{ "schema_version": "1.6.1", "id": "RPI-CVE-2026-68415", "modified": "2026-10-09T11:39:44Z", "upstream": [ "CVE-2026-68415" ], "published": "2026-08-10T12:08:00Z", "summary": "xfrm: clear mode callbacks after failed mode setup", "details": "In the Linux kernel, the following vulnerability has been resolved:\n\nxfrm: clear mode callbacks after failed mode setup\n\nxfrm_state_gc_task can run long after a failed IPTFS state setup. In the\nreproduced case, __xfrm_init_state() cached x->mode_cbs, IPTFS setup\nreturned -ENOMEM before publishing mode_data, and the temporary module\nreference from xfrm_get_mode_cbs() was dropped immediately. The dead state\nthen kept x->mode_cbs until deferred GC ran after xfrm_iptfs had been\nunloaded.\n\nClear x->mode_cbs when mode init or clone fails before publishing\nmode_data. Those states never installed mode-specific state or the\nlong-term IPTFS module pin, so deferred GC has nothing mode-specific to\ndestroy and must not retain a callback table pointer past the temporary\nlookup reference.\n\nThe buggy scenario involves two paths, with each column showing the order\nwithin that path:\n\nfailed setup path:\n1. cache x->mode_cbs\n2. mode setup fails before mode_data\n3. drop the temporary module ref\n4. dead state keeps x->mode_cbs cached\n\nGC/unload path:\n1. xfrm_state_put() queues GC work\n2. xfrm_iptfs unloads later\n3. xfrm_state_gc_task runs\n4. GC dereferences stale x->mode_cbs\n\nThis also covers the failed clone path where clone_state() returns before\npublishing mode_data.\n\nValidation reproduced this kernel report:\nKernel panic - not syncing: Fatal exception\nCONFIG_FAULT_INJECTION_STACKTRACE_FILTER=y\nfailslab_stacktrace_filter matched xfrm_iptfs frames\nack_error=-12\nFAULT_INJECTION: forcing a failure\nBUG: unable to handle page fault\nWorkqueue: events xfrm_state_gc_task\nRIP: xfrm_state_gc_task+0x142/0x650\nModules linked in: esp4_offload xfrm_user [last unloaded: xfrm_iptfs]\nKernel panic - not syncing: Fatal exception", "severity": [ { "type": "CVSS_V3", "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H" } ], "affected": [ { "package": { "ecosystem": "Raspberry Pi OS:13", "name": "linux" }, "versions": [ "1:6.18.29-1+rpt1", "1:6.18.33-1+rpt1", "1:6.18.34-1+rpt1", "1:6.18.39-1+rpt1" ], "ecosystem_specific": { "binaries": [ { "binary_name": "linux-image-6.18.29+rpt-rpi-2712", "binary_version": "1:6.18.29-1+rpt1" }, { "binary_name": "linux-image-6.18.29+rpt-rpi-v6", "binary_version": "1:6.18.29-1+rpt1" }, { "binary_name": "linux-image-6.18.29+rpt-rpi-v7", "binary_version": "1:6.18.29-1+rpt1" }, { "binary_name": "linux-image-6.18.29+rpt-rpi-v8", "binary_version": "1:6.18.29-1+rpt1" }, { "binary_name": "linux-image-6.18.29+rpt-rpi-v8-rt", "binary_version": "1:6.18.29-1+rpt1" }, { "binary_name": "linux-image-6.18.33+rpt-rpi-2712", "binary_version": "1:6.18.33-1+rpt1" }, { "binary_name": "linux-image-6.18.33+rpt-rpi-v6", "binary_version": "1:6.18.33-1+rpt1" }, { "binary_name": "linux-image-6.18.33+rpt-rpi-v7", "binary_version": "1:6.18.33-1+rpt1" }, { "binary_name": "linux-image-6.18.33+rpt-rpi-v8", "binary_version": "1:6.18.33-1+rpt1" }, { "binary_name": "linux-image-6.18.33+rpt-rpi-v8-rt", "binary_version": "1:6.18.33-1+rpt1" }, { "binary_name": "linux-image-6.18.34+rpt-rpi-2712", "binary_version": "1:6.18.34-1+rpt1" }, { "binary_name": "linux-image-6.18.34+rpt-rpi-v6", "binary_version": "1:6.18.34-1+rpt1" }, { "binary_name": "linux-image-6.18.34+rpt-rpi-v7", "binary_version": "1:6.18.34-1+rpt1" }, { "binary_name": "linux-image-6.18.34+rpt-rpi-v8", "binary_version": "1:6.18.34-1+rpt1" }, { "binary_name": "linux-image-6.18.34+rpt-rpi-v8-rt", "binary_version": "1:6.18.34-1+rpt1" }, { "binary_name": "linux-image-6.18.39+rpt-rpi-2712", "binary_version": "1:6.18.39-1+rpt1" }, { "binary_name": "linux-image-6.18.39+rpt-rpi-v6", "binary_version": "1:6.18.39-1+rpt1" }, { "binary_name": "linux-image-6.18.39+rpt-rpi-v7", "binary_version": "1:6.18.39-1+rpt1" }, { "binary_name": "linux-image-6.18.39+rpt-rpi-v8", "binary_version": "1:6.18.39-1+rpt1" }, { "binary_name": "linux-image-6.18.39+rpt-rpi-v8-rt", "binary_version": "1:6.18.39-1+rpt1" } ] }, "database_specific": { "per_version": { "1:6.18.29-1+rpt1": { "linux_commit": "66318c5b5cd8145a662303af12ed39ac07eb27cc", "upstream_version": "6.18.29", "assessment": "commit