From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-89793][MODERATE 7.0] ublk: clear VM_MAYWRITE on read-only ublk char device mmap Date: Wed, 16 Sep 2026 13:53:04 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-89793 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: be41733c24be58e2a1ef718c80fafdfb98a40d5e List-Id: CVE: CVE-2026-89793 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: ublk: clear VM_MAYWRITE on read-only ublk char device mmap Commit: be41733c24be58e2a1ef718c80fafdfb98a40d5e Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=be41733c24be58e2a1ef718c80fafdfb98a40d5e Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89793 Analysis date: Wed, 16 Sep 2026 13:53:04 -0400 ActionableScore: 6 ActionableScore lower bound: 5 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-2026-89793 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89793 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-2026-89793 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-89793 CHECK WITH IMPACT FROM ORIG NN LOW Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H';CWE-732;CWE-284;CWE-269;Other CVSS 'AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H';BEST CVSS score: '7.8';DESCR 'A protection flaw in ublk_ch_mmap leaves VM_MAYWRITE set on a command buffer that is intended to be mapped read only. A local ublk daemon can map the buffer with PROT_READ and then use mprotect to upgrade the mapping to PROT_WRITE, gaining a constrained write primitive over kernel-written ublksrv_io_desc entries. The attacker can modify fields such as addr, op_flags, nr_sectors, and start_sector while the buffer participates in block I/O processing. This can corrupt request semantics, disrupt the affected ublk device, damage data, or cause a denial of service. A direct arbitrary kernel write or reliable privilege escalation is not established, but the writable kernel-allocated ABI buffer creates a plausible memory corruption boundary that requires review of all later kernel consumers. For the CVSS the PR:L is selected because triggering requires a local process with access to the affected ublk character device, but it does not necessarily require full root privileges. The issue is not directly network reachable. The paranoid score uses CIA:HHH because controlled corruption of kernel-owned I/O descriptors may have broader confidentiality, integrity, and availability consequences if any later kernel path trusts the modified fields.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6) SKIP CVE-2026-89793 UNKNOWN SKIP The Fixes patch not applied yet, so unlikely that actual: 3fee8d7599e1 YES NO NO unknown MAYBE READ KASAN INIT SIMPLEFIX HARDWARE IO_URING KPANIC INCREASED_TO_MODERATE7_FROM_LOW_BASED_ON_GUESSCVSS_AND_ACTIONABLESCOREHIGHER4 KPANIC INCREASED_TO_HIGH_BASED_ON_ACTIONABLESCOREHIGHERTHAN5 DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP - - checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=6 ActionableScoreLower=5 PrimitiveClass=CLASS_A RealUAF=NO HostAdminRequired=NO NamespaceScopedPrivilege=UNKNOWN ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 5 * Conservative score: 5. No floor or cap applied. * Raw paranoid score: 6 * Paranoid score: 6. No floor or cap applied. * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: 2. Signal breakdown Conservative signals: * +2 Strong corruption primitive. An unprivileged daemon obtains a writable mapping of a kernel-owned command buffer and can directly modify active `ublksrv_io_desc` fields, including addresses, flags, lengths, and sectors. * +1 Local unprivileged trigger. The supplied evidence explicitly identifies an unprivileged daemon and provides a working `mmap` plus `mprotect` sequence. * +1 Integrity plausible. The attacker can make unauthorized modifications to kernel-written I/O descriptors that define block request semantics. * +1 Important storage path. The corrupted descriptors participate in the ublk block-I/O path. * Total: 5. Paranoid addition: * +1 Availability realistic. Corruption of active address, operation, size, and sector fields can plausibly invalidate requests or disrupt the affected block device. This is not counted as a demonstrated host-wide kernel crash. * Total: 6. Material non-awards: * No UAF, stale lifetime, or repeated refcount operation is established. * No arbitrary or directed write to an attacker-selected kernel address is established. * No controlled reclaim, callback control, credential modification, or demonstrated LPE is established. * Confidentiality is not awarded because no read-back or protected-data disclosure path is shown. * Remote reachability is not established. * The write-protection bypass is not scored as a second independent +2 authorization signal because it is the mechanism enabling the already-counted descriptor corruption. * No reliable-kernel-crash point is awarded because the evidence demonstrates descriptor corruption, not a confirmed panic. * No high-control API bonus is added separately because attacker influence over the descriptors is already intrinsic to the +2 corruption primitive. 