Hi, I have a basic doubt in compiler code generation for a target CPU. In my opinion, it is compiler that generates assembly code for target cpu (e.g -mcpu=750, for PowerPC CPU 750). I could see target option provided by powerpc assembler(-m601,-m403,-m405,-m750,...). Whats the effect of specifying this option to assembler? One more doubt is, I'm using GCC 4.1.0 and the GNU assembler 2.19.1 build for PowerPC family(powerpc-eabi). GCC 4.1.0 support code generation for target CPU 750. When I invoked my assembler with --target-help, it listed few CPU type including 750CL. Currently, I'm porting an application to GCC. There I could see some inline assembly code in C source with an instruction "mfspr %r3, %sprg2". GNU assembler throws error for the register "sprg2". Does this mean, the assembler doesn't support the register "sprg2"? Or it could interpret throught different name? How to know the target register names manipulated by the assembler? In my tool chain installation directory, I found ppc-asm.h file which lists the constant values for each register defined by macro. But I couldn't see any register like "sprg". Please clarify this. best regards, Jeffi -- View this message in context: http://old.nabble.com/Basic-doubt-in-code-generation-for-target-CPU-tp26492010p26492010.html Sent from the gcc - Help mailing list archive at Nabble.com.
