Bit-exact simulation models reproduce precise nvidia gpu special functions

SFU-Sim: Bit-Exact Reverse Engineering of NVIDIA MUFU and Validation of PTX Paths

Hardware Architecture

Summary

Graphics cards use special instructions to calculate things like sine and cosine, but the exact output bits are not public. The authors created SFU-Sim, a tool that copies the exact bit output of these special GPU instructions by carefully studying real hardware without needing internal details. They tested it exhaustively on different NVIDIA GPU models and found it matches perfectly. This tool can help simulate GPU computations exactly or verify new hardware designs.

What this means in practice

  • For gpu simulator developers: Integrate SFU-Sim to provide bit-exact outputs of NVIDIA special function instructions in instruction-level GPU simulators.
  • For hardware verification engineers: Use SFU-Sim as a precise arithmetic reference to verify correctness of new hardware implementations matching NVIDIA GPU numerical behavior.

Authors

Haozhi Xie, Tao Su

Abstract

Reproducing the numerical behavior of a GPU requires predicting the exact output of special function instructions. Vendor documents give only the function semantics and error bounds. These do not fix the output bit pattern. We present SFU-Sim, which builds bit-exact models of NVIDIA multi-function unit (MUFU) instructions by black-box reverse engineering on real GPUs. We first model a single bare MUFU instruction. We then add the surrounding operations found in the disassembly to model full PTX paths. The model covers 12 common MUFU forms and two BF16 forms on the tested Blackwell GPU. Using SIN/COS as a complete case study, we show how we analyze the PTX-to-SASS compilation path and recover the segment structure, the equivalent coefficient tables, and the finite-width arithmetic. On NVIDIA GPUs of three architectures (Ampere, Ada Lovelace, and Blackwell), we exhaustively compare every tested PTX path over its specified input domain. FP16 and FP32 paths cover all $2^{16}$ and $2^{32}$ input encodings, respectively. The model output matches the hardware bit for bit, with zero mismatches. SFU-Sim computes each output directly from its input. It needs neither a lookup table of all inputs nor access to a physical GPU. It can serve as a bit-exact special function component for instruction simulators. It can also serve as an arithmetic reference for hardware design and functional verification that target the same numerical behavior. SFU-Sim and its validation programs are open source at https://github.com/hzX40000000/SFU-Sim