{"id":"CVE-2026-93240","published":"2026-09-24T16:17:19.827","lastModified":"2026-09-25T13:17:18.003","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nmemcg: make the v1 soft limit knob inert\n\nThe v1 soft limit has been deprecated since v6.12 and nobody has reported\ndepending on it.  Start the removal by decoupling the interface from the\nimplementation: keep memory.soft_limit_in_bytes, but ignore writes to it\nand always report the maximum value on read similar to what\nmemory.kmem.limit_in_bytes already does.\n\nWrites are still parsed, so malformed input keeps returning -EINVAL.  The\nknob now also behaves the same everywhere: it used to return -EOPNOTSUPP\non PREEMPT_RT, where soft limit reclaim has always been disabled.\n\nThis also fixes the syzbot report linked below.  Soft limit reclaim is the\nonly caller that runs shrink_lruvec() from kswapd against a specific\nmemcg, so it is the only way to reach lru_gen_shrink_lruvec() and in turn\nset_mm_walk(), which warns when called from kswapd.","cvssScore":null,"cvssSeverity":null,"cvssVector":null,"cwes":[],"vendors":[],"products":[],"references":[{"url":"https://git.kernel.org/stable/c/0a90e268cce7023f34087bfa4affb42e29ff5f45","tags":[]},{"url":"https://git.kernel.org/stable/c/5ec8f629a300cc40e746d6178f724bdff7683fa7","tags":[]},{"url":"https://git.kernel.org/stable/c/72fb67f6e0e5d78e4f34cf9929d1afdb0fd54203","tags":[]},{"url":"https://git.kernel.org/stable/c/a3417097fb107cea3358b19bcbb4eb655fd67f8c","tags":[]}],"exploitRefs":[],"hasPoc":false,"ai":null}