> Google tells me there is there are GNU utilities, but it seems to only su= pport assembly, and not C. You'll have to be clearer about what you're talking about. "Gnu tools" for software development typically come in three parts: Gnu utilities ("binutils"): Assembler, linker. and assorted minor utilities= .. Gnu Compiler Collection ("gcc"): Assorted compilers: Ada, Fortran, Pascal, = C++, and most famously: C Gnu libraries: runtime support for the compilers, along with IO and math an= d stuff. C is somewhat unique in how little library support is needed to h= ave a functioning compiler (and, IMO, uniquely successful because of this.)= Gnu licenses are rather incompatible with embedded applications, so for ma= ny microcontrolers libraries from assorted 3rd parties (frequently includin= g the chip vendors) are used (For Atmel AVR there is "avr-libc", and many A= RMs use "newlib.") None of the gnu utilities, AFAIK, support the 8-bit PICs. Both 16 and 32bi= t PIC cpus use the gnu assembler and C compiler, but I believe if you use t= he downloads from Microchip, they use proprietary libraries owned by Microc= hip. (MPIDE, the Arduino-compatible environment for the ChipKit 32bit PIC = boards, is currently bundling the compiler with "newlib" instead of the Mic= rochip libraries. There are some interesting implications: 1) Many of the languages supported by the compiler collection do not have r= untime libraries for many targets. So AFAIK, there's no Ada or Fortran for= AVR, even though you can theoretically build the relevant compiler. 2) The build process for creating a cross compiler is complicated. You hav= e to build a "partial compiler" and binutils for a target so that you can b= uild the libraries needed for a full compiler. 3) There are some "default" libraries provided with gcc for things like sof= tware floating point. They're generic, written in C, large, and not particu= larly optimized for any given architecture. If you want faster optimized S= W floating point, you have to look elsewhere. This is one of the things p= rovided by avr-libc for AVRs. As a result, 32bit floating point on a 16MHz= AVR is almost as fast as it is on a 48MHz ARM chip. :-( 4) Licensing issues. binutils and gcc are shining models of open source. = Libraries: frequently not so much :-( BillW --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .