basic floating point support in Newlib

[Eric Haw-Luen Tsai(蔡浩倫)]

Dear Sir,
 
I'm porting Newlib to our RISC CPU.  I successfully built a couple of libraries such as, libc, libg, libm, libgloss and libnosys.  When I tried to use libc and libm, I got undefined references, such as muldf3, adddf3 and floatsidf.  These look like arithmetic functions for double that my program, compiled by our GCC, tries to use.  I can't find the source code for these functions in Newlib.  So, Is there any equivalent functions in Newlib?
 
Regards,
 
Eric Tsai