section .text:
:
: 00400816 :
: knapsack():
743737 :400816: push %rbp
Reported-by: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Signed-off-by: Taeung Song
---
tools/perf/builtin-annotate.c | 4 +--
tools/perf/builtin-report.c | 13 ++---
struct sym_hist has addr[] but it should have
not only number of samples but also the sample period.
So use new struct symhist_entry for that.
Cc: Namhyung Kim
Cc: Jiri Olsa
Signed-off-by: Taeung Song
---
tools/perf/ui/browsers/annotate.c | 6 +++---
tools/perf/ui/gtk/annotate.c | 4
ine_samples (Arnaldo)
- Fix a case that can't switch 'the total period view' to
'the number of samples view' on the annotate TUI browser (Arnaldo)
- Calculate the percentage with 'period', not number of samples
Taeung Song (9):
perf annotate: Introduce struct
On 07/13/2017 05:13 AM, Arnaldo Carvalho de Melo wrote:
Em Wed, Jul 12, 2017 at 07:14:14AM +0900, Taeung Song escreveu:
Add --show-nr-samples option to perf-annotate
so that it corresponds with perf-report.
Additionally h->sum is properly renamed h->nr_samples.
Please first do a
On 07/13/2017 05:16 AM, Arnaldo Carvalho de Melo wrote:
Em Wed, Jul 12, 2017 at 07:14:19AM +0900, Taeung Song escreveu:
Cc: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Signed-off-by: Taeung Song
Humm, this is cycling thru different things to show on a column, I think
'toggle'
head_us, period_us)
366 HPP_PERCENT_FNS(overhead_guest_sys, period_guest_sys)
367 HPP_PERCENT_FNS(overhead_guest_us, period_guest_us)
368 HPP_PERCENT_ACC_FNS(overhead_acc, period)
Thanks,
Taeung
On 07/13/2017 05:13 AM, Arnaldo Carvalho de Melo wrote:
Em Wed, Jul 12, 2017 at 07:14:14AM +0900, Tae
On 07/13/2017 05:10 AM, Arnaldo Carvalho de Melo wrote:
Em Wed, Jul 12, 2017 at 07:14:10AM +0900, Taeung Song escreveu:
Currently the toggle total period view on the annotate TUI
shows the number of samples, not period like below.
$ perf annotate --show-total-period
Percent
Hi Arnaldo :)
On 07/13/2017 05:09 AM, Arnaldo Carvalho de Melo wrote:
Em Wed, Jul 12, 2017 at 07:14:04AM +0900, Taeung Song escreveu:
Currently the --show-total-period option of perf-annotate
is different from perf-report's.
For example, perf-report ordinarily shows period and numb
(84813704
event count)
--
:
:
:
: Disassembly of section .text:
:
: 00400816 :
: knapsack():
743737 :400816: push %rbp
Cc: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Signed-off-by: Taeung Song
Add --show-nr-samples option to perf-annotate
so that it corresponds with perf-report.
Additionally h->sum is properly renamed h->nr_samples.
Cc: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Signed-off-by: Taeung Song
---
tools/perf/builtin-annotate.c | 2 ++
tools/perf/ui/gtk/anno
otate stdio and TUI browser.
And support --show-nr-samples into perf-anntate
so that it correpond with perf-report's.
The code is available on 'perf/ann-fix-period' branch at
git://github.com/taeung/linux-perf.git
Thanks,
Taeung
Taeung Song (4):
perf annotate: Fix wrong --sh
ow.
Event count│
│
│
│Disassembly of section .text:
│
│00109a90 <_mcount@@GLIBC_2.2.5>:
│ sub$0x38,%rsp
2204022 │ mov%rax,(%rsp)
2207405 │ mov%rcx,0x8(%rsp)
Cc: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Signed-off-by: Taeung Song
---
tools/perf/ui/browsers/annotate.c | 17 -
1 file changed, 16 insertions(+), 1 deletion(-)
diff --git a/tools/perf/ui/browsers/annotate.c
b/tools/perf/ui/browsers/annotate.c
index 0ddc3b2..237903d
Hi Namhyung :)
On 07/05/2017 03:07 PM, Namhyung Kim wrote:
Hi Taeung,
On Wed, Jul 5, 2017 at 2:47 PM, Taeung Song wrote:
Hi,
Two problems of perf-annotate were mentioned in recent PATCH reviews
by Milian and Namhyung.
