kernel/trace/trace.c | 3 + kernel/trace/trace_events_hist.c | 110 ++++++++++++++++++++++++++++++++++++-- 2 files changed, 105 insertions(+), 8 deletions(-)
Hi,
Here are a couple of patches to add .percent and .graph histogram
value suffixes to show the value in percentage and in bar-graph.
This will help us to check the trend of the histogram instantly
without the post processing tool.
Here shows the example of the percentage and the bar graph of
the runtime of the running tasks.
/sys/kernel/tracing # echo hist:keys=pid:vals=runtime.percent,runtime.graph:sort
=pid >> events/sched/sched_stat_runtime/trigger
/sys/kernel/tracing # sleep 10
/sys/kernel/tracing # cat events/sched/sched_stat_runtime/hist
# event histogram
#
# trigger info: hist:keys=pid:vals=hitcount,runtime.percent,runtime.graph:sort=pid:size=2048 [active]
#
{ pid: 8 } hitcount: 11 runtime: 4.11 runtime: #
{ pid: 9 } hitcount: 4 runtime: 1.28 runtime:
{ pid: 14 } hitcount: 10 runtime: 2.22 runtime:
{ pid: 15 } hitcount: 1 runtime: 0.07 runtime:
{ pid: 16 } hitcount: 21 runtime: 3.35 runtime: #
{ pid: 57 } hitcount: 6 runtime: 2.41 runtime: #
{ pid: 61 } hitcount: 42 runtime: 9.79 runtime: ####
{ pid: 66 } hitcount: 5 runtime: 0.69 runtime:
{ pid: 147 } hitcount: 36 runtime: 45.33 runtime: ####################
{ pid: 8548 } hitcount: 9 runtime: 17.25 runtime: #######
{ pid: 8549 } hitcount: 8 runtime: 13.43 runtime: #####
Totals:
Hits: 153
Entries: 11
Dropped: 0
Thank you,
---
Masami Hiramatsu (Google) (2):
tracing: Add .percent suffix option to histogram values
tracing: Add .graph suffix option to histogram value
kernel/trace/trace.c | 3 +
kernel/trace/trace_events_hist.c | 110 ++++++++++++++++++++++++++++++++++++--
2 files changed, 105 insertions(+), 8 deletions(-)
--
Masami Hiramatsu (Google) <mhiramat@kernel.org>
On Mon, 1 Aug 2022 12:00:40 +0900
"Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:
> Here are a couple of patches to add .percent and .graph histogram
> value suffixes to show the value in percentage and in bar-graph.
>
> This will help us to check the trend of the histogram instantly
> without the post processing tool.
>
> Here shows the example of the percentage and the bar graph of
> the runtime of the running tasks.
>
> /sys/kernel/tracing # echo hist:keys=pid:vals=runtime.percent,runtime.graph:sort
> =pid >> events/sched/sched_stat_runtime/trigger
> /sys/kernel/tracing # sleep 10
> /sys/kernel/tracing # cat events/sched/sched_stat_runtime/hist
> # event histogram
> #
> # trigger info: hist:keys=pid:vals=hitcount,runtime.percent,runtime.graph:sort=pid:size=2048 [active]
> #
>
> { pid: 8 } hitcount: 11 runtime: 4.11 runtime: #
> { pid: 9 } hitcount: 4 runtime: 1.28 runtime:
> { pid: 14 } hitcount: 10 runtime: 2.22 runtime:
> { pid: 15 } hitcount: 1 runtime: 0.07 runtime:
> { pid: 16 } hitcount: 21 runtime: 3.35 runtime: #
> { pid: 57 } hitcount: 6 runtime: 2.41 runtime: #
> { pid: 61 } hitcount: 42 runtime: 9.79 runtime: ####
> { pid: 66 } hitcount: 5 runtime: 0.69 runtime:
> { pid: 147 } hitcount: 36 runtime: 45.33 runtime: ####################
> { pid: 8548 } hitcount: 9 runtime: 17.25 runtime: #######
> { pid: 8549 } hitcount: 8 runtime: 13.43 runtime: #####
>
> Totals:
> Hits: 153
> Entries: 11
> Dropped: 0
>
Hi Masami,
Thanks for this as well. But as it's a new feature, and I freeze new
features when the merge window starts, I'll add this to my queue for the
next window.
-- Steve
On Mon, 1 Aug 2022 16:58:40 -0400
Steven Rostedt <rostedt@goodmis.org> wrote:
> On Mon, 1 Aug 2022 12:00:40 +0900
> "Masami Hiramatsu (Google)" <mhiramat@kernel.org> wrote:
>
> > Here are a couple of patches to add .percent and .graph histogram
> > value suffixes to show the value in percentage and in bar-graph.
> >
> > This will help us to check the trend of the histogram instantly
> > without the post processing tool.
> >
> > Here shows the example of the percentage and the bar graph of
> > the runtime of the running tasks.
> >
> > /sys/kernel/tracing # echo hist:keys=pid:vals=runtime.percent,runtime.graph:sort
> > =pid >> events/sched/sched_stat_runtime/trigger
> > /sys/kernel/tracing # sleep 10
> > /sys/kernel/tracing # cat events/sched/sched_stat_runtime/hist
> > # event histogram
> > #
> > # trigger info: hist:keys=pid:vals=hitcount,runtime.percent,runtime.graph:sort=pid:size=2048 [active]
> > #
> >
> > { pid: 8 } hitcount: 11 runtime: 4.11 runtime: #
> > { pid: 9 } hitcount: 4 runtime: 1.28 runtime:
> > { pid: 14 } hitcount: 10 runtime: 2.22 runtime:
> > { pid: 15 } hitcount: 1 runtime: 0.07 runtime:
> > { pid: 16 } hitcount: 21 runtime: 3.35 runtime: #
> > { pid: 57 } hitcount: 6 runtime: 2.41 runtime: #
> > { pid: 61 } hitcount: 42 runtime: 9.79 runtime: ####
> > { pid: 66 } hitcount: 5 runtime: 0.69 runtime:
> > { pid: 147 } hitcount: 36 runtime: 45.33 runtime: ####################
> > { pid: 8548 } hitcount: 9 runtime: 17.25 runtime: #######
> > { pid: 8549 } hitcount: 8 runtime: 13.43 runtime: #####
> >
> > Totals:
> > Hits: 153
> > Entries: 11
> > Dropped: 0
> >
>
> Hi Masami,
>
> Thanks for this as well. But as it's a new feature, and I freeze new
> features when the merge window starts, I'll add this to my queue for the
> next window.
Yeah, that's no problem. Please queue it for the next window :)
Thank you!
>
> -- Steve
--
Masami Hiramatsu (Google) <mhiramat@kernel.org>
© 2016 - 2026 Red Hat, Inc.