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

Groups > linux.kernel > #1167160 > unrolled thread

[GIT PULL 0/8] perf/core improvements and fixes

Started byArnaldo Carvalho de Melo <acme@kernel.org>
First post2015-06-17 23:30 +0200
Last post2015-06-19 08:30 +0200
Articles 6 — 2 participants

Back to article view | Back to linux.kernel

Contents

  [GIT PULL 0/8] perf/core improvements and fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-06-17 23:30 +0200
    [PATCH 4/8] perf probe: Show usage even if the last event is skipped Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-06-17 23:30 +0200
    Re: [GIT PULL 0/8] perf/core improvements and fixes Ingo Molnar <mingo@kernel.org> - 2015-06-18 09:50 +0200
      [RFC] hotkey for disabling/enabling events in 'perf top' TUI was Re:  [GIT PULL 0/8] perf/core improvements and fixes Arnaldo Carvalho de Melo <acme@kernel.org> - 2015-06-18 22:20 +0200
        Re: [RFC] hotkey for disabling/enabling events in 'perf top' TUI was  Re: [GIT PULL 0/8] perf/core improvements and fixes Ingo Molnar <mingo@kernel.org> - 2015-06-18 23:00 +0200
          Re: [RFC] hotkey for disabling/enabling events in 'perf top' TUI was  Re: [GIT PULL 0/8] perf/core improvements and fixes Ingo Molnar <mingo@kernel.org> - 2015-06-19 08:30 +0200

#1167160 — [GIT PULL 0/8] perf/core improvements and fixes

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2015-06-17 23:30 +0200
Subject[GIT PULL 0/8] perf/core improvements and fixes
Message-ID<pCt4e-51d-5@gated-at.bofh.it>
Hi Ingo,

	Please consider pulling, this is on top of perf-core-for-mingo, that is
still outstanding,

Thanks!

- Arnaldo

The following changes since commit b031220d520238075bd99513a420e65cf37866ad:

  perf probe: Fix to return error if no probe is added (2015-06-16 11:39:51 -0300)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-2

for you to fetch changes up to 5d484f99aed547e235f2229653c95392a1bc3692:

  perf top: Allow disabling/enabling events dynamicly (2015-06-17 16:50:52 -0300)

----------------------------------------------------------------
perf/core improvements and fixes:

User visible:

- Allow disabling/enabling events dynamicly in 'perf top':
  a 'perf top' session can instantly become a 'perf report'
  one, i.e. going from dynamic analysis to a static one,
  returning to a dynamic one is possible, to toogle the
  modes, just press CTRL+z. (Arnaldo Carvalho de Melo)

- Greatly speed up 'perf probe --list' by caching debuginfo
  (Masami Hiramatsu)

- Fix 'perf trace' race condition at the end of started
  workloads (Sukadev Bhattiprolu)

- Fix a problem when opening old perf.data with different
  byte order (Wang Nan)

Infrastructure:

- Ignore .config-detected in .gitignore (Wang Nan)

- Move libtraceevent dynamic list to separated LDFLAGS
  variable (Wang Nan)

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

----------------------------------------------------------------
Arnaldo Carvalho de Melo (2):
      perf evlist: Add toggle_enable() method
      perf top: Allow disabling/enabling events dynamicly

Masami Hiramatsu (2):
      perf probe: Show usage even if the last event is skipped
      perf probe: Speed up perf probe --list by caching debuginfo

Sukadev Bhattiprolu (1):
      perf trace: Fix race condition at the end of started workloads

Wang Nan (3):
      perf tools: Ignore .config-detected in .gitignore
      perf tools: Fix a problem when opening old perf.data with different byte order
      perf tools: Move libtraceevent dynamic list to separated LDFLAGS variable

 tools/perf/.gitignore          |  1 +
 tools/perf/Makefile.perf       |  8 ++--
 tools/perf/builtin-top.c       | 52 ++++++++++++++++++--------
 tools/perf/ui/browsers/hists.c |  2 +
 tools/perf/util/evlist.c       | 18 ++++++++-
 tools/perf/util/evlist.h       |  2 +
 tools/perf/util/probe-event.c  | 83 +++++++++++++++++++++++++++++++-----------
 tools/perf/util/session.c      | 50 ++++++++++++++++++-------
 8 files changed, 160 insertions(+), 56 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1167161 — [PATCH 4/8] perf probe: Show usage even if the last event is skipped

