From: Ian Rogers <irogers@google.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Namhyung Kim <namhyung@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>,
Kan Liang <kan.liang@linux.intel.com>,
James Clark <james.clark@linaro.org>,
Xu Yang <xu.yang_2@nxp.com>,
"Masami Hiramatsu (Google)" <mhiramat@kernel.org>,
Collin Funk <collin.funk1@gmail.com>,
Howard Chu <howardchu95@gmail.com>,
Weilin Wang <weilin.wang@intel.com>,
Andi Kleen <ak@linux.intel.com>,
"Dr. David Alan Gilbert" <linux@treblig.org>,
Thomas Richter <tmricht@linux.ibm.com>,
Tiezhu Yang <yangtiezhu@loongson.cn>,
Gautam Menghani <gautam@linux.ibm.com>,
Thomas Falcon <thomas.falcon@intel.com>,
Chun-Tse Shao <ctshao@google.com>,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org
Subject: Re: [PATCH v7 00/16] New perf ilist app
Date: Wed, 23 Jul 2025 14:30:47 -0700 [thread overview]
Message-ID: <CAP-5=fXCCKn5EREFCrS5JpRf_Y2ueWAd0cXy4fBfLQcP3XmqJA@mail.gmail.com> (raw)
In-Reply-To: <aIE24fDAuIqDXX3h@x1>
On Wed, Jul 23, 2025 at 12:24 PM Arnaldo Carvalho de Melo
<acme@kernel.org> wrote:
>
> On Wed, Jul 23, 2025 at 04:11:13PM -0300, Arnaldo Carvalho de Melo wrote:
> > On Wed, Jul 23, 2025 at 04:08:55PM -0300, Arnaldo Carvalho de Melo wrote:
> > > On Wed, Jul 23, 2025 at 11:00:18AM -0700, Namhyung Kim wrote:
> > > > > I think there is some follow up for "make install" for scripts like
> > > > > these, but I'm keen for the python API to move forward.
>
> > > > I'll review the series today so that we can get some part of it, at
> > > > least. Basically I think we need a wrapper script like perf-ilist to
> > > > run this easily (maybe with documentation).
>
> > > I just tried, with the series applied:
>
> > > root@number:~# perf ilist
> > > perf: 'ilist' is not a perf-command. See 'perf --help'.
>
> > > Did you mean this?
> > > list
> > > root@number:~#
>
> > > Now trying to figure out why it is not running.
>
> > So it is not wired up like 'perf archive', trying it directly:
>
> > root@number:~# ~acme/git/perf-tools-next/tools/perf/python/ilist.py
> > Traceback (most recent call last):
> > File "/home/acme/git/perf-tools-next/tools/perf/python/ilist.py", line 11, in <module>
> > from textual import on
> > ModuleNotFoundError: No module named 'textual'
> > root@number:~#
>
> > I thought there was some discussion about catching this exception and
> > providing guidance, lemme try...
>
> root@number:~# sudo dnf install python-textual
> Updating and loading repositories:
> google-chrome 100% | 6.1 KiB/s | 1.3 KiB | 00m00s
> Copr repo for PyCharm owned by phracek 100% | 2.0 KiB/s | 2.1 KiB | 00m01s
> RPM Fusion for Fedora 42 - Nonfree - NVIDIA Driver 100% | 2.6 KiB/s | 2.5 KiB | 00m01s
> RPM Fusion for Fedora 42 - Nonfree - Steam 100% | 5.8 KiB/s | 2.4 KiB | 00m00s
> google-chrome 100% | 7.2 KiB/s | 3.2 KiB | 00m00s
> Copr repo for PyCharm owned by phracek 100% | 15.7 KiB/s | 4.8 KiB | 00m00s
> Repositories loaded.
> Package Arch Version Repository Size
> Installing:
> python3-textual noarch 1.0.0-1.fc42 fedora 6.6 MiB
> Installing dependencies:
> python3-linkify-it-py noarch 2.0.3-4.fc42 fedora 110.4 KiB
> python3-markdown-it-py noarch 3.0.0-8.fc42 fedora 496.3 KiB
> python3-markdown-it-py+linkify noarch 3.0.0-8.fc42 fedora 9.0 KiB
> python3-markdown-it-py+plugins noarch 3.0.0-8.fc42 fedora 9.0 KiB
> python3-mdit-py-plugins noarch 0.4.2-2.fc42 fedora 289.4 KiB
> python3-mdurl noarch 0.1.2-9.fc42 fedora 41.4 KiB
> python3-platformdirs noarch 4.2.2-4.fc42 fedora 162.0 KiB
> python3-pygments noarch 2.18.0-4.fc42 fedora 10.6 MiB
> python3-rich noarch 13.9.4-2.fc42 fedora 2.5 MiB
> python3-uc-micro-py noarch 1.0.3-4.fc42 fedora 13.1 KiB
>
> Transaction Summary:
> Installing: 11 packages
>
> Total size of inbound packages is 5 MiB. Need to download 5 MiB.
