We decided to hide NULL metric units rather than showing it as "(null)",
but on hybrid systems if the process doesn't hit a PMU you get an empty
string metric unit instead. To make it consistent also remove empty
strings.
Note that metric-threshold is already hidden in this case without this
change.
Where a process only runs on cpu_core and never hits cpu_atom:
Before:
$ perf stat -j -- true
...
{"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00, "metric-value" : "0.000000", "metric-unit" : ""}
{"counter-value" : "6326.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 293786, "pcnt-running" : 100.00, "metric-value" : "3.553394", "metric-unit" : "of all branches", "metric-threshold" : "good"}
...
After:
...
{"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00}
{"counter-value" : "5778.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 282240, "pcnt-running" : 100.00, "metric-value" : "3.226797", "metric-unit" : "of all branches", "metric-threshold" : "good"}
...
Signed-off-by: James Clark <james.clark@linaro.org>
---
tools/perf/util/stat-display.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
index a5d72f4a515c..9b7fd985a42a 100644
--- a/tools/perf/util/stat-display.c
+++ b/tools/perf/util/stat-display.c
@@ -506,7 +506,7 @@ static void print_metric_json(struct perf_stat_config *config __maybe_unused,
struct outstate *os = ctx;
FILE *out = os->fh;
- if (unit) {
+ if (unit && strlen(unit)) {
json_out(os, "\"metric-value\" : \"%f\", \"metric-unit\" : \"%s\"", val, unit);
if (thresh != METRIC_THRESHOLD_UNKNOWN) {
json_out(os, "\"metric-threshold\" : \"%s\"",
--
2.34.1
Hello,
On Fri, Oct 25, 2024 at 10:03:05AM +0100, James Clark wrote:
> We decided to hide NULL metric units rather than showing it as "(null)",
> but on hybrid systems if the process doesn't hit a PMU you get an empty
> string metric unit instead. To make it consistent also remove empty
> strings.
>
> Note that metric-threshold is already hidden in this case without this
> change.
>
> Where a process only runs on cpu_core and never hits cpu_atom:
> Before:
> $ perf stat -j -- true
> ...
> {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00, "metric-value" : "0.000000", "metric-unit" : ""}
> {"counter-value" : "6326.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 293786, "pcnt-running" : 100.00, "metric-value" : "3.553394", "metric-unit" : "of all branches", "metric-threshold" : "good"}
> ...
I guess you're talking about "metric-unit", not plain "unit", right?
Then please update the subject line to reduce the config.
Ian, can you please review?
Thanks,
Namhyung
>
> After:
> ...
> {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00}
> {"counter-value" : "5778.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 282240, "pcnt-running" : 100.00, "metric-value" : "3.226797", "metric-unit" : "of all branches", "metric-threshold" : "good"}
> ...
>
> Signed-off-by: James Clark <james.clark@linaro.org>
> ---
> tools/perf/util/stat-display.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
> index a5d72f4a515c..9b7fd985a42a 100644
> --- a/tools/perf/util/stat-display.c
> +++ b/tools/perf/util/stat-display.c
> @@ -506,7 +506,7 @@ static void print_metric_json(struct perf_stat_config *config __maybe_unused,
> struct outstate *os = ctx;
> FILE *out = os->fh;
>
> - if (unit) {
> + if (unit && strlen(unit)) {
> json_out(os, "\"metric-value\" : \"%f\", \"metric-unit\" : \"%s\"", val, unit);
> if (thresh != METRIC_THRESHOLD_UNKNOWN) {
> json_out(os, "\"metric-threshold\" : \"%s\"",
> --
> 2.34.1
>
On Tue, Oct 29, 2024 at 5:42 PM Namhyung Kim <namhyung@kernel.org> wrote:
>
> Hello,
>
> On Fri, Oct 25, 2024 at 10:03:05AM +0100, James Clark wrote:
> > We decided to hide NULL metric units rather than showing it as "(null)",
> > but on hybrid systems if the process doesn't hit a PMU you get an empty
> > string metric unit instead. To make it consistent also remove empty
> > strings.
> >
> > Note that metric-threshold is already hidden in this case without this
> > change.
> >
> > Where a process only runs on cpu_core and never hits cpu_atom:
> > Before:
> > $ perf stat -j -- true
> > ...
> > {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00, "metric-value" : "0.000000", "metric-unit" : ""}
> > {"counter-value" : "6326.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 293786, "pcnt-running" : 100.00, "metric-value" : "3.553394", "metric-unit" : "of all branches", "metric-threshold" : "good"}
> > ...
>
> I guess you're talking about "metric-unit", not plain "unit", right?
> Then please update the subject line to reduce the config.
>
> Ian, can you please review?
It'd be nice to see the stack trace for when metric-unit is "" as I'm
not seeing the logic in stat-shadow.c. If we know the caller than it
seems logical the unit can be passed as NULL rather than "".