Currently perf-annotate has a '--show-total-period' option
and
Hi,
Two problems of perf-annotate were mentioned in recent PATCH reviews
by Milian and Namhyung.
Currently perf-annotate has a '--show-total-period' option
and a 't' key "Toggle total period view" on TUI browser.
However, they actually show the number of samples, not period(Raw number
of event
I'm late..
On 06/28/2017 06:53 PM, Milian Wolff wrote:
On Wednesday, June 28, 2017 5:18:08 AM CEST Taeung Song wrote:
Hi,
The --source-only option and new source code TUI view can show
the result of performance analysis based on full source code per
symbol(function). (Namhyung Kim to
Thank you for detailed review.
On 06/30/2017 12:18 AM, Namhyung Kim wrote:
Resend with a proper mail address..
On Fri, Jun 30, 2017 at 12:14 AM, Namhyung Kim wrote:
Hi Taeung,
Thanks for doing this work!
On Wed, Jun 28, 2017 at 12:14 PM, Taeung Song wrote:
The option can show the result
Hi,
On 06/29/2017 04:11 PM, Namhyung Kim wrote:
Hello,
On Wed, Jun 28, 2017 at 11:53:22AM +0200, Milian Wolff wrote:
On Wednesday, June 28, 2017 5:18:08 AM CEST Taeung Song wrote:
Hi,
The --source-only option and new source code TUI view can show
the result of performance analysis based on
On 06/28/2017 06:53 PM, Milian Wolff wrote:
On Wednesday, June 28, 2017 5:18:08 AM CEST Taeung Song wrote:
Hi,
The --source-only option and new source code TUI view can show
the result of performance analysis based on full source code per
symbol(function). (Namhyung Kim told me this idea and
Hi, Milian :)
Thank you for your comment.
On 06/28/2017 06:44 PM, Milian Wolff wrote:
On Wednesday, June 28, 2017 5:14:12 AM CEST Taeung Song wrote:
Unlike asm code level, the new source code view
provides a additional viewpoint based on source code level
for performance analysis.
Of course
Additionally,
the code is available on 'annotate/src-only' branch at
git://github.com/taeung/linux-perf.git
Thanks,
Taeung
On 06/28/2017 12:18 PM, Taeung Song wrote:
Hi,
The --source-only option and new source code TUI view can show
the result of performance analysis based on f
t to listen to your
opinions
about this new feature, if you don't mind.
If you give some feedback, I'd appreciate it! :)
Thanks,
Taeung
Taeung Song (4):
perf annotate: Add --source-only option
perf annotate: Add new source code view to the annotate TUI browser
perf annotate: Fold
2 return p - ptr;
0.00 : 373 }
Suggested-by: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Cc: Masami Hiramatsu
Cc: Adrian Hunter
Cc: Wang Nan
Cc: Jin Yao
Cc: Andi Kleen
Cc: Kim Phillips
Cc: David Ahern
Signed-off-by: Taeung Song
---
tools/perf/builtin-annotate.c |
er
Cc: Wang Nan
Cc: Jin Yao
Cc: Andi Kleen
Cc: Kim Phillips
Cc: David Ahern
Signed-off-by: Taeung Song
---
tools/perf/ui/browsers/annotate.c | 153 +-
tools/perf/util/annotate.c| 6 +-
tools/perf/util/annotate.h| 7 ++
3 files changed,
Cc: Namhyung Kim
Cc: Jiri Olsa
Signed-off-by: Taeung Song
---
tools/perf/ui/browsers/annotate.c | 17 -
1 file changed, 16 insertions(+), 1 deletion(-)
diff --git a/tools/perf/ui/browsers/annotate.c
b/tools/perf/ui/browsers/annotate.c
index 200ee0c..bb5a933 100644
--- a/tools
Suggested-by: Namhyung Kim
Cc: Milian Wolff
Cc: Jiri Olsa
Cc: Masami Hiramatsu
Cc: Adrian Hunter
Cc: Wang Nan
Cc: Jin Yao
Cc: Andi Kleen
Cc: Kim Phillips
Cc: David Ahern
Signed-off-by: Taeung Song
---
tools/perf/ui/browser.h | 1 +
tools/perf/ui/browsers/annotate.c | 146
Commit-ID: 8c1cedb4466809f9d741a4088314783cb88680a9
Gitweb: http://git.kernel.org/tip/8c1cedb4466809f9d741a4088314783cb88680a9
Author: Taeung Song