> After this operation, 21 MiB extra will be used (install 21 MiB, remove 0 B).
> Is this ok [y/N]: y
> <SNIP>
> [13/13] Installing python3-textual-0:1.0.0-1.fc42.noarch 100% | 10.1 MiB/s | 6.8 MiB | 00m01s
> Complete!
> root@number:~# ~acme/git/perf-tools-next/tools/perf/python/ilist.py
> ╭────────────────────────────────────────────────────────────────── Traceback (most recent call last) ──────────────────────────────────────────────────────────────────╮
> │ /home/acme/git/perf-tools-next/tools/perf/python/ilist.py:470 in compose │
> │ │
> │ 467 │ │ │ return tree │
> │ 468 │ │ │
> │ 469 │ │ yield Header(id="header") │
> │ ❱ 470 │ │ yield Horizontal(Vertical(metric_event_tree(), id="events"), │
> │ 471 │ │ │ │ │ │ Vertical(Label("event name", id="event_name"), │
> │ 472 │ │ │ │ │ │ │ │ Static("description", markup=False, id="event_descript │
> │ 473 │ │ │ │ │ │ │ │ )) │
> │ │
> │ ╭───────────────────────────────────────────────── locals ─────────────────────────────────────────────────╮ │
> │ │ self = IListApp(title='Interactive Perf List', classes={'-dark-mode'}, pseudo_classes={'dark', 'focus'}) │ │
> │ ╰──────────────────────────────────────────────────────────────────────────────────────────────────────────╯ │
> │ │
> │ /home/acme/git/perf-tools-next/tools/perf/python/ilist.py:433 in metric_event_tree │
> │ │
> │ 430 │ │ │ """Create tree of PMUs and metricgroups with events or metrics under.""" ╭─────────── locals ────────────╮ │
> │ 431 │ │ │ tree: Tree[TreeValue] = Tree("Root", id="root") │ pmus = TreeNode('PMUs', None) │ │
> │ 432 │ │ │ pmus = tree.root.add("PMUs") │ tree = Tree(id='root') │ │
> │ ❱ 433 │ │ │ for pmu in perf.pmus(): ╰───────────────────────────────╯ │
> │ 434 │ │ │ │ pmu_name = pmu.name().lower() │
> │ 435 │ │ │ │ pmu_node = pmus.add(pmu_name) │
> │ 436 │ │ │ │ try: │
> ╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
> AttributeError: module 'perf' has no attribute 'pmus'
> root@number:~#
>
> Ok, that was with the 'python-perf3' rpm package on fedora 42, trying
> with the new one...
>
> root@number:~# export PYTHONPATH=/tmp/build/perf-tools-next/python
> root@number:~# ~acme/git/perf-tools-next/tools/perf/python/ilist.py
>
> Cool stuff!
>
> Lots of flashing lights! :-)
>
> Interesting to quickly browse all those events, I like it.
>
> I searched for "wakeup" and stumbled on ftrace:wakeup failures, but that
> should be just a minor adjustment, some exception list:
>
> root@number:~# ls -la /sys/kernel/tracing/events/ftrace/wakeup/
> total 0
> drwxr-xr-x. 1 root root 0 Jul 23 16:04 .
> drwxr-xr-x. 1 root root 0 Jul 23 16:04 ..
> -r--r-----. 1 root root 0 Jul 23 16:04 format
> -r--r-----. 1 root root 0 Jul 23 16:19 hist
> root@number:~#
> root@number:~# ls -la /sys/kernel/tracing/events/sched/sched_wakeup/
> total 0
> drwxr-xr-x. 1 root root 0 Jul 23 16:04 .
> drwxr-xr-x. 1 root root 0 Jul 23 16:04 ..
> -rw-r-----. 1 root root 0 Jul 23 16:19 enable
> -rw-r-----. 1 root root 0 Jul 23 16:19 filter
> -r--r-----. 1 root root 0 Jul 23 16:04 format
> -r--r-----. 1 root root 0 Jul 23 16:19 hist
> -r--r-----. 1 root root 0 Jul 23 16:04 id
> -rw-r-----. 1 root root 0 Jul 23 16:19 trigger
> root@number:~#
>
> Do you know how to take text screen shots in textual?
So there's a built-in screen shot thing. Press ctrl+p to bring up the
inbuilt command palette, then select "Save screenshot" which will save
an SVG screenshot. It should be possible to force a lowly terminal and
then grab the characters, I tried but didn't get very far with it.
There are console recorder apps. My preference has just been to grab
regular screenshots and then cut out a PNG. Attaching any kind of
non-standard character or image to an LKML post I think would cause it
to be bounced. I've posted some images here:
ilist working first on events:
https://fosstodon.org/@irogers/114677152263351891
ilist metrics working on hybrid:
https://fosstodon.org/@irogers/114831660998075850
ilist running on WSL on AMD:
https://fosstodon.org/@irogers/114849910913098482
> Apart from the super minor nits, thanks for working on this:
>
> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Thanks! I think there are still improvements to be had like improving
the layout of the screen, deduplicating PMUs, etc. nothing that can't
be incrementally added. Probably the biggest thing is what to do for
"make install" and packaging it [1]. I think it is a very nice way to
discover what perf can inform you about, and I think everyone I've
talked to prefers doing it with this app rather than trying to get the
same data from regular `perf list`. For example, I was surprised how
much data I could get from WSL.
Thanks,
Ian
[1] (some thoughts on packaging in reply to Namhyung)
https://lore.kernel.org/lkml/CAP-5=fX1j_MdzjqLTYeNGFTWLaQp8Dmbi093QN98b2pNgFcpBA@mail.gmail.com/
> - Arnaldo
prev parent reply other threads:[~2025-07-23 21:30 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-14 16:43 [PATCH v7 00/16] New perf ilist app Ian Rogers
2025-07-14 16:43 ` [PATCH v7 01/16] perf python: Add more exceptions on error paths Ian Rogers
2025-07-23 16:48 ` Arnaldo Carvalho de Melo
2025-07-23 18:12 ` Namhyung Kim
2025-07-23 18:21 ` Ian Rogers
2025-07-23 22:24 ` Ian Rogers
2025-07-14 16:43 ` [PATCH v7 02/16] perf jevents: Add common software event json Ian Rogers
2025-07-23 16:52 ` Arnaldo Carvalho de Melo
2025-07-23 18:24 ` Namhyung Kim
2025-07-23 22:34 ` Ian Rogers
2025-07-24 0:10 ` Namhyung Kim
2025-07-24 1:47 ` Ian Rogers
2025-07-24 16:58 ` Namhyung Kim
2025-07-14 16:43 ` [PATCH v7 03/16] perf parse-events: Remove non-json software events Ian Rogers
2025-07-24 0:20 ` Namhyung Kim
2025-07-14 16:43 ` [PATCH v7 04/16] perf tp_pmu: Factor existing tracepoint logic to new file Ian Rogers
2025-07-23 16:57 ` Arnaldo Carvalho de Melo
2025-07-14 16:43 ` [PATCH v7 05/16] perf tp_pmu: Add event APIs Ian Rogers
2025-07-23 18:57 ` Arnaldo Carvalho de Melo
2025-07-14 16:43 ` [PATCH v7 06/16] perf list: Remove tracepoint printing code Ian Rogers
2025-07-14 16:43 ` [PATCH v7 07/16] perf list: Skip ABI PMUs when printing pmu values Ian Rogers
2025-07-14 16:43 ` [PATCH v7 08/16] perf python: Improve the tracepoint function if no libtraceevent Ian Rogers
2025-07-14 16:43 ` [PATCH v7 09/16] perf python: Add basic PMU abstraction and pmus sequence Ian Rogers
2025-07-14 16:43 ` [PATCH v7 10/16] perf python: Add function returning dictionary of all events on a PMU Ian Rogers
2025-07-14 16:43 ` [PATCH v7 11/16] perf ilist: Add new python ilist command Ian Rogers
2025-07-21 7:32 ` Gautam Menghani
2025-07-21 13:41 ` Ian Rogers
2025-07-23 18:33 ` Falcon, Thomas
2025-07-23 21:33 ` Ian Rogers
2025-07-14 16:44 ` [PATCH v7 12/16] perf python: Add parse_metrics function Ian Rogers
2025-07-14 16:44 ` [PATCH v7 13/16] perf python: Add evlist metrics function Ian Rogers
2025-07-14 16:44 ` [PATCH v7 14/16] perf python: Add evlist compute_metric Ian Rogers
2025-07-14 16:44 ` [PATCH v7 15/16] perf python: Add metrics function Ian Rogers
2025-07-14 16:44 ` [PATCH v7 16/16] perf ilist: Add support for metrics Ian Rogers
2025-07-23 15:32 ` [PATCH v7 00/16] New perf ilist app Ian Rogers
2025-07-23 18:00 ` Namhyung Kim
2025-07-23 18:15 ` Ian Rogers
2025-07-23 19:08 ` Arnaldo Carvalho de Melo
2025-07-23 19:11 ` Arnaldo Carvalho de Melo
2025-07-23 19:24 ` Arnaldo Carvalho de Melo
2025-07-23 21:30 ` Ian Rogers [this message]
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='CAP-5=fXCCKn5EREFCrS5JpRf_Y2ueWAd0cXy4fBfLQcP3XmqJA@mail.gmail.com' \
--to=irogers@google.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=collin.funk1@gmail.com \
--cc=ctshao@google.com \
--cc=gautam@linux.ibm.com \
--cc=howardchu95@gmail.com \
--cc=james.clark@linaro.org \
--cc=jolsa@kernel.org \
--cc=kan.liang@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=linux@treblig.org \
--cc=mark.rutland@arm.com \
--cc=mhiramat@kernel.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=thomas.falcon@intel.com \
--cc=tmricht@linux.ibm.com \
--cc=weilin.wang@intel.com \
--cc=xu.yang_2@nxp.com \
--cc=yangtiezhu@loongson.cn \
/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).