CVE-2026-99183

Source
https://cve.org/CVERecord?id=CVE-2026-99183
Import Source
https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-99183.json
JSON Data
https://api.osv.dev/v1/vulns/CVE-2026-99183
Downstream
Published
2026-10-28T10:35:36Z
Modified
2026-11-08T02:31:04Z
Summary
accel/rocket: search every core slot when looking up a scheduler
Details

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.

Database specific
{
    "cna_assigner":  "Linux",
    "osv_generated_from":  "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/99xxx/CVE-2026-99183.json"
}
References

Affected packages

Git / git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git

Affected ranges

Type
GIT
Repo
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
Events
Introduced
0810d5ad88a18f1e6d549853a388ad0316f74e36
Fixed
16c95870c95f2a3d78c81b9fbc0645f4b7178c11
Fixed
c137756469d5bcc79e97beb406ae4d14b9585463
Fixed
89c2c777df17eb0201c8e22afabbf3691f2a28fe

Database specific

source
"https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-99183.json"

Linux / Kernel

Package

Name
Kernel

Affected ranges

Type
ECOSYSTEM
Events
Introduced
6.18.0
Fixed
6.18.61
Type
ECOSYSTEM
Events
Introduced
6.19.0
Fixed
7.2.13

Database specific

source
"https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-99183.json"