From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-64438][MODERATE 7.0] crypto: qat - fix VF2PF work teardown race in adf_disable_sriov() [ Upstream Date: Sat, 25 Jul 2026 07:30:48 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-64438 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: 5 X-AL-KERNEL-ActionableScore-Lower: 4 X-AL-KERNEL-Commit: 218c2836b3987f3fa1d9eac505462cded0821e4c List-Id: CVE: CVE-2026-64438 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: crypto: qat - fix VF2PF work teardown race in adf_disable_sriov() [ Upstream Commit: 218c2836b3987f3fa1d9eac505462cded0821e4c Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=218c2836b3987f3fa1d9eac505462cded0821e4c Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64438 Analysis date: Sat, 25 Jul 2026 07:30:48 -0400 ActionableScore: 5 ActionableScore lower bound: 4 Actionable bucket: Strong Important candidate / Actionable Moderate at minimum Manual review required: YES Summary: A race in Intel QAT SR-IOV teardown can leave queued VF2PF response work holding a freed per-VF state pointer, allowing a VF or guest-mediated device event during PF reconfiguration to trigger a host kernel use-after-free and crash. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-64438 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64438 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-2026-64438 MODERATE 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:L/UI:N/S:C/C:H/I:H/A:H';*CWE-416;CWE-362;CWE-667;Other CVSS 'AV:A/AC:H/PR:L/UI:N/S:C/C:N/I:N/A:H';BEST CVSS score: '8';DESCR 'A race in the Intel QAT SR-IOV teardown path can leave queued VF2PF response work holding a raw pointer to per-VF state after that state has been freed. A VF or guest that can generate VF2PF messages may race with PF SR-IOV disable or reconfiguration and cause the PF workqueue to dereference freed memory in adf_iov_send_resp. For the CVSS the PR:L is used because the higher risk scenario is control of a VF or guest facing QAT interface rather than full host administrator access. The issue is not Internet network reachable. It is reachable through the SR-IOV VF to PF device channel, so AV:A is used. Impact is at least denial of service via host kernel crash and in the paranoid case may include confidentiality or integrity impact because the bug class is use-after-free in kernel driver state.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Strong Important candidate / Actionable Moderate at minimum (with actual score 5) YES LOCK DANGER NULLPTR INIT RACE UAF DMAorINTERRUPT HARDWARE KPANIC KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS KPANIC DECREASED_TO_MODERATE70_BASED_ON_ACTIONABLESCORELOWERTHAN7 NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=5 ActionableScoreLower=4 ## 1. ActionableScore * Conservative score: 4 * Paranoid score: 5 * Final recommended bucket: **Strong Important candidate / Actionable Moderate at minimum**:: ## 2. Signal breakdown Triggered signals: * Firmware-mediated or device-mediated external influence: +1 The triggering input comes through the SR-IOV VF2PF device channel rather than a normal local syscall or Internet network path. * Memory corruption, weak primitive: +1 The patch and commit explicitly describe a use-after-free of `struct adf_accel_vf_info`, but there is no demonstrated attacker-controlled reclaim, type confusion, arbitrary write, or callback hijack. * Real lifetime corruption: +1 Queued PF-side work can retain a raw pointer to per-VF state after `adf_disable_sriov()` frees that state. * Reliable kernel crash / strong DoS: +1 KASAN shows a concrete crash path in `mutex_lock()` from `adf_iov_send_resp()` via the QAT PF response workqueue. * Privileged kernel/device-management memory corruption path: +1 The bug is in SR-IOV teardown and interrupt/workqueue synchronization for a hardware crypto accelerator PF driver. Negative / limiting signals: * Hard or unreliable race / special timing required: -1 Exploitation requires overlap between VF2PF work scheduling or execution and PF SR-IOV teardown. Paranoid-only uplift: * Reduced-privilege guest/VF control concern: +1 In virtualized deployments, a tenant or guest controlling a QAT VF may generate VF2PF activity while the PF lifecycle is controlled externally. This does not prove LPE, but it makes auto-closing risky. Race-UAF cap applied: * This is a race-dependent UAF without demonstrated reclaim, object replacement, type confusion, callback dispatch, or arbitrary write. Conservative score is capped at 4 and paranoid score at 5. ## 3. Reachability analysis The bug is not Internet network reachable. It is reachable through the QAT SR-IOV VF to PF communication path when QAT SR-IOV is enabled and VF2PF interrupts can queue PF-side response work. A realistic attacker is most plausibly a guest or tenant with access to a QAT VF, while the host PF is undergoing SR-IOV disable, reset, or reconfiguration. If only a fully trusted host administrator can both generate the VF2PF activity and trigger teardown, the practical risk is lower. If QAT VFs are assigned to less trusted guests, namespaces or containers do not directly matter as much as the VF assignment and device access boundary. The path is hardware and configuration dependent, but SR-IOV VF assignment is a real deployment model for QAT. The race condition makes exploitation timing-dependent, but the commit shows a concrete UAF crash and the fix adds gating, IRQ synchronization, and workqueue flushing, which indicates a real lifetime hazard. Call-site confidence: high. The patch includes the queuing path, worker path, interrupt gating, teardown path, and the synchronization fix. ## 4. Severity interpretation This is more than an ordinary Moderate correctness bug because it is a real kernel lifetime bug with an explicit UAF in an asynchronous workqueue path. The realistic demonstrated impact is host kernel crash / DoS. The patch does not show a controlled reclaim, arbitrary write, function pointer control, or clear privilege-escalation primitive, so conservative scoring should not treat it as a proven Important LPE. The paranoid interpretation should still keep it out of auto-close because it crosses a guest/VF to host/PF device boundary and involves freed kernel driver state. It is best handled as Actionable Moderate at minimum, with Strong Important candidate status depending on whether untrusted guests can access QAT VFs in the evaluated environment. ## 5. One-sentence report phrase A race in Intel QAT SR-IOV teardown can leave queued VF2PF response work holding a freed per-VF state pointer, allowing a VF or guest-mediated device event during PF reconfiguration to trigger a host kernel use-after-free and crash. ## 6. Manual review recommendation MANUAL CHECK REQUIRED Manual review is required because this is an explicit race-dependent UAF in a host PF driver path reachable through SR-IOV VF activity. The available evidence supports DoS strongly and LPE only weakly, but guest-to-host device-boundary exposure makes auto-closing unsafe. ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: crypto: qat - fix VF2PF work teardown race in adf_disable_sriov() [ Upstream Commit: 218c2836b3987f3fa1d9eac505462cded0821e4c Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=218c2836b3987f3fa1d9eac505462cded0821e4c Changed files: drivers/crypto/qat/qat_common/adf_accel_devices.h drivers/crypto/qat/qat_common/adf_common_drv.h drivers/crypto/qat/qat_common/adf_isr.c drivers/crypto/qat/qat_common/adf_pf2vf_msg.c drivers/crypto/qat/qat_common/adf_sriov.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=218c2836b3987f3fa1d9eac505462cded0821e4c ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-64438 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-64438 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:L/UI:N/S:C/C:N/I:N/A:H The Best / paranoid CVSS vector: AV:A/AC:H/PR:L/UI:N/S:C/C:H/I:H/A:H The Best / paranoid CVSS score: 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: 4 Paranoid ActionableScore: 5 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).