what are the parameters for __cudaRegisterFatBinary and __cudaRegisterFunction functions?

Question

I came across two functions

__cudaRegisterFatBinary()

and

__cudaRegisterFunction()

I understand nvcc injects it into the source code to get the handle to the compiled cubin and to register the program with the runtime.

Can some one explain me or point me towards where i can find information on each of the parameters of the functions? specifically I want to know more about the __cudafatcudabinaryrec pointer used in the first function. And about host function and device function pointers in the second function.

Thank you

Answer 1

Prototypes are in cudart.h.

void** __cudaRegisterFatBinary(void *fatCubin);

void __cudaRegisterFunction(void **fatCubinHandle, const char *hostFun, char *deviceFun,
                            const char *deviceName, int thread_limit, uint3 *tid,
                            uint3 *bid, dim3 *bDim, dim3 *gDim, int *wSize);

But please note that these functions are not meant to be called directly by user code.

Answer 2

I know this is a very old thread, but I just want to share some of my discoveries. I reverse engineered some part of the executable that NVCC produce. So I am not sure about the correctness and use at your own risk. I am using cuda 8.0 RC, so I am not sure if other versions changed anything.

The __cuRegisterFatBinary takes a void * as input. It points to the executable and in my example, I got the following.

B1 43 62 46 01 00 00 00 70 15 40 00 00 00 00 00 00 00 00 00 00 00 00 00

The sequence of hex follows the format

struct {
    uint32_t magic; // Always 0x466243b1
    uint32_t seq;   // Sequence number of the cubin
    uint64_t ptr;   // The pointer to the real cubin
    uint64_t data_ptr;    // Some pointer related to the data segment
}

So if you follow the address in field ptr, you will be able to find the real fat binary which follows the definition that you can find in fatbinary.h in your cuda include directory. There are some header information. If you search for the next occurrence of 0x7F + 'ELF' (the elf magic), you will be able to extract the cubin file there.