Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1517151 > unrolled thread

[PATCH 0/4] perf tools: Assorted fixes for hierarchy mode

Started byNamhyung Kim <namhyung@kernel.org>
First post2016-11-08 14:10 +0100
Last post2016-11-09 14:20 +0100
Articles 13 — 3 participants

Back to article view | Back to linux.kernel

Contents

  [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Namhyung Kim <namhyung@kernel.org> - 2016-11-08 14:10 +0100
    [PATCH 4/4] perf hists: Fix column length on --hierarchy Namhyung Kim <namhyung@kernel.org> - 2016-11-08 14:20 +0100
    [PATCH 2/4] perf hist browser: Show folded sign properly on --hierarchy Namhyung Kim <namhyung@kernel.org> - 2016-11-08 14:20 +0100
      Re: [PATCH 2/4] perf hist browser: Show folded sign properly on  --hierarchy Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-11-09 15:30 +0100
        Re: [PATCH 2/4] perf hist browser: Show folded sign properly on  --hierarchy Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-11-09 15:30 +0100
    Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Markus Trippelsdorf <markus@trippelsdorf.de> - 2016-11-08 14:30 +0100
      Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-11-09 14:20 +0100
        Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Markus Trippelsdorf <markus@trippelsdorf.de> - 2016-11-09 14:20 +0100
    Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Markus Trippelsdorf <markus@trippelsdorf.de> - 2016-11-08 14:50 +0100
      Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Namhyung Kim <namhyung@kernel.org> - 2016-11-08 16:10 +0100
        Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Markus Trippelsdorf <markus@trippelsdorf.de> - 2016-11-08 16:20 +0100
          Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Markus Trippelsdorf <markus@trippelsdorf.de> - 2016-11-09 14:20 +0100
          Re: [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-11-09 14:20 +0100

#1517151 — [PATCH 0/4] perf tools: Assorted fixes for hierarchy mode

FromNamhyung Kim <namhyung@kernel.org>
Date2016-11-08 14:10 +0100
Subject[PATCH 0/4] perf tools: Assorted fixes for hierarchy mode
Message-ID<sBek2-1kJ-21@gated-at.bofh.it>
Hello,

This patches fix problems in hierarchy output Markus reported some
time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
in my tree:

  git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git

Any feedbacks are welcomed.

Thanks,
Namhyung


Cc: Markus Trippelsdorf <markus@trippelsdorf.de>

Namhyung Kim (4):
  perf hist browser: Fix indentation of folded sign on --hierarchy
  perf hist browser: Show folded sign properly on --hierarchy
  perf hist browser: Fix column indentation on --hierarchy
  perf hists: Fix column length on --hierarchy

 tools/perf/ui/browsers/hists.c | 35 ++++++++++++++++++++++++-----------
 tools/perf/util/hist.c         | 12 ++++++------
 2 files changed, 30 insertions(+), 17 deletions(-)

-- 
2.10.1

[toc] | [next] | [standalone]

#1517152 — [PATCH 4/4] perf hists: Fix column length on --hierarchy

FromNamhyung Kim <namhyung@kernel.org>
Date2016-11-08 14:20 +0100
Subject[PATCH 4/4] perf hists: Fix column length on --hierarchy
Message-ID<sBetH-1nU-3@gated-at.bofh.it>
In reply to#1517151
Markus reported that there's a weird behavior on perf top --hierarch
regarding the column length.  Looking at the code, I found a debious
code which affects the symtoms.  When --hierarchy option is used, the
last column length might be inaccurate since it skips to update the
length on leaf entries.  I cannot remember why it did and looks like a
leftover from previous version during the development.  Anyway updating
the column length often is not harmful.  So let's move the code out.

Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/util/hist.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index e1be4132054d..6770a9645609 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -1601,18 +1601,18 @@ static void hists__hierarchy_output_resort(struct hists *hists,
 		if (prog)
 			ui_progress__update(prog, 1);
 
+		hists->nr_entries++;
+		if (!he->filtered) {
+			hists->nr_non_filtered_entries++;
+			hists__calc_col_len(hists, he);
+		}
+
 		if (!he->leaf) {
 			hists__hierarchy_output_resort(hists, prog,
 						       &he->hroot_in,
 						       &he->hroot_out,
 						       min_callchain_hits,
 						       use_callchain);
-			hists->nr_entries++;
-			if (!he->filtered) {
-				hists->nr_non_filtered_entries++;
-				hists__calc_col_len(hists, he);
-			}
-
 			continue;
 		}
 
-- 
2.10.1

[toc] | [prev] | [next] | [standalone]

#1517156 — [PATCH 2/4] perf hist browser: Show folded sign properly on --hierarchy

FromNamhyung Kim <namhyung@kernel.org>
Date2016-11-08 14:20 +0100
Subject[PATCH 2/4] perf hist browser: Show folded sign properly on --hierarchy
Message-ID<sBetH-1nU-13@gated-at.bofh.it>
In reply to#1517151
When horizontall scrolling is used in hierarchy mode, the folded signed
disappears at the right most column.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/ui/browsers/hists.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
index fe5677ccbc22..7722ad311318 100644
--- a/tools/perf/ui/browsers/hists.c
+++ b/tools/perf/ui/browsers/hists.c
@@ -1381,8 +1381,14 @@ static int hist_browser__show_hierarchy_entry(struct hist_browser *browser,
 		}
 
 		perf_hpp_list__for_each_format(entry->hpp_list, fmt) {
-			ui_browser__write_nstring(&browser->b, "", 2);
-			width -= 2;
+			if (first) {
+				ui_browser__printf(&browser->b, "%c ", folded_sign);
+				width -= 2;
+				first = false;
+			} else {
+				ui_browser__write_nstring(&browser->b, "", 2);
+				width -= 2;
+			}
 
 			/*
 			 * No need to call hist_entry__snprintf_alignment()
-- 
2.10.1

[toc] | [prev] | [next] | [standalone]

#1518202 — Re: [PATCH 2/4] perf hist browser: Show folded sign properly on --hierarchy

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2016-11-09 15:30 +0100
SubjectRe: [PATCH 2/4] perf hist browser: Show folded sign properly on --hierarchy
Message-ID<sBC30-8sr-13@gated-at.bofh.it>
In reply to#1517156
Em Tue, Nov 08, 2016 at 10:08:31PM +0900, Namhyung Kim escreveu:
> When horizontall scrolling is used in hierarchy mode, the folded signed
> disappears at the right most column.

Humm, this indeed shows the '+' folded signal after pressing ->, but it
moves from the first to the third column :-\

- Arnaldo
 
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> ---
>  tools/perf/ui/browsers/hists.c | 10 ++++++++--
>  1 file changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
> index fe5677ccbc22..7722ad311318 100644
> --- a/tools/perf/ui/browsers/hists.c
> +++ b/tools/perf/ui/browsers/hists.c
> @@ -1381,8 +1381,14 @@ static int hist_browser__show_hierarchy_entry(struct hist_browser *browser,
>  		}
>  
>  		perf_hpp_list__for_each_format(entry->hpp_list, fmt) {
> -			ui_browser__write_nstring(&browser->b, "", 2);
> -			width -= 2;
> +			if (first) {
> +				ui_browser__printf(&browser->b, "%c ", folded_sign);
> +				width -= 2;
> +				first = false;
> +			} else {
> +				ui_browser__write_nstring(&browser->b, "", 2);
> +				width -= 2;
> +			}
>  
>  			/*
>  			 * No need to call hist_entry__snprintf_alignment()
> -- 
> 2.10.1

[toc] | [prev] | [next] | [standalone]

#1518203 — Re: [PATCH 2/4] perf hist browser: Show folded sign properly on --hierarchy

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2016-11-09 15:30 +0100
SubjectRe: [PATCH 2/4] perf hist browser: Show folded sign properly on --hierarchy
Message-ID<sBC30-8sr-15@gated-at.bofh.it>
In reply to#1518202
Em Wed, Nov 09, 2016 at 11:28:11AM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Tue, Nov 08, 2016 at 10:08:31PM +0900, Namhyung Kim escreveu:
> > When horizontall scrolling is used in hierarchy mode, the folded signed
> > disappears at the right most column.
> 
> Humm, this indeed shows the '+' folded signal after pressing ->, but it
> moves from the first to the third column :-\
> 
> > +++ b/tools/perf/ui/browsers/hists.c
> > @@ -1381,8 +1381,14 @@ static int hist_browser__show_hierarchy_entry(struct hist_browser *browser,
> >  		}
> >  
> >  		perf_hpp_list__for_each_format(entry->hpp_list, fmt) {
> > -			ui_browser__write_nstring(&browser->b, "", 2);
> > -			width -= 2;

Also why move this invariant to both branches?

