[PATCH] perf ui/tui: Fix a small callchain display bug

[PATCH] perf ui/tui: Fix a small callchain display bug

[Namhyung Kim]

The currently when perf TUI report shows callchain, the first level
chains have bogus '+' sign even though only the last one has children.

Since they are on a single line of the chain, toggling intermediate
entries has no effect.  Fix it to show '+' sign at the last entry
only.  Note that non-first level callchain entries don't have this
problem.

Before:

  Children      Self  Command  Shared Object      Symbols
---------------------------------------------------------------------------
-   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
   + cpuidle_wrap_enter
   + cpuidle_enter_tk
   + cpuidle_idle_call
   + cpu_idle

After:

  Children      Self  Command  Shared Object      Symbols
---------------------------------------------------------------------------
-   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
     cpuidle_wrap_enter
     cpuidle_enter_tk
     cpuidle_idle_call
   + cpu_idle

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/ui/browsers/hists.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
index 045c1e16ac59..1818d1275d02 100644
--- a/tools/perf/ui/browsers/hists.c
+++ b/tools/perf/ui/browsers/hists.c
@@ -228,8 +228,10 @@ static void callchain_node__init_have_children(struct callchain_node *node)
 {
 	struct callchain_list *chain;
 
-	list_for_each_entry(chain, &node->val, list)
+	if (!list_empty(&node->val)) {
+		chain = list_entry(node->val.prev, struct callchain_list, list);
 		chain->ms.has_children = !RB_EMPTY_ROOT(&node->rb_root);
+	}
 
 	callchain_node__init_have_children_rb_tree(node);
 }
-- 
2.0.0

Re: [PATCH] perf ui/tui: Fix a small callchain display bug

[Ingo Molnar]

* Namhyung Kim <namhyung@kernel.org> wrote:

> The currently when perf TUI report shows callchain, the first level
> chains have bogus '+' sign even though only the last one has children.
> 
> Since they are on a single line of the chain, toggling intermediate
> entries has no effect.  Fix it to show '+' sign at the last entry
> only.  Note that non-first level callchain entries don't have this
> problem.
> 
> Before:
> 
>   Children      Self  Command  Shared Object      Symbols
> ---------------------------------------------------------------------------
> -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
>    + cpuidle_wrap_enter
>    + cpuidle_enter_tk
>    + cpuidle_idle_call
>    + cpu_idle
> 
> After:
> 
>   Children      Self  Command  Shared Object      Symbols
> ---------------------------------------------------------------------------
> -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
>      cpuidle_wrap_enter
>      cpuidle_enter_tk
>      cpuidle_idle_call
>    + cpu_idle
> 
> Cc: Frederic Weisbecker <fweisbec@gmail.com>
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>

Nice fix!

Acked-by: Ingo Molnar <mingo@kernel.org>

Thanks,

	Ingo

Re: [PATCH] perf ui/tui: Fix a small callchain display bug

[Arnaldo Carvalho de Melo]

Em Wed, Aug 13, 2014 at 03:02:41PM +0900, Namhyung Kim escreveu:
> The currently when perf TUI report shows callchain, the first level
> chains have bogus '+' sign even though only the last one has children.
> 
> Since they are on a single line of the chain, toggling intermediate
> entries has no effect.  Fix it to show '+' sign at the last entry
> only.  Note that non-first level callchain entries don't have this
> problem.
> 
> Before:
> 
>   Children      Self  Command  Shared Object      Symbols
> ---------------------------------------------------------------------------

Please avoid starting a line with --- inside the changeset comment log as this
confuses scripts, that takes that as the end of the comment, like in ...
(see below)

> -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
>    + cpuidle_wrap_enter
>    + cpuidle_enter_tk
>    + cpuidle_idle_call
>    + cpu_idle
> 
> After:
> 
>   Children      Self  Command  Shared Object      Symbols
> ---------------------------------------------------------------------------
> -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
>      cpuidle_wrap_enter
>      cpuidle_enter_tk
>      cpuidle_idle_call
>    + cpu_idle
> 
> Cc: Frederic Weisbecker <fweisbec@gmail.com>
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> ---

... here ^ ^ ^ ^ ^ 

:-)