ancestry (strak rule)", "upstream_fixed": "6.18.42" }, "1:6.18.33-1+rpt1": { "linux_commit": "95b85bebbedcaedfa7ca79116ed38b7376fba412", "upstream_version": "6.18.33", "assessment": "commit ancestry (strak rule)", "upstream_fixed": "6.18.42" }, "1:6.18.34-1+rpt1": { "linux_commit": "c8c7494100e99ee05b11aaa4f0588a223a63d1af", "upstream_version": "6.18.34", "assessment": "commit ancestry (strak rule)", "upstream_fixed": "6.18.42" }, "1:6.18.39-1+rpt1": { "linux_commit": "60ea684a8ace97bb0db1a16e20753bdd6ab371ff", "upstream_version": "6.18.39", "assessment": "commit ancestry (strak rule)", "upstream_fixed": "6.18.42" } } } } ], "references": [ { "type": "ADVISORY", "url": "https://www.cve.org/CVERecord?id=CVE-2026-68415" }, { "type": "FIX", "url": "https://git.kernel.org/stable/c/9845a35986a658816f7752f7ebd7c455a4c7dfdf" }, { "type": "FIX", "url": "https://git.kernel.org/stable/c/c37a079230128a5237f45fb4e181bc069a5c2955" }, { "type": "FIX", "url": "https://git.kernel.org/stable/c/2538bd3cd1ff5af655908469544ac7b7ae259386" } ], "database_specific": { "source": "https://git.kernel.org/pub/scm/linux/security/vulns.git", "upstream_modified": "2026-08-17T05:24:08Z", "severity": { "source": "kernel CNA", "cvss_v3_score": 7.8, "cvss_v3_severity": "High", "scenario": "AV:L - The vulnerable paths are reached only through the XFRM configuration interface \u2014 netlink XFRM_MSG_NEWSA/XFRM_MSG_MIGRATE handled by xfrm_user.c \u2014 which requires a local socket on the machine; no remote peer data reaches this code.\nAC:L - The attacker fully controls SA creation and the failing allocation size (XFRMA_IPTFS_REORDER_WINDOW up to 65535 slots, ~1MB order-8 kmalloc), so the -ENOMEM setup failure is attacker-inducible and repeatable, and unlimited poisoned dead states can be armed so the stale x->mode_cbs dereference in xfrm_state_gc_task is hit deterministically once xfrm_iptfs is unloaded.\nPR:L - xfrm_user_rcv_msg() gates on netlink_net_capable(skb, CAP_NET_ADMIN), which is evaluated against the net namespace's user_ns, so any unprivileged local user obtains it with `unshare -Urn`; the IPTFS module is auto-loaded on demand by xfrm_get_mode_cbs().\nUI:N - The attacker performs the entire sequence with netlink messages from its own process; no victim action, mount, or file open is needed to arm or trigger the dead state's deferred GC.\nS:U - The stale callback dereference occurs in kernel context within the same security authority; no VM, IOMMU, or sandbox boundary is crossed.\nC:H - xfrm_state_gc_task loads x->mode_cbs->destroy_state from module memory that was freed by module unload and can be reoccupied by attacker-influenced allocations, giving a stale-pointer read/indirect-call primitive that can be leveraged to disclose arbitrary kernel memory.\nI:H - Calling a function pointer read out of freed module text/rodata is a control-flow hijack primitive; combined with the missing module reference that lets the memory be recycled, this permits arbitrary kernel-state modification rather than a benign fault.\nA:H - The commit's reproducer shows the direct outcome is \"BUG: unable to handle page fault\" in the events workqueue running xfrm_state_gc_task, escalating to \"Kernel panic - not syncing: Fatal exception\" \u2014 a full system crash." } } }