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2015-06-17 23:30 +0200
Subject[PATCH 4/8] perf probe: Show usage even if the last event is skipped
Message-ID<pCt4g-51d-45@gated-at.bofh.it>
In reply to#1167160
From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>

When the last part of converted events are blacklisted or out-of-text,
those are skipped and perf probe doesn't show usage examples.  This
fixes it to show the example even if the last part of event list is
skipped.

E.g. without this patch, events are added, but suddenly end:

  # perf probe vfs_*
  vfs_caches_init_early is out of .text, skip it.
  vfs_caches_init is out of .text, skip it.
  Added new events:
    probe:vfs_fallocate  (on vfs_*)
    probe:vfs_open       (on vfs_*)
  ...
    probe:vfs_dentry_acceptable (on vfs_*)
    probe:vfs_load_quota_inode (on vfs_*)
  #

With this fix:

  # perf probe vfs_*
  vfs_caches_init_early is out of .text, skip it.
  vfs_caches_init is out of .text, skip it.
  Added new events:
    probe:vfs_fallocate  (on vfs_*)
  ...
    probe:vfs_load_quota_inode (on vfs_*)

  You can now use it in all perf tools, such as:

	perf record -e probe:vfs_load_quota_inode -aR sleep 1

Note that this can be reproduced ONLY IF the vfs_caches_init* is the
last part of matched symbol list. I've checked this happens on
"3.19.0-generic #18-Ubuntu" kernel binary.

Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naohiro Aota <naota@elisp.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20150616115057.19906.5502.stgit@localhost.localdomain
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/probe-event.c | 35 +++++++++++++++++------------------
 1 file changed, 17 insertions(+), 18 deletions(-)

diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 85c8207c25cc..65a1c8252270 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -2157,7 +2157,8 @@ static bool kprobe_blacklist__listed(unsigned long address)
 	return !!kprobe_blacklist__find_by_address(&kprobe_blacklist, address);
 }
 
-static int perf_probe_event__sprintf(struct perf_probe_event *pev,
+static int perf_probe_event__sprintf(const char *group, const char *event,
+				     struct perf_probe_event *pev,
 				     const char *module,
 				     struct strbuf *result)
 {
@@ -2170,7 +2171,7 @@ static int perf_probe_event__sprintf(struct perf_probe_event *pev,
 	if (!place)
 		return -EINVAL;
 
-	ret = e_snprintf(buf, 128, "%s:%s", pev->group, pev->event);
+	ret = e_snprintf(buf, 128, "%s:%s", group, event);
 	if (ret < 0)
 		goto out;
 
@@ -2195,13 +2196,14 @@ out:
 }
 
 /* Show an event */
-static int show_perf_probe_event(struct perf_probe_event *pev,
+static int show_perf_probe_event(const char *group, const char *event,
+				 struct perf_probe_event *pev,
 				 const char *module, bool use_stdout)
 {
 	struct strbuf buf = STRBUF_INIT;
 	int ret;
 
-	ret = perf_probe_event__sprintf(pev, module, &buf);
+	ret = perf_probe_event__sprintf(group, event, pev, module, &buf);
 	if (ret >= 0) {
 		if (use_stdout)
 			printf("%s\n", buf.buf);
@@ -2253,7 +2255,8 @@ static int __show_perf_probe_events(int fd, bool is_kprobe,
 								is_kprobe);
 			if (ret < 0)
 				goto next;
-			ret = show_perf_probe_event(&pev, tev.point.module,
+			ret = show_perf_probe_event(pev.group, pev.event,
+						    &pev, tev.point.module,
 						    true);
 		}
 next:
@@ -2438,7 +2441,7 @@ static int __add_probe_trace_events(struct perf_probe_event *pev,
 	int i, fd, ret;
 	struct probe_trace_event *tev = NULL;
 	char buf[64];
-	const char *event, *group;
+	const char *event = NULL, *group = NULL;
 	struct strlist *namelist;
 	bool safename;
 