Thanks for the fix, checking with 'P' after 'E' to expand everything,
before and after, to then do a:

diff -u perf.hist.0 perf.hist.1

- Arnaldo

>  tools/perf/ui/browsers/hists.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
> index 045c1e16ac59..1818d1275d02 100644
> --- a/tools/perf/ui/browsers/hists.c
> +++ b/tools/perf/ui/browsers/hists.c
> @@ -228,8 +228,10 @@ static void callchain_node__init_have_children(struct callchain_node *node)
>  {
>  	struct callchain_list *chain;
>  
> -	list_for_each_entry(chain, &node->val, list)
> +	if (!list_empty(&node->val)) {
> +		chain = list_entry(node->val.prev, struct callchain_list, list);
>  		chain->ms.has_children = !RB_EMPTY_ROOT(&node->rb_root);
> +	}
>  
>  	callchain_node__init_have_children_rb_tree(node);
>  }
> -- 
> 2.0.0

Re: [PATCH] perf ui/tui: Fix a small callchain display bug

[Arnaldo Carvalho de Melo]

Em Wed, Aug 13, 2014 at 10:19:09AM +0200, Ingo Molnar escreveu:
> * Namhyung Kim <namhyung@kernel.org> wrote:
> > Before:

> >   Children      Self  Command  Shared Object      Symbols
> > ---------------------------------------------------------------------------
> > -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
> >    + cpuidle_wrap_enter
> >    + cpuidle_enter_tk
> >    + cpuidle_idle_call
> >    + cpu_idle

> > After:

> >   Children      Self  Command  Shared Object      Symbols
> > ---------------------------------------------------------------------------
> > -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
> >      cpuidle_wrap_enter
> >      cpuidle_enter_tk
> >      cpuidle_idle_call
> >    + cpu_idle
> > 
> > Cc: Frederic Weisbecker <fweisbec@gmail.com>
> > Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> 
> Nice fix!

Indeed, longstanding problem, kudos to Namhyung for deving into TUI +
callchain land to fix this!

- Arnaldo
 
> Acked-by: Ingo Molnar <mingo@kernel.org>
> 
> Thanks,
> 
> 	Ingo

[tip:perf/core] perf hists browser: Fix a small callchain display bug

[tip-bot for Namhyung Kim]

Commit-ID:  82162b5ae3d152fd7d887b36213f5b6785fe1294
Gitweb:     http://git.kernel.org/tip/82162b5ae3d152fd7d887b36213f5b6785fe1294
Author:     Namhyung Kim <namhyung@kernel.org>
AuthorDate: Wed, 13 Aug 2014 15:02:41 +0900
Committer:  Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Wed, 13 Aug 2014 17:12:56 -0300

perf hists browser: Fix a small callchain display bug

The currently when perf TUI report shows callchain, the first level
chains have bogus '+' sign even though only the last one has children.

Since they are on a single line of the chain, toggling intermediate
entries has no effect.  Fix it to show '+' sign at the last entry only.
Note that non-first level callchain entries don't have this problem.

  Before:
  ---------------------------------------------------------------------------
    Children      Self  Command  Shared Object      Symbols
  -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
     + cpuidle_wrap_enter
     + cpuidle_enter_tk
     + cpuidle_idle_call
     + cpu_idle

  After:
  ---------------------------------------------------------------------------
    Children      Self  Command  Shared Object      Symbols
  -   40.70%     0.00%  swapper  [kernel.kallsyms]  [k] cpuidle_wrap_enter
       cpuidle_wrap_enter
       cpuidle_enter_tk
       cpuidle_idle_call
     + cpu_idle

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/1407909761-10822-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/ui/browsers/hists.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
index 045c1e1..1818d12 100644
--- a/tools/perf/ui/browsers/hists.c
+++ b/tools/perf/ui/browsers/hists.c
@@ -228,8 +228,10 @@ static void callchain_node__init_have_children(struct callchain_node *node)
 {
 	struct callchain_list *chain;
 
-	list_for_each_entry(chain, &node->val, list)
+	if (!list_empty(&node->val)) {
+		chain = list_entry(node->val.prev, struct callchain_list, list);
 		chain->ms.has_children = !RB_EMPTY_ROOT(&node->rb_root);
+	}
 
 	callchain_node__init_have_children_rb_tree(node);
 }