From mboxrd@z Thu Jan 1 00:00:00 1970 From: AL-KERNEL To: kernel-cve@kernelcve.org Subject: [CVE-2026-90047][MODERATE 7.0] drm/xe: Don't hand out the flat CCS storage as usable VRAM Date: Wed, 16 Sep 2026 14:33:38 -0400 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-AL-KERNEL-CVE: CVE-2026-90047 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: c96477e0cabf55bfbdf078078e9de1c8024f8060 List-Id: CVE: CVE-2026-90047 Priority: MODERATE 7.0 AL-KERNEL base severity: MODERATE KPANIC flag: YES Patch: drm/xe: Don't hand out the flat CCS storage as usable VRAM Commit: c96477e0cabf55bfbdf078078e9de1c8024f8060 Upstream patch: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c96477e0cabf55bfbdf078078e9de1c8024f8060 Original CVE announcement: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90047 Analysis date: Wed, 16 Sep 2026 14:33:38 -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-2026-90047 is available here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90047 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-2026-90047 MODERATE https://www.kernelcve.org/list/?q=CVE-2026-90047 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-682;*CWE-787;*CWE-664;Other CVSS 'AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:H';BEST CVSS score: '7';DESCR 'drm/xe can expose part of flat CCS compression metadata storage as usable VRAM because the CCS boundary is rounded in the wrong direction. Compression hardware may then write metadata into memory that the VRAM allocator already handed out for normal objects, including GPU page tables. This can corrupt GPU address translation state and lead to a black screen or repeated compositor failure, as seen when a Mesa VM level 3 page table landed on the affected page. For the CVSS the PR:L is selected because practical triggering and impact are tied to local GPU use through a graphics session or render node access rather than full administrator privileges. The issue is not network reachable and depends on affected Intel Xe hardware layout and allocator placement, so AC:H is appropriate. Impact is at least local denial of service and integrity corruption of GPU memory. For the paranoid score, choose the highest still defensible interpretation supported by the memory corruption class and patch context, with preference for manual review sensitivity over autoclosed false negatives.';YES REQUIRES MANUAL CHECK; ,and ActionableScore result is Actionable Moderate / Strong Important candidate requiring manual review (with actual score 6) YES OOB HARDWARE LINUS KPANIC KPANIC INCREASED_TO_HIGH_BASED_ON_GUESSCVSS DECREASED_TO_MODERATE7_BASED_ON_FALSEPOSCHECKOFKP YES NO checked ====================================================================== ACTIONABLESCORE ANALYSIS ====================================================================== ActionableScore=6 ActionableScoreLower=5 PrimitiveClass=CLASS_B RealUAF=NO HostAdminRequired=NO NamespaceScopedPrivilege=NO ImportantCandidate=YES ManualReview=REQUIRED AutoDowngradeAllowed=NO 1. ActionableScore * Raw conservative score: 5 * Conservative score: 5. No floor or cap changed the score. * Raw paranoid score: 6 * Paranoid score: 6. No floor or cap changed the score. * Final recommended bucket: **Actionable Moderate / Strong Important candidate requiring manual review**:: 2. Signal breakdown Conservative positive signals: * +1 Constrained kernel buffer overwrite. The driver exposes part of flat CCS compression metadata storage as usable VRAM, and compression hardware writes fixed metadata into memory handed out by the VRAM allocator. * +1 High value device memory corruption. The overwritten allocator page can contain active GPU objects, and the supplied evidence shows a Mesa VM level 3 page table was corrupted. * +1 Memory layout invariant restoration. The patch fixes a wrong boundary calculation where a limit meaning usable memory ends here was rounded up instead of rounded down. * +1 Reliable strong DoS. The observed consequence is compositor submission failure, repeated gdm restart, and black screen. * +1 Local below host admin reachability. Practical impact is tied to local GPU use through a graphics session or render node style workload, not host administrator control. Paranoid additional signal: * +1 Unresolved active translation state leverage. The corrupted object can be a GPU page table, so the issue affects active GPU address translation state, although controlled target selection, controlled payload, arbitrary host memory access, or privilege escalation are not demonstrated. Non awarded signals: * Strong LPE leverage +2 not awarded. The metadata payload is hardware generated and no controlled reclaim, arbitrary write, callback control, credential corruption, or demonstrated privilege escalation is established. * Remote reachability not awarded. This is not network triggered. * Real UAF not awarded. The bug is an allocator boundary and reserved memory exposure issue, not a stale post lifetime object use. * Extra availability not awarded. The black screen and compositor restart are the same DoS consequence already counted. * Confidentiality not awarded. No read back or information disclosure path is established. * CLASS_A not selected. The corruption is real and affects active GPU memory, but attacker controlled payload or directed host kernel corruption is not established. Negative signals: * No hardware rarity penalty applied. The issue is hardware specific to affected Intel Xe layouts, but common GPU hardware specificity alone is not enough for a rare or physical only penalty. * No hard race penalty applied. The failure depends on layout and allocation placement, not on a narrow attacker timed race. 3. Reachability analysis The vulnerable condition is created by the driver during VRAM layout setup. The wrong rounded boundary lets the VRAM allocator hand out a page that partly belongs to flat CCS storage. The compression hardware can then write metadata into that page independently of GPU submissions or page table entries. For DRM and display roles, the most relevant attacker or trigger context is a local graphics workload. An ordinary logged in user, compositor, Mesa process, or render node user can cause GPU allocations that may land in the exposed page. A DRM master or compositor may be the immediate process affected in a desktop session. Host administrator privilege is not positively required for the practical failure mode described. Namespace scoped privilege is not relevant. There is prior hardware and driver setup, namely an affected Intel Xe platform and drm/xe using flat CCS. This is not a public network exposure. The condition is hardware and layout dependent, and the exact corrupted object depends on allocator placement. The supplied evidence shows repeatable cold boot placement on one Battlemage G21 system, but not a generic attacker directed target selection primitive. 