From: Li Wang <li.wang@linux.dev>
To: Andrea Cervesato <andrea.cervesato@suse.com>
Cc: ltp@lists.linux.it
Subject: Re: [LTP] [PATCH v3 3/4] tst_kconfig: add module mappings for vsock/vsock_loopback
Date: Wed, 14 Oct 2026 13:01:12 +0800 [thread overview]
Message-ID: <as8MmB6eIyIh3tT2@linux.dev> (raw)
In-Reply-To: <iDTNt1jXwFqCtxRdYVwoga28j8OKFAmDRyJ6@localhost.localdomain>
On Mon, Oct 12, 2026 at 03:51:32PM +0000, Andrea Cervesato wrote:
> Hi Li,
>
> > The test needs root only to load the modules; the reproducer runs as
> > an ordinary nonroot process in the initial namespace, which can use
> > loopback VSOCK on nonprivileged ports. Drop privileges to nobody after
> > module loading and regain root in cleanup() so the modules can be
> > unloaded.
>
> Please drop the root _only_ if we are root, we shouldn'f force the test
> to be root always, even when we are user.
After thinking it over, I'd prefer to leave this patch unchanged:
The test does need root, but not for the reproducer itself -- it needs
root for setup and cleanup:
- loading vsock_loopback requires CAP_SYS_MODULE;
- unloading it in cleanup() also requires CAP_SYS_MODULE.
Without .needs_root = 1, a non-root run would skip the automatic
module load in tst_kconfig, bind() would then fail with
EADDRNOTAVAIL and the test would abort with TBROK instead of being
skipped cleanly.
Keeping .needs_root = 1 and dropping privileges to nobody in setup()
is exactly what makes the reproducer run as an ordinary non-root
process while still allowing the modules to be loaded and unloaded.
--
Regards,
Li Wang
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next prev parent reply other threads:[~2026-10-14 5:01 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-10 3:36 [LTP] [PATCH v3 0/4] tst_kconfig: automatic module load/unload for CONFIG_*=m Li Wang
2026-10-10 3:36 ` [LTP] [PATCH v3 1/4] lib: add module is loaded check Li Wang
2026-10-10 9:38 ` [LTP] " linuxtestproject.agent
2026-10-12 15:48 ` [LTP] [PATCH v3 1/4] " Andrea Cervesato via ltp
2026-10-14 3:18 ` Li Wang
2026-10-10 3:36 ` [LTP] [PATCH v3 2/4] tst_kconfig: load and unload kernel modules for CONFIG_*=m Li Wang
2026-10-12 16:09 ` Andrea Cervesato via ltp
2026-10-14 3:36 ` Li Wang
2026-10-10 3:36 ` [LTP] [PATCH v3 3/4] tst_kconfig: add module mappings for vsock/vsock_loopback Li Wang
2026-10-12 15:51 ` Andrea Cervesato via ltp
2026-10-14 3:43 ` Li Wang
2026-10-14 5:01 ` Li Wang [this message]
2026-10-10 3:36 ` [LTP] [PATCH v3 4/4] testcases: drop manual tst_modprobe() now handled by tst_kconfig Li Wang
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=as8MmB6eIyIh3tT2@linux.dev \
--to=li.wang@linux.dev \
--cc=andrea.cervesato@suse.com \
--cc=ltp@lists.linux.it \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.