mips处理器构架（MIPS processor architecture）.docVIP

mips处理器构架（MIPS processor architecture） The MIPS architecture is one kind of RISC architecture There are 32 general registers 1 MIPS32 architecture, including $0 (no matter how you set, save the data in the register is 0) and $31 (return address save function calls JAL) has a special use, other registers can be used as a general-purpose register for any instruction. Although the specified register hardware no mandatory rules of use, in actual use, the usage of these registers will follow a series of conventions. These conventions and hardware does not, but if you want to use other peoples code, the compiler and the operating system, youd better follow these conventions. Register number mnemonic usage 0 zero always returns a value of 0 1 at temporary variable assembler 2-3 V0, the V1 function call return results 4-7 a0-a3 function call parameters 8-15 t0-t7 24-25 T8-T9 temporary variables, functions can be used without preservation and restoration 16-25 s0-s7 function register variables. The sub function must save and restore the used variables before the function returns, which calls the function know these register values did not change. 26,27 K0, K1 is usually an interrupt or exception handler as the preservation of some system parameters 28 GP global pointer. Some operating system maintenance more convenient access to the pointer static and extern variable. 29 SP stack pointer 30 s8/fp ninth register variables. The sub function can be used as the frame pointer Return to 31 RA sub function 2 MIPS32 if FPA (floating-point coprocessor), there will be 32 floating-point registers, according to assembly language agreed to $f0~$f31, only even numbered MIPS32 in the practical use of the floating-point registers, the odd number is: in double precision floating-point, 32 bit floating-point registers before the store even numbered the rest can not put aside. For example, in double precision floating-point operations, the remaining part of the $1 $0 storage, so the MIPS32 can