From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-90326][MODERATE 7.0] Revert "blk-cgroup: synchronize pd_free_fn() from blkg_free_workfn() and blkcg_deactivate_policy()" This reverts commit 81c1188 which is Date: Fri, 18 Sep 2026 05:08:46 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-90326 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: 4 X-AL-KERNEL-Commit: b5dae1cd0d8368b4338430ff93403df67f0b8bcc List-Id: CVE: CVE-2026-90326 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: Revert "blk-cgroup: synchronize pd_free_fn() from blkg_free_workfn() and blkcg_deactivate_policy()" This reverts commit 81c1188 which is Commit: b5dae1cd0d8368b4338430ff93403df67f0b8bcc Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=b5dae1cd0d8368b4338430ff93403df67f0b8bcc Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90326 Analysis date: Fri, 18 Sep 2026 05:08:46 -0400 ActionableScore: 6 ActionableScore lower bound: 4 Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review Manual review required: YES Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90326 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90326 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-2026-90326 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-90326 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:H/I:H/A:H';CWE-362;*CWE-416;*CWE-401;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:N/I:N/A:H';BEST CVSS score: '7';DESCR 'A race in blk cgroup policy activation can occur when blkcg_activate_policy() rolls back after a partial blkg_policy_data allocation failure while another thread destroys the same blkg. The rollback path may access freed blkg policy data, including the online state, which makes this a use after free candidate rather than a simple resource leak. The same lifetime mismatch can also allocate policy data for a blkg that is already being destroyed, leaving the new allocation unreachable. For the CVSS the PR:L is used in the paranoid interpretation because a delegated local actor or reduced-capability admin may be able to participate in cgroup teardown or block policy changes in some deployments. The issue is not network reachable and requires local control plane activity on block cgroup or IO scheduler state. Impact is at least local denial of service via kernel crash. In the worst case it may allow confidentiality or integrity impact due to use after free and should be reviewed manually.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5) YES LOCK INIT RACE LEAK UAF HARDWARE KPANIC INCREASED_TO_MODERATE7_FROM_LOW_BASED_ON_GUESSCVSS_AND_ACTIONABLESCOREHIGHER4 YES NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=6 ActionableScoreLower=4 PrimitiveClass=CLASS_B RealUAF=YES HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 4 * Conservative score: 4 * Raw paranoid score: 6 * Paranoid score: 6 * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: No floor or cap changed the score. 2. Signal breakdown Conservative signals: * +2 Real UAF / active lifetime violation. The commit explicitly describes blkg1 and blkg1->pd being freed by blkg_free_workfn() while blkcg_activate_policy() rollback later accesses blkg1->pd->online. * +1 Reliable kernel crash / strong DoS. A kernel use-after-free in block-cgroup policy rollback can realistically produce an Oops or crash. * +1 Concrete persistent kernel resource/reference leak. The commit separately states that blkcg_activate_policy() can allocate a new pd for a blkg already being destroyed, leaving that pd unreachable. * +1 Important block/storage path. The bug is in blk-cgroup policy activation for block queues and IO scheduler state. * -1 Privileged local control-plane uncertainty. Triggering appears to involve IO scheduler switching and blkcg deletion, which are privileged control-plane operations, but the complete privilege boundary is not fully resolved from the supplied input. Paranoid additional signal: * +1 Attacker-influenced active-object lifetime manipulation. In a stronger but still defensible interpretation, a local actor with delegated cgroup/block-management capability may be able to repeatedly drive policy activation and blkcg teardown, making the blkg/pd lifetime race more actionable. Material non-awarded signals: * Controlled reclaim is not demonstrated. * Arbitrary or directed kernel write is not demonstrated. * Confidentiality and integrity impact are not awarded directly because the patch establishes UAF and leak, but not a disclosure path, controlled replacement, credential overwrite, or arbitrary R/W primitive. * Remote reachability is not established. * Extra availability point is not awarded because the same crash/DoS is already counted once. * Structured stale-object leverage is not separately awarded because the supplied stale operation is mainly rollback access to freed pd state, without a distinct post-reuse field-consumption primitive. * Weak LPE is not separately awarded merely because this is a UAF. * Concrete repeated release/refcount +1 is not awarded. The first lifetime-ending free is established via blkg_free_workfn(), and the stale reference survives into rollback, but the supplied patch text only directly establishes access to freed pd->online and an unreachable allocation leak, not a concrete second put/free/refcount mutation on the exact same dangling reference. * Paranoid repeated-refcount +1 is withheld for the same reason. * FLOOR 5 from repeated release/refcount is not applied because the exact second lifetime/refcount mutation is not established. 