From: Shuah Khan <skhan@linuxfoundation.org>
To: Kuniyuki Iwashima <kuniyu@google.com>, Shuah Khan <shuah@kernel.org>
Cc: Kuniyuki Iwashima <kuni1840@gmail.com>,
linux-kselftest@vger.kernel.org,
Shuah Khan <skhan@linuxfoundation.org>
Subject: Re: [PATCH v2] selftests: harness: Support KCOV.
Date: Thu, 6 Nov 2025 14:42:06 -0700 [thread overview]
Message-ID: <84faddcb-07c5-4031-a56d-6785d49bf663@linuxfoundation.org> (raw)
In-Reply-To: <20251028024339.2028774-1-kuniyu@google.com>
On 10/27/25 20:42, Kuniyuki Iwashima wrote:
> While writing a selftest with kselftest_harness.h, I often want to
> check which paths are actually exercised.
>
> Let's support generating KCOV coverage data.
>
> We can specify the output directory via the KCOV_OUTPUT environment
> variable, and the number of instructions to collect via the KCOV_SLOTS
> environment variable.
>
> # KCOV_OUTPUT=$PWD/kcov KCOV_SLOTS=$((4096 * 2)) \
> ./tools/testing/selftests/net/af_unix/scm_inq
>
> Both variables can also be specified as the make variable.
>
> # make -C tools/testing/selftests/ \
> KCOV_OUTPUT=$PWD/kcov KCOV_SLOTS=$((4096 * 4)) \
> kselftest_override_timeout=60 TARGETS=net/af_unix run_tests
>
> The coverage data can be simply decoded with addr2line:
>
> $ cat kcov/* | sort | uniq | addr2line -e vmlinux | grep unix
> net/unix/af_unix.c:1056
> net/unix/af_unix.c:3138
> net/unix/af_unix.c:3834
> net/unix/af_unix.c:3838
> net/unix/af_unix.c:311 (discriminator 2)
> ...
>
> or more nicely with a script embedded in vock [0]:
>
> $ cat kcov/* | sort | uniq > local.log
> $ python3 ~/kernel/tools/vock/report.py \
> --kernel-src ./ --vmlinux ./vmlinux \
> --mode local --local-log local.log --filter unix
> ...
> ------------------------------- Coverage Report --------------------------------
> 📄 net/unix/af_unix.c (276 lines)
> ...
> 942 | static int unix_setsockopt(struct socket *sock, int level, int optname,
> 943 | sockptr_t optval, unsigned int optlen)
> 944 | {
> ...
> 961 | switch (optname) {
> 962 | case SO_INQ:
> 963 > if (sk->sk_type != SOCK_STREAM)
> 964 | return -EINVAL;
> 965 |
> 966 > if (val > 1 || val < 0)
> 967 | return -EINVAL;
> 968 |
> 969 > WRITE_ONCE(u->recvmsg_inq, val);
> 970 | break;
>
> Link: https://github.com/kzall0c/vock/blob/f3d97de9954f9df758c0ab287ca7e24e654288c7/report.py #[0]
> Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
> ---
> v2: Support TEST()
> v1: https://lore.kernel.org/linux-kselftest/20251017084022.3721950-1-kuniyu@google.com/
> ---
> Documentation/dev-tools/kselftest.rst | 41 ++++++
> tools/testing/selftests/Makefile | 14 ++-
> tools/testing/selftests/kselftest_harness.h | 133 +++++++++++++++++++-
> 3 files changed, 178 insertions(+), 10 deletions(-)
>
I am okay with kcov support being added. You probably have to fix a few things.
kernel test robot is still complaining on v2.
thanks,
-- Shuah
next prev parent reply other threads:[~2025-11-06 21:42 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-28 2:42 [PATCH v2] selftests: harness: Support KCOV Kuniyuki Iwashima
2025-11-06 8:50 ` kernel test robot
2025-11-06 21:42 ` Shuah Khan [this message]
2025-11-07 7:08 ` Kuniyuki Iwashima
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=84faddcb-07c5-4031-a56d-6785d49bf663@linuxfoundation.org \
--to=skhan@linuxfoundation.org \
--cc=kuni1840@gmail.com \
--cc=kuniyu@google.com \
--cc=linux-kselftest@vger.kernel.org \
--cc=shuah@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