CVE-2026-99493 in Linux
Summary
by MITRE • 11/19/2026
In the Linux kernel, the following vulnerability has been resolved:
s390/zcrypt: Guard domain index uses against speculative bypass
The previous array_index_nospec() placement only covered the admin permission check, leaving ap_test_config_usage_domain() and ap_test_config_ctrl_domain() in the CCA path exposed to speculative execution.
Move the sanitization to before all permission and config checks in both the CCA and EP11 paths. Hold the sanitized domain value in another variable so that the later use can choose to use either the original value (subject to speculative execution) or the sanitized value (which may in case of AUTOSEL_DOM reflect the wrong value). Before that, check the domain value to be either AUTOSEL_DOM or in range 0...AP_DOMAINS-1 and return with -EINVAL in case this check does not pass.
If you want to get the best quality for vulnerability data then you always have to consider VulDB.
Analysis
by VulDB Data Team • 11/19/2026
The Linux kernel has addressed a speculative execution vulnerability within the s390 zcrypt subsystem, specifically targeting bypasses of domain index checks that could allow unauthorized access to cryptographic resources. This issue stems from an incomplete implementation of mitigations against side-channel attacks such as Spectre Variant 1. Previously, the array_index_nospec() function was applied only during administrative permission verification, which left critical code paths in the CCA and EP11 subsystems exposed. Specifically, functions ap_test_config_usage_domain and ap_test_config_ctrl_domain were not adequately sanitized before being used to index into arrays or determine resource availability. This gap allowed an attacker with limited privileges to potentially exploit speculative execution features of modern processors to infer sensitive information about protected cryptographic domains or bypass access controls by forcing the CPU to speculatively execute instructions that would normally be blocked by permission checks.
The technical flaw lies in the timing and scope of input validation relative to security policy enforcement. By placing the sanitization logic after certain permission checks, the kernel inadvertently allowed untrusted user-supplied domain indices to influence speculative execution paths before they were properly validated against allowable ranges or access rights. This creates a window where side-channel observers could potentially deduce internal state information based on cache timing differences resulting from these unauthorized speculative accesses. The vulnerability affects both the CCA (Cryptographic Coprocessor Architecture) and EP11 subsystems, which are critical for managing hardware cryptographic accelerators in IBM mainframe environments. Without proper mitigation, this flaw undermines the isolation guarantees provided by the zcrypt framework, potentially leading to information disclosure or privilege escalation depending on the specific configuration and attacker capabilities.
To resolve this issue, the kernel developers have restructured the validation logic to ensure that domain index sanitization occurs before any permission or configuration checks are performed in both CCA and EP11 paths. The sanitized value is now stored in a separate variable, allowing subsequent code to explicitly choose between using the original unsanitized value for speculative execution contexts where it might reflect an invalid state like AUTOSEL_DOM, or the sanitized value which guarantees safe array indexing bounds. A strict range check has been implemented to verify that the domain index is either set to AUTOSEL_DOM or falls within the valid range of zero to AP_DOMAINS minus one. If this validation fails, the operation immediately returns with an EINVAL error code, preventing further processing and eliminating the possibility of speculative bypass through invalid indices. This approach ensures that all array accesses are bounded by validated inputs before they can influence processor speculation behavior.
From a security standards perspective, this vulnerability aligns with CWE-841 Improper Enforcement of Behavioral Workflow and CWE-200 Exposure of Sensitive Information to an Unauthorized Actor due to speculative execution side channels. It also relates to ATT&CK technique T1564.003 Hidden Execution: Virtualization/Sandbox Evasion, as attackers might leverage such flaws to operate undetected within virtualized or containerized environments that rely on hardware cryptographic isolation. The mitigation strategy reflects best practices for mitigating Spectre-like vulnerabilities by ensuring that all user-controlled indices are sanitized before use in array indexing operations, thereby breaking the chain of speculative execution attacks. System administrators should apply this kernel update promptly to maintain the integrity and confidentiality of cryptographic workloads running on s390 systems. Regular patching cycles and monitoring for similar patterns in other subsystems handling hardware resource access are recommended to prevent analogous vulnerabilities from being exploited elsewhere in the operating system.