← Back to CVE intelligence
CVE intelligence

CVE-2026-99183

Vulnerability intelligence

Published Oct 28, 2026Sources checked Nov 20, 2026
N/ACVSS not listedScore unavailable
What this means

Review the available evidence

CVE-2026-99183. It is not in the current CISA KEV record we collected. That does not prove exploitation has not occurred.

What to do next

Remediation and response

Microsoft Security Response Center guidance

Vendor revision: Nov 20, 2026

Release Notes

Vendor-listed releases

  • 0:6.18.62-1.1.azl4

Ubuntu Security Notices guidance

Vendor revision: Nov 19, 2026
Vendor remediation details

This update corrects flaws in the following subsystems: - S390 architecture; - x86 architecture; - Block layer subsystem; - Cryptographic API; - Compute Acceleration Framework; - FireWire subsystem; - GPIO subsystem; - GPU drivers; - Microsoft Hyper-V drivers; - I2C subsystem; - Multiple devices driver; - Network drivers; - STMicroelectronics network drivers; - Power supply drivers; - S/390 drivers; - SCSI subsystem; - SPI subsystem; - TCM subsystem; - UFS subsystem; - File systems infrastructure; - Ext4 file system; - File system notification infrastructure; - NTFS3 file system; - OCFS2 file system; - SMB network file system; - Tracing file system; - XFS file system; - Kernel futex primitives; - Key management; - Key Management; - VMware vSockets driver; - io_uring subsystem; - IPC subsys

CISA KEVNot listedBased on the latest collected catalog
EPSS0.2%Estimated 30-day exploitation probability
Ransomware useNot markedCISA KEV ransomware field
Threat actors0Source-linked actor relationships
Overview

What is CVE-2026-99183?

In the Linux kernel, the following vulnerability has been resolved:

accel/rocket: search every core slot when looking up a scheduler

sched_to_core() walks rdev->cores[] up to rdev->num_cores, and rocket_remove() decrements num_cores for every core it removes. Unbind a core that is not the last one and the cores behind it fall outside the search, so sched_to_core() returns NULL for a core that is still bound and still running jobs.

Neither caller checks the result:

rocket_job_run(): rocket_fence_create(core), core->dev rocket_job_timedout(): dev_err(core->dev, "NPU job timed out")

Unbinding the middle core of the three on an RK3588 while three clients are submitting to all of them faults twice, once from the surviving core's job queue and once from its reset work:

KASAN: null-ptr-deref in range [0x0000000000000220-0x0000000000000227] Workqueue: fdad0000.npu drm_sched_run_job_work [gpu_sched] pc : rocket_job_run+0x234/0x838 [rocket] Call trace: rocket_job_run+0x234/0x838 [rocket] drm_sched_run_job_work+0x2cc/0xad8 [gpu_sched] process_one_work+0x640/0x14f0

KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] Workqueue: rocket-reset-2 drm_sched_job_timedout [gpu_sched] pc : rocket_job_timedout+0xf0/0x1e0 [rocket] Call trace: rocket_job_timedout+0xf0/0x1e0 [rocket] drm_sched_job_timedout+0x188/0x6a0 [gpu_sched]

Both are the third core: the workqueue names are its device and its core->index, and it was left at slot 2 while num_cores had dropped to 2.

Search all the slots that were allocated, the way find_core_for_dev() now does. A core that is still bound is then found, and the two callers get the pointer they already assume they have.

This does not make unbinding one core out of several safe. An open client keeps an entity pointing at the scheduler of the core that went away: drm_sched reports it as not ready for every job that lands on it, and the client waits in dma_fence_default_wait for a fence that will never signal.

Stopping the NULL dereference is what belongs in a fix; the rest wants more thought.

Source: NVD