arXiv:2610.36612v1 Announce Type: new Abstract: Hardware malware detectors (HMDs) classify hardware performance-counter measurements to identify malicious execution. Faults in their decision logic or parameters can weaken detection and allow malware to be classified as benign. We investigate this vulnerability through structural fault injection across a design space comprising 18 classifiers, four counter budgets, and AdaBoost and Bagging ensembles. The simulation evaluates 453,174 decision stuck-at and selected parameter bit-flip events on a 7,400-instance. Primary comparisons use 128 completed configurations. These completed-subset results are distinguished from the simulation summaries covering 140 configurations with enumerable fault sites. Fault outcomes are classified by their security consequences, including evasion, false-alarm increases, and metric silence. A fault is critical when malware recall decreases by at least five percentage points. Tree-based models exhibit a lower mean critical-fault fraction than function-based models, 5.3% versus 14.3%; exhaustive-only analysis preserves this ordering. Moreover, 72.3% of evaluated tree faults are metric-silent, changing both recall and false-positive rate by less than 0.1 percentage points without necessarily preserving individual predictions. Among the 27 completed matched comparisons, Bagging reduces criticality in 26, whereas AdaBoost increases it in two. Finally, an exact minimax selection method identifies protection candidates under additive unit costs and ideal correction. For a representative J48 detector, protecting 16 of 115 structural units reduces predicted maximum recall loss from 60.1 to 3.3 percentage points. These results characterize structural-model sensitivity and potential protection benefits; faulted RTL correspondence and physical hardening overhead remain unvalidated.