> > +			if (first) {
> > +				ui_browser__printf(&browser->b, "%c ", folded_sign);
> > +				width -= 2;
> > +				first = false;
> > +			} else {
> > +				ui_browser__write_nstring(&browser->b, "", 2);
> > +				width -= 2;
> > +			}
> >  
> >  			/*
> >  			 * No need to call hist_entry__snprintf_alignment()
> > -- 
> > 2.10.1

[toc] | [prev] | [next] | [standalone]

#1517166

FromMarkus Trippelsdorf <markus@trippelsdorf.de>
Date2016-11-08 14:30 +0100
Message-ID<sBeDo-1rd-15@gated-at.bofh.it>
In reply to#1517151
On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
> Hello,
> 
> This patches fix problems in hierarchy output Markus reported some
> time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
> in my tree:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
> 
> Any feedbacks are welcomed.

It looks perfect now. Many thanks for your fixes.

-- 
Markus

[toc] | [prev] | [next] | [standalone]

#1518135

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2016-11-09 14:20 +0100
Message-ID<sBAXg-7ME-21@gated-at.bofh.it>
In reply to#1517166
Em Tue, Nov 08, 2016 at 02:21:17PM +0100, Markus Trippelsdorf escreveu:
> On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
> > Hello,
> > 
> > This patches fix problems in hierarchy output Markus reported some
> > time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
> > in my tree:
> > 
> >   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
> > 
> > Any feedbacks are welcomed.
> 
> It looks perfect now. Many thanks for your fixes.

Ok, I'll take that as a Tested-by: Markus, ok?

- Arnaldo

[toc] | [prev] | [next] | [standalone]

#1518141

FromMarkus Trippelsdorf <markus@trippelsdorf.de>
Date2016-11-09 14:20 +0100
Message-ID<sBAXg-7ME-19@gated-at.bofh.it>
In reply to#1518135
On 2016.11.09 at 10:11 -0300, Arnaldo Carvalho de Melo wrote:
> Em Tue, Nov 08, 2016 at 02:21:17PM +0100, Markus Trippelsdorf escreveu:
> > On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
> > > Hello,
> > > 
> > > This patches fix problems in hierarchy output Markus reported some
> > > time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
> > > in my tree:
> > > 
> > >   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
> > > 
> > > Any feedbacks are welcomed.
> > 
> > It looks perfect now. Many thanks for your fixes.
> 
> Ok, I'll take that as a Tested-by: Markus, ok?

Sure, feel free. 
Thanks.

-- 
Markus

[toc] | [prev] | [next] | [standalone]

#1517196

FromMarkus Trippelsdorf <markus@trippelsdorf.de>
Date2016-11-08 14:50 +0100
Message-ID<sBeWK-1y1-7@gated-at.bofh.it>
In reply to#1517151
On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
> 
> This patches fix problems in hierarchy output Markus reported some
> time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
> in my tree:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
> 
> Any feedbacks are welcomed.

By the way, I hope that:
https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/commit/?h=perf/core&id=8a06b0be6507f97f3aa92ca814335b8b65fd3de2
doesn't fall through the cracks.

-- 
Markus

[toc] | [prev] | [next] | [standalone]

#1517274

FromNamhyung Kim <namhyung@kernel.org>
Date2016-11-08 16:10 +0100
Message-ID<sBgc9-2wv-29@gated-at.bofh.it>
In reply to#1517196
Hello,

On Tue, Nov 8, 2016 at 10:43 PM, Markus Trippelsdorf
<markus@trippelsdorf.de> wrote:
> On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
>>
>> This patches fix problems in hierarchy output Markus reported some
>> time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
>> in my tree:
>>
>>   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
>>
>> Any feedbacks are welcomed.
>
> By the way, I hope that:
> https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/commit/?h=perf/core&id=8a06b0be6507f97f3aa92ca814335b8b65fd3de2
> doesn't fall through the cracks.