AuthorDate: Mon, 8 May 2017 20:07:30 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Mon, 19 Jun 2017 15:14:58 -0300
perf config
Commit-ID: 4f1fd74283582f3f5c34d1c9ed55117d775b4a20
Gitweb: http://git.kernel.org/tip/4f1fd74283582f3f5c34d1c9ed55117d775b4a20
Author: Taeung Song
AuthorDate: Sat, 17 Jun 2017 12:46:37 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Mon, 19 Jun 2017 22:05:54 -0300
perf config
Commit-ID: dfe1c6d7efa8ead6878b73216d4c891a28207528
Gitweb: http://git.kernel.org/tip/dfe1c6d7efa8ead6878b73216d4c891a28207528
Author: Taeung Song
AuthorDate: Sat, 17 Jun 2017 12:46:42 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Mon, 19 Jun 2017 22:05:55 -0300
perf config
#x27; with -1 instead of 0 for more compact code in cmd_config()
- Add a error message when perf_config_set__new() failed in cmd_config()
v2:
- there is no need to consider empty config file (Arnaldo)
Taeung Song (5):
perf config: Check error cases of {show_spec, set}_config()
perf config: Ref
Currently only section->from_system_config is being checked multiple times.
items->from_system_config should be also checked, so fix it.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
Currently set_config() can be repeatedly called for each
input config on the below case:
$ perf config kmem.default=slab report.children=false ...
But it's a waste, so finally write changed configs at a time.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/bu
To simplify the code related to 'ret' variable in cmd_config(),
initialize 'ret' with -1 instead of 0.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 30 +-
1 file changed, 13 insertions(+), 17 deleti
show_spec_config() and set_config() can be called multiple times
in the loop in cmd_config().
However, The error cases of them wasn't checked, so fix it.
Reported-by: Arnaldo Carvalho de Melo
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c
.
(i.e. by perf_config_set__delete() at the end of cmd_config())
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/config.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c
index 8d724f0..1cc5d80
ng Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/builtin-config.c b/tools/perf/builtin-config.c
index f62cc2d..34ebe83 100644
--- a/tools/perf/builtin-config.c
+++ b/tools/perf/builtin-config.c
@@ -56,7
auto-generated.
[report]
children = false
NOTE:
The config set will be freed at the tail end.
(i.e. by perf_config_set__delete() at the end of cmd_config())
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/config.c | 5 +
1 file changed, 1 insertio
Currently set_config() can be repeatedly called for each
input config on the below case:
$ perf config kmem.default=slab report.children=false ...
But it's a waste, so finally write changed configs at a time.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/bu
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/tools/perf/builtin-config.c b/tools/perf/builtin-config.c
index 6a4d8cd..e9fe673 100644
--- a/tools/perf/builtin-config.c
+++ b/tools
show_spec_config() and set_config()
in the loop in cmd_config() and if the return value means a error, handle it.
Cc: Jiri Olsa
Cc: Namhyung Kim
Reported-by: Arnaldo Carvalho de Melo
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 17 +++--
1 file changed, 15 insertions(+), 2
fig_set__new() failed in cmd_config()
v2:
- there is no need to consider empty config file (Arnaldo)
Taeung Song (7):
perf config: Handle errors from {show_spec, set}_config() in the loop
perf config: Refactor the code using 'ret' variable in cmd_config()
perf config: Return -1 if sho
If the config isn't found in the config set,
show_spec_config() cannot show the config.
