From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-97494][MODERATE 7.0] drm/amdgpu: validate and share PSP fw_pri_buf copies via psp_copy_fw [ Upstream Date: Thu, 24 Sep 2026 14:25:04 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-97494 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: 25bec2425c3d50ba9a92ad87bbf8628be29e5c32 List-Id: CVE: CVE-2026-97494 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: drm/amdgpu: validate and share PSP fw_pri_buf copies via psp_copy_fw [ Upstream Commit: 25bec2425c3d50ba9a92ad87bbf8628be29e5c32 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=25bec2425c3d50ba9a92ad87bbf8628be29e5c32 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97494 Analysis date: Thu, 24 Sep 2026 14:25:04 -0400 ActionableScore: 5 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-2026-97494 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97494 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-2026-97494 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-97494 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-787;*CWE-20;CWE-1284;Other CVSS 'AV:L/AC:H/PR:H/UI:N/S:U/C:H/I:H/A:H';BEST CVSS score: '7';DESCR 'An out of bounds write can occur in the AMDGPU PSP firmware loading path because firmware sizes were copied into the fixed 1 MiB fw_pri_buf without a common upper bound check. An oversized PSP system driver, Secure OS, KDB, TOC, RL, or TA image can make memcpy write beyond the destination buffer and corrupt adjacent kernel memory. The immediate result may be a kernel crash or GPU initialization failure. Because this is a concrete kernel memory corruption primitive, confidentiality and integrity impact and local privilege escalation cannot be excluded without reviewing the allocation layout and attacker control over the firmware contents. For the CVSS the PR:L value in the paranoid score represents environments where a reduced capability local service, container root, or GPU management process can trigger PSP loading or recovery after a malformed image has become available to the system. AC:H applies because reliable exploitation depends on firmware delivery, affected GPU hardware, memory layout, and the ability to convert the overflow into controlled corruption. The issue is not directly network reachable. Standard installations normally obtain AMDGPU firmware from an administrator controlled firmware directory, which significantly limits practical exposure.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 5) SKIP CVE-2026-97494 UNKNOWN SKIP No affected files built, so skip this CVE NO - - unknown MAYBE INIT ERRORPATH HARDWARE KPANIC INCREASED_TO_MODERATE7_FROM_LOW_BASED_ON_GUESSCVSS_AND_ACTIONABLESCOREHIGHER4 - - checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=5 ActionableScoreLower=4 PrimitiveClass=CLASS_A RealUAF=NO HostAdminRequired=UNKNOWN NamespaceScopedPrivilege=NO ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 4 * Conservative score: 4 * Raw paranoid score: 5 * Paranoid score: 5 * No floor or cap was applied. * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: 2. Signal breakdown Conservative signals: * +2 Strong corruption primitive. An image-controlled length is passed to `memcpy()` with a fixed 1 MiB destination. An oversized image therefore produces a concrete kernel OOB write, and the copied image bytes determine the overflow payload. * +1 Reliable kernel crash or strong DoS. Corruption beyond `fw_pri_buf` can realistically crash the kernel or make GPU and PSP initialization fail. * +1 Broadly exercised vulnerable path. PSP firmware loading is a normal AMDGPU initialization and recovery operation on supported AMD hardware, rather than an experimental feature. * Total: 4. Additional paranoid signal: * +1 Weak exploitation leverage. The overflow carries bytes from the supplied image rather than a fixed zero pattern, making useful corruption of an adjacent allocation technically plausible. Allocation placement, a useful target, and attacker access to image provisioning remain unresolved. * Total: 5. Not awarded: * No UAF, stale object, repeated release, or refcount mutation is established. * Controlled target selection, arbitrary write, callback influence, credential corruption, and controlled reclaim are not demonstrated. * No confidentiality or separate integrity point is awarded merely because adjacent kernel memory is overwritten. * Local unprivileged reachability is not established. * Remote reachability is not established. * No host-admin penalty is applied because the complete firmware and TA provisioning boundary is not shown. * No hardware rarity penalty is applied because AMDGPU hardware is not inherently rare. * No additional availability point is awarded for the same possible crash. * No separate memory-layout-invariant point is awarded because it would duplicate the OOB-write primitive. 3. Reachability analysis The vulnerable operation is reached when AMDGPU copies PSP system driver, Secure OS, KDB, TOC, RL, or TA images into `fw_pri_buf`. The supplied patch establishes that image size and contents control the copy, but it does not establish who can provision each image or whether signature and firmware-loading policy prevent an ordinary user from supplying a malformed image. Standard system firmware files are commonly administrator-managed, but that deployment expectation is insufficient to mark `HostAdminRequired=YES`. Conversely, the patch does not prove that an ordinary logged-in user, DRM master, compositor, delegated graphics service, or container can supply an oversized image, so `HostAdminRequired=NO` and a local-unprivileged bonus are not justified. No relevant user or network namespace capability path is present. No network trigger is visible. Supported AMD GPU and PSP firmware loading are prerequisites. This is not a race, and no timing penalty applies. 4. Primitive and severity interpretation The exact primitive is a kernel OOB write caused by copying `bin_size` bytes into a fixed `PSP_1_MEG` buffer without checking that `bin_size` is between 1 and 1 MiB. The overflow contents originate from the firmware image, providing a stronger primitive than a fixed-value overwrite, although the attacker-accessible source of that image remains unresolved. This is not a UAF and does not involve stale storage, cleanup-only dereferencing, or repeated lifetime mutation. It is classified as CLASS_A because a direct payload-bearing kernel overwrite is established. The demonstrated practical consequences are memory corruption, possible kernel crash, and GPU or PSP initialization failure. Privilege escalation is plausible only if an attacker can supply the image and arrange a useful adjacent target. It is not demonstrated. `ImportantCandidate=YES` and `AutoDowngradeAllowed=NO` follow from the concrete OOB write and unresolved image-control boundary. The lack of a proven below-host-admin trigger keeps the conservative score below the Important range, while the payload-bearing overwrite warrants actionable manual review. 5. One-sentence report phrase An oversized AMDGPU PSP firmware image can overflow the fixed 1 MiB private firmware buffer with image-derived bytes, creating a concrete kernel OOB write whose practical exploitability depends on who can provision or trigger loading of the malformed image. 6. Manual review recommendation MANUAL CHECK REQUIRED Review all firmware and TA image origins, signature and header validation, `fw_pri_buf` allocation context, adjacent allocation layout, and whether a DRM master or delegated GPU service can initiate loading of attacker-influenced images. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: drm/amdgpu: validate and share PSP fw_pri_buf copies via psp_copy_fw [ Upstream Commit: 25bec2425c3d50ba9a92ad87bbf8628be29e5c32 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=25bec2425c3d50ba9a92ad87bbf8628be29e5c32 Changed files: drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h drivers/gpu/drm/amd/amdgpu/psp_v11_0.c drivers/gpu/drm/amd/amdgpu/psp_v12_0.c drivers/gpu/drm/amd/amdgpu/psp_v13_0.c drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c drivers/gpu/drm/amd/amdgpu/psp_v14_0.c drivers/gpu/drm/amd/amdgpu/psp_v3_1.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=25bec2425c3d50ba9a92ad87bbf8628be29e5c32 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-97494 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-97494 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:H/I:H/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: 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).