QtMRA32

ARMv7 32 bit emulator

Instructions

CMP <RN>, <const or register>
MOV <Rd>, <const or Register>
ADD <Rd>, <Rn>, <Rm or const>
SUB <Rd>, <Rn>, <Rm or const>
MUL <Rd>, <Rn>, <Rm>

MLA <Rd>, <Rn>, <Rm>, <Rs> # MULTIPLY accumalte
B <BRANCH_NAME> # this is a comment
branch: 
ORR <Rd>, <Rn>, <Rm or const> # bitwise ORR not logical OR
AND <Rd>, <Rn>, <Rm or const> # bitwise AND not logical AND

also added comments with the "#" they follow the MIPS and numbers are like this

Conditionally Executed Instructions

this will follow the CPSR flags which set by adding a "S" prefix or CMP

EQ <Rn>, <Rm or const> (Z_FLAG = 1)
NE <Rn>, <Rm or const> (Z_FLAG = 0)
LE <Rn>, <Rm or const> (N_FLAG =  1 OR Z_FLAG = 1)
GE <Rn>, <Rm or const> (N_FLAG = 0 OR Z_FLAG = 1)
LT <Rn>, <Rm or const> (N_FLAG = 1)
GT <Rn>, <Rm or const> (N_FLAG = 0)

you can edit the flags

Registers

R0-R12 general purpose registers

R7 sys call register

R13 stack pointer (when i impl it)

R14 Link register (when I impl. it)

R15 being the PC

Other Technical Details

can only do Integers and positive numbers. and numbers up to 2^12 or 4096

if a number is not a power of 2, and is greater than 4096. then do "ORR" to the nearest power of 2

example:

MOV R0, 5000
ORR R0, R0, 4096

if this results in a "wierd" number, its one of the few numbers the ROR works with, if it does. you dont need the ORR

floating point numbers will be coming soon.

(using my own floating point lib I used for MIPS)

as of right now uses big endian. will implement little endian eventually

(not a full 1:1 emulation of an ARM CPU)