4. Primitive and severity interpretation The exact primitive is device generated overwrite of memory that the VRAM allocator incorrectly treats as usable, while that memory is actually owned by flat CCS compression metadata storage. This is active device memory corruption rather than a UAF. It is not fixed stale storage in the UAF sense, but it is constrained because the payload is compression metadata and the target depends on VRAM allocation placement. The strongest demonstrated consequence is corruption of GPU address translation state, specifically a Mesa VM level 3 page table, leading to compositor fault and black screen. The speculative consequence is broader GPU memory integrity impact if other active GPU objects or page tables are placed in the exposed page. There is no demonstrated host kernel arbitrary read or write, no controlled reclaim, and no proven privilege escalation. PrimitiveClass is CLASS_B because a real active memory ownership and allocator boundary violation exists, with corruption of active GPU translation state, but without a demonstrated strong attacker controlled corruption primitive. ImportantCandidate is YES because GPU page table corruption is more significant than ordinary crash only behavior and should not be automatically downgraded. AutoDowngradeAllowed is NO because automatic closure would lose an unresolved active memory corruption primitive involving allocator owned VRAM and GPU page tables. Likely operational severity is actionable Moderate at minimum, with Strong Important review justified for affected Intel Xe systems. 5. One-sentence report phrase A drm/xe VRAM boundary calculation bug can expose flat CCS compression metadata storage as usable VRAM, allowing compression hardware to overwrite allocator owned GPU memory such as page tables and causing at least local display DoS with unresolved GPU memory integrity impact. 6. Manual review recommendation MANUAL CHECK REQUIRED Reason: concrete allocator owned VRAM corruption and observed GPU page table corruption are established, but attacker control over target selection and any privilege escalation path remain unresolved. 7. Call-site confidence Call-site confidence: high ====================================================================== UPSTREAM PATCH SUMMARY ====================================================================== Patch: drm/xe: Don't hand out the flat CCS storage as usable VRAM Commit: c96477e0cabf55bfbdf078078e9de1c8024f8060 Upstream URL: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=c96477e0cabf55bfbdf078078e9de1c8024f8060 Commit description: get_flat_ccs_offset() reads the base of the flat CCS storage from the hardware, scales it by the number of enabled L3 nodes, and rounds the result up to 128K. Everything below that offset is then handed to the VRAM allocator as usable memory. Rounding a limit that means "usable memory ends here" upwards publishes whatever lies between the real base and the rounded one as free memory, and that memory belongs to the compression hardware. The scaled value has no reason to be 128K aligned, and on a Battlemage G21 with 16 GiB it is not: flat CCS base: raw 0x3fafff800, rounded 0x3fb000000 so the last 2 KiB of page 0x3fafff000 is CCS storage, in the allocator's pool. Whatever is allocated there gets that tail overwritten by the compression hardware, which needs no page-table entry, no buffer object and no GPU submission to do it, and does it before userspace exists. On this machine a Mesa VM's level-3 page table landed on that page on every cold boot. It lost the entry covering the compositor's batch-buffer heap, so the compositor's first submission faulted fetching its batch and gdm restarted it forever: a black screen on an otherwise working machine. Restarting gdm cleared it because the next VM's page tables were allocated somewhere else. Round down instead, to the page size the allocator works in. On this machine that excludes exactly one page. Reading the reserved page afterwards shows what had been writing it: [369] 0xcccc000000000000 [371] 0xcc77000000000000 [373] 0xcccc000000000000 [375] 0xcc77000000000000 compression metadata, two bytes per sixteen, sitting where the driver used to hand out memory. The assertion that should have caught this compares the offset against GSMBASE - ccs_size for equality. That value is 128K aligned, so it agrees with the rounded-up offset precisely when the base is not aligned - the check cannot fail in the case it exists to catch, and is compiled out unless CONFIG_DRM_XE_DEBUG is set. Replace it with one that can fail: CCS storage must not run into GSM. [ And this was a debug session from hell, enormously helped by an AI doing much of the grunt-work. I'd like to call it my tireless helper, but the AI several times stated flat out that this was impossible and unsolvable and that we should just write a report about it. I suspect those things have been trained by people who may not be quite as stubborn as I am. But while the AI was ready to give up several times, it did keep adding debug code and analyzing it faithfully when I pushed. So credit where credit is due and I let the AI write the commit message above. This is basically a one-liner fixing a bogus "round_up()" to a "round_down()", but there were 24 patches adding more and more debug information to this, and 18 kernel boot to finally narrow it down to this. - Linus ] Fixes: 3717339 ("drm/xe/vram: fix ccs offset calculation") Cc: stable@kernel.org Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman Changed files: drivers/gpu/drm/xe/xe_vram.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=c96477e0cabf55bfbdf078078e9de1c8024f8060 ====================================================================== DETAILED REPORT METHODOLOGY ====================================================================== The original Linux kernel CVE announcement for CVE-2026-90047 can be found here: https://lore.kernel.org/linux-cve-announce/?q=CVE-2026-90047 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:L/UI:N/S:U/C:N/I:L/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: 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).