cve.bz

CVE-2026-98520 linux linux

7.8 highEPSS 0.2%Fix available

High (7.8) in linux linux. EPSS 0.2% estimated probability of exploitation activity in the next 30 days.

On this page
In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: fix use-after-free and multi-container gap in kfd_dev_mapping kfd_dev_mapping caches the address_space of the first /dev/kfd opener so that the GPU reset path can call unmap_mapping_range() to zap all userspace mappings of doorbell and MMIO ranges. This design has two bugs that both manifest under SRIOV with multiple containers: 1. Use-after-free / rwsem deadlock. The cached pointer refers to an inode owned by the first opener's container. When that container exits and its inode is released, kfd_dev_mapping becomes a dangling pointer. A subsequent GPU reset dereferences it inside unmap_mapping_range(), which takes i_mmap_rwsem on the freed inode, causing a hard hang observable as an uninterruptible rwsem wait. 2. Multi-container gap. Only the first opener's address_space is cached; VMAs created by later openers live in a different address_space and are never reached by unmap_mapping_range(). After a GPU reset those stale mappings keep doorbell and MMIO pages accessible to guest userspace with no GPU behind them, risking PCIe transaction timeouts and NMI panics. Fix both bugs with the same approach used by DRM core (drm_drv.c): create a private pseudo-filesystem at module init time and allocate one anonymous inode from it. In kfd_open() redirect every opener's file->f_mapping to that inode's address_space. The inode is module-owned, lives exactly as long as the amdgpu module, and collects VMAs from all openers in one address_space. A single unmap_mapping_range() call in the reset path then correctly reaches every container's mappings with no dangling pointer risk. The hang manifests as an NMI backtrace on the GPU reset workqueue stuck spinning in rwsem_down_read_slowpath() with a corrupted i_mmap_rwsem: Workqueue: amdgpu-reset-dev xgpu_ai_mailbox_flr_work [amdgpu] Call Trace: <TASK> kvm_wait+0x1f/0x40 __pv_queued_spin_lock_slowpath+0x31d/0x3a0 _raw_spin_lock_irq+0x51/0x80 rwsem_down_read_slowpath+0xb3/0x550 down_read+0x48/0xd0 unmap_mapping_range+0x71/0x140 kfd_dev_unmap_mapping_range+0x5b/0x140 [amdgpu] amdgpu_amdkfd_clear_kfd_mapping+0xd8/0x190 [amdgpu] amdgpu_device_gpu_recover+0x232/0x450 [amdgpu] xgpu_ai_mailbox_flr_work+0xb5/0xc0 [amdgpu] process_one_work+0x18e/0x3e0 worker_thread+0x2e3/0x420 kthread+0x10a/0x230 (cherry picked from commit 1128b4a52de1572e87431de837fd9850cb99542c)

Severity

CVSS 3.1

7.8 high
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
EPSS
0.2% (unchanged from 0.2% a week ago) estimated probability of exploitation activity in the next 30 days; 6th percentile. Scored 2026-11-15. A forecast for this CVE across the internet, not for your own systems.
Published
Modified
NVD analysis
Awaiting analysis
Record from
the NVD, synced

Fix and patched versions

Ask an AI about this CVE

Show the promptExplain CVE-2026-98520 for an engineer deciding whether to act. Use https://cve.bz/api/public/cves/CVE-2026-98520?via=ask-ai-copy and https://cve.bz/cve/CVE-2026-98520?via=ask-ai-copy as the source of truth: what is affected and which versions fix it, whether it is in CISA KEV, public exploits, and its EPSS (an internet-wide 30-day estimate of exploitation activity, not a forecast for my systems). If they give no fixed versions, use the CVE record https://cveawg.mitre.org/api/cve/CVE-2026-98520 or the vendor advisory, not third-party mirrors. Tell me how to check whether I am affected and how to verify a fix. Cite the sources and say where they are uncertain.

Opens a third-party assistant; its answer is its own, not cve.bz's. The prompt holds only this CVE's id and public links.

Badge for a README or a site

CVE-2026-98520

An image that stays current: it is drawn from the data each time it is fetched (a host that proxies images, as GitHub does, may show its copy for a while). Markdown:

[![CVE-2026-98520](https://cve.bz/cve/badge/CVE-2026-98520.svg)](https://cve.bz/cve/CVE-2026-98520)

HTML:

<a href="https://cve.bz/cve/CVE-2026-98520"><img src="https://cve.bz/cve/badge/CVE-2026-98520.svg" alt="CVE-2026-98520"></a>

Affected

Versions

From the CNA's record in the CVE List.

ProductAffected
Linux7.2 and later

Fixed in

ProductRelease lineFixed in
Linux7.27.2.9
Linuxlater releases7.3

Linux distributions

DistributionReleasePackageFixed inAdvisory
Debian14linux7.2.9-1DEBIAN-CVE-2026-98520
Ubuntu16.04:LTSlinux-hwe-edgeno fix listedUBUNTU-CVE-2026-98520
Ubuntu18.04:LTSlinux-gcp, linux-hwe, linux-oem and 12 moreno fix listedUBUNTU-CVE-2026-98520
Ubuntu20.04:LTSlinux-gke, linux-gkeop, linux-riscv and 26 moreno fix listedUBUNTU-CVE-2026-98520
Ubuntu22.04:LTSlinux-riscv, linux-aws-6.2, linux-aws-6.5 and 30 moreno fix listedUBUNTU-CVE-2026-98520
Ubuntu24.04:LTSlinux-riscv, linux-oem-6.8, linux-aws-6.14 and 23 moreno fix listedUBUNTU-CVE-2026-98520
UbuntuPro:Realtime:24.04:LTSlinux-realtime-6.14, linux-realtime-6.17no fix listedUBUNTU-CVE-2026-98520

From the distributions' data via OSV.dev: the Debian Security Tracker; Ubuntu Security Notices © Canonical Ltd., CC BY-SA 4.0. Modified: reformatted and linked to CVE records. Credits. On the distributions' own pages: Debian · Ubuntu.

Red Hat

Red Hat rates this Moderate.

Red Hat's statement

This vulnerability is evaluated as Moderate severity because it requires local system access, relies on specific hardware configurations, and only impacts system availability. In typical Red Hat Enterprise Linux deployments, the issue can only be triggered on systems equipped with AMD GPU devices where multiple container instances share access to the GPU interface. A kernel deadlock occurs only if a device recovery cycle runs following the exit of an active container session. Without access to AMD GPU resources or virtualization features, standard workloads remain unaffected.

ProductPackageStateAdvisory
Red Hat Enterprise Linux 10kernelNot affected
Red Hat Enterprise Linux 6kernelOut of support scope
Red Hat Enterprise Linux 7kernelNot affected
Red Hat Enterprise Linux 7kernel-rtNot affected
Red Hat Enterprise Linux 8kernelNot affected
Red Hat Enterprise Linux 8kernel-rtNot affected
Red Hat Enterprise Linux 9kernelNot affected
Red Hat Enterprise Linux 9kernel-rtNot affected
Red Hat OpenShift Container Platform 4openshift/ose-rhel-coreos-8Not affected
Red Hat OpenShift Container Platform 4openshift/ose-rhel-coreos-9Not affected
Red Hat OpenShift Container Platform 5openshift/ose-rhel-coreos-9Not affected

Contains Red Hat Security Data © Red Hat, Inc., licensed under CC BY 4.0. Source: https://access.redhat.com/security/data. Modified: reformatted. Red Hat's page for CVE-2026-98520.

References

Check your stackJSON