{"id":"CVE-2026-98367","published":"2026-10-06T09:18:30.970","lastModified":"2026-10-07T07:17:11.103","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nRDMA/siw: Clear association under lock if siw_qp_modify fails in siw_accept\n\nWe need to clear cep before release state_lock as siw_qp_llp_close and\nsiw_qp_modify->siw_qp_llp_close did.\n\nOtherwise if siw_qp_modify() fails in siw_accept(), the QP's state_lock\nis released before the error path cleanup. A concurrent ibv_modify_qp()\ntransitioning the QP to ERROR can race in this window:\n\n  siw_accept()                       ibv_modify_qp(ERROR)\n  ----------------------             ----------------------\n  siw_qp_modify() fails\n  up_write(&qp->state_lock)\n                                     down_write(&qp->state_lock)\n                                     nextstate_from_idle():\n\t\t\t\t     if (qp->cep)\n                                       siw_cep_put(qp->cep) <- frees cep\n                                       qp->cep = NULL\n  goto error\n    cep->qp = NULL                   <- UAF\n\nClear qp->cep and drop the association reference taken by siw_cep_get(),\nall under the write lock held from the initial down_write(&qp->state_lock).\nThread B therefore sees qp->cep == NULL, skips its own put, and cannot free\nthe cep before siw_accept() is done with it.","cvssScore":7.8,"cvssSeverity":"HIGH","cvssVector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","cwes":[],"vendors":[],"products":[],"references":[{"url":"https://git.kernel.org/stable/c/030306bbb9273af80f14d7af20129661964cd9a7","tags":[]},{"url":"https://git.kernel.org/stable/c/32cd87f54dd1070020e664ccb0312a9f0fea79b4","tags":[]},{"url":"https://git.kernel.org/stable/c/9dcc0f4e488b70cff81e0e5717a498c929cf5de3","tags":[]},{"url":"https://git.kernel.org/stable/c/ad50d19f3d1ce052b3a146143581e930a1efb33e","tags":[]},{"url":"https://git.kernel.org/stable/c/bfdc744bf20ae4c3ef2e470298de5237c5c9a13c","tags":[]},{"url":"https://git.kernel.org/stable/c/df2584750314336edcbcc21fb388e04b260f35b7","tags":[]},{"url":"https://git.kernel.org/stable/c/e3f039082856adab7e195dea1af45d93dd6a3f1c","tags":[]},{"url":"https://git.kernel.org/stable/c/f11e09fe2fc3a11ccdf8f932b68181b0bb1d2078","tags":[]}],"exploitRefs":[],"hasPoc":false,"ai":{"summary":"The Linux kernel's RDMA/siw implementation has a race condition where the connection endpoint (cep) reference is not properly cleared under a lock when QP modification fails during acceptance. This can lead to use-after-free conditions when concurrent operations try to modify the QP state, potentially allowing attackers to corrupt memory or escalate privileges.","exploitability":"Exploitation requires concurrent access to a RDMA QP with specific timing windows, making it moderately difficult but possible with controlled network conditions and concurrent thread manipulation.","blast_radius":"The vulnerability could affect systems running RDMA-enabled applications that handle concurrent QP modifications, potentially leading to kernel memory corruption or privilege escalation in high-traffic RDMA environments.","remediation":"Apply the kernel patch for Linux kernel version 6.10 or later which fixes the siw_qp_modify failure handling in siw_accept().","detection":"No reliable host or network indicator is derivable from the published description.","tags":["rdma","kernel","memory-corruption"],"model":"garnet-coder","analyzedAt":"2026-10-14T09:02:20.685Z"}}