Thanks,
Ian
> Thanks,
> Namhyung
>
> >
> > After:
> > ...
> > {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00}
> > {"counter-value" : "5778.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 282240, "pcnt-running" : 100.00, "metric-value" : "3.226797", "metric-unit" : "of all branches", "metric-threshold" : "good"}
> > ...
> >
> > Signed-off-by: James Clark <james.clark@linaro.org>
> > ---
> > tools/perf/util/stat-display.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
> > index a5d72f4a515c..9b7fd985a42a 100644
> > --- a/tools/perf/util/stat-display.c
> > +++ b/tools/perf/util/stat-display.c
> > @@ -506,7 +506,7 @@ static void print_metric_json(struct perf_stat_config *config __maybe_unused,
> > struct outstate *os = ctx;
> > FILE *out = os->fh;
> >
> > - if (unit) {
> > + if (unit && strlen(unit)) {
> > json_out(os, "\"metric-value\" : \"%f\", \"metric-unit\" : \"%s\"", val, unit);
> > if (thresh != METRIC_THRESHOLD_UNKNOWN) {
> > json_out(os, "\"metric-threshold\" : \"%s\"",
> > --
> > 2.34.1
> >
On 30/10/2024 2:45 am, Ian Rogers wrote:
> On Tue, Oct 29, 2024 at 5:42 PM Namhyung Kim <namhyung@kernel.org> wrote:
>>
>> Hello,
>>
>> On Fri, Oct 25, 2024 at 10:03:05AM +0100, James Clark wrote:
>>> We decided to hide NULL metric units rather than showing it as "(null)",
>>> but on hybrid systems if the process doesn't hit a PMU you get an empty
>>> string metric unit instead. To make it consistent also remove empty
>>> strings.
>>>
>>> Note that metric-threshold is already hidden in this case without this
>>> change.
>>>
>>> Where a process only runs on cpu_core and never hits cpu_atom:
>>> Before:
>>> $ perf stat -j -- true
>>> ...
>>> {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00, "metric-value" : "0.000000", "metric-unit" : ""}
>>> {"counter-value" : "6326.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 293786, "pcnt-running" : 100.00, "metric-value" : "3.553394", "metric-unit" : "of all branches", "metric-threshold" : "good"}
>>> ...
>>
>> I guess you're talking about "metric-unit", not plain "unit", right?
>> Then please update the subject line to reduce the config.
>>
Yep I'll update it.
>> Ian, can you please review?
>
> It'd be nice to see the stack trace for when metric-unit is "" as I'm
> not seeing the logic in stat-shadow.c. If we know the caller than it
> seems logical the unit can be passed as NULL rather than "".
>
> Thanks,
> Ian
>
Here's the stack:
print_metric_json() (stat-display.c:516)
printout() (stat-display.c:912)
print_counter_aggrdata() (stat-display.c:1110)
print_counter() (stat-display.c:1224)
evlist__print_counters() (stat-display.c:1734)
print_counters() (builtin-stat.c:1016)
cmd_stat() (builtin-stat.c:2872)
run_builtin() (perf/perf.c:351)
handle_internal_command() (perf.c:404)
run_argv() (perf.c:448)
main() (perf.c:560)
The empty string is from printout():
pm(config, os, METRIC_THRESHOLD_UNKNOWN, /*format=*/NULL, /*unit=*/"",
/*val=*/0);
Changing it to NULL seems to work, so is probably a bit neater. I can do
that if you think that makes sense?
There's another one for --metric-only that I didn't run into, but it
could also make sense to change:
if (config->metric_only) {
pm(config, os, METRIC_THRESHOLD_UNKNOWN, "", "", 0);
>> Thanks,
>> Namhyung
>>
>>>
>>> After:
>>> ...
>>> {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00}
>>> {"counter-value" : "5778.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 282240, "pcnt-running" : 100.00, "metric-value" : "3.226797", "metric-unit" : "of all branches", "metric-threshold" : "good"}
>>> ...
