From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-97428][MODERATE REGULAR] drm/amdgpu: harden FRU PIA parsing with bounded helpers [ Upstream Date: Thu, 24 Sep 2026 21:21:43 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-97428 X-AL-KERNEL-Priority: MODERATE REGULAR X-AL-KERNEL-Severity: MODERATE REGULAR X-AL-KERNEL-Base-Severity: MODERATE X-AL-KERNEL-KPANIC: NO X-AL-KERNEL-ActionableScore: 3 X-AL-KERNEL-ActionableScore-Lower: 1 X-AL-KERNEL-Commit: 813311512505e95e4e3c898ff94b11886a15e494 List-Id: CVE: CVE-2026-97428 Priority: MODERATE REGULAR AL-KERNEL base severity: MODERATE KPANIC flag: NO Patch: drm/amdgpu: harden FRU PIA parsing with bounded helpers [ Upstream Commit: 813311512505e95e4e3c898ff94b11886a15e494 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=813311512505e95e4e3c898ff94b11886a15e494 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97428 Analysis date: Thu, 24 Sep 2026 21:21:43 -0400 ActionableScore: 3 ActionableScore lower bound: 1 Actionable bucket: Borderline / manual review recommended Manual review required: YES Summary: No one-sentence report phrase was found. ====================================================================== ABOUT THIS REPORT ====================================================================== The original Linux kernel CVE announcement for CVE-2026-97428 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97428 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 REGULAR 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-2026-97428 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-97428 CHECK WITH IMPACT FROM ORIG NN MODERATE Maybe valid. Check manually. Hints by AL-KERNEL: The best (paranoid) CVSS is 'AV:P/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H';CWE-125;CWE-126;*CWE-20;Other CVSS 'AV:P/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:L';BEST CVSS score: '6.1';DESCR 'The AMDGPU FRU Product Info Area parser can read beyond the allocated EEPROM buffer because the original TLV handling checks that a field header exists but does not limit each copy to the number of bytes remaining in the buffer. Truncated or malformed FRU data can therefore cause a heap out of bounds read while manufacturer, product, serial, or FRU identifier fields are parsed. The issue may expose adjacent kernel heap data through device information fields or cause a kernel fault if the read reaches inaccessible memory. For the CVSS the PR:N value is used because exploitation through a malicious physical EEPROM does not require an authenticated user account or local process on the target. AV:P is used because reliable triggering normally requires physical replacement or modification of the GPU FRU EEPROM. The base impact is limited information disclosure and availability degradation. The paranoid impact allows high confidentiality loss and a kernel crash, but integrity remains unaffected because the available primitive is an out of bounds read rather than a write. Public Internet reachability is not expected, although platforms that expose FRU programming through a BMC or another management path require separate review.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Borderline / manual review recommended (with actual score 2) SKIP CVE-2026-97428 UNKNOWN SKIP No affected files built, so skip this CVE NO - - unknown MAYBE READ OOB HARDWARE DECREASED_TO_MODERATEREG_BASED_ON_ACTIONABLESCORELESSTHAN4 - - checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=3 ActionableScoreLower=1 PrimitiveClass=CLASS_C RealUAF=NO HostAdminRequired=NO NamespaceScopedPrivilege=NO ImportantCandidate=NO ManualReview=RECOMMENDED AutoDowngradeAllowed=YES 1. ActionableScore * Raw conservative score: 1 * Conservative score: 1 * Raw paranoid score: 3 * Paranoid score: 3 * No floor or cap was applied. * Final recommended bucket: **Borderline / manual review recommended**:: 2. Signal breakdown Conservative signals: * +1 Firmware-mediated external influence. Malformed or truncated FRU EEPROM data directly controls the TLV lengths consumed by the parser. * +0 Limited source OOB read. The destination copy remains bounded, and no write beyond the destination, UAF, or lifetime corruption is established. * +0 Confidentiality. Bytes can be copied from beyond the EEPROM buffer into FRU fields, but the supplied context does not show a concrete user-visible read-back path. * +0 Availability. The patch establishes an OOB read but does not document an actual Oops, panic, or host crash. * +0 Physical-only penalty. Reliable exploitation likely involves EEPROM modification or device replacement, but exclusive physical reachability is