* [PATCH 1/2] trace_stat: keep original order
@ 2009-03-25 8:58 Lai Jiangshan
2009-03-25 13:30 ` Frederic Weisbecker
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Lai Jiangshan @ 2009-03-25 8:58 UTC (permalink / raw)
To: Ingo Molnar; +Cc: Steven Rostedt, Frederic Weisbecker, LKML
Impact: make trace_stat files show items with the original order
trace_stat tracer reverse the items, it makes the output
looks a little ugly.
Example, when we read trace_stat/workqueues, we get cpu#7's stat.
at first, and then cpu#6... cpu#0.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
---
diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c
index f71b85b..c08e7f1 100644
--- a/kernel/trace/trace_stat.c
+++ b/kernel/trace/trace_stat.c
@@ -125,23 +125,21 @@ static int stat_seq_init(struct tracer_stat_session *session)
INIT_LIST_HEAD(&new_entry->list);
new_entry->stat = stat;
- list_for_each_entry(iter_entry, &session->stat_list, list) {
+ list_for_each_entry_reverse(iter_entry, &session->stat_list,
+ list) {
/* Insertion with a descendent sorting */
- if (ts->stat_cmp(new_entry->stat,
- iter_entry->stat) > 0) {
+ if (ts->stat_cmp(iter_entry->stat,
+ new_entry->stat) >= 0) {
- list_add_tail(&new_entry->list,
- &iter_entry->list);
- break;
-
- /* The current smaller value */
- } else if (list_is_last(&iter_entry->list,
- &session->stat_list)) {
list_add(&new_entry->list, &iter_entry->list);
break;
}
}
+
+ /* The current larger value */
+ if (list_empty(&new_entry->list))
+ list_add(&new_entry->list, &session->stat_list);
}
exit:
mutex_unlock(&session->stat_mutex);
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 1/2] trace_stat: keep original order
2009-03-25 8:58 [PATCH 1/2] trace_stat: keep original order Lai Jiangshan
@ 2009-03-25 13:30 ` Frederic Weisbecker
2009-03-25 14:36 ` Steven Rostedt
2009-03-25 17:33 ` [tip:tracing/ftrace] " Lai Jiangshan
2 siblings, 0 replies; 5+ messages in thread
From: Frederic Weisbecker @ 2009-03-25 13:30 UTC (permalink / raw)
To: Lai Jiangshan; +Cc: Ingo Molnar, Steven Rostedt, LKML
On Wed, Mar 25, 2009 at 04:58:39PM +0800, Lai Jiangshan wrote:
>
> Impact: make trace_stat files show items with the original order
>
> trace_stat tracer reverse the items, it makes the output
> looks a little ugly.
>
> Example, when we read trace_stat/workqueues, we get cpu#7's stat.
> at first, and then cpu#6... cpu#0.
>
> Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
> ---
> diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c
> index f71b85b..c08e7f1 100644
> --- a/kernel/trace/trace_stat.c
> +++ b/kernel/trace/trace_stat.c
> @@ -125,23 +125,21 @@ static int stat_seq_init(struct tracer_stat_session *session)
> INIT_LIST_HEAD(&new_entry->list);
> new_entry->stat = stat;
>
> - list_for_each_entry(iter_entry, &session->stat_list, list) {
> + list_for_each_entry_reverse(iter_entry, &session->stat_list,
> + list) {
>
> /* Insertion with a descendent sorting */
> - if (ts->stat_cmp(new_entry->stat,
> - iter_entry->stat) > 0) {
> + if (ts->stat_cmp(iter_entry->stat,
> + new_entry->stat) >= 0) {
>
> - list_add_tail(&new_entry->list,
> - &iter_entry->list);
> - break;
> -
> - /* The current smaller value */
> - } else if (list_is_last(&iter_entry->list,
> - &session->stat_list)) {
> list_add(&new_entry->list, &iter_entry->list);
> break;
> }
> }
> +
> + /* The current larger value */
> + if (list_empty(&new_entry->list))
> + list_add(&new_entry->list, &session->stat_list);
> }
> exit:
> mutex_unlock(&session->stat_mutex);
>
>
Nice!
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 1/2] trace_stat: keep original order
2009-03-25 8:58 [PATCH 1/2] trace_stat: keep original order Lai Jiangshan
2009-03-25 13:30 ` Frederic Weisbecker
@ 2009-03-25 14:36 ` Steven Rostedt
2009-03-26 2:19 ` Lai Jiangshan
2009-03-25 17:33 ` [tip:tracing/ftrace] " Lai Jiangshan
2 siblings, 1 reply; 5+ messages in thread
From: Steven Rostedt @ 2009-03-25 14:36 UTC (permalink / raw)
To: Lai Jiangshan; +Cc: Ingo Molnar, Frederic Weisbecker, LKML
On Wed, 2009-03-25 at 16:58 +0800, Lai Jiangshan wrote:
> Impact: make trace_stat files show items with the original order
>
> trace_stat tracer reverse the items, it makes the output
> looks a little ugly.
>
> Example, when we read trace_stat/workqueues, we get cpu#7's stat.
> at first, and then cpu#6... cpu#0.
>
This doesn't reverse the order of the annotated branch profiler, does
it?
