CVE-2026-43296
Workaround SQM/PSE Stalls in Linux Kernel
Publication date: 2026-05-08
Last updated on: 2026-05-08
Assigner: kernel.org
Description
Description
CVSS Scores
EPSS Scores
| Probability: | |
| Percentile: |
Meta Information
Affected Vendors & Products
Helpful Resources
Exploitability
| CWE ID | Description |
|---|---|
| CWE-UNKNOWN |
Attack-Flow Graph
AI Powered Q&A
Can you explain this vulnerability to me?
This vulnerability exists in the Linux kernel's octeontx2-af component related to the NIX SQ manager (SQM) sticky mode. The sticky mode causes stalls when multiple submission queues (SQs) share a shared memory queue (SMQ) and transmit concurrently. Additionally, the PSE (Packet Scheduling Engine) may deadlock during transitions between sticky and non-sticky transmissions. There is also a credit drop issue when certain clocks are gated. The vulnerability is addressed by disabling the SQM sticky operation, disabling the sticky to non-sticky transition path that can cause deadlocks, and preventing credit drops by keeping the control-flow clock enabled.
How can this vulnerability impact me? :
This vulnerability can cause stalls and deadlocks in the network transmission process within the Linux kernel, specifically affecting the octeontx2-af component. These stalls and deadlocks can lead to degraded network performance or loss of forward progress under load. Additionally, credit drops may occur, potentially causing further transmission issues. The overall impact is reduced reliability and performance of network packet processing.
What immediate steps should I take to mitigate this vulnerability?
To mitigate this vulnerability in the Linux kernel related to octeontx2-af SQM/PSE stalls, the following immediate steps should be taken:
- Disable SQM sticky operation by clearing TM6 (bit 15) and TM11 (bit 14).
- Disable the sticky to non-sticky transition path that can deadlock PSE by clearing TM5 (bit 23).
- Prevent credit drops by keeping the control-flow clock enabled by setting TM9 (bit 21).
These changes are applied via the NIX_AF_SQM_DBG_CTL_STATUS configuration, which ensures SQM/PSE maintain forward progress under load without credit loss, at the cost of disabling sticky optimizations.