From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-90015][MODERATE 7.0] xhci: fix lost bounce buffers on TDs spanning several ring segments Date: Wed, 16 Sep 2026 07:05:10 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-90015 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: c8124b28f12dbdd126118e63d0ebf8093a01fb81 List-Id: CVE: CVE-2026-90015 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: xhci: fix lost bounce buffers on TDs spanning several ring segments Commit: c8124b28f12dbdd126118e63d0ebf8093a01fb81 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c8124b28f12dbdd126118e63d0ebf8093a01fb81 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90015 Analysis date: Wed, 16 Sep 2026 07:05:10 -0400 ActionableScore: 5 ActionableScore lower bound: 4 Actionable bucket: Actionable Moderate Manual review required: NO Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90015 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90015 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: NO A detailed explanation of the methodology and priority rules is included at the end of this message. ====================================================================== AL-KERNEL CLASSIFICATION RESULT ====================================================================== CVE-2026-90015 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-90015 CHECK WITH IMPACT FROM ORIG NN LOW Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H';*CWE-664;*CWE-772;*CWE-200;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:L';BEST CVSS score: '5.8';DESCR 'An xHCI bulk TD that spans three or more ring segments can use bounce buffers at several link TRBs, but the old state tracking kept only the last bounced segment. Earlier bounced segments may then be skipped on completion, so IN transfer data is not copied back into the URB buffer and stale previous contents remain while the URB may still complete as successful. The same skipped segments also leave DMA mappings active, creating a resource leak that can accumulate across repeated large and fragmented bulk transfers. For the CVSS the PR:L is used because reliable triggering normally requires a local user or local process that can drive sufficiently large and fragmented USB bulk IO, or equivalent access to a USB workload on the victim. The issue is not network reachable. Impact is primarily data integrity loss and denial of service, with limited confidentiality concern if stale buffer contents become visible to the caller.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate (with actual score 5) YES USB LEAK DMAorINTERRUPT HARDWARE KPANIC MEMORY NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=5 ActionableScoreLower=4 PrimitiveClass=CLASS_C RealUAF=NO HostAdminRequired=NO NamespaceScopedPrivilege=NO ImportantCandidate=NO ManualReview=RECOMMENDED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 4 * Conservative score: 4. No floor or cap applied. * Raw paranoid score: 5 * Paranoid score: 5. No floor or cap applied. * Final recommended bucket: **Actionable Moderate**:: 2. Signal breakdown Conservative signals: * +1 Concrete persistent kernel resource/reference leak. Earlier bounced ring segments are not DMA unmapped, leaking one DMA mapping per dropped bounce. * +1 Integrity plausible. For IN transfers, earlier bounce data is not copied back into the URB buffer, while the URB can complete successfully with the requested actual_length. This can silently leave stale data in the destination buffer. * +1 Important filesystem/storage path. The demonstrated scenario involves USB mass storage backing dm-verity, where stale hash block data is cached and detected as corrupted metadata. * +1 Broad/default/common subsystem exposure. xHCI and USB bulk transfers are common, and the commit states that any sufficiently large and fragmented bulk transfer can hit the condition. Paranoid additional signal: * +1 Confidentiality plausible. The stale hole in the destination buffer may expose previous buffer contents to the URB consumer, although the evidence supports only limited stale-data exposure, not arbitrary disclosure. Non-awarded signals: * +0 Real UAF not awarded. No freed object is later dereferenced or released. * +0 Controlled reclaim not demonstrated. * +0 Arbitrary write not demonstrated. * +0 Strong LPE or strong exploitation leverage not awarded. * +0 Reliable kernel crash not awarded. The demonstrated issue is silent data corruption plus DMA mapping leak, not panic. * +0 Extra Availability not awarded. Resource exhaustion from accumulated DMA mappings is plausible but not separately demonstrated beyond the concrete leak already counted. * +0 Race/timing penalty not applied. This is not a race and does not require a narrow timing window. * +0 Hardware rarity penalty not applied. xHCI USB host controllers are common and the vulnerable path is not rare hardware only. * +0 Concrete repeated release/refcount not awarded. There is no first release followed by a second lifetime mutation on the same dangling reference. * +0 FLOOR 5 for UAF not applicable because RealUAF=NO. 3. Reachability analysis The trigger is local or device-adjacent through sufficiently large and fragmented USB bulk I/O on an affected xHCI path. A local process that can cause reads or other bulk transfers through an already configured USB storage workload may reach the vulnerable behavior without host-admin-equivalent privilege, so HostAdminRequired=NO for the evaluated practical path. The complete trigger is not namespace-scoped. No network namespace or namespaced capability path is involved. There may be prior setup, such as a USB mass storage device, dm-verity configuration, or a workload that produces large fragmented scatterlist requests. That setup affects exposure but does not make the vulnerable code path host-admin-only once the device and filesystem are in use. No direct network reachability is established. No special race is required. The stated condition is a sufficiently large TD spanning three or more ring segments with unaligned boundaries relative to wMaxPacketSize. 