But it return 0 on the situation though, so fix it.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
To simplify the code related to 'ret' variable
in cmd_config(), initialize 'ret' with -1 instead of 0
and use goto statement 'out_err'.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 30 +-
On 05/09/2017 01:41 AM, Arnaldo Carvalho de Melo wrote:
Em Mon, May 08, 2017 at 08:07:49PM +0900, Taeung Song escreveu:
When setting config key-value pairs,
if user or system config files are nonexistent,
currently perf-config's setting didn't work right.
"currently doesn
On 05/09/2017 01:37 AM, Arnaldo Carvalho de Melo wrote:
Em Mon, May 08, 2017 at 08:07:40PM +0900, Taeung Song escreveu:
Currently set_config() can be repeatedly called for each
input config on the below case:
$ perf config kmem.default=slab report.children=false ...
But it's a wast
Hi Arnaldo,
On 05/09/2017 01:39 AM, Arnaldo Carvalho de Melo wrote:
Em Mon, May 08, 2017 at 08:07:45PM +0900, Taeung Song escreveu:
Currently section's from_system_config was checked twice.
you mean:
"The item->from_system_config is not being checked when iterating over
s
Currently set_config() can be repeatedly called for each
input config on the below case:
$ perf config kmem.default=slab report.children=false ...
But it's a waste, so finally write changed configs at a time.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/bu
t__init() failed,
it'll be freed at the tail end (i.e. by perf_config_set__delete()
at the end of cmd_config()).
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/config.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/tools/perf/util/conf
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 44 +++-
1 file changed, 23 insertions(+), 21 deletions(-)
diff --git a/tools/perf/builtin-config.c b/tools/perf/builtin-config.c
index 80668fa..7545966 100644
Currently section's from_system_config was checked twice.
So adjust it in order to correctly check not only section's
but also item's from_system_config.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 2 +-
1 file changed, 1
eed to consider empty config file (Arnaldo)
Taeung Song (4):
perf config: Invert a if statement to reduce nesting in cmd_config()
perf config: Finally write changed configs on config file at a time
perf config: Correctly check whether it is from system config
perf config: Automatically create
Hi SeongJae,
Nice to see you :)
On 05/07/2017 07:36 PM, SeongJae Park wrote:
An example in perf-probe documentation for pattern of function name
based probe addition is not providing example command for the case.
This commit fixes the example to give appropriate example command.
Signed-off-by:
Commit-ID: 4341ec6b3db4c3e903d6c44958722918baec1e59
Gitweb: http://git.kernel.org/tip/4341ec6b3db4c3e903d6c44958722918baec1e59
Author: Taeung Song
AuthorDate: Wed, 26 Apr 2017 21:21:02 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 2 May 2017 18:23:12 -0300
perf config
Hi Arnaldo,
On 05/03/2017 12:12 AM, Arnaldo Carvalho de Melo wrote:
Em Wed, Apr 26, 2017 at 09:21:03PM +0900, Taeung Song escreveu:
If existent config files contains nothing,
the sections list in config_set can be empty.
So check not only NULL pointer of config_set but
also the list in
On 04/27/2017 08:30 PM, Steven Rostedt wrote:
On Thu, 27 Apr 2017 11:52:14 +0900
Taeung Song wrote:
On 04/27/2017 11:47 AM, Steven Rostedt wrote:
On Thu, 27 Apr 2017 08:46:21 +0900
Taeung Song wrote:
Even though PyRun_String() failed,
just 0 will be returned but we need to return -1
Even though PyRun_String() failed,
just 0 will be returned but we need to return -1
that means error status, so fix it.
Signed-off-by: Taeung Song
---
plugin_python.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/plugin_python.c b/plugin_python.c
index 2997679..c283ed7
On 04/27/2017 11:47 AM, Steven Rostedt wrote:
On Thu, 27 Apr 2017 08:46:21 +0900
Taeung Song wrote:
Even though PyRun_String() failed,
just 0 will be returned but we need to return -1
that means error status, so fix it.
