linux-kernel - Re: [PATCH 0/3] tools lib traceevent: Fixing the API to be less policy driven

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 29 Aug 2019 17:37:51 -0300
From:   Arnaldo Carvalho de Melo <arnaldo.melo@...il.com>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     linux-kernel@...r.kernel.org, linux-trace-devel@...r.kernel.org,
        Ingo Molnar <mingo@...nel.org>, Jiri Olsa <jolsa@...hat.com>,
        Namhyung Kim <namhyung@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Patrick McLean <chutzpah@...too.org>,
        Tzvetomir Stoyanov <tstoyanov@...are.com>
Subject: Re: [PATCH 0/3] tools lib traceevent: Fixing the API to be less
 policy driven

Em Thu, Aug 29, 2019 at 02:00:10PM -0400, Steven Rostedt escreveu:
> On Mon, 05 Aug 2019 16:43:12 -0400
> Steven Rostedt <rostedt@...dmis.org> wrote:
> 
> > Hi Arnaldo and Jiri,
> 
> Hi Arnaldo,
> 
> I think these fell through the cracks.

Got it now, pushed to my perf/core branch, soon to be pushed upstream.

Thanks,

- Arnaldo
 
> -- Steve
> 
> > 
> > We are still working on getting libtraceevent ready to be a stand alone
> > library. Part of that is to audit all the interfaces. We noticed
> > that the most the tep_print_*() interfaces define policy and limit
> > the way an application can display data. Instead of fixing this later
> > and being stuck with a limiting API that we must maintain for backward
> > compatibility, we removed and replaced most of it. perf was only affected
> > by a single function that was removed. These functions are replaced
> > by a more flexible one that allows the user to place what they want
> > where they want it (timestamps, event info, latency format, COMM, PID, etc).
> > 
> > The other noticeable perf change, is that we changed the location to
> > where the plugins are loaded from:
> > 
> >  ${HOME}/.traceevent/plugins
> > 
> > to
> > 
> >  ${HOME}/.local/lib/traceevent/plugins
> > 
> > As Patrick McLean (Gentoo package maintainer) informed us of the
> > XGD layout.
> > 
> > Should we have something the warns people if they have plugins in
> > the old directory. Should we move them on install? Currently, we
> > just ignore them.
> > 
> > Anyway, please add these patches to tip.
> > 
> > Thanks!
> > 
> > -- Steve
> > 
> > 
> > Tzvetomir Stoyanov (3):
> >       tools/lib/traceevent, tools/perf: Changes in tep_print_event_* APIs
> >       tools/lib/traceevent: Remove tep_register_trace_clock()
> >       tools/lib/traceevent: Change user's plugin directory
> > 
> > ----
> >  tools/lib/traceevent/Makefile            |   6 +-
> >  tools/lib/traceevent/event-parse-api.c   |  40 ----
> >  tools/lib/traceevent/event-parse-local.h |   6 -
> >  tools/lib/traceevent/event-parse.c       | 333 +++++++++++++++++--------------
> >  tools/lib/traceevent/event-parse.h       |  30 +--
> >  tools/lib/traceevent/event-plugin.c      |   2 +-
> >  tools/perf/builtin-kmem.c                |   3 +-
> >  tools/perf/util/sort.c                   |   3 +-
> >  tools/perf/util/trace-event-parse.c      |   2 +-
> >  9 files changed, 207 insertions(+), 218 deletions(-)

-- 

- Arnaldo