From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-74378][MODERATE 7.0] RDMA/rxe: Fix TOCTOU heap overflow in get_srq_wqe [ Upstream Date: Sat, 15 Aug 2026 20:25:53 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-74378 X-AL-KERNEL-Priority: MODERATE 7.0 X-AL-KERNEL-Severity: MODERATE 7.0 X-AL-KERNEL-Base-Severity: MODERATE X-AL-KERNEL-KPANIC: YES X-AL-KERNEL-ActionableScore: 6 X-AL-KERNEL-ActionableScore-Lower: 5 X-AL-KERNEL-Commit: 3cfa2a3adc51b7c57729961a03446962ff10e3d2 List-Id: CVE: CVE-2026-74378 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: RDMA/rxe: Fix TOCTOU heap overflow in get_srq_wqe [ Upstream Commit: 3cfa2a3adc51b7c57729961a03446962ff10e3d2 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3cfa2a3adc51b7c57729961a03446962ff10e3d2 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74378 Analysis date: Sat, 15 Aug 2026 20:25:53 -0400 ActionableScore: 6 ActionableScore lower bound: 5 Actionable bucket: Strong Important candidate / Actionable Moderate at minimum Manual review required: YES Summary: A TOCTOU bug in RDMA RXE `get_srq_wqe()` lets local userspace modify `num_sge` between validation and memcpy size calculation, causing a kernel heap overflow in the SRQ WQE copy path with DoS and plausible privilege escalation impact. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-74378 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74378 The original announcement does not normally provide a security severity estimate, CVSS assessment, or enough information to determine whether the reported kernel bug represents a practically relevant security issue. This report was generated by AL-KERNEL, an AI-assisted Linux kernel vulnerability analysis system developed by Alexander Larkin. It combines an autonomous classifier with LLM-assisted technical analysis and a separate ActionableScore mechanism. The purpose of this report is to prioritize Linux kernel CVEs before manual review, identify cases that require prompt investigation, and support automatic closure of issues that are unlikely to have meaningful security impact. Published priority for this report: MODERATE 7.0 Manual review required: YES A detailed explanation of the methodology and priority rules is included at the end of this message. ====================================================================== AL-KERNEL CLASSIFICATION RESULT ====================================================================== CVE-2026-74378 MODERATE CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H';CWE-367;CWE-122;*CWE-787;CWE-362;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H';BEST CVSS score: '7.8';DESCR 'RDMA rxe has a TOCTOU heap overflow in get_srq_wqe because num_sge is read from a userspace mapped SRQ WQE, checked against max_sge, and then read again to compute the memcpy size. A local userspace process can race a second thread to change num_sge between the check and use, causing a copy larger than the destination srq_wqe buffer and corrupting kernel heap memory. For the CVSS the PR:L is used because reliable triggering requires local code able to create or access rxe RDMA objects and the userspace mapped SRQ buffer. The issue is not directly network reachable by a remote peer alone, although RDMA traffic may drive the responder path once the local malicious process has prepared the SRQ. Impact is at least local denial of service via kernel crash and may allow privilege escalation because the bug is a kernel heap overflow with attacker influenced size and source data.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 7) YES OOB DANGER SIMPLEFIX NETWORK DMAorINTERRUPT KPANIC MEMORY KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=6 ActionableScoreLower=5 ## 1. ActionableScore * Conservative score: 5 * Paranoid score: 6 * Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**:: ## 2. Signal breakdown Triggered signals: * Local unprivileged trigger: +1 A local userspace process with access to RXE RDMA objects and the userspace mapped SRQ buffer can modify `wqe->dma.num_sge`. * Strong LPE plausibility: +2 The bug is a kernel heap overflow where userspace can influence the copied WQE data and the copy size. This is stronger than a NULL dereference or pure DoS. * Memory corruption, strong corruption primitive: +2 The commit explicitly describes a heap buffer overflow in `memcpy()` into `qp->resp.srq_wqe`. * Reliable kernel crash / strong DoS: +1 A kernel heap overflow in this path can realistically crash the kernel even if privilege escalation is not proven. * Local high-control device data path: +1 Although RDMA APIs should not automatically receive this bonus, here the patch shows userspace influence over a mapped WQE buffer and the size used by a kernel copy. Subtractors: * Hard or unreliable race / special timing required: -1 in conservative score The issue is a TOCTOU race between validation and use. However, the race is local and attacker-controlled through shared userspace mapped memory, so the paranoid score does not subtract this. * Rare / non-default subsystem exposure: -1 RXE / Soft RoCE and RDMA userspace access are not universally enabled by default and require the relevant device and configuration. Not counted: * Remote reachable: +0 A remote RDMA peer alone does not appear sufficient. Local userspace control of the mapped SRQ buffer is required. * Firmware-mediated external influence: +0 No firmware-fed or device-firmware controlled input is involved. * Real lifetime corruption: +0 This is not a UAF, double-free, refcount, or stale lifetime bug. ## 3. Reachability analysis The practical trigger is local. The attacker needs local code execution with access to RXE RDMA resources and the userspace mapped shared receive queue buffer. In typical systems this means a user allowed to use `/dev/infiniband/uverbs*` or equivalent RDMA userspace interfaces. Namespaces and containers may matter if RDMA device access is delegated into a container. In that case, a containerized or reduced-privilege user with RDMA access could potentially trigger a host kernel bug, so treating this as only fully privileged would be too optimistic. The path is not broadly default-enabled on all Linux systems because RXE / Soft RoCE and RDMA userspace access must be present and usable. It is also not directly network-triggerable by packet traffic alone, although RDMA traffic may help drive the responder path once a local malicious process has prepared the SRQ state. ## 4. Severity interpretation This is not