[avr-libc-dev] [bug #3485] Using float arithmetic without linking with -lm result in incorrect code

[Joerg Wunsch]

Update of bug #3485 (project avr-libc):

                Category:           Documentation => Library                
              Item Group:           Documentation => libm code              
             Open/Closed:                  Closed => Open                   
           Fixed Release:                   1.6.2 => None                   

    _______________________________________________________

Follow-up Comment #5:

Well, let's keep this one open until all of our libgcc routines
have really been integrated into libgcc (__fixunssfsi etc.).
*This* is the actual reason for the incorrect code, and I've
already talked to Dmitry, he wants to look into what needs to
be done to contribute that back to the GCC project.

Unless any of the functions from <math.h> is needed, linking
against libm.a ought to be entirely optional, and should not
cause different code to be generated.

    _______________________________________________________

Reply to this item at:

  <http://savannah.nongnu.org/bugs/?3485>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.nongnu.org/