From: Tom Zanussi <tom.zanussi@linux.intel.com>
To: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>,
linux-kernel@vger.kernel.org, Ingo Molnar <mingo@kernel.org>,
Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Subject: Re: [PATCH v2 00/12] tracing: Unifying dynamic event interface
Date: Wed, 28 Nov 2018 17:42:22 -0600 [thread overview]
Message-ID: <ff8ca3062cad3747a664e6888790cc3c58e24c7e.camel@linux.intel.com> (raw)
In-Reply-To: <20181128163101.d86844338bcba3fc3b0580f1@kernel.org>
Hi Masami,
On Wed, 2018-11-28 at 16:31 +0900, Masami Hiramatsu wrote:
> Ping?
>
> Hi Tom,
>
> This series, especially [09/12] tracing: Remove unneeded
> synth_event_mutex
> will effect your current working series. Please tell me your opinion.
>
Sorry for the delay in reviewing this - I completely forgot about it in
my inbox.
It's all very nice, and the mutex updates along with the dyn_event
management make that part of the code much cleaner, thanks!
In any case, you can have my
Reviewed-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Tested-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Thanks,
Tom
> Thank you,
>
> On Mon, 5 Nov 2018 17:59:46 +0900
> Masami Hiramatsu <mhiramat@kernel.org> wrote:
>
> > Hi,
> >
> > This is v2 series of unifying dynamic event interface on ftrace.
> > Currently ftrace has 3 dynamic event interfaces, kprobes, uprobes
> > and synthetic. This series unifies those dynamic event interfaces
> > to "dynamic_events" so that we can add other dynamic events easily
> > on same interface, e.g. function events.
> > The older interfaces are left on the tracefs for backward
> > compatibility.
> >
> > dynamic_events syntax has no difference from kprobe_events and
> > uprobe_events. You can use same syntax for dynamic_events
> > interface.
> > For synthetic events, similar to the probe events, dynamic_events
> > adds "s:[GROUP/]" prefix, where the "GROUP/" must be "synthetic/".
> >
> > s:[synthetic/]<event-name> type arg [type arg]...
> >
> > E.g.
> >
> > $ echo 'wakeup_latency u64 lat pid_t pid char' > synthetic_events
> >
> > is same as
> >
> > $ echo 's:wakeup_latency u64 lat pid_t pid char' > dynamic_events
> >
> > Or
> >
> > $ echo 's:synthetic/wakeup_latency u64 lat pid_t pid char' >
> > dynamic_events
> >
> > This series modifies synthetic event interface behavior a bit,
> > reorder lock dependency and related cleanups so that we can
> > integrate
> > the synthetic event to dynamic_events interface.
> >
> > In this version, I changed the generic '!' erase command, which
> > now supports entire line style like other interfaces. So you can
> > delete events via dynamic_events as below
> >
> > $ cat dynamic_events | while read line; \
> > do echo "!$line" >> dynamic_events; done
> >
> > Also, the big change will be removing dyn_event_mutex and
> > synth_event_mutex because all those parts are protected by
> > event_mutex.
> >
> > Changes from v2 are here;
> >
> > New patches:
> > - Reorder event_mutex and synth_event_mutex to solve
> > AB-BA deadlock correctly. ([2/12])
> > - Simplify creation and deletion of synthetic event. ([3/12])
> > - Retern -ENOENT if there is no synthetic event when deleting
> > ([4/12])
> > - Integrate similar probe argument parsers ([5/12])
> > - Use dyn_event framework for synthetic events ([9/12])
> > - Remove synth_event_mutex ([10/12])
> > - Remove unused APIs ([11/12])
> >
> > Modified patches:
> > [6/12] - [8/12]
> > - Generalize delete event and export as dyn_event_release_all().
> > - Add match operation for find deleting event.
> > - Reorder event_mutex and dyn_event_mutex to solve lock dependency
> > issue.
> > - Pass const char **argv for create operation and use -ECANCELED
> > to
> > signal for trying next dyn_event_operations.
> > - Remove dyn_event_mutex.
> >
> > [12/12]
> > - Accept entire line, but instead of checking the given entire
> > line
> > strictly, simply checking the event and group name.
> >
> > Tom, thanks for your Ack for v1 series. Since I changed many things
> > from v1 (not only minor change), I decided to not add your Ack for
> > this version. Anyway, what I've added in this version are related
> > to
> > synthetic events. I need your review for those.
> > (especially removing synth_event_mutex)
> >
> > You can try it from my git tree.
