Thank you for your contribution! Sashiko AI review found 2 potential issues to consider: - [Low] ve: commit message omits override_ve() VE-switching refusal rules - [Low] ve: feature_capable() ignores overrided_ve while ve_capable() honours it --
commit 5277cbabde9ed3d2f2cb580c3e77aa427fe5adf8 Author: Vladimir Riabchun <[email protected]> ve: Introduce override_ve/revert_ve helpers This adds override_ve() and revert_ve(), which mirror override_creds() and revert_creds(). It also adds a task->overrided_ve field that ve_capable() and ve_capable_noaudit() now consult. > ve: Introduce override_ve/revert_ve helpers > > These helpers mirror override_creds/revert_creds and are supposed to > be used together in cases when VE creds may be stored and later used > by the different VE task. > > To keep task_ve consistent for other users, introduce a separate > task->overrided_ve field. Nobody should race on it since its > updates are supposed to go with override_creds, and non-current > processes shouldn't do this. > > ve_capable and friends now use overrided_ve field if it > is non-NULL. > > This field is not supposed to be inherited via fork, so a check > for non-NULL overrided_ve was added in fork path. > > https://virtuozzo.atlassian.net/browse/VSTOR-142405 > Signed-off-by: Vladimir Riabchun <[email protected]> [Severity: Low] As reported by Konstantin Khorenko on v5 [1], the message doesn't describe what override_ve() allows and refuses. A task inside a container that asks for any VE other than its own gets ERR_PTR(-EPERM) and a ratelimited pr_err. Only a host (super-VE) task may switch to an arbitrary VE. The message also doesn't say that dup_task_struct() now warns if a task forks while an override is active. Could the message spell out these rules? > diff --git a/kernel/capability.c b/kernel/capability.c > --- a/kernel/capability.c > +++ b/kernel/capability.c > @@ -406,10 +407,11 @@ bool ve_capable(int cap) > bool ve_capable_noaudit(int cap) > { > struct cred *cred; > + struct ve_struct *ve = current->overrided_ve ?: get_exec_env(); > bool ret; > > rcu_read_lock(); > - cred = get_exec_env()->init_cred; > + cred = ve->init_cred; [Severity: Low] Should feature_capable() be updated as well? It still tests get_exec_env()->features and then calls ve_capable(). Inside an override_ve() section the feature bit is then read from the task's real VE, while the capability check uses overrided_ve->init_cred. If the real VE lacks the feature bit, the fallback capable(cap) runs and the override is ignored. No in-tree caller was found that reaches this from an override section, so this is an inconsistency in semantics rather than a demonstrated failure. [1] https://lore.kernel.org/r/179448942112.4060562.12799497421042932533.b4-review@b4 -- Sashiko AI review ยท https://sashiko.vzint.dev/#/patchset/[email protected]?part=2 _______________________________________________ Devel mailing list [email protected] https://lists.openvz.org/mailman/listinfo/devel
