Why the different encodings?

Why are there different encodings of instructions?

What's the idea/background/etc for their co-existence?

Can different encodings be mixed in the code? (Not ARM encodings with Thumb encodings- without ARM/Thumb mode change,

but, like A1 and A2 or T1 and T2)?

I'm trying to put together a gdb stub, and for single stepping the machine code needs to be partially decoded.

How can one tell apart the encoding of an instruction in a machine code program (binary)?

Oh, and an additional question: what do the bit values in parenthesis mean?

For the case when cond is 0b1111, see Unconditional instructions on page A5-216.

t = UInt(Rt); n = UInt(Rn); imm32 = Zeros(32); // Zero offset

if t == 15 || n == 15 then UNPREDICTABLE;

Encoding T1 ARMv6T2, ARMv7

LDREX<c> <Rt>, [<Rn>{, #<imm>}]

Encoding A1 ARMv6*, ARMv7

LDREX<c> <Rt>, [<Rn>]

1 1 0 1 0 0 0 0 1 0 1 Rn Rt (1) (1) (1) (1) imm8

15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

1

1514131211 10 9 8 7 6 5 4 3 2 1 0

cond 0 0 0 1 1 0 0 1 Rn Rt (1) (1) (1) (1) 1 0 0 1 (1) (1) (1) (1)