30

Doing K&R using Xcode, in the Functions section I entered the code for their example of the power function as follows.

#include <stdio.h>

int power(int m, int n);

int main()
{
    int i;

    for (i=0; 1<10; ++i)
        printf("%d %d %d\n", i, power(2, i), power(-3, i));

    return 0;
}

When I try to run it, the following error appears:

Undefined symbols for architecture x86_64:
  "_power", referenced from:
      _main in main.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)

I have read a lot of the answers to this question, but cannot see how it applies to my situation with such a little program. How can I fix this problem?

Improve this question
12
  • 1
    Because the power function doesn't exist? Commented Dec 4, 2015 at 4:20
  • power(int m, int n) function not defined. Commented Dec 4, 2015 at 4:20
  • check here stackoverflow.com/questions/11704448/… Commented Dec 4, 2015 at 4:22
  • I thought it was in <stdio.h> How do I define it if it isn't there? Commented Dec 4, 2015 at 4:23
  • 1
    1<10 should be i<10 Commented Dec 4, 2015 at 4:27

7 Answers 7

Reset to default
14

Why do I get "clang: error: linker command failed with exit code 1"?

You just declared the function. There is not any definition in the code. At the time of the linking process, the compiler (here Clang) cannot link power function to its definition, so the linker throws the error in this kind of situation. If you define

int power(int x, int y)
{
    /* Do calculation */
}

Then the linker can link your declaration of power function to its definition, and you will not get any error.

For an integer number I have made a function:

#include <stdio.h>

int power(int base, int exp);

int main()
{
    int i;

    for (i=0; i<10; ++i)
        printf("%d %d %d\n", i, power(2, i), power(-3, i));

    return 0;
}

int power(int base, int exp)
{
    int result = 1;
    while (exp)
    {
        if (exp & 1)
            result *= base;
        exp >>= 1;
        base *= base;
    }
    return result;
}

Compile this with gcc file.c.

Improve this answer
Sign up to request clarification or add additional context in comments.

6 Comments

using a double as a loop condition is not a great idea due to potential floating point rounding errors (assuming you meant i < 10)
ohh I will consider your suggestion and update the answer. Thanks :-)
This answer would be better if you explained why OP's code doesn't work. And probably should mention that the math library needs to be linked in with -lm.
@kaylum ...if using an old, naff compiler
@M.M Are you saying newer compilers don't need -lm? I didn't know that.
|
4

You missed the definition of function int power (int base, int n) which is given after your main ends on the next page of the book.

When you declare prototype of a function you need to define what it should do you just declared the power function and never defined that, that's why you got error.

Include the following definition, your code will compile the way you wants it to be.

int power (int base, int n) {
    int i, p;
    p = 1;
    for (i=1; i<=n; ++i)
        p = p*base;
    return p;
}

Pre-edit answer

Now this is not relevant, but useful

I think you want to use function pow() defined in math.h.

double pow(double a, double b)

The C library function pow(double a, double b) returns a raised to the power of b. This function returns a double value so to print that correct specifier will be "%lf".

In this case you just need to include header file

#include <math.h>

In your program.

There is no need to give function declaration int power(int m, int n);

The error you are having is due to giveing I as on of the parameter to pow() because when you will compile your code (after including math.h and using pow() replacing i with any integer numbers will compile your code and will give proper output.

printf("%lf %lf %lf\n", i, pow(2, 3), pow(3, 2));

This will give you proper result but when you compile it with

for (i=0; i<10; ++i) {
    printf("%lf %lf %lf\n", i, pow(2, i), pow(-3, i));
}

It throws same error so I think pow() takes only constants as input so it won't run in for loop.

And if you don't want to include math.h you can simply declare

extern double pow (double base, double exponent);

That will link correctly with the library code without using the math.h include file, here is an example.

int main() {
    extern double pow (double base, double exponent);
    printf("%lf", pow(8.0, 8.0));
    return 0;
}

For more on pow() you can check man page on Linux i.e. man pow.

Improve this answer

1 Comment

a lot said..nothing to be added +1
1

There is a standard library function which does just that...

#include <math.h>
double pow(double x, double y)

You have to link it explicitly because the default linker, that is ld invoked when no other options given, doesn't link with standard math library.

So you have to do it like...

gcc file.c -lm

Improve this answer

2 Comments

Re "the default linker, that is ld": For GCC only, presumably? The signature for the (real - not executable gcc on macOS which is Clang by default) GCC is "ld returned 1 exit status" whereas it is "linker command failed with exit code 1" for Clang (in this question).
Using executable 'gcc' would invoke Clang on that system, not GCC (also fairly clear from the error message: "clang: error: linker command...").
1

You can define the power function

int power(int m, int n) {
    // Implement the function body
}

And then your issue will get fixed.

You're getting an error, because there isn't any deceleration for the defined function. So add the deceleration as shown above.

Improve this answer

5 Comments

function Declarations has absolutely nothing to do with linking. Even if you omit every kind of declarations, and compile without -Werror, and supply all the libraries as linker flags, it will link just fine. Declarations fir functions are just a new thing(they were nonexistant before ansiC, and only made mandatory in c99 which nobody enables during compilation) which helps compiler to do some accounting beforehand. Only type declarations are a must.
Tried that ANjaNA but no success. However, when I read K&R again it actually said that pow(x,y) was held in the standard library. Again, that did not work immediately, but the following program did.
#include <stdio.h> double pow(double m, double n); int main() { double i; for (i=0; i<5; ++i) printf("%.0f %.0f %.0f\n", i, pow(2,i), pow(-3,i)); return 0; }
It's not looking right but the answer was using Double value instead of Int values. I don't know why exactly, but I'm happy (sort of)
Thank you everyone. If I even become one quarter as smart as you lot I will be a happy man. Yes, I left out the second part of the program because it was on the next page and I didn't realise it was required. Well I do now.
0

First, you get the error because the compiler can not find the definition of the power function that you are using. Even if you write

int power(int m, int n);

There is an error because you are not defining the function. There is missing {} at the end of the definition and even if you are putting it at the end of the definition, you will get an error because you are not returning nothing for a int function. So, at least if you want to define a function, you have to proceed like this:

int power(int m, int n){return 0};

Then, you will be able to use your function power() in the main function. But, you are doing nothing in the power() function, so you will get nothing back of calling it. If you want to compute the power of a number, you can use the function pow() that is present in the cmath library. A straightforward way of doing it is something like this:

#include <stdio.h>  // printf
#include <cmath>    // pow()
#include <iostream> // cout

void main()
{
    for (int i = 0; i < 10; i++)
        std::cout << i << " " << pow(2, i) << " " << pow(-3, i) << std::endl;
}

I included the iostream to use a different way of printing out using the object cout defined in the std namespace. Note that the pow() function has some requirements for its definition, so be careful with the types you are using. You can take a look at pow for more details.

Improve this answer

Comments

0

The function should be completed. I met the same question today only because I didn’t define function main in my code.

Improve this answer

1 Comment

As it’s currently written, your answer is unclear. Please edit to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers in the help center.
-1

Try command s(save) before clanging. Had the same problem, now it works

Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.