首页 > 代码库 > ptxas fatal : Unresolved extern function Error 255

ptxas fatal : Unresolved extern function Error 255

2024-08-14 21:51:48   222人阅读

This question already has an answer here:

  • External calls are not supported - CUDA 1 answer

I am trying to understand how to decouple CUDA __device__ codes in separate header files.

I have three files.

File: 1: int2.cuh#ifndef INT2_H_#define INT2_H_#include "cuda.h"#include "cuda_runtime.h"#include "device_launch_parameters.h"__global__ void kernel();__device__ int k2(int k);int launchKernel(int dim);#endif /* INT2_H_ */File 2: int2.cu#include "int2.cuh"#include "cstdio"__global__ void kernel() {    int tid = threadIdx.x;    printf("%d\n", k2(tid));}__device__ int k2(int i) {    return i * i;}int launchKernel(int dim) {    kernel<<<1, dim>>>();    cudaDeviceReset();    return 0;}File 3: CUDASample.cuinclude <stdio.h>#include <stdlib.h>#include "int2.cuh"#include "iostream"using namespace std;static const int WORK_SIZE = 256;__global__ void sampleCuda() {    int tid = threadIdx.x;//    printf("%d\n", k2(tid)); //Can not call k2    printf("%d\n", tid * tid);}int main(void) {    int var;    var = launchKernel(16);    kernel<<<1, 16>>>();    cudaDeviceReset();    sampleCuda<<<1, 16>>>();    cudaDeviceReset();    return 0;}

 

The code works file. I can call the sampleCuda() kernel (in same file), call the C function launchKernel() (in other file), and call kernel() directly (in other file).

The problem I am facing is calling the __device__ function from sampleCuda() kernel. then it shows the following error. However, the same function is callable in kernel().

10:58:11 **** Incremental Build of configuration Debug for project CUDASample ****make all Building file: ../src/CUDASample.cuInvoking: NVCC Compiler/Developer/NVIDIA/CUDA-6.5/bin/nvcc -G -g -O0 -gencode arch=compute_20,code=sm_20  -odir "src" -M -o "src/CUDASample.d" "../src/CUDASample.cu"/Developer/NVIDIA/CUDA-6.5/bin/nvcc -G -g -O0 --compile --relocatable-device-code=false -gencode arch=compute_20,code=compute_20 -gencode arch=compute_20,code=sm_20  -x cu -o  "src/CUDASample.o" "../src/CUDASample.cu"../src/CUDASample.cu(18): warning: variable "var" was set but never used../src/CUDASample.cu(8): warning: variable "WORK_SIZE" was declared but never referenced../src/CUDASample.cu(18): warning: variable "var" was set but never used../src/CUDASample.cu(8): warning: variable "WORK_SIZE" was declared but never referencedptxas fatal   : Unresolved extern function ‘_Z2k2i‘make: *** [src/CUDASample.o] Error 25510:58:14 Build Finished (took 2s.388ms)

 




上面这个错误的解决方法都是：需要分开编译文件，对应的编译参数为--relocatable-device-code={true,false}将false 改为 true
nsight 的设置的位置为：
         In Nsight Eclipse, the option is available as the radio button "Separate compilation" under Project > Properties > Build > Settings > CUDA > switch to advanced mode 里面勾选就可以了。 
VS的设置位置为：刚才突然想写。。找不到了。。具体google 搜索吧，best wishes....

ptxas fatal : Unresolved extern function Error 255


联系
我们