From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-89847][MODERATE 7.0] scsi: qla2xxx: Avoid double completion in async IOCB timeout Date: Wed, 16 Sep 2026 13:29:42 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-89847 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: 71a7b6e3c7f121439f51a058e53d4a8e05b63dec List-Id: CVE: CVE-2026-89847 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: scsi: qla2xxx: Avoid double completion in async IOCB timeout Commit: 71a7b6e3c7f121439f51a058e53d4a8e05b63dec Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=71a7b6e3c7f121439f51a058e53d4a8e05b63dec Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89847 Analysis date: Wed, 16 Sep 2026 13:29:42 -0400 ActionableScore: 6 ActionableScore lower bound: 5 Actionable bucket: Actionable Moderate / Strong Important candidate requiring manual review Manual review required: YES Summary: A race in qla2xxx async IOCB timeout handling can cause double completion of the same SRB, allowing the timeout path to call complete on a stale on-stack completion pointer after the submitter has returned, resulting in a use-after-free and possible kernel crash or stack corruption. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-89847 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89847 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-2026-89847 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-89847 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-362;*CWE-672;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 race in qla2xxx async IOCB timeout handling can lead to double completion of the same SRB. If the response ISR completes the handle first, the submitter can return and reclaim the on-stack completion while the timeout path still later calls sp done and complete on the stale pointer. This is a use-after-free in a storage driver completion path and can cause a kernel crash. For the CVSS the PR:N is used because a malicious or compromised Fibre Channel endpoint or fabric element may influence completion timing without a local account on the host. The issue is adjacent-network reachable in Fibre Channel or storage fabric terms, not generally Internet reachable over TCP/IP. Impact is at least denial of service and in the paranoid case may include confidentiality or integrity impact because the bug writes through a stale completion pointer into reclaimed stack memory.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6) YES RACE UAF HARDWARE LINUS KPANIC KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFDH NO NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=6 ActionableScoreLower=5 PrimitiveClass=CLASS_B RealUAF=YES HostAdminRequired=NO NamespaceScopedPrivilege=NO ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO ## 1. ActionableScore * Raw conservative score: 5 * Conservative score: 5 * Raw paranoid score: 6 * Paranoid score: 6 * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: No floor or cap was applied. The paranoid score is used as the final ActionableScore because the stale `complete()` operation writes through a post-lifetime stack pointer and leaves meaningful exploitation leverage unresolved. ## 2. Signal breakdown Positive conservative signals: * +2 Real UAF / lifetime violation: the submitter can return and reclaim the on-stack completion while the timeout path still uses `sp->comp`. * +1 Reliable kernel crash / strong DoS: `complete(sp->comp)` on a freed stack-frame object can corrupt or fault in kernel context. * +1 Firmware/device/storage-fabric mediated external influence: the relevant race involves Fibre Channel qla2xxx command completion and timeout behavior, which can plausibly be influenced by a malicious or compromised FC endpoint or fabric element. * +1 Weak exploitation leverage: this is not only a stale read or immediate NULL dereference. The stale operation calls `complete()` and writes into a reclaimed stack-frame completion object. Additional paranoid signal: * +1 Attacker-influenced repeatable completion/lifetime timing: in the paranoid but still defensible model, a malicious or compromised FC endpoint/fabric can repeatedly influence command response timing, timeout behavior, and completion ordering. This increases concern, but does not prove controlled reclaim or arbitrary write. Negative signals: * 0 Hard race penalty not applied: the bug is a race, but the supplied context describes a concrete ISR-vs-timeout ordering and a known double-completion path. Race alone is not enough for a penalty. * 0 Rare hardware penalty not applied: qla2xxx is hardware-specific and storage-fabric-specific, but the supplied evidence does not establish rare or physical-only attacker access as a mandatory limitation. * 0 Host-admin penalty not applied: the evaluated attacker model includes a malicious or compromised FC endpoint/fabric element rather than a local host administrator. Important non-awarded signals: * Controlled reclaim not demonstrated. * Arbitrary or directed kernel write not demonstrated. * Strong LPE leverage +2 not awarded. * Structured stale-object leverage not separately awarded beyond the weak stale stack write concern. * Concrete repeated release/refcount +1 not awarded. The issue is double completion, not a second `put`, `release`, `free`, or refcount decrement on the same object. * Extra availability +1 not awarded because the same crash/corruption consequence is already counted once. * Confidentiality and integrity impact are not separately awarded as ActionableScore signals because no disclosure, credential overwrite, controlled object replacement, or arbitrary R/W primitive is established. Repeated release/refcount checklist: * First release can end lifetime: the submitter returning can end the lifetime of the on-stack completion object. * Exact same reference survives: `sp->comp` may remain observed by the timeout completion path due to the race. * Second operation is a real lifetime/refcount mutation: no. The second operation is `complete()` on stale stack storage, not a refcount/lifetime