Signed-off-by: Taeung Song
---
plugin_python.c | 5 +++--
1 file
Even though PyRun_String() failed,
just 0 will be returned but we need to return -1
that means error status, so fix it.
Signed-off-by: Taeung Song
---
plugin_python.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/plugin_python.c b/plugin_python.c
index 2997679
On 04/27/2017 02:54 AM, Steven Rostedt wrote:
On Wed, 26 Apr 2017 23:56:02 +0900
Taeung Song wrote:
Want to send another patch on top of Federico's?
Yes! I want to do it.
After you update 'master' branch on your repo
I just updated my master branch.
I got it !
I&
Currently the FILTER_TRIVIAL_FALSE hasn't break statement
so if the trivial type is FALSE, it'll be hit always.
So add break statement at the FALSE case
to correctly remove trivial filters.
Reported-by: Namhyung Kim
Signed-off-by: Taeung Song
---
parse-filter.c | 1 +
1 file
Hi Steven :)
On 04/26/2017 08:11 AM, Steven Rostedt wrote:
On Mon, 24 Apr 2017 22:43:31 +0900
Taeung Song wrote:
Currently the return type of load_plugin() in plugin_python.c is void type,
but it is different from the argument type of trace_util_load_plugins().
So fix it for the below
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 19 ---
1 file changed, 8 insertions(+), 11 deletions(-)
diff --git a/tools/perf/builtin-config.c b/tools/perf/builtin-config.c
index 55f04f8..80668fa 100644
--- a/tools/perf/builtin
fig --user report.children=false
$ cat ~/.perfconfig
# this file is auto-generated.
[report]
children = false
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/config.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/tools/perf/util/
Currently set_config() can be repeatedly called for each
input config on the below case:
$ perf config kmem.default=slab report.children=false ...
But it's a waste, so finally write changed configs at a time.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/bu
Currently no bugs in the checking code.
But adjust it to correctly check item->from_system_config,
not section's from_system_config.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 50 -
1 file changed, 27 insertions(+), 23 deletions(-)
diff --git a/tools/perf/builtin-config.c b/tools/perf/builtin-config.c
index ad0a112..f4596ef 100644
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 19 ---
1 file changed, 8 insertions(+), 11 deletions(-)
diff --git a/tools/perf/builtin-config.c b/tools/perf/builtin-config.c
index 55f04f8..80668fa 100644
--- a/tools/perf/builtin
If existent config files contains nothing,
the sections list in config_set can be empty.
So check not only NULL pointer of config_set but
also the list in config_set.
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-config.c | 4 ++--
tools/perf/util/config.c
that skip the error message.
After:
$ perf config -l
Nothing configured, please check your /home/taeung/.perfconfig
$ perf config report.queue-size
Nothing configured, please check your /home/taeung/.perfconfig
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/bu
Hi all, :)
This is simple patchset for perf-config
to fix small bugs and refactor code.
I'd appreciate some feedback on this patchset.
