decuda：cubin的反编译工具

#include <cuda_runtime_api.h> #include <cuda.h> #include <algorithm> #include <cassert> #define SAFE_CALL(x) assert((x)==CUDA_SUCCESS); int main() { CUmodule m = 0; CUdevice dev = 0; CUcontext ctx = 0; /// Init cuda SAFE_CALL(cuInit(0)); /// Get device SAFE_CALL(cuDeviceGet(&dev, 0)); /// Create context SAFE_CALL(cuCtxCreate(&ctx, 0, dev)); SAFE_CALL(cuModuleLoad(&m, "test.cubin")); CUfunction hfunc; SAFE_CALL(cuModuleGetFunction(&hfunc, m, "my_kernel")); int width = 16; int size = width*4; uint32_t *data=0; CUdeviceptr gdata; SAFE_CALL(cuMemAlloc(&gdata, size)); SAFE_CALL(cuMemsetD32(gdata, 0, size/4)); data = (uint32_t*)malloc(size); SAFE_CALL(cuFuncSetBlockShape(hfunc, 16, 1, 1)); SAFE_CALL(cuFuncSetSharedSize(hfunc, 0)); SAFE_CALL(cuParamSetSize(hfunc, 8)); // 8+4 SAFE_CALL(cuParamSetv(hfunc, 0, &gdata, sizeof(gdata))); //SAFE_CALL(cuParamSeti(hfunc, 4, 1)); //SAFE_CALL(cuParamSeti(hfunc, 8, 2)); //SAFE_CALL(cuParamSeti(hfunc, 12, 3)); //SAFE_CALL(cuParamSeti(hfunc, 16, 0)); //SAFE_CALL(cuParamSeti(hfunc, 20, 0)); int shared_size = 0; SAFE_CALL(cuLaunch(hfunc)); //SAFE_CALL(cuLaunchGrid(hfunc, 3, 4)); SAFE_CALL(cuMemcpyDtoH((void*)data, gdata, size)); for(int x=0; x<width; ++x) printf("%08x ", data[x]); printf("\n"); for(int x=0; x<width; ++x) printf("%f ", *(float*)(&data[x])); printf("\n"); return 0; }