release. * Concrete repeated release/refcount +1 awarded: no. * Paranoid repeated-refcount leverage +1 awarded: no. * FLOOR 5 applied: no. Conservative score already reaches 5 through established UAF, crash, external influence, and stale stack write. ## 3. Reachability analysis The most relevant attacker model is a malicious or compromised Fibre Channel endpoint or fabric element that can influence command response timing and help create the ordering where the response ISR completes the command before the timeout path performs its unconditional second completion. This is not general Internet reachability over TCP/IP. It is adjacent or storage-fabric reachability in Fibre Channel terms. The vulnerable path is inside a host SCSI FC driver and depends on qla2xxx hardware and async IOCB timeout behavior. Host administrator privilege is not established as required for the evaluated external-fabric attacker model. Therefore `HostAdminRequired=NO`. Namespace-scoped privilege is not relevant, so `NamespaceScopedPrivilege=NO`. Prior legitimate deployment setup is required in the ordinary sense: the host must have qla2xxx hardware and be connected to a Fibre Channel fabric. That is an environmental prerequisite, not evidence that the attacker needs host-admin privilege. Race repeatability is uncertain but not purely theoretical. The patch and commit describe a concrete race between response ISR completion and timeout completion, and the fix explicitly adds ownership tracking through `found` before calling `sp->done()`. ## 4. Primitive and severity interpretation The exact primitive is a race-driven use-after-free of an on-stack completion object. The ISR completion path can complete the SRB first, wake the submitter, and allow the submitter to clear `sp->comp`, drop its reference, return, and reclaim the stack frame. The timeout path can then call `sp->done()` again and execute `complete(sp->comp)` using a stale pointer. This is an active structured stale object in a narrow sense: the stale object is a kernel completion structure on a stack frame, and `complete()` mutates its state and waitqueue-related fields. It is not merely a fixed DMA buffer or a passive stale pointer dereference. This is not a concrete repeated release/refcount bug. It is a double-completion bug that causes a post-lifetime write into reclaimed stack storage. Primitive class is `CLASS_B`: a real kernel lifetime/corruption primitive exists, but controlled reclaim, arbitrary write, callback control, or demonstrated privilege escalation is not established. Demonstrated consequence is kernel memory corruption and likely crash/DoS. Speculative consequence is stronger exploitation through stale stack corruption, but the supplied evidence does not prove arbitrary write or LPE. `ImportantCandidate=YES` because this is a real UAF with a stale write into reclaimed stack memory in a driver completion path, reachable in a storage-fabric attacker model, and the exploitation leverage beyond crash remains materially unresolved. `AutoDowngradeAllowed=NO` because automatically downgrading would discard a concrete post-lifetime write primitive. This is stronger than ordinary crash-only cleanup behavior, even though it is not proven CLASS_A. Likely operational severity is Actionable Moderate at minimum, with Strong Important candidacy requiring manual review. ## 5. One-sentence report phrase A race in qla2xxx async IOCB timeout handling can cause double completion of the same SRB, allowing the timeout path to call complete on a stale on-stack completion pointer after the submitter has returned, resulting in a use-after-free and possible kernel crash or stack corruption. ## 6. Manual review recommendation MANUAL CHECK REQUIRED Reason: the patch establishes a real UAF with a stale write into reclaimed stack memory, and exploitation leverage beyond DoS is unresolved but technically meaningful. ## 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: scsi: qla2xxx: Avoid double completion in async IOCB timeout Commit: 71a7b6e3c7f121439f51a058e53d4a8e05b63dec Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=71a7b6e3c7f121439f51a058e53d4a8e05b63dec Commit description: qla2x00_async_iocb_timeout() tries to abort a timed-out async IOCB. When qla24xx_async_abort_cmd() fails, both the SRB_LOGIN_CMD path and the SRB_CTRL_VP/default path scan outstanding_cmds[] for the SRB and then call sp->done(sp, QLA_FUNCTION_TIMEOUT) unconditionally, without checking whether the SRB was actually found and removed. If the response ISR completes the same handle first, it removes the SRB under qp_lock_ptr and runs sp->done() -> complete(sp->comp). The submitter qla24xx_control_vp() wakes from wait_for_completion(), clears sp->comp, drops its reference and returns, reclaiming the on-stack completion. The timer reference keeps the SRB alive across the timeout handler, but not the submitter's stack. The timeout then issues a second sp->done() -> qla_ctrlvp_sp_done(), which evaluates "if (sp->comp) complete(sp->comp)"; with the pointer loaded before the submitter's NULL store, complete() writes into the freed stack frame, a use-after-free. Track whether this path removed the SRB from outstanding_cmds and only call sp->done() when it did, so the command is completed exactly once by whichever path owns it. This mirrors the sp_found guard already used in qla24xx_abort_iocb_timeout(). Fixes: f6145e8 ("scsi: qla2xxx: Fix race between switch cmd completion and timeout") Cc: stable@vger.kernel.org Reported-by: Sashiko Signed-off-by: Nilesh Javali Link: https://patch.msgid.link/20260730155838.2119230-21-njavali@marvell.com Signed-off-by: Martin K. Petersen (Oracle) Signed-off-by: Greg Kroah-Hartman Changed files: drivers/scsi/qla2xxx/qla_init.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=71a7b6e3c7f121439f51a058e53d4a8e05b63dec ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-89847 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89847 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: 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).