not positively established because a platform management or firmware programming path may exist. Additional paranoid signals: * +1 Confidentiality plausible. Adjacent heap bytes may be copied into manufacturer, product, serial, or FRU identifier fields. If these fields are exposed through sysfs, debugfs, logs, or management interfaces, this becomes a limited kernel information disclosure. * +1 Reliable kernel fault plausible. A source OOB read crossing into inaccessible memory can fault in kernel context, although no observed crash is supplied. * No Integrity point. The copies are bounded by destination size, and no OOB write or other state-modification primitive is shown. * No remote-reachability point. Network delivery to this parser is not established. * No strong corruption or exploitation-leverage point. Controlled reclaim, arbitrary write, type confusion, callback influence, and privilege escalation are not demonstrated. 3. Reachability analysis The vulnerable input comes from the AMD GPU FRU EEPROM and is parsed when the driver obtains product information. An ordinary local user is not shown to be able to rewrite or substitute this EEPROM. A malicious physical device or modified EEPROM can trigger the path without an authenticated host account, so host-admin privilege is not inherently required for that attacker model. The operation is not namespace-scoped, and user or network namespaces do not provide control over the EEPROM parser. No prior host configuration beyond use of a supported AMD GPU and execution of the FRU-reading path is established. Direct network reachability is absent. A BMC, service processor, firmware update facility, or vendor-specific EEPROM programming interface could alter the practical attack vector, but none is shown in the supplied context. There is no race or timing requirement. 4. Primitive and severity interpretation The exact primitive is a bounded-size heap source OOB read. The old code checked that a TLV header existed but selected the copy length from the untrusted TLV length and destination size without also limiting it to the bytes remaining in `pia`. A malformed final field could therefore make `memcpy` read beyond the allocated EEPROM buffer. This is not a UAF and does not involve stale storage or repeated lifetime mutation. The destination buffers remain bounded, so no kernel overwrite is established. The demonstrated primitive is therefore CLASS_C rather than CLASS_A or CLASS_B. The practical impact is most likely malformed device information or a limited internal OOB read. Information disclosure requires confirmation that the affected FRU fields are exposed to an attacker. A kernel fault is technically plausible but not demonstrated. ImportantCandidate is NO because there is no write, lifetime corruption, authorization bypass, or established substantial disclosure. AutoDowngradeAllowed is YES because the unresolved questions concern the observable effect of a limited read rather than a strong corruption primitive. 5. One-sentence report phrase Malformed AMDGPU FRU TLV data can cause a bounded heap source out-of-bounds read, with possible limited information disclosure or kernel fault if copied FRU fields are externally exposed or the read crosses inaccessible memory. 6. Manual review recommendation MANUAL CHECK RECOMMENDED Confirm whether the populated FRU fields are readable through sysfs, debugfs, logs, ioctl output, or platform management interfaces, and whether any non-physical mechanism can modify the EEPROM contents. 7. Call-site confidence Call-site confidence: medium ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: drm/amdgpu: harden FRU PIA parsing with bounded helpers [ Upstream Commit: 813311512505e95e4e3c898ff94b11886a15e494 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=813311512505e95e4e3c898ff94b11886a15e494 Changed files: drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.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=813311512505e95e4e3c898ff94b11886a15e494 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-97428 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97428 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:P/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:L The Best / paranoid CVSS vector: AV:P/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H The Best / paranoid CVSS score: 6.1 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: 1 Paranoid ActionableScore: 3 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: NO Published priority for this report (same as in Subject): MODERATE REGULAR 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).