>>>
>>> Signed-off-by: James Clark <james.clark@linaro.org>
>>> ---
>>> tools/perf/util/stat-display.c | 2 +-
>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
>>> index a5d72f4a515c..9b7fd985a42a 100644
>>> --- a/tools/perf/util/stat-display.c
>>> +++ b/tools/perf/util/stat-display.c
>>> @@ -506,7 +506,7 @@ static void print_metric_json(struct perf_stat_config *config __maybe_unused,
>>> struct outstate *os = ctx;
>>> FILE *out = os->fh;
>>>
>>> - if (unit) {
>>> + if (unit && strlen(unit)) {
>>> json_out(os, "\"metric-value\" : \"%f\", \"metric-unit\" : \"%s\"", val, unit);
>>> if (thresh != METRIC_THRESHOLD_UNKNOWN) {
>>> json_out(os, "\"metric-threshold\" : \"%s\"",
>>> --
>>> 2.34.1
>>>
On Wed, Oct 30, 2024 at 2:36 AM James Clark <james.clark@linaro.org> wrote:
>
>
>
> On 30/10/2024 2:45 am, Ian Rogers wrote:
> > On Tue, Oct 29, 2024 at 5:42 PM Namhyung Kim <namhyung@kernel.org> wrote:
> >>
> >> Hello,
> >>
> >> On Fri, Oct 25, 2024 at 10:03:05AM +0100, James Clark wrote:
> >>> We decided to hide NULL metric units rather than showing it as "(null)",
> >>> but on hybrid systems if the process doesn't hit a PMU you get an empty
> >>> string metric unit instead. To make it consistent also remove empty
> >>> strings.
> >>>
> >>> Note that metric-threshold is already hidden in this case without this
> >>> change.
> >>>
> >>> Where a process only runs on cpu_core and never hits cpu_atom:
> >>> Before:
> >>> $ perf stat -j -- true
> >>> ...
> >>> {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00, "metric-value" : "0.000000", "metric-unit" : ""}
> >>> {"counter-value" : "6326.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 293786, "pcnt-running" : 100.00, "metric-value" : "3.553394", "metric-unit" : "of all branches", "metric-threshold" : "good"}
> >>> ...
> >>
> >> I guess you're talking about "metric-unit", not plain "unit", right?
> >> Then please update the subject line to reduce the config.
> >>
>
> Yep I'll update it.
>
> >> Ian, can you please review?
> >
> > It'd be nice to see the stack trace for when metric-unit is "" as I'm
> > not seeing the logic in stat-shadow.c. If we know the caller than it
> > seems logical the unit can be passed as NULL rather than "".
> >
> > Thanks,
> > Ian
> >
>
>
> Here's the stack:
>
> print_metric_json() (stat-display.c:516)
> printout() (stat-display.c:912)
> print_counter_aggrdata() (stat-display.c:1110)
> print_counter() (stat-display.c:1224)
> evlist__print_counters() (stat-display.c:1734)
> print_counters() (builtin-stat.c:1016)
> cmd_stat() (builtin-stat.c:2872)
> run_builtin() (perf/perf.c:351)
> handle_internal_command() (perf.c:404)
> run_argv() (perf.c:448)
> main() (perf.c:560)
>
> The empty string is from printout():
>
> pm(config, os, METRIC_THRESHOLD_UNKNOWN, /*format=*/NULL, /*unit=*/"",
> /*val=*/0);
>
> Changing it to NULL seems to work, so is probably a bit neater. I can do
> that if you think that makes sense?
>
> There's another one for --metric-only that I didn't run into, but it
> could also make sense to change:
>
> if (config->metric_only) {
> pm(config, os, METRIC_THRESHOLD_UNKNOWN, "", "", 0);
Yeah, it looks like all the callers should be passing NULL. I feel
more comfortable doing this than ignoring the empty string which is
something a json metric may accidentally do. Could you send a patch?
Thanks,
Ian
> >> Thanks,
> >> Namhyung
> >>
> >>>
> >>> After:
> >>> ...
> >>> {"counter-value" : "<not counted>", "unit" : "", "event" : "cpu_atom/branch-misses/", "event-runtime" : 0, "pcnt-running" : 0.00}
> >>> {"counter-value" : "5778.000000", "unit" : "", "event" : "cpu_core/branch-misses/", "event-runtime" : 282240, "pcnt-running" : 100.00, "metric-value" : "3.226797", "metric-unit" : "of all branches", "metric-threshold" : "good"}
> >>> ...
> >>>
> >>> Signed-off-by: James Clark <james.clark@linaro.org>
> >>> ---
> >>> tools/perf/util/stat-display.c | 2 +-
> >>> 1 file changed, 1 insertion(+), 1 deletion(-)
> >>>
> >>> diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
> >>> index a5d72f4a515c..9b7fd985a42a 100644
> >>> --- a/tools/perf/util/stat-display.c
> >>> +++ b/tools/perf/util/stat-display.c
> >>> @@ -506,7 +506,7 @@ static void print_metric_json(struct perf_stat_config *config __maybe_unused,
> >>> struct outstate *os = ctx;
> >>> FILE *out = os->fh;
> >>>
> >>> - if (unit) {
> >>> + if (unit && strlen(unit)) {
> >>> json_out(os, "\"metric-value\" : \"%f\", \"metric-unit\" : \"%s\"", val, unit);
> >>> if (thresh != METRIC_THRESHOLD_UNKNOWN) {
> >>> json_out(os, "\"metric-threshold\" : \"%s\"",
> >>> --
> >>> 2.34.1
> >>>
>
© 2016 - 2026 Red Hat, Inc.