From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-64535][IMPORTANT] nvmet-tcp: Fix potential UAF when ddgst mismatch Date: Mon, 27 Jul 2026 02:58:20 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-64535 X-AL-KERNEL-Priority: IMPORTANT X-AL-KERNEL-Severity: IMPORTANT X-AL-KERNEL-Base-Severity: IMPORTANT X-AL-KERNEL-KPANIC: YES X-AL-KERNEL-ActionableScore: 8 X-AL-KERNEL-ActionableScore-Lower: 6 X-AL-KERNEL-Commit: 96fe2513df590e74b04253a45089cae75569570e List-Id: CVE: CVE-2026-64535 Priority: IMPORTANT AL-KERNEL base severity: IMPORTANT KPANIC flag: YES Patch: nvmet-tcp: Fix potential UAF when ddgst mismatch Commit: 96fe2513df590e74b04253a45089cae75569570e Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=96fe2513df590e74b04253a45089cae75569570e Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64535 Analysis date: Mon, 27 Jul 2026 02:58:20 -0400 ActionableScore: 8 ActionableScore lower bound: 6 Actionable bucket: Strong Important candidate / Actionable Moderate at minimum Manual review required: YES Summary: A remote NVMe/TCP initiator can trigger a double `percpu_ref_put()` and potential UAF by causing a data-digest mismatch on a non-final `H2C_DATA` PDU during R2T transfer, leading at least to DoS and requiring manual review for possible memory-corruption impact. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-64535 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64535 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: IMPORTANT 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-2026-64535 IMPORTANT CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H';*CWE-416;CWE-415;CWE-703;Other CVSS 'AV:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7.5';DESCR 'A potential use after free in the NVMe TCP target path can occur when data digest is enabled and a digest mismatch happens on a non-final H2C_DATA PDU during an R2T based transfer. The digest error path can call nvmet_req_uninit without marking the command as completed or failed, so later queue teardown may treat the same command as still needing data and call nvmet_req_uninit again. This creates a double percpu_ref_put against a single reference acquisition and can lead to premature lifetime release of NVMe target request or queue state. For the CVSS the PR:N is used because a remote NVMe TCP host can trigger the protocol path if it can reach and connect to the target without any local account on the victim. AV:A is used because NVMe TCP targets are normally deployed on isolated storage or data center networks rather than the public Internet. Impact is at least denial of service via target or kernel crash. In the worst still defensible interpretation, the double reference drop is a UAF class issue and may allow confidentiality or integrity impact, so manual review is required.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 8) YES REMOTE INIT NVME UAF NETWORK LINUS TEST KPANIC KPANIC INCREASED_TO_MODERATE7_BASED_ON_GUESSCVSS KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHEREQTHAN7 NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=8 ActionableScoreLower=6 ## 1. ActionableScore * Conservative score: 6 * Paranoid score: 8 * Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**:: ## 2. Signal breakdown Conservative signals: * Remote reachable / network-triggerable: +2. NVMe/TCP target is reachable over TCP by an initiator that can connect to the exposed target port. * Generic memory corruption: +2. The bug is a double `nvmet_req_uninit()` / double `percpu_ref_put()` against one reference acquisition, described as potential UAF. * Real lifetime corruption: +1. The issue is a request / queue lifetime bug caused by failing to mark the command completed before teardown. * Reliable kernel crash / strong DoS: +1. Premature reference release can realistically crash or destabilize the NVMe target path. * Important filesystem/storage path: +1. NVMe/TCP target is an important storage data path. * Rare AND difficult-to-reach configuration: -1. Trigger requires data digest enabled, R2T-based transfer, non-final `H2C_DATA`, and digest mismatch. Paranoid additions / interpretation: * Weak LPE concern: +1. Refcount lifetime corruption can sometimes become exploitable beyond DoS, but no reclaim or controlled reuse primitive is shown. * Confidentiality / integrity plausible: +1 combined under capped interpretation. The UAF class justifies manual review, but the patch does not demonstrate arbitrary read/write. * Paranoid score is capped at 8 to avoid over-counting the same double-ref lifetime primitive. ## 3. Reachability analysis A remote NVMe/TCP initiator can trigger the vulnerable path if it can connect to the target and negotiate/use data digest. No local account on the victim is required, so PR is effectively none for the network-facing target service. This is not usually Internet-exposed, but it is realistic inside storage or data-center networks. Containers and namespaces are not central to triggering unless they can reach the NVMe/TCP target network path. Call-site confidence: high. The commit message explicitly identifies `nvmet_tcp_try_recv_ddgst()`, `nvmet_req_uninit()`, `nvmet_tcp_uninit_data_in_cmds()`, and the second uninit condition. ## 4. Severity interpretation This is stronger than an ordinary Moderate because it is network-triggerable within the NVMe/TCP storage network and involves a real reference-lifetime bug, not just a NULL dereference or validation cleanup. Realistic impact is at least DoS through target or kernel crash. Theoretical impact may extend to privilege escalation or broader memory corruption because the primitive is potential UAF / double reference release, but no attacker-controlled reclaim, overwrite, or callback control is demonstrated in the patch. ## 5. One-sentence report phrase A remote NVMe/TCP initiator can trigger a double `percpu_ref_put()` and potential UAF by causing a data-digest mismatch on a non-final `H2C_DATA` PDU during R2T transfer, leading at least to DoS and requiring manual review for possible memory-corruption impact. ## 6. Manual review recommendation MANUAL CHECK REQUIRED The issue is network-triggerable in reachable NVMe/TCP deployments and involves a concrete lifetime corruption primitive with potential UAF, even though practical exploitation beyond DoS is not demonstrated. ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: nvmet-tcp: Fix potential UAF when ddgst mismatch Commit: 96fe2513df590e74b04253a45089cae75569570e Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=96fe2513df590e74b04253a45089cae75569570e Commit description: Shivam Kumar found via vulnerability testing: When data digest is enabled on an NVMe/TCP connection and a digest mismatch occurs on a non-final H2C_DATA PDU during an R2T-based data transfer, the digest error handler in nvmet_tcp_try_recv_ddgst() calls nvmet_req_uninit() — which performs percpu_ref_put() on the submission queue — but does NOT mark the command as completed. It does not set cqe->status, does not modify rbytes_done, and does not clear any flag. When the subsequent fatal error triggers queue teardown, nvmet_tcp_uninit_data_in_cmds() iterates all commands, checks nvmet_tcp_need_data_in() for each one, and finds that the already-uninited command still appears to need data (because rbytes_done < transfer_len and cqe->status == 0). It therefore calls nvmet_req_uninit() a second time on the same command — a double percpu_ref_put against a single percpu_ref_get. Reported-by: Shivam Kumar Reviewed-by: Christoph Hellwig Signed-off-by: Sagi Grimberg Signed-off-by: Keith Busch Signed-off-by: Greg Kroah-Hartman Changed files: drivers/nvme/target/tcp.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=96fe2513df590e74b04253a45089cae75569570e ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-64535 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64535 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:A/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS vector: AV:A/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 7.5 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: 6 Paranoid ActionableScore: 8 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: IMPORTANT KPANIC detected for this report: YES Published priority for this report (same as in Subject): IMPORTANT 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).