_output_fields(const struct option *opt __maybe_unused,
if (change == REMOVE) {
output[j].fields &= ~all_output_options[i].field;
output[j].user_set_fields &= ~all_output_options[i].field;
+ output[j].user_unset_fields |= all_output_options[i].field;
} else {
output[j].fields |= all_output_options[i].field;
output[j].user_set_fields |= all_output_options[i].field;
+ output[j].user_unset_fields &= ~all_output_options[i].field;
}
output[j].user_set = true;
output[j].wildcard_set = true;
--
2.17.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] perf script: Fix --call-trace for Intel PT
2020-05-27 18:02 [PATCH] perf script: Fix --call-trace for Intel PT Adrian Hunter
@ 2020-05-28 14:32 ` Arnaldo Carvalho de Melo
0 siblings, 0 replies; 2+ messages in thread
From: Arnaldo Carvalho de Melo @ 2020-05-28 14:32 UTC (permalink / raw)
To: Adrian Hunter; +Cc: Jiri Olsa, linux-kernel
Em Wed, May 27, 2020 at 09:02:50PM +0300, Adrian Hunter escreveu:
> Make process_attr() respect -F-ip, noting also that the condition in
> process_attr() (callchain_param.record_mode != CALLCHAIN_NONE) is always
> true so test the sample type directly.
>
> Example:
>
> Before:
>
> $ perf record -e intel_pt//u uname
> Linux
> [ perf record: Woken up 1 times to write data ]
> [ perf record: Captured and wrote 0.033 MB perf.data ]
> $ perf script --call-trace | head -5
> uname 30992 [006] 41758.313696574: cbr: 42 freq: 4219 MHz (156%) 0 [unknown] ([unknown] )
> uname 30992 [006] 41758.313696907: _start 7f71792c4100 _start+0x0 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
> uname 30992 [006] 41758.313699574: _dl_start 7f71792c4103 _start+0x3 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
> uname 30992 [006] 41758.313699907: _dl_start 7f71792c4e18 _dl_start+0x28 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
> uname 30992 [006] 41758.313701574: _dl_start 7f71792c5128 _dl_start+0x338 (/usr/lib/x86_64-linux-gnu/ld-2.31.so )
>
> After:
>
> $ perf script --call-trace | head -5
> uname 30992 [006] 41758.313696574: cbr: 42 freq: 4219 MHz (156%)
> uname 30992 [006] 41758.313696907: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _start
> uname 30992 [006] 41758.313699574: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _dl_start
> uname 30992 [006] 41758.313699907: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _dl_start
> uname 30992 [006] 41758.313701574: (/usr/lib/x86_64-linux-gnu/ld-2.31.so ) _dl_start
Thanks, tested and applied.
- Arnaldo
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> Fixes: f288e8e1aa4f ("perf script: Enable IP fields for callchains")
> ---
> tools/perf/builtin-script.c | 19 +++++++++++++++----
> 1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
> index 5bdd1a393399..3249ead2deef 100644
> --- a/tools/perf/builtin-script.c
> +++ b/tools/perf/builtin-script.c
> @@ -167,6 +167,7 @@ static struct {
> u64 fields;
> u64 invalid_fields;
> u64 user_set_fields;
> + u64 user_unset_fields;
> } output[OUTPUT_TYPE_MAX] = {
>
> [PERF_TYPE_HARDWARE] = {
> @@ -2132,10 +2133,18 @@ static int process_attr(struct perf_tool *tool, union perf_event *event,
> sample_type = perf_evlist__combined_sample_type(evlist);
> callchain_param_setup(sample_type);
>
> - /* Enable fields for callchain entries, if it got enabled. */
> - if (callchain_param.record_mode != CALLCHAIN_NONE) {
> - output[output_type(evsel->core.attr.type)].fields |= PERF_OUTPUT_IP |
> - PERF_OUTPUT_SYM;
> + /* Enable fields for callchain entries */
> + if (symbol_conf.use_callchain &&
> + (sample_type & PERF_SAMPLE_CALLCHAIN ||
> + sample_type & PERF_SAMPLE_BRANCH_STACK ||
> + (sample_type & PERF_SAMPLE_REGS_USER &&
> + sample_type & PERF_SAMPLE_STACK_USER))) {
> + int type = output_type(evsel->core.attr.type);
> +
> + if (!(output[type].user_unset_fields & PERF_OUTPUT_IP))
> + output[type].fields |= PERF_OUTPUT_IP;
> + if (!(output[type].user_unset_fields & PERF_OUTPUT_SYM))
> + output[type].fields |= PERF_OUTPUT_SYM;
> }
> set_print_ip_opts(&evsel->core.attr);
> return 0;
> @@ -2704,9 +2713,11 @@ static int parse_output_fields(const struct option *opt __maybe_unused,
> if (change == REMOVE) {
> output[j].fields &= ~all_output_options[i].field;
> output[j].user_set_fields &= ~all_output_options[i].field;
> + output[j].user_unset_fields |= all_output_options[i].field;
> } else {
> output[j].fields |= all_output_options[i].field;
> output[j].user_set_fields |= all_output_options[i].field;
> + output[j].user_unset_fields &= ~all_output_options[i].field;
> }
> output[j].user_set = true;
> output[j].wildcard_set = true;
> --
> 2.17.1
>
--
- Arnaldo
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2020-05-28 14:32 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-27 18:02 [PATCH] perf script: Fix --call-trace for Intel PT Adrian Hunter
2020-05-28 14:32 ` Arnaldo Carvalho de Melo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).