Rigel: Reverse-Engineering the Metal 4.1 Tensor Compute Path on the Apple M4 Max GPU
Apple's Metal 4.1 exposes a tensor compute path: the Metal Performance Primitives (MPP) matmul2d operation over cooperative_tensor fragments, whose interface is documented but whose hardware behavior is deliberately hidden. The specification states which data-type rows are supported, never whether they are hardware-accelerated, where the operation physically executes, what its accumulator width is, or how it partitions matrix fragments across threads. We present Rigel, an empirical characterization of this path on a single Apple M4 Max (a pre-neural-accelerator generation). Using a checksum-gated, provenance-tracked microbenchmark harness, Rigel recovers eleven facts the v4.1 specification hides or contradicts. The headline finding: the Metal 4.1 fp8 (E4M3) matmul2d is emulated, not accelerated: it sustains 0.94x the throughput of fp16 despite reading half the operand bytes, so on M4 it is a memory-footprint feature, not a performance feature. We further show, via a three-signal triangulation (throughput ceiling, comparison against simdgroup_matrix, and per-rail power attribution), that matmul2d executes entirely on the GPU shader cores with no dedicated matrix datapath and no evidence of Apple Neural Engine routing; that it accumulates in >=fp32; and we reconstruct the opaque 8x8 cooperative_tensor fragment layout Apple documents nowhere. Acting on the characterization, a hand-fused GEMM + bias + GELU kernel beats the decomposed path by +6.5-12.9% in the cache-resident regime. All findings are reproducible from committed MIT-licensed code and per-cell CSVs.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Identification of Metallic Objects using Spectral Magnetic Polarizability Tensor Signatures: Object Classification
The early detection of terrorist threat objects, such as guns and knives, through improved metal detection, has the potential to reduce the number of attacks and improve public safety and security. To achieve this, there…
ClassificationObjectRigel: Self-Distilled Score Adaptation for Image and Video Captioning Evaluation
Automatic evaluation of image and video captioning is essential for benchmarking multimodal systems, although standard evaluation metrics show limited alignment with human judgments. Recent approaches using large languag…
Video CaptioningMasterful: A Training Platform for Computer Vision Models
Masterful is a software platform to train deep learning computer vision models. Data and model architecture are inputs to the platform, and the output is a trained model. The platform's primary goal is to maximize a trai…
ManagementLSP Framework: A Compensatory Model for Defeating Trigger Reverse Engineering via Label Smoothing Poisoning
Deep neural networks are vulnerable to backdoor attacks. Among the existing backdoor defense methods, trigger reverse engineering based approaches, which reconstruct the backdoor triggers via optimizations, are the most …
Backdoor Attackbackdoor defenseIdentification of Metallic Objects using Spectral MPT Signatures: Object Characterisation and Invariants
The early detection of terrorist threats, such as guns and knives, through improved metal detection, has the potential to reduce the number of attacks and improve public safety and security. To achieve this, there is con…