@@ -2500,15 +2503,12 @@ static int __add_probe_trace_events(struct perf_probe_event *pev,
 		/* Add added event name to namelist */
 		strlist__add(namelist, event);
 
-		/* Trick here - save current event/group */
-		event = pev->event;
-		group = pev->group;
-		pev->event = tev->event;
-		pev->group = tev->group;
-		show_perf_probe_event(pev, tev->point.module, false);
-		/* Trick here - restore current event/group */
-		pev->event = (char *)event;
-		pev->group = (char *)group;
+		/* We use tev's name for showing new events */
+		show_perf_probe_event(tev->group, tev->event, pev,
+				      tev->point.module, false);
+		/* Save the last valid name */
+		event = tev->event;
+		group = tev->group;
 
 		/*
 		 * Probes after the first probe which comes from same
@@ -2522,11 +2522,10 @@ static int __add_probe_trace_events(struct perf_probe_event *pev,
 		warn_uprobe_event_compat(tev);
 
 	/* Note that it is possible to skip all events because of blacklist */
-	if (ret >= 0 && tev->event) {
+	if (ret >= 0 && event) {
 		/* Show how to use the event. */
 		pr_info("\nYou can now use it in all perf tools, such as:\n\n");
-		pr_info("\tperf record -e %s:%s -aR sleep 1\n\n", tev->group,
-			 tev->event);
+		pr_info("\tperf record -e %s:%s -aR sleep 1\n\n", group, event);
 	}
 
 	strlist__delete(namelist);
-- 
2.1.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

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

#1167637

FromIngo Molnar <mingo@kernel.org>
Date2015-06-18 09:50 +0200
Message-ID<pCCKd-1XC-5@gated-at.bofh.it>
In reply to#1167160
* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:

> Hi Ingo,
> 
> 	Please consider pulling, this is on top of perf-core-for-mingo, that is
> still outstanding,
> 
> Thanks!
> 
> - Arnaldo
> 
> The following changes since commit b031220d520238075bd99513a420e65cf37866ad:
> 
>   perf probe: Fix to return error if no probe is added (2015-06-16 11:39:51 -0300)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-2
> 
> for you to fetch changes up to 5d484f99aed547e235f2229653c95392a1bc3692:
> 
>   perf top: Allow disabling/enabling events dynamicly (2015-06-17 16:50:52 -0300)
> 
> ----------------------------------------------------------------
> perf/core improvements and fixes:
> 
> User visible:
> 
> - Allow disabling/enabling events dynamicly in 'perf top':
>   a 'perf top' session can instantly become a 'perf report'
>   one, i.e. going from dynamic analysis to a static one,
>   returning to a dynamic one is possible, to toogle the
>   modes, just press CTRL+z. (Arnaldo Carvalho de Melo)

Nice!! :-)

Btw., it would be nice if the status line carried information about whether 
collection is 'frozen' or running, at a glance. A hint might also suggest how to 
unfreeze the session - in case someone pressed Ctrl-Z to suspend the perf top 
session ...

Also, there's now a GUI inconsistency with perf report: which will now exit on 
Ctrl-Z. It should probably print a warning in the status line instead, that 
freezing/unfreezing only works in 'perf top'.

