It turns out that, actually, the can_attach(), cancel_attach() and attach() code is broken -- we're incrementing a ref on the old_css of a task in can_attach(). Then we decrement the ref on a *different* css (because the task has been migrated). This is clearly a bad thing. Should we make cgroup_migrate() deal with the accounting for us (by getting it to grab a ref before can_attach() and dropping it after the attach succeeds or fails?).
-- Aleksa Sarai (cyphar) www.cyphar.com -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