3. Reachability analysis The trigger is local control-plane activity involving blk-cgroup policy activation during IO scheduler switching and concurrent blkcg deletion. The affected object can be associated with a block queue and blkcg/blkg state. The supplied trace-style scenario uses loop0->queue and concurrent cgroup deletion, but does not fully prove whether an ordinary unprivileged user can perform the complete trigger sequence. Host-admin-equivalent privilege is likely in ordinary deployments because switching an IO scheduler on a block device is normally an administrative operation. However, the input does not fully resolve delegated cgroup management, reduced-capability service accounts, containerized administration, or exact capability namespace checks for all required stages. Therefore HostAdminRequired is UNKNOWN rather than YES. NamespaceScopedPrivilege is UNKNOWN. Cgroup and block control-plane operations may be delegated in some environments, but the supplied input does not prove that the complete trigger chain is available with attacker-obtainable namespaced capabilities. No network trigger is shown. No special hardware requirement is established beyond a block device/queue using blk-cgroup policies. The bug is a race, but no hard-race penalty is applied because the relevant lifecycle operations can plausibly be repeated by a capable local actor. The ENOMEM rollback condition makes practical triggering more conditional, but not enough by itself to apply a strong negative penalty under the supplied evidence. 4. Primitive and severity interpretation The exact primitive is a race-induced use-after-free of blk-cgroup policy data. blkcg_activate_policy() can enter rollback after partially allocating blkg_policy_data, while concurrent blkcg deletion frees the corresponding blkg and pd. Rollback then accesses freed pd state. This is an active structured kernel object lifetime violation, not merely a NULL dereference or passive bookkeeping bug. The patch also establishes a separate resource leak: pd may be newly allocated for a blkg that is already being destroyed and then become unreachable. That leak is concrete, but by itself would not justify Important severity. This is PrimitiveClass CLASS_B. It is not CLASS_A because controlled reclaim, arbitrary R/W, callback influence, credential manipulation, demonstrated LPE, or a concrete security-boundary bypass is not established. It is stronger than CLASS_C because an actual UAF/lifetime violation is directly described. Demonstrated consequence is local kernel memory safety violation with realistic DoS and a concrete memory leak. Speculative consequence is possible exploitation beyond crash due to active kernel object lifetime corruption, but no complete exploitation primitive is proven. ImportantCandidate=YES because this is a real active-object UAF in block-cgroup lifetime handling with unresolved exploitation leverage, plus a related lifecycle leak. Manual review is required to determine whether the rollback path can be shaped into stronger corruption or LPE in delegated cgroup/block-management scenarios. AutoDowngradeAllowed=NO because automatically closing this as ordinary Moderate would discard a concrete kernel UAF in active blkcg/blkg policy lifecycle code. The issue is not proven Important, but it is actionable and should not be downgraded solely due to absence of a public exploit or controlled reclaim. Likely operational severity is Actionable Moderate with Strong Important review potential. 5. One-sentence report phrase A race between blkcg policy activation rollback and concurrent blkg destruction can cause a real use-after-free of blkg_policy_data and a separate unreachable pd leak, making this a local block-cgroup lifetime corruption issue that requires manual review despite no proven arbitrary R/W or LPE. 6. Manual review recommendation MANUAL CHECK REQUIRED Reason: the patch directly establishes a real kernel UAF in active blk-cgroup object lifetime handling, with unresolved exploitation leverage and possible delegated local control-plane reachability. 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: Revert "blk-cgroup: synchronize pd_free_fn() from blkg_free_workfn() and blkcg_deactivate_policy()" This reverts commit 81c1188 which is Commit: b5dae1cd0d8368b4338430ff93403df67f0b8bcc Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=b5dae1cd0d8368b4338430ff93403df67f0b8bcc Commit description: It is reported to cause problems, as only 2 of the 3 patch series were applied to the stable branches. Reported-by: Mike Cloaked Reported-by: Eric Biggers Cc: Yu Kuai Cc: Tejun Heo Cc: Christoph Hellwig Cc: Jens Axboe Cc: Sasha Levin Link: https://bugzilla.kernel.org/show_bug.cgi?id=217174 Link: https://lore.kernel.org/r/ZAuPkCn49urWBN5P@sol.localdomain Signed-off-by: Greg Kroah-Hartman Signed-off-by: Greg Kroah-Hartman Changed files: block/blk-cgroup.c include/linux/blkdev.h 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=b5dae1cd0d8368b4338430ff93403df67f0b8bcc ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90326 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90326 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:H/UI:N/S:U/C:N/I:N/A:H The Best / paranoid CVSS vector: AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 7 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: 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).