4. Primitive and severity interpretation The exact primitive is lost per-segment bounce-buffer handling for a multi-segment xHCI bulk TD. Earlier bounced segments are skipped during completion because only the last bounced segment is tracked in td->bounce_seg. This is fixed stale storage and resource-leak behavior, not an active structured-object UAF. The stale data remains in the URB destination buffer because the expected copy-back from bounce buffer to URB buffer is skipped. The DMA mapping for skipped bounced segments also remains active. There is no concrete repeated release/refcount operation. There is no first lifetime operation that may free an object followed by a second lifetime mutation through the same dangling reference. PrimitiveClass=CLASS_C because the demonstrated behavior is silent data corruption and resource leakage without a stronger independently useful kernel memory-corruption primitive. The demonstrated consequence is incorrect data returned to upper layers and a persistent DMA mapping leak. The speculative consequence is limited stale-data disclosure or resource exhaustion after repeated triggering. Privilege escalation is not supported by the patch. ImportantCandidate=NO because no Class A primitive, RealUAF, controlled reclaim, arbitrary write, authorization bypass, or demonstrated LPE is established. AutoDowngradeAllowed=NO because the issue has concrete silent storage-path integrity impact plus a persistent DMA mapping leak, so automatic closure as ordinary low-risk cleanup would likely miss a security-relevant data-integrity condition. Likely operational severity is actionable Moderate: not an Important-class memory corruption issue, but not safe to dismiss automatically. 5. One-sentence report phrase An xHCI bulk TD spanning multiple ring segments can lose earlier bounce buffers, causing successful IN transfers to return stale data and leaking DMA mappings, with practical impact including storage data corruption and possible limited stale-data exposure. 6. Manual review recommendation MANUAL CHECK RECOMMENDED Manual review is recommended because the issue affects storage data integrity and may expose stale buffer contents, but the patch does not establish UAF, arbitrary write, controlled reclaim, authorization bypass, or privilege escalation. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: xhci: fix lost bounce buffers on TDs spanning several ring segments Commit: c8124b28f12dbdd126118e63d0ebf8093a01fb81 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c8124b28f12dbdd126118e63d0ebf8093a01fb81 Commit description: When a TD reaches a link TRB with data that is not aligned to the endpoint's wMaxPacketSize, xhci_align_td() stages the unalignable tail through the bounce buffer of the ring segment holding that link TRB. xhci_unmap_td_bounce_buffer() later unmaps it and, for IN transfers, copies the data back into the URB's buffer. The enqueue path records the segment that was bounced in td->bounce_seg, under the assumption that a TD never spans more than two ring segments. That assumption does not hold: a TD large enough to span three or more segments crosses several link TRBs and can be bounced at each of them. Only the last one survives in td->bounce_seg, so every earlier bounce buffer is neither copied back nor DMA unmapped. The URB still completes with actual_length equal to the requested length and no error, so the transfer looks successful while a wMaxPacketSize sized hole in the destination buffer silently keeps its previous contents. It also leaks a DMA mapping per dropped bounce. Any sufficiently large and fragmented bulk transfer can hit this. It was found with a USB mass storage device behind xHCI backing a dm-verity target with 512 byte hash blocks, where the stale data is detected rather than silently consumed. The device enumerates as SuperSpeed, so wMaxPacketSize is 1024, while dm-bufio issues one 512 byte bio per hash block. verity_prefetch_io() makes the block layer merge hundreds of them into a single request of up to 512 scatterlist entries of 512 bytes each. At 256 TRBs per ring segment such a TD spans three segments, and every segment boundary falls on an odd multiple of 512, i.e. unaligned to wMaxPacketSize. dm-bufio then caches a hash block holding stale data and dm-verity declares the metadata block corrupted: device-mapper: verity: 8:2: metadata block 10850 is corrupted A reproducer running this under qemu is available at https://github.com/baloo/xhci-verity The bounce state (bounce_buf, bounce_dma, bounce_len, bounce_offs) already lives on the ring segment, so there is nothing extra to track. Keep recording the last bounced segment in td->bounce_seg and, on completion, walk the segments from td->start_seg up to it, unmapping every segment that still has a pending bounce. Stopping at td->bounce_seg rather than td->end_seg matters: a bounce implies the TD continues past that segment's link TRB, so bounce_seg is always strictly before end_seg, and a later TD may already have started in end_seg and been bounced there. Walking that far would copy a foreign bounce buffer into this URB and unmap it twice. It also keeps the walk correct if a TD ever wraps the whole ring so that end_seg == start_seg. [mn: Add ring->num_segs check to prevent unlikely infinite for loop.] Fixes: f9c589e ("xhci: TD-fragment, align the unsplittable case with a bounce buffer") Cc: stable@vger.kernel.org Suggested-by: Michal Pecio Signed-off-by: Arthur Gautier Signed-off-by: Mathias Nyman Link: https://patch.msgid.link/20260831090448.95644-4-mathias.nyman@linux.intel.com [Michal: solved context conflict due to xhci_td_cleanup() type] Signed-off-by: Michal Pecio Signed-off-by: Greg Kroah-Hartman Changed files: drivers/usb/host/xhci-ring.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=c8124b28f12dbdd126118e63d0ebf8093a01fb81 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90015 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90015 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:L/I:L/A:L The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS score: 5.8 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).