> 
> - Greatly speed up 'perf probe --list' by caching debuginfo
>   (Masami Hiramatsu)
> 
> - Fix 'perf trace' race condition at the end of started
>   workloads (Sukadev Bhattiprolu)
> 
> - Fix a problem when opening old perf.data with different
>   byte order (Wang Nan)
> 
> Infrastructure:
> 
> - Ignore .config-detected in .gitignore (Wang Nan)
> 
> - Move libtraceevent dynamic list to separated LDFLAGS
>   variable (Wang Nan)
> 
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> 
> ----------------------------------------------------------------
> Arnaldo Carvalho de Melo (2):
>       perf evlist: Add toggle_enable() method
>       perf top: Allow disabling/enabling events dynamicly
> 
> Masami Hiramatsu (2):
>       perf probe: Show usage even if the last event is skipped
>       perf probe: Speed up perf probe --list by caching debuginfo
> 
> Sukadev Bhattiprolu (1):
>       perf trace: Fix race condition at the end of started workloads
> 
> Wang Nan (3):
>       perf tools: Ignore .config-detected in .gitignore
>       perf tools: Fix a problem when opening old perf.data with different byte order
>       perf tools: Move libtraceevent dynamic list to separated LDFLAGS variable
> 
>  tools/perf/.gitignore          |  1 +
>  tools/perf/Makefile.perf       |  8 ++--
>  tools/perf/builtin-top.c       | 52 ++++++++++++++++++--------
>  tools/perf/ui/browsers/hists.c |  2 +
>  tools/perf/util/evlist.c       | 18 ++++++++-
>  tools/perf/util/evlist.h       |  2 +
>  tools/perf/util/probe-event.c  | 83 +++++++++++++++++++++++++++++++-----------
>  tools/perf/util/session.c      | 50 ++++++++++++++++++-------
>  8 files changed, 160 insertions(+), 56 deletions(-)

Pulled, thanks a lot Arnaldo!

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

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

#1168270 — [RFC] hotkey for disabling/enabling events in 'perf top' TUI was Re: [GIT PULL 0/8] perf/core improvements and fixes

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2015-06-18 22:20 +0200
Subject[RFC] hotkey for disabling/enabling events in 'perf top' TUI was Re: [GIT PULL 0/8] perf/core improvements and fixes
Message-ID<pCOs1-2aq-9@gated-at.bofh.it>
In reply to#1167637
Em Thu, Jun 18, 2015 at 09:40:10AM +0200, Ingo Molnar escreveu:
> * Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> > User visible:

> > - Allow disabling/enabling events dynamicly in 'perf top':
> >   a 'perf top' session can instantly become a 'perf report'
> >   one, i.e. going from dynamic analysis to a static one,
> >   returning to a dynamic one is possible, to toogle the
> >   modes, just press CTRL+z. (Arnaldo Carvalho de Melo)
 
> Nice!! :-)
 
> Btw., it would be nice if the status line carried information about whether 
> collection is 'frozen' or running, at a glance. A hint might also suggest how to 
> unfreeze the session - in case someone pressed Ctrl-Z to suspend the perf top 
> session ...

Right, and I think we better find other hotkey and make Ctrl+Z work like
with other tools, i.e. suspend:

[acme@zoo linux]$ mutt

[1]+  Stopped                 mutt
[acme@zoo linux]$ vim

[2]+  Stopped                 vim
[acme@zoo linux]$ 

The perf TUI should work like that as well...

Ideas?

We already have:

h/?/F1        Show this window
UP/DOWN/PGUP
PGDN/SPACE    Navigate
q/ESC/CTRL+C  Exit browser

For multiple event sessions:

TAB/UNTAB     Switch events

For symbolic views (--sort has sym):

->            Zoom into DSO/Threads & Annotate current symbol
<-            Zoom out
a             Annotate current symbol
C             Collapse all callchains
d             Zoom into current DSO
D             Show some developer debug info
E             Expand all callchains
F             Toggle percentage of filtered entries
H             Display column headers
i             Show header information
P             Print histograms to perf.hist.N
r             Run available scripts
s             Switch to another data file in PWD
t             Zoom into current Thread
V             Verbose (DSO names in callchains, etc)
z             Toggle zeroing of samples 
/             Filter symbol by name 
 
> Also, there's now a GUI inconsistency with perf report: which will now exit on 
> Ctrl-Z. It should probably print a warning in the status line instead, that 
> freezing/unfreezing only works in 'perf top'.

I'll fix that.

- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

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

#1168289 — Re: [RFC] hotkey for disabling/enabling events in 'perf top' TUI was Re: [GIT PULL 0/8] perf/core improvements and fixes

FromIngo Molnar <mingo@kernel.org>
Date2015-06-18 23:00 +0200
SubjectRe: [RFC] hotkey for disabling/enabling events in 'perf top' TUI was Re: [GIT PULL 0/8] perf/core improvements and fixes
Message-ID<pCP4K-2Tq-13@gated-at.bofh.it>
In reply to#1168270
* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:

> Em Thu, Jun 18, 2015 at 09:40:10AM +0200, Ingo Molnar escreveu:
> > * Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> > > User visible:
> 
> > > - Allow disabling/enabling events dynamicly in 'perf top':
> > >   a 'perf top' session can instantly become a 'perf report'
> > >   one, i.e. going from dynamic analysis to a static one,
> > >   returning to a dynamic one is possible, to toogle the
> > >   modes, just press CTRL+z. (Arnaldo Carvalho de Melo)
>  
> > Nice!! :-)
>  
> > Btw., it would be nice if the status line carried information about whether 
> > collection is 'frozen' or running, at a glance. A hint might also suggest how to 
> > unfreeze the session - in case someone pressed Ctrl-Z to suspend the perf top 
> > session ...
> 
> Right, and I think we better find other hotkey and make Ctrl+Z work like
> with other tools, i.e. suspend:
> 
> [acme@zoo linux]$ mutt
> 
> [1]+  Stopped                 mutt
> [acme@zoo linux]$ vim
> 
> [2]+  Stopped                 vim
> [acme@zoo linux]$ 
> 
> The perf TUI should work like that as well...
> 
> Ideas?
> 
> We already have:
> 
> h/?/F1        Show this window
> UP/DOWN/PGUP
> PGDN/SPACE    Navigate
> q/ESC/CTRL+C  Exit browser
> 
> For multiple event sessions:
> 
> TAB/UNTAB     Switch events
> 
> For symbolic views (--sort has sym):
> 
> ->            Zoom into DSO/Threads & Annotate current symbol
> <-            Zoom out
> a             Annotate current symbol
> C             Collapse all callchains
> d             Zoom into current DSO
> D             Show some developer debug info
> E             Expand all callchains
> F             Toggle percentage of filtered entries
> H             Display column headers
> i             Show header information
> P             Print histograms to perf.hist.N
> r             Run available scripts
> s             Switch to another data file in PWD
> t             Zoom into current Thread
> V             Verbose (DSO names in callchains, etc)
> z             Toggle zeroing of samples 
> /             Filter symbol by name 

Is 'f' (for 'freeze') still available?

Thanks,

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

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

#1168558 — Re: [RFC] hotkey for disabling/enabling events in 'perf top' TUI was Re: [GIT PULL 0/8] perf/core improvements and fixes

FromIngo Molnar <mingo@kernel.org>
Date2015-06-19 08:30 +0200
SubjectRe: [RFC] hotkey for disabling/enabling events in 'perf top' TUI was Re: [GIT PULL 0/8] perf/core improvements and fixes
Message-ID<pCXYl-7Dj-9@gated-at.bofh.it>
In reply to#1168289
* Arnaldo Carvalho de Melo <acme@kernel.org> wrote:

> Em Thu, Jun 18, 2015 at 10:58:13PM +0200, Ingo Molnar escreveu:
> > * Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> > > > Btw., it would be nice if the status line carried information about whether 
> > > > collection is 'frozen' or running, at a glance. A hint might also suggest how to 
> > > > unfreeze the session - in case someone pressed Ctrl-Z to suspend the perf top 
> > > > session ...
> > > 
> > > Right, and I think we better find other hotkey and make Ctrl+Z work like
> > > with other tools, i.e. suspend:
> > > 
> > > [acme@zoo linux]$ mutt
> > > 
> > > [1]+  Stopped                 mutt
> > > [acme@zoo linux]$ vim
> > > 
> > > [2]+  Stopped                 vim
> > > [acme@zoo linux]$ 
> > > 
> > > The perf TUI should work like that as well...
> > > 
> > > Ideas?
> > > 
> > > We already have:
> > > 
> > > h/?/F1        Show this window
> > > UP/DOWN/PGUP
> <SNIP>
> > > z             Toggle zeroing of samples 
> > > /             Filter symbol by name 
> > 
> > Is 'f' (for 'freeze') still available?
> 
> It is available in both the 'report'/'top' (aka the "hists" browser) and in the 
> annotate browser, so I'll go with it, and leave CTRL+z alone, then make it it 
> suspend.

Sounds good to me!

Thanks,

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web