On Sun, 2005-11-27 at 10:32 +0200, Dotan Cohen wrote: > On 11/27/05, Matthew Miller <mattdm@xxxxxxxxxx> wrote: > > On Sun, Nov 27, 2005 at 12:57:29AM +0200, Dotan Cohen wrote: > > > > gcc -lm 1_3.c > > > Thank you, it compiled now. I just spacebar'ed about 100 times in the > > > gcc man page looking for the "m" option that you mention. I did find > > > the "l" option (search the library named library when linking), but > > > not the "m" option. > > > > Right -- "m" isn't an option, but actually the name of the library. Well, > > actually, the library is "libm.so", but the -l assumes a prefix of lib and a > > suffix of .so. > > > > Why doesn't it include the math library when I put at the top of the file: > # include <math.h> > ? It did. Rather, it allowed you to call the functions from within the math library. However, since the source code only had prototypes for those functions, you have to tell GCC where it finds the actual *implementation* of those functions, so it can create appropriate output code. That's my understanding of it, anyway. -- Peter Gordon (codergeek42) GnuPG Public Key: 0xDA3634D7
Attachment:
signature.asc
Description: This is a digitally signed message part