3. Reachability analysis The trigger is local and requires a process with access to the affected ublk character-device mapping. The supplied evidence positively states that an unprivileged daemon can map the buffer read-only and upgrade it with `mprotect`, so host-admin-equivalent privilege is not required for the vulnerable operation itself. Access to the relevant ublk device or file descriptor remains a prerequisite. Device creation or delegation may involve prior privileged setup, but that does not make the daemon performing the corruption host administrator. The supplied context does not establish that an arbitrary account without ublk access can trigger the flaw. No namespace capability check is shown, so complete namespace-scoped reachability is unresolved. No special hardware, network access, or timing race is required. The PoC indicates a deterministic and repeatable `mmap` followed by `mprotect` sequence. 4. Primitive and severity interpretation The exact primitive is an attacker-controlled writable mapping of a kernel-allocated, kernel-written command buffer that was intended to remain read-only. This permits constrained but direct modification of active structured I/O descriptors. This is broader than passive metadata corruption because the affected fields influence addresses, operation flags, transfer sizes, and block-sector selection. It is nevertheless not an arbitrary kernel write because the destination is confined to the mapped command-buffer region. There is no lifetime violation or stale storage. The issue is classified as CLASS_A because attacker-controlled corruption of active kernel descriptors is directly established. The demonstrated consequence is unauthorized descriptor modification. Device disruption and data-integrity damage are plausible consequences. Host-wide memory corruption, information disclosure, and privilege escalation remain unproven. ImportantCandidate is YES because the descriptor consumers must be inspected to determine whether corrupted fields can reach unsafe memory operations or cross a stronger storage boundary. AutoDowngradeAllowed is NO because automatically reducing this to an ordinary configuration defect would discard a concrete controlled kernel-state corruption primitive. 5. One-sentence report phrase An unprivileged ublk daemon can use mprotect to convert an intended read-only command-buffer mapping into a writable mapping and corrupt active kernel-written I/O descriptors, creating a concrete storage-integrity primitive with unresolved memory-safety and availability consequences. 6. Manual review recommendation MANUAL CHECK REQUIRED Trace every kernel consumer of `addr`, `op_flags`, `nr_sectors`, and `start_sector` to determine whether concurrent descriptor modification can produce an out-of-bounds access, attacker-directed memory operation, cross-request data corruption, or host-wide crash. 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: ublk: clear VM_MAYWRITE on read-only ublk char device mmap Commit: be41733c24be58e2a1ef718c80fafdfb98a40d5e Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=be41733c24be58e2a1ef718c80fafdfb98a40d5e Commit description: ublk_ch_mmap() rejects mmap requests with VM_WRITE set, but never clears VM_MAYWRITE on the resulting read-only mapping. This allows a userspace daemon to mmap the per-queue command buffer PROT_READ, then upgrade it to PROT_WRITE via mprotect(), since VM_MAYWRITE was never cleared. The command buffer holds struct ublksrv_io_desc entries that are kernel-written ABI; a writable mapping lets an unprivileged daemon process corrupt fields such as addr, op_flags, nr_sectors, and start_sector. Same bug class as the drm/panthor and drm/vc4 VM_MAYWRITE fixes, and the 2026-08-13 ptp/vmclock fix ( a5edadb ). Verified via mprotect() PoC: before the fix, a PROT_READ mapping can be upgraded to PROT_READ|PROT_WRITE and a write into the command buffer corrupts io_desc fields (confirmed under KASAN). After the fix, mprotect() returns -EACCES. Fixes: 3fee8d7599e1 ("ublk_drv: add io_uring based userspace block driver") Cc: stable@vger.kernel.org Signed-off-by: Kanishka De Silva Reviewed-by: Ming Lei Link: https://patch.msgid.link/20260830070133.559-1-kpskanna1915@gmail.com Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman Changed files: drivers/block/ublk_drv.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=be41733c24be58e2a1ef718c80fafdfb98a40d5e ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-89793 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-89793 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:L/PR:L/UI:N/S:U/C:L/I:L/A:H The Best / paranoid CVSS vector: AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H The Best / paranoid CVSS score: 7.8 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: 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).