-- Steve
> Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
> ---
> diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c
> index f71b85b..c08e7f1 100644
> --- a/kernel/trace/trace_stat.c
> +++ b/kernel/trace/trace_stat.c
> @@ -125,23 +125,21 @@ static int stat_seq_init(struct tracer_stat_session *session)
> INIT_LIST_HEAD(&new_entry->list);
> new_entry->stat = stat;
>
> - list_for_each_entry(iter_entry, &session->stat_list, list) {
> + list_for_each_entry_reverse(iter_entry, &session->stat_list,
> + list) {
>
> /* Insertion with a descendent sorting */
> - if (ts->stat_cmp(new_entry->stat,
> - iter_entry->stat) > 0) {
> + if (ts->stat_cmp(iter_entry->stat,
> + new_entry->stat) >= 0) {
>
> - list_add_tail(&new_entry->list,
> - &iter_entry->list);
> - break;
> -
> - /* The current smaller value */
> - } else if (list_is_last(&iter_entry->list,
> - &session->stat_list)) {
> list_add(&new_entry->list, &iter_entry->list);
> break;
> }
> }
> +
> + /* The current larger value */
> + if (list_empty(&new_entry->list))
> + list_add(&new_entry->list, &session->stat_list);
> }
> exit:
> mutex_unlock(&session->stat_mutex);
>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* [tip:tracing/ftrace] trace_stat: keep original order
2009-03-25 8:58 [PATCH 1/2] trace_stat: keep original order Lai Jiangshan
2009-03-25 13:30 ` Frederic Weisbecker
2009-03-25 14:36 ` Steven Rostedt
@ 2009-03-25 17:33 ` Lai Jiangshan
2 siblings, 0 replies; 5+ messages in thread
From: Lai Jiangshan @ 2009-03-25 17:33 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, hpa, mingo, fweisbec, srostedt, tglx, laijs, mingo
Commit-ID: 220ba351dfa57eca4bec5ce0098a276446a47958
Gitweb: http://git.kernel.org/tip/220ba351dfa57eca4bec5ce0098a276446a47958
Author: Lai Jiangshan <laijs@cn.fujitsu.com>
AuthorDate: Wed, 25 Mar 2009 16:58:39 +0800
Committer: Ingo Molnar <mingo@elte.hu>
CommitDate: Wed, 25 Mar 2009 18:32:34 +0100
trace_stat: keep original order
Impact: make trace_stat files show items with the original order
trace_stat tracer reverse the items, it makes the output
looks a little ugly.
Example, when we read trace_stat/workqueues, we get cpu#7's stat.
at first, and then cpu#6... cpu#0.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Steven Rostedt <srostedt@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <49C9F23F.5040307@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
kernel/trace/trace_stat.c | 18 ++++++++----------
1 files changed, 8 insertions(+), 10 deletions(-)
diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c
index 8c129dd..acdebd7 100644
--- a/kernel/trace/trace_stat.c
+++ b/kernel/trace/trace_stat.c
@@ -125,23 +125,21 @@ static int stat_seq_init(struct tracer_stat_session *session)
INIT_LIST_HEAD(&new_entry->list);
new_entry->stat = stat;
- list_for_each_entry(iter_entry, &session->stat_list, list) {
+ list_for_each_entry_reverse(iter_entry, &session->stat_list,
+ list) {
/* Insertion with a descendent sorting */
- if (ts->stat_cmp(new_entry->stat,
- iter_entry->stat) > 0) {
+ if (ts->stat_cmp(iter_entry->stat,
+ new_entry->stat) >= 0) {
- list_add_tail(&new_entry->list,
- &iter_entry->list);
- break;
-
- /* The current smaller value */
- } else if (list_is_last(&iter_entry->list,
- &session->stat_list)) {
list_add(&new_entry->list, &iter_entry->list);
break;
}
}
+
+ /* The current larger value */
+ if (list_empty(&new_entry->list))
+ list_add(&new_entry->list, &session->stat_list);
}
exit:
mutex_unlock(&session->stat_mutex);
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 1/2] trace_stat: keep original order
2009-03-25 14:36 ` Steven Rostedt
@ 2009-03-26 2:19 ` Lai Jiangshan
0 siblings, 0 replies; 5+ messages in thread
From: Lai Jiangshan @ 2009-03-26 2:19 UTC (permalink / raw)
To: Steven Rostedt; +Cc: Ingo Molnar, Frederic Weisbecker, LKML
Steven Rostedt wrote:
> On Wed, 2009-03-25 at 16:58 +0800, Lai Jiangshan wrote:
>> Impact: make trace_stat files show items with the original order
>>
>> trace_stat tracer reverse the items, it makes the output
>> looks a little ugly.
>>
>> Example, when we read trace_stat/workqueues, we get cpu#7's stat.
>> at first, and then cpu#6... cpu#0.
>>
>
> This doesn't reverse the order of the annotated branch profiler, does
> it?
>
The annotated ftrace_branch_data are orderless. So it's OK
with or without this fix.
(From code, not test) If this fix applied,
__start_annotated_branch_profile is shown at the first, and then
__start_annotated_branch_profile + 1... and
__stop_annotated_branch_profile is shown at the last.
Lai
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2009-03-26 2:22 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-03-25 8:58 [PATCH 1/2] trace_stat: keep original order Lai Jiangshan
2009-03-25 13:30 ` Frederic Weisbecker
2009-03-25 14:36 ` Steven Rostedt
2009-03-26 2:19 ` Lai Jiangshan
2009-03-25 17:33 ` [tip:tracing/ftrace] " Lai Jiangshan
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox