From: Hagen Paul Pfeifer <hagen@jauu.net>
To: linux-kernel@vger.kernel.org
Cc: Jiri Olsa <jolsa@kernel.org>, Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: perf script, libperf: python binding bug (bytearrays vs. strings)
Date: Sun, 27 Sep 2020 09:43:12 +0200 [thread overview]
Message-ID: <20200927074312.GA3664097@laniakea> (raw)
Hallo Jiri, Arnaldo,
after updating Debian (probably with the advent of Python 3.8.5, guessing)
I get a wired behavior with python scripting. The error is that the python type
for prev_comm and next_comm are not strings anymore, rather bytearrays. Which
are incompatible types and scripts will not work anymore. NOTE: common_comm is
still fine (see swapper & mutex-thread-co), so they must be treated internal
differently compared to prev_comm and next_comm and possibly show a way to solve
this problem!
After bisecting the kernel (perf) even back to v5.6 the problem still exist.
Compiling perf with PYTHON=python2 do not show any problems - no problems in
the Python2 world. So I assume with Python 3.8.5 (or other helper library)
something changed internally. I assume the cause exists in perf forever but
is now triggered with the new Python3 version.
How to reproduce:
make PYTHON=python3
./perf record -e sched:sched_switch -a -- sleep 5
./perf script --gen-script py
./perf script -s ./perf-script.py
[..]
sched__sched_switch 7 563231.759525792 0 swapper prev_comm=bytearray(b'swapper/7\x00\x00\x00\x00\x00\x00\x00'), prev_pid=0, prev_prio=120, prev_state=, next_comm=bytearray(b'mutex-thread-co\x00'), next_pid=3447985, next_prio=120
Sample: {addr=0, cpu=7, datasrc=84410401, datasrc_decode=N/A|SNP N/A|TLB N/A|LCK N/A, ip=18446744072189289569, period=1, phys_addr=0, pid=0, tid=0, time=563231759525792, transaction=0, values=[(0, 0)], weight=0}
sched__sched_switch 7 563231.759582596 3447985 mutex-thread-co prev_comm=bytearray(b'mutex-thread-co\x00'), prev_pid=3447985, prev_prio=120, prev_state=, next_comm=bytearray(b'swapper/7\x00\x00\x00\x00\x00\x00\x00'), next_pid=0, next_prio=120
Sample: {addr=0, cpu=7, datasrc=84410401, datasrc_decode=N/A|SNP N/A|TLB N/A|LCK N/A, ip=18446744072189289569, period=1, phys_addr=0, pid=3447983, tid=3447985, time=563231759582596, transaction=0, values=[(0, 0)], weight=0}
See =bytearray(b'swapper/7\x00\x00\x00\x00\x00\x00\x00') - should be swapper/7
Note: the byte array has the length of 16 - exactly like the kernel
(TASK_COMM_LEN). I assume this is somehow copied directly into the variables
and not stringified anymore.
Even worse: I discovered bytearrays which are not correctly "memseted":
bytearray(b'chrome\x00sandbox\x00\x00')
chrome should be the comm name, but is 'chromesandbox' somehow. See the null
bytes in between.
Jiri, Arnaldo - I tried to fix this. But the Python binding magic for the
automatically generated events are hard to get comfy.
Hagen
PS: assume this fix is also kernel stable relevant.
next reply other threads:[~2020-09-27 7:43 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-27 7:43 Hagen Paul Pfeifer [this message]
2020-09-27 9:20 ` perf script, libperf: python binding bug (bytearrays vs. strings) Hagen Paul Pfeifer
2020-09-28 10:08 ` Jiri Olsa
2020-09-28 10:43 ` Hagen Paul Pfeifer
2020-09-28 13:39 ` Jiri Olsa
2020-09-28 19:19 ` Arnaldo Carvalho de Melo
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=20200927074312.GA3664097@laniakea \
--to=hagen@jauu.net \
--cc=acme@redhat.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.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