The code is also avaiable at 'config/refactoring' branch on
git://github.com/taeung/linux-perf.git
Thanks,
Taeung
Taeung Song (7):
perf c
*)’
int trace_util_load_plugins(struct pevent *pevent, const char *suffix,
^
Signed-off-by: Taeung Song
---
plugin_python.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/plugin_python.c b/plugin_python.c
index da07d27..d736f50 100644
--- a/plugin_python.c
+++ b/plugi
On 04/20/2017 07:35 PM, Jiri Olsa wrote:
On Thu, Apr 20, 2017 at 07:17:34PM +0900, Taeung Song wrote:
Hi Jiri,
On 04/20/2017 06:24 PM, Jiri Olsa wrote:
Recent commit broke command name strip in perf_event__get_comm_ids
function. It replaced left to right search for '\n' with rt
("perf tools: Refactor the code to strip command name with
{l,r}trim()")
Signed-off-by: Jiri Olsa
Cc: David Ahern
Cc: Namhyung Kim
Cc: Peter Zijlstra
Cc: Taeung Song
Cc: Jin Yao
Link: http://lkml.kernel.org/n/tip-51mt8hxaig74zlu42s3rv...@git.kernel.org
---
tools/perf/util/event.c
Commit-ID: 4597cf0664d2fad785509dedfed22f8fe8951ebb
Gitweb: http://git.kernel.org/tip/4597cf0664d2fad785509dedfed22f8fe8951ebb
Author: Taeung Song
AuthorDate: Sat, 8 Apr 2017 09:52:24 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 11 Apr 2017 16:22:22 -0300
perf annotate
Commit-ID: 986a5bc028a84d487c354a529730b48682d1fb41
Gitweb: http://git.kernel.org/tip/986a5bc028a84d487c354a529730b48682d1fb41
Author: Taeung Song
AuthorDate: Sat, 8 Apr 2017 09:52:25 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 11 Apr 2017 16:22:23 -0300
perf annotate
Commit-ID: bdd97ca63faa374c98314d53c0bcaedb473c5a33
Gitweb: http://git.kernel.org/tip/bdd97ca63faa374c98314d53c0bcaedb473c5a33
Author: Taeung Song
AuthorDate: Fri, 7 Apr 2017 23:24:21 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 11 Apr 2017 15:23:26 -0300
perf tools
Commit-ID: b07c40df1f4e6f937271921cb116d570bb9c4a31
Gitweb: http://git.kernel.org/tip/b07c40df1f4e6f937271921cb116d570bb9c4a31
Author: Taeung Song
AuthorDate: Fri, 7 Apr 2017 23:24:18 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 11 Apr 2017 08:45:10 -0300
perf stat
Commit-ID: aa4beb10a94358bf2474d1fc9c4ccde34660cc9d
Gitweb: http://git.kernel.org/tip/aa4beb10a94358bf2474d1fc9c4ccde34660cc9d
Author: Taeung Song
AuthorDate: Fri, 7 Apr 2017 23:24:20 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 11 Apr 2017 08:45:10 -0300
perf pmu
Commit-ID: e21600fd41106f7a0ca124cec2404b2b3562768d
Gitweb: http://git.kernel.org/tip/e21600fd41106f7a0ca124cec2404b2b3562768d
Author: Taeung Song
AuthorDate: Fri, 7 Apr 2017 23:24:19 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 11 Apr 2017 08:45:10 -0300
perf ui browser
Hi David,
On 04/09/2017 03:48 AM, David Ahern wrote:
On 4/7/17 8:52 PM, Taeung Song wrote:
After reading command name from /proc//status,
use ltrim() and rtrim() to strip command name, not using
just while loop, isspace() and etc.
Cc: David Ahern
Cc: Don Zickus
Cc: Jiri Olsa
Cc: Namhyung
After reading command name from /proc//status,
use ltrim() and rtrim() to strip command name, not using
just while loop, isspace() and etc.
Cc: David Ahern
Cc: Don Zickus
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/event.c | 11 ++-
1 file changed
Cc: Andi Kleen
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/pmu.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index 362051e..11c7525 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf
When parsing disassemble lines for source line number,
use a stripped line instead of raw line.
Cc: Andi Kleen
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/annotate.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/perf/util
When parsing disassemble lines,
use ltrim() and rtrim() to strip them,
not using just while loop and isspace().
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/annotate.c | 42 +++---
1 file changed, 7 insertions(+), 35
main purpose of PATCH (Arnaldo)
Taeung Song (5):
perf annotate: Refactor the code to parse disassemble lines with
{l,r}trim()
perf annotate: Use stripped line instead of raw disassemble line
perf ui browser: Refactor the code to parse color configs with ltrim()
perf pmu: Refactor wordwr
When parsing {fore, back} ground color configs,
use ltrim() instead of just while loop and isspace().
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/ui/browser.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/ui/browser.c b/tools/perf
On 04/08/2017 12:04 AM, Arnaldo Carvalho de Melo wrote:
Em Fri, Apr 07, 2017 at 11:24:17PM +0900, Taeung Song escreveu:
When parsing disassemble lines,
use ltrim() and rtrim() to strip them,
not using just while loop and isspace().