What do you mean?  It's already in the tip/perf/core so will be merged
to the mainline eventually.

Thanks,
Namhyung

[toc] | [prev] | [next] | [standalone]

#1517286

FromMarkus Trippelsdorf <markus@trippelsdorf.de>
Date2016-11-08 16:20 +0100
Message-ID<sBglQ-2zG-33@gated-at.bofh.it>
In reply to#1517274
On 2016.11.09 at 00:05 +0900, Namhyung Kim wrote:
> Hello,
> 
> On Tue, Nov 8, 2016 at 10:43 PM, Markus Trippelsdorf
> <markus@trippelsdorf.de> wrote:
> > On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
> >>
> >> This patches fix problems in hierarchy output Markus reported some
> >> time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
> >> in my tree:
> >>
> >>   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
> >>
> >> Any feedbacks are welcomed.
> >
> > By the way, I hope that:
> > https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/commit/?h=perf/core&id=8a06b0be6507f97f3aa92ca814335b8b65fd3de2
> > doesn't fall through the cracks.
> 
> What do you mean?  It's already in the tip/perf/core so will be merged
> to the mainline eventually.

Ok. I was just wondering, because it sits there for two weeks already...

-- 
Markus

[toc] | [prev] | [next] | [standalone]

#1518133

FromMarkus Trippelsdorf <markus@trippelsdorf.de>
Date2016-11-09 14:20 +0100
Message-ID<sBAXf-7ME-1@gated-at.bofh.it>
In reply to#1517286
On 2016.11.09 at 10:10 -0300, Arnaldo Carvalho de Melo wrote:
> Em Tue, Nov 08, 2016 at 04:10:23PM +0100, Markus Trippelsdorf escreveu:
> > On 2016.11.09 at 00:05 +0900, Namhyung Kim wrote:
> > > On Tue, Nov 8, 2016 at 10:43 PM, Markus Trippelsdorf
> > > <markus@trippelsdorf.de> wrote:
> > > > On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
> > > >> This patches fix problems in hierarchy output Markus reported some
> > > >> time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
> > > >> in my tree:
> 
> > > >>   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
> 
> > > >> Any feedbacks are welcomed.
> 
> > > > By the way, I hope that:
> > > > https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/commit/?h=perf/core&id=8a06b0be6507f97f3aa92ca814335b8b65fd3de2
> > > > doesn't fall through the cracks.
> 
> > > What do you mean?  It's already in the tip/perf/core so will be merged
> > > to the mainline eventually.
> 
> > Ok. I was just wondering, because it sits there for two weeks already...
> 
> If you think something qualifies for perf/urgent, i.e. to go to a kernel
> that is in -rc stage, v4.9-rc4 now, for instance, please point that out
> and I'll consider it.

Because "perf top --hierarchy" is new in 4.9, I think all fixes for that
feature qualify for perf/urgent by default.

-- 
Markus

[toc] | [prev] | [next] | [standalone]

#1518137

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2016-11-09 14:20 +0100
Message-ID<sBAXf-7ME-3@gated-at.bofh.it>
In reply to#1517286
Em Tue, Nov 08, 2016 at 04:10:23PM +0100, Markus Trippelsdorf escreveu:
> On 2016.11.09 at 00:05 +0900, Namhyung Kim wrote:
> > On Tue, Nov 8, 2016 at 10:43 PM, Markus Trippelsdorf
> > <markus@trippelsdorf.de> wrote:
> > > On 2016.11.08 at 22:08 +0900, Namhyung Kim wrote:
> > >> This patches fix problems in hierarchy output Markus reported some
> > >> time ago.  The code is available on the 'perf/hierarchy-fix-v1' branch
> > >> in my tree:

> > >>   git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git

> > >> Any feedbacks are welcomed.

> > > By the way, I hope that:
> > > https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/commit/?h=perf/core&id=8a06b0be6507f97f3aa92ca814335b8b65fd3de2
> > > doesn't fall through the cracks.

> > What do you mean?  It's already in the tip/perf/core so will be merged
> > to the mainline eventually.

> Ok. I was just wondering, because it sits there for two weeks already...

If you think something qualifies for perf/urgent, i.e. to go to a kernel
that is in -rc stage, v4.9-rc4 now, for instance, please point that out
and I'll consider it.

- Arnaldo

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web