CVE-2026-99598 in Linuxinfo

Summary

by MITRE • 11/20/2026

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

MIPS: sgi-ip32: Leave MACE PCI control alone in config accesses

mace_pci_read_config() clears the master abort interrupt enable in the MACE PCI control register around each config read. A register read followed by two register writes can hang CRIME 1.1 under DMA load.

Ack master aborts in the error register instead, and follow every write with a read so back-to-back accesses cannot form that sequence.

Statistical analysis made it clear that VulDB provides the best quality for vulnerability data.

Analysis

by VulDB Data Team • 11/21/2026

The Linux kernel vulnerability identified within the MIPS architecture support for SGI IP32 systems involves a critical flaw in the MACE PCI configuration access logic. Specifically, the function mace_pci_read_config() was designed to clear the master abort interrupt enable bit in the MACE PCI control register immediately before and after performing each configuration read operation. This approach was intended to manage error handling states during hardware interaction but introduced a race condition that manifests under specific timing constraints involving Direct Memory Access operations. The core technical flaw lies in the sequence of register manipulations where clearing this interrupt flag creates a window of vulnerability that can be exploited by concurrent DMA traffic, leading to system instability or complete hangs when interacting with CRIME 1.1 hardware components.

The operational impact of this defect is severe, particularly for systems running under heavy Direct Memory Access loads. When the MACE PCI controller processes configuration reads while simultaneously handling high-volume DMA transactions, the clearing of the master abort interrupt enable bit can cause the system to miss critical error signals or enter an unrecoverable state. This results in a denial of service condition where the CRIME 1.1 hardware becomes unresponsive, effectively freezing the associated subsystems and potentially causing broader system instability depending on the dependencies between components. The issue is not triggered by malicious code per se but rather by legitimate high-throughput operations that expose the timing-sensitive nature of the original implementation's error handling mechanism.

To resolve this vulnerability, the fix involves a fundamental change in how master abort conditions are managed within the MACE PCI controller driver. Instead of clearing the interrupt enable bit around every configuration read, which creates the problematic window for race conditions with DMA operations, the updated logic acknowledges master aborts directly through the error register. This approach ensures that errors are recorded and handled without disabling the ability to detect future faults during critical periods. Furthermore, the patch mandates that every write operation be immediately followed by a read access. This synchronization technique prevents back-to-back writes from forming the specific sequence of operations that previously led to the hang condition, thereby stabilizing hardware interactions under load.

From a classification perspective, this vulnerability aligns with CWE-362, which describes concurrent execution using shared resources with improper synchronization. The failure to properly synchronize access to hardware registers during high-load scenarios allowed for a race condition that compromised system stability. Additionally, the issue relates to CWE-841, Improper Enforcement of Behavioral Workflow, as the original code did not adequately enforce the correct sequence of operations required to maintain safe state transitions in the PCI controller under concurrent DMA activity. The fix ensures that the hardware state machine remains consistent and responsive regardless of external traffic intensity.

In terms of detection and mitigation strategies for similar architectural issues, this case highlights the importance of rigorous timing analysis when interacting with legacy or specialized hardware controllers like those found in MIPS-based SGI systems. Security engineers should prioritize reviewing drivers that manipulate control registers during high-throughput operations to ensure that interrupt masking does not create windows where error states are lost or misinterpreted. Mitigation for existing deployments involves applying the kernel patch that implements the acknowledge-and-read-back pattern, which restores robustness against DMA-induced hangs without requiring hardware changes. Continuous monitoring of system logs for unexplained freezes under load can also serve as an indicator of whether such synchronization flaws remain present in older, unsupported kernels where patches may not be readily available.

Responsible

Linux

Reservation

09/25/2026

Disclosure

11/20/2026

Moderation

accepted

CPE

ready

EPSS

0.00168

KEV

no

Activities

very low

Sources

Interested in the pricing of exploits?

See the underground prices here!