Re: [PATCH v1 1/6] perf report: Sort child tasks by tid

From: Namhyung Kim <namhyung@kernel.org>
To: Ian Rogers <irogers@google.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	 Ingo Molnar <mingo@redhat.com>,
	Mark Rutland <mark.rutland@arm.com>,
	 Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@kernel.org>,
	 Adrian Hunter <adrian.hunter@intel.com>,
	Oliver Upton <oliver.upton@linux.dev>,
	 Yang Jihong <yangjihong1@huawei.com>,
	linux-kernel@vger.kernel.org,  linux-perf-users@vger.kernel.org,
	bpf@vger.kernel.org
Subject: Re: [PATCH v1 1/6] perf report: Sort child tasks by tid
Date: Fri, 16 Feb 2024 12:25:38 -0800	[thread overview]
Message-ID: <CAM9d7cjb7XDH6JgbKg07RJ4ypzPm12+mpBDE7mGXyUP3SJ8V1g@mail.gmail.com> (raw)
In-Reply-To: <CAP-5=fXeLX8i8sK8EVquDqnV31rtum_K4TLtw5nG=nfL9-PKvQ@mail.gmail.com>

On Wed, Feb 14, 2024 at 9:42 AM Ian Rogers <irogers@google.com> wrote:
>
> On Wed, Feb 14, 2024 at 9:24 AM Arnaldo Carvalho de Melo
> <acme@kernel.org> wrote:
> >
> > On Tue, Feb 13, 2024 at 10:37:03PM -0800, Ian Rogers wrote:
> > > Commit 91e467bc568f ("perf machine: Use hashtable for machine
> > > threads") made the iteration of thread tids unordered. The perf report
> > > --tasks output now shows child threads in an order determined by the
> > > hashing. For example, in this snippet tid 3 appears after tid 256 even
> > > though they have the same ppid 2:
> > >
> > > ```
> > > $ perf report --tasks
> > > %      pid      tid     ppid  comm
> > >          0        0       -1 |swapper
> > >          2        2        0 | kthreadd
> > >        256      256        2 |  kworker/12:1H-k
> > >     693761   693761        2 |  kworker/10:1-mm
> > >    1301762  1301762        2 |  kworker/1:1-mm_
> > >    1302530  1302530        2 |  kworker/u32:0-k
> > >          3        3        2 |  rcu_gp
> > > ...
> > > ```
> > >
> > > The output is easier to read if threads appear numerically
> > > increasing. To allow for this, read all threads into a list then sort
> > > with a comparator that orders by the child task's of the first common
> > > parent. The list creation and deletion are created as utilities on
> > > machine.  The indentation is possible by counting the number of
> > > parents a child has.
> > >
> > > With this change the output for the same data file is now like:
> > > ```
> > > $ perf report --tasks
> > > %      pid      tid     ppid  comm
> > >          0        0       -1 |swapper
> > >          1        1        0 | systemd
> > >        823      823        1 |  systemd-journal
> > >        853      853        1 |  systemd-udevd
> > >       3230     3230        1 |  systemd-timesyn
> > >       3236     3236        1 |  auditd
> > >       3239     3239     3236 |   audisp-syslog
> > >       3321     3321        1 |  accounts-daemon
> >
> >
> > Since we're adding extra code for sorting wouldn't be more convenient to
> > have this done in an graphically hierarchical output?
> >
> > But maybe to make this honour asking for a CSV output the above is
> > enough? Or can we have both, i.e. for people just doing --tasks, the
> > hirarchical way, for CSV, then like above, with the comma separator.
> >
> > But then perf stat has -x to ask for CSV that is used by the more
> > obscure --exclude-other option :-\
> >
> > Maybe we need a --csv that is consistent accross all tools.
>
> I've no objection to a graphical/CSV output, I was in this code as I
> was restructuring it for memory savings. Fixing the output ordering
> was a side-effect, the "graphical" sorting/indentation is mentioned as
> it is carrying forward a behavior from the previous code but done in a
> somewhat different way. Let's have other output things as follow up
> work.

Agreed, maybe a good project for GSoC students..

Thanks,
Namhyung