OpenVZ / Virtuozzo kernel development (devel@openvz.org)
 help / color / mirror / Atom feed
From: no-reply@sashiko.virtuozzo.com
To: Vladimir Riabchun <vladimir.riabchun@virtuozzo.com>,
	vasileios.almpanis@virtuozzo.com, ptikhomirov@virtuozzo.com,
	Konstantin Khorenko <khorenko@virtuozzo.com>
Cc: devel@openvz.org
Subject: Re: [Devel] [PATCH hci-8.1 v7 3/6] ve: Defer ve creds drop until ve is completely destroyed
Date: Mon, 16 Nov 2026 17:26:22 +0000	[thread overview]
Message-ID: <010701a1e5654773-01417b02-9d42-4a4d-80bf-4ef9e94fa6e2-000000@eu-central-1.amazonses.com> (raw)
In-Reply-To: <20261116-ovl-use-orig-ve-v7-3-37daa56081ed@virtuozzo.com>

Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Low] ve: init_cred now outlives VE stop, changing
  NULL-means-not-running checks undocumented
--

commit fb3c64a0ae176947a1145ebda4ba570e97c3e935
Author: Vladimir Riabchun <vladimir.riabchun@virtuozzo.com>

ve: Defer ve creds drop until ve is completely destroyed

This moves the put of ve->init_cred out of ve_drop_context() into a new
ve_drop_cred(), which ve_destroy() calls. That way the creds stay valid
while users of a dead VE, such as an overlayfs superblock, still need them.

> diff --git a/kernel/ve/ve.c b/kernel/ve/ve.c
> --- a/kernel/ve/ve.c
> +++ b/kernel/ve/ve.c
> @@ -378,7 +378,10 @@ static void ve_drop_context(struct ve_struct *ve)
>  	rcu_assign_pointer(ve->ve_nsproxy, NULL);
>  	synchronize_rcu();
>  	put_nsproxy(ve_nsproxy);
> +}
>  
> +static void ve_drop_cred(struct ve_struct *ve)
> +{
>  	put_cred(ve->init_cred);
>  	ve->init_cred = NULL;
>  }

[Severity: Low]
As reported by Konstantin Khorenko on v5 [1], init_cred now stays non-NULL
after a VE stops or fails to start. Is this behavioral change intended to be
left undocumented?

Several users treat init_cred == NULL as "VE not running". These are
ve_capable(), ve_capable_noaudit(), netlink_ve_capable(), ve_init_user_ns()
and sget_fc(). For a dead VE they now take the container branch. For
example, ve_init_user_ns() returns the container user_ns instead of
init_user_ns, which changes the early return in mount_too_revealing(). The
ve0 fallback in ve_capable() is likewise no longer taken.

The patch only renames the comments from "ve isn't running" to "ve isn't
started". The commit message does not say that these checks change for a
dead VE, and it is unclear whether these users were audited. Could the
message describe the new semantics, with NULL meaning only "never
started"?

[1] https://lore.kernel.org/r/179448942112.4060562.14182200426614413151.b4-review@b4

-- 
Sashiko AI review · https://sashiko.vzint.dev/#/patchset/20261116-ovl-use-orig-ve-v7-0-37daa56081ed@virtuozzo.com?part=3

_______________________________________________
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel

  reply	other threads:[~2026-11-16 17:27 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-16 17:18 [Devel] [PATCH hci-8.1 v7 0/6] Fix kernel crash in overlayfs Vladimir Riabchun
2026-11-16 17:18 ` [Devel] [PATCH hci-8.1 v7 1/6] ve: Remove STOPPED state Vladimir Riabchun
2026-11-16 17:23   ` no-reply
2026-11-16 17:35     ` Vladimir Riabchun
2026-11-16 17:18 ` [Devel] [PATCH hci-8.1 v7 2/6] ve: Introduce override_ve/revert_ve helpers Vladimir Riabchun
2026-11-16 17:25   ` no-reply
2026-11-16 17:18 ` [Devel] [PATCH hci-8.1 v7 3/6] ve: Defer ve creds drop until ve is completely destroyed Vladimir Riabchun
2026-11-16 17:26   ` no-reply [this message]
2026-11-16 17:18 ` [Devel] [PATCH hci-8.1 v7 4/6] ovl: Keep creator VE in overlayfs superblock Vladimir Riabchun
2026-11-16 17:27   ` no-reply
2026-11-16 17:18 ` [Devel] [PATCH hci-8.1 v7 5/6] ovl: Fix leaking fd references in ovl_copyfile Vladimir Riabchun
2026-11-16 17:27   ` no-reply
2026-11-16 17:18 ` [Devel] [PATCH hci-8.1 v7 6/6] ve/overlayfs: Override VE together with creds Vladimir Riabchun
2026-11-16 17:32   ` no-reply
2026-11-16 17:45     ` Vladimir Riabchun
2026-11-16 17:43 ` [Devel] [PATCH hci-8.1 v7 0/6] Fix kernel crash in overlayfs Konstantin Khorenko
2026-11-16 20:22   ` Vladimir Riabchun

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=010701a1e5654773-01417b02-9d42-4a4d-80bf-4ef9e94fa6e2-000000@eu-central-1.amazonses.com \
    --to=no-reply@sashiko.virtuozzo.com \
    --cc=devel@openvz.org \
    --cc=khorenko@virtuozzo.com \
    --cc=ptikhomirov@virtuozzo.com \
    --cc=vasileios.almpanis@virtuozzo.com \
    --cc=vladimir.riabchun@virtuozzo.com \
    /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