> >
> > https://github.com/mhiramat/linux/tree/unify-dynamic-events-v2
> >
> > Thank you,
> >
> > ---
> >
> > Masami Hiramatsu (12):
> > tracing/uprobes: Add busy check when cleanup all uprobes
> > tracing: Lock event_mutex before synth_event_mutex
> > tracing: Simplify creation and deletion of synthetic event
> > tracing: Integrate similar probe argument parsers
> > tracing: Add unified dynamic event framework
> > tracing/kprobes: Use dyn_event framework for kprobe events
> > tracing/uprobes: Use dyn_event framework for uprobe events
> > tracing: Use dyn_event framework for synthetic events
> > tracing: Remove unneeded synth_event_mutex
> > tracing: Remove orphaned trace_add/remove_event_call
> > functions
> > tracing: Add generic event-name based remove event method
> > selftests/ftrace: Add testcases for dynamic event
> >
> >
> > Documentation/trace/kprobetrace.rst | 3
> > Documentation/trace/uprobetracer.rst | 4
> > include/linux/trace_events.h | 4
> > kernel/trace/Kconfig | 6
> > kernel/trace/Makefile | 1
> > kernel/trace/trace.c | 12 +
> > kernel/trace/trace_dynevent.c | 217
> > ++++++++++++
> > kernel/trace/trace_dynevent.h | 119 +++++++
> > kernel/trace/trace_events.c | 12 -
> > kernel/trace/trace_events_hist.c | 322
> > ++++++++++--------
> > kernel/trace/trace_kprobe.c | 357
> > ++++++++++----------
> > kernel/trace/trace_probe.c | 74 ++++
> > kernel/trace/trace_probe.h | 9 -
> > kernel/trace/trace_uprobe.c | 305
> > ++++++++---------
> > .../ftrace/test.d/dynevent/add_remove_kprobe.tc | 30 ++
> > .../ftrace/test.d/dynevent/add_remove_synth.tc | 27 ++
> > .../ftrace/test.d/dynevent/clear_select_events.tc | 50 +++
> > .../ftrace/test.d/dynevent/generic_clear_event.tc | 49 +++
> > 18 files changed, 1094 insertions(+), 507 deletions(-)
> > create mode 100644 kernel/trace/trace_dynevent.c
> > create mode 100644 kernel/trace/trace_dynevent.h
> > create mode 100644
> > tools/testing/selftests/ftrace/test.d/dynevent/add_remove_kprobe.tc
> > create mode 100644
> > tools/testing/selftests/ftrace/test.d/dynevent/add_remove_synth.tc
> > create mode 100644
> > tools/testing/selftests/ftrace/test.d/dynevent/clear_select_events.
> > tc
> > create mode 100644
> > tools/testing/selftests/ftrace/test.d/dynevent/generic_clear_event.
> > tc
> >
> > --
> > Masami Hiramatsu (Linaro) <mhiramat@kernel.org>
>
>
next prev parent reply other threads:[~2018-11-28 23:42 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-05 8:59 [PATCH v2 00/12] tracing: Unifying dynamic event interface Masami Hiramatsu
2018-11-05 9:00 ` [PATCH v2 01/12] tracing/uprobes: Add busy check when cleanup all uprobes Masami Hiramatsu
2018-12-04 17:43 ` Steven Rostedt
2018-12-07 2:19 ` Masami Hiramatsu
2018-11-05 9:00 ` [PATCH v2 02/12] tracing: Lock event_mutex before synth_event_mutex Masami Hiramatsu
2018-11-05 9:01 ` [PATCH v2 03/12] tracing: Simplify creation and deletion of synthetic event Masami Hiramatsu
2018-11-05 9:01 ` [PATCH v2 04/12] tracing: Integrate similar probe argument parsers Masami Hiramatsu
2018-11-05 9:02 ` [PATCH v2 05/12] tracing: Add unified dynamic event framework Masami Hiramatsu
2018-11-05 9:02 ` [PATCH v2 06/12] tracing/kprobes: Use dyn_event framework for kprobe events Masami Hiramatsu
2018-11-05 9:03 ` [PATCH v2 07/12] tracing/uprobes: Use dyn_event framework for uprobe events Masami Hiramatsu
2018-11-05 9:03 ` [PATCH v2 08/12] tracing: Use dyn_event framework for synthetic events Masami Hiramatsu
2018-11-05 9:04 ` [PATCH v2 09/12] tracing: Remove unneeded synth_event_mutex Masami Hiramatsu
2018-11-05 9:04 ` [PATCH v2 10/12] tracing: Remove orphaned trace_add/remove_event_call functions Masami Hiramatsu
2018-12-04 18:51 ` Steven Rostedt
2018-12-07 2:22 ` Masami Hiramatsu
2018-11-05 9:04 ` [PATCH v2 11/12] tracing: Add generic event-name based remove event method Masami Hiramatsu
2018-11-05 9:05 ` [PATCH v2 12/12] selftests/ftrace: Add testcases for dynamic event Masami Hiramatsu
2018-11-28 7:31 ` [PATCH v2 00/12] tracing: Unifying dynamic event interface Masami Hiramatsu
2018-11-28 23:42 ` Tom Zanussi [this message]
2018-11-29 3:46 ` Steven Rostedt
2018-11-29 5:20 ` Masami Hiramatsu
2018-11-29 15:08 ` Tom Zanussi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ff8ca3062cad3747a664e6888790cc3c58e24c7e.camel@linux.intel.com \
--to=tom.zanussi@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=mingo@kernel.org \
--cc=ravi.bangoria@linux.vnet.ibm.com \
--cc=rostedt@goodmis.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).