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
On 04/08/2017 12:06 AM, Arnaldo Carvalho de Melo wrote:
Em Fri, Apr 07, 2017 at 11:24:18PM +0900, Taeung Song escreveu:
To strip csv output, use ltrim() instead of
just while loop and isspace() at print_metric_{only}_csv().
Applied.
Thank you!
- Taeung
On 04/08/2017 12:04 AM, Arnaldo Carvalho de Melo wrote:
Em Fri, Apr 07, 2017 at 12:01:30PM -0300, Arnaldo Carvalho de Melo escreveu:
Em Fri, Apr 07, 2017 at 11:24:17PM +0900, Taeung Song escreveu:
@@ -1148,9 +1136,9 @@ static int symbol__parse_objdump_line(struct symbol *sym,
struct map
Cc: Andi Kleen
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/pmu.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index 362051e..11c7525 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf
After reading command name from /proc//status,
use ltrim() and rtrim() to strip command name, not using
just while loop, isspace() and etc.
Cc: David Ahern
Cc: Don Zickus
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/event.c | 11 ++-
1 file changed
When parsing disassemble lines,
use ltrim() and rtrim() to strip them,
not using just while loop and isspace().
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/util/annotate.c | 49 ++
1 file changed, 10 insertions(+), 39
When parsing {fore, back} ground color configs,
use ltrim() instead of just while loop and isspace().
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/ui/browser.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/ui/browser.c b/tools/perf
To strip csv output, use ltrim() instead of
just while loop and isspace() at print_metric_{only}_csv().
Cc: Andi Kleen
Cc: Jiri Olsa
Cc: Namhyung Kim
Signed-off-by: Taeung Song
---
tools/perf/builtin-stat.c | 10 ++
1 file changed, 2 insertions(+), 8 deletions(-)
diff --git a/tools
Hi, :)
It is to simply refactor the code about stip strings
with ltrim() and rtrim().
I'd appreciate some feedback on this PATCHset.
Thanks,
Taeung
Taeung Song (5):
perf annotate: Refactor the code to parse disassemble lines with
{l,r}trim()
perf stat: Refactor the code to stri
Commit-ID: 99094a5e941fe88d95cbd594e6a41bee24003ecb
Gitweb: http://git.kernel.org/tip/99094a5e941fe88d95cbd594e6a41bee24003ecb
Author: Taeung Song
AuthorDate: Tue, 28 Mar 2017 21:12:05 +0900
Committer: Arnaldo Carvalho de Melo
CommitDate: Tue, 4 Apr 2017 21:08:00 -0300
perf annotate
calc_percent()) has two alternative logic
when using '-l' option or not as below.
888 if (src_line) {
...
903 } else {
...
914 }
On 03/29/2017 01:15 AM, Taeung Song wrote:
On 03/29/2017 12:28 AM, Arnaldo Carvalho de Melo wrote:
On 03/29/2017 12:28 AM, Arnaldo Carvalho de Melo wrote:
Em Tue, Mar 28, 2017 at 11:21:33PM +0900, Taeung Song escreveu:
On 03/28/2017 10:48 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Mar 28, 2017 at 09:12:05PM +0900, Taeung Song escreveu:
The option 'show-total-period' works fi
On 03/28/2017 09:09 PM, Taeung Song wrote:
Good morning, Arnaldo :)
On 03/28/2017 03:26 AM, Arnaldo Carvalho de Melo wrote:
Em Mon, Mar 27, 2017 at 04:10:38PM +0900, Taeung Song escreveu:
If running 'perf annotate --stdio -l --show-total-period',
you can see a problem showing on
On 03/28/2017 11:21 PM, Taeung Song wrote:
On 03/28/2017 10:48 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Mar 28, 2017 at 09:12:05PM +0900, Taeung Song escreveu:
The option 'show-total-period' works fine without a option '-l'.
But if running 'perf annotate --std
101 - 200 of 884 matches
Mail list logo