an ordinary Moderate memory leak or validation cleanup. It is an explicit TOCTOU heap overflow from userspace-controlled shared memory into a kernel heap object. Realistic exploitation evidence supports at least local DoS and manual review for LPE. Theoretical privilege escalation is plausible because the primitive is a heap overflow with attacker-influenced source data and size, but the patch does not by itself demonstrate a reliable object replacement, allocator grooming result, or full exploit chain. Overall this should be treated as Actionable Moderate at minimum and a Strong Important candidate for kernel CVE triage. ## 5. One-sentence report phrase A TOCTOU bug in RDMA RXE `get_srq_wqe()` lets local userspace modify `num_sge` between validation and memcpy size calculation, causing a kernel heap overflow in the SRQ WQE copy path with DoS and plausible privilege escalation impact. ## 6. Manual review recommendation MANUAL CHECK REQUIRED Reason: the patch fixes an explicit kernel heap overflow caused by a userspace-controlled TOCTOU race, and the primitive is strong enough to require manual LPE assessment rather than autoclosing as DoS-only. ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: RDMA/rxe: Fix TOCTOU heap overflow in get_srq_wqe [ Upstream Commit: 3cfa2a3adc51b7c57729961a03446962ff10e3d2 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3cfa2a3adc51b7c57729961a03446962ff10e3d2 Changed files: drivers/infiniband/sw/rxe/rxe_resp.c Diff excerpt: Not included in this email. See the upstream URL for the full patch. Full patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=3cfa2a3adc51b7c57729961a03446962ff10e3d2 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-74378 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-74378 The original CVE announcement normally does not include a security-level estimate. In particular, it may not contain a CVSS assessment, an impact level, or enough information to determine whether the reported bug is a practically relevant security issue. One purpose of this parallel CVE list is to provide that missing technical and prioritization information. The original goal of the AL-KERNEL project was to prioritize Linux kernel CVE analysis automatically before manual review. The system can also help identify non-security issues that may be suitable for automatic closure. This report was generated by AL-KERNEL, an AI-assisted Linux kernel vulnerability analysis system developed by Alexander Larkin. The first analysis stage combines an autonomous classifier with additional LLM-based analysis. The autonomous classifier runs locally on a CPU and is based on a backpropagation neural network. Together, these mechanisms produce a technical vulnerability description, identify likely weakness types, estimate CVSS severity, and provide input for ActionableScore. Two CVSS estimates are retained because incomplete kernel vulnerability information often permits more than one defensible interpretation: Conservative CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 7.8 The conservative vector represents a lower-impact interpretation. The Best/paranoid vector intentionally represents a plausible upper-bound interpretation and should not automatically be treated as demonstrated real-world impact. CVSS may also need to be adjusted for a particular Linux deployment, because actual reachability, privileges, enabled kernel configuration, hardware, namespaces, exposed device nodes, and other environmental conditions can differ significantly between systems. A separate ActionableScore mechanism evaluates practical remediation urgency. Its analysis may include reachability, attack prerequisites, subsystem exposure, memory-corruption characteristics, denial-of-service reliability, and possible confidentiality, integrity, or privilege-escalation impact. Conservative ActionableScore: 5 Paranoid ActionableScore: 6 The final base severity is taken directly from the second tab-separated field of the AL-KERNEL classification result. ActionableScore does not replace or independently override that final AL-KERNEL decision, and if ActionableScore adjusted impact level of ALKERNEL, then you would see self-readable flags above like INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7. For an AL-KERNEL result of MODERATE, this report uses the following additional presentation split: ActionableScore below 5 -> MODERATE REGULAR ActionableScore 5 or more -> MODERATE 7.0 The distinction between MODERATE REGULAR and MODERATE 7.0 makes it possible to identify Moderate issues that should receive manual analysis and fixes before lower-priority MODERATE REGULAR issues. In many cases, MODERATE REGULAR fixes may wait for a later rebase or routine update. There is one override in which MODERATE REGULAR becomes MODERATE 7.0 even when the ActionableScore is below 5. When the AL-KERNEL result contains the KPANIC flag, a MODERATE result is always presented as MODERATE 7.0. The KPANIC flag selected with few regexps without usage of AI at all, so it helps to detect cases when Kernel Crash happens and similar (to filter False-Negative results from the LLM usage). KPANIC indicates that a reliable kernel crash, kernel panic, or similarly serious kernel availability impact was identified by the classification workflow. AL-KERNEL base severity for this report: MODERATE KPANIC detected for this report: YES Published priority for this report (same as in Subject): MODERATE 7.0 These results are intended to support engineering triage. They are machine-generated estimates, and cases marked for manual review should be validated by a human security engineer before final disposition. For more info read docs linked from here: https://kernelcve.org/ (and you can submit you own patch there to generate such a report for non-existant CVE-id yet). Note that in many cases this AI tool selects higher severity, than real is (means you can expect Importants instead of Moderate 7.0 or Moderates 7.0 instead of regular Moderates). If you see such cases, please use reply email interface to add additional manual analyses info to this particular CVE. And please, please, let me know when you see Lows instead of Importants or Important instead of Low (because particular for such cases I need to tune this AI tool to make it better for this one and next similar). My contact email for such notifications is alexanjelausa@gmail.com (and both send reply to CVE record itself too and see "reply" button below for howto reply).