Applying with these fixes on my side:
- Feature lines: "ve: ve generic structures" on 1-3, "ve: CRIU
  support" on 4 and 6 (instead of "ve/overlayfs", which does not
  exist), "fix ms/overlayfs" on 5
- patch 2 message: "ve_capable and friends" -> ve_capable() and
  ve_capable_noaudit(), plus one sentence that override_ve() refuses a
  CT task any VE other than its own with -EPERM
- patch 3 message: init_cred == NULL means the VE was never started
- patch 5 message: the two mainline commits cited by 12-char sha and
  subject instead of GitHub links, first paragraph reworded
- patch 6 message: the compile-error sentence limited to
  ovl_override_creds() callers, with the ovl_creds() users named;
  "returns" -> "saves", "would be handled further" -> "will be
  handled"

On Mon, 16 Nov 2026 20:20:38 +0100, Vladimir Riabchun wrote:
> Vladimir Riabchun (6):
>       ve: Remove STOPPED state
>       ve: Introduce override_ve/revert_ve helpers
>       ve: Defer ve creds drop until ve is completely destroyed
>       ovl: Keep creator VE in overlayfs superblock
>       ovl: Fix leaking fd references in ovl_copyfile
>       ve/overlayfs: Override VE together with creds

The code is fine: the rebase onto the ve_name fix, the plain
ve_name(new_ve) print, the out_put_real label in ovl_copyfile() and
the "true" from ovl_lower_positive() all hold. The texts above are
what was left from my earlier rounds.

I did not add the Reviewed-by of Vasileios Almpanis: he gave it on v5,
and patches 2 and 6 were redesigned in v6. Vasileios, if you are fine
with v8, please reply with it.

The comments "ve isn't started" in kernel/capability.c and
net/netlink/af_netlink.c stay as they are; "ve was never started"
would be more exact, but it is not worth another round.

Thanks.

-- 
Konstantin Khorenko,
Virtuozzo Linux Kernel Team

_______________________________________________
Devel mailing list
[email protected]
https://lists.openvz.org/mailman/listinfo/devel

Reply via email to