What do the E and R prefixes stand for in the names of Intel 32-bit and 64-bit registers?

Question

The 32-bit register names start with an E and the 64-bit ones start with an R. What do the E and R stand for? Is there a reason for choosing these alphabets?

Also, in 64-bit registers, too, we can see in any low-level debugger such as Windbg that the right-most bits are still referred to by the same name as that of the 64-bit register except the name starts with an E. For e.g. the right-most 32 bits of the RAX register in a 64-bit system is referred to as EAX.

So, do the E and R stand for something? And also, why the suffix X for register?

Answer 1

• R just stands for "register". The AMD64 ISA extension added 8 additional general-purpose registers, named R8 through R15. The 64-bit extended versions of the original 8 registers had an R prefix added to them for symmetry.

• E stands for "extended" or "enhanced". (Wikipedia says "extended".) They are the "extended" versions of the 16-bit registers, in that they offer 16 additional bits for 32 bits total.

• X is also for "extended"—or perhaps it implies 16 as in hexadecimal.^* The X-suffixed registers are the 16-bit extended versions of the 8-bit registers. For 8-bit registers, the L suffix means "low", and the H suffix means "high".

Therefore, taking one particular register as an example, you have the 8-bit AL and AH registers, which are the low and high bytes of the 16-bit AX register, which is the low word of the 32-bit EAX register, which is the low double-word of the 64-bit RAX register.

| 63 - 32 | 31 - 16 | 15 - 8 | 7 - 0 | ====================================== .         .         | AH     | AL    | .         .         | AX             | .         | EAX                      |  | RAX                                | ====================================== | 63 - 32 | 31 - 16 | 15 - 8 | 7 - 0 | 

__
_{* X was used in the mnemonics (such as LXI and DCX) on the 8080 for instructions that treated a pair of otherwise-separate 8-bit registers as a 16-bit integer, similar to how AX represents the AH:AL pair. Thus, another possible interpretation is that X means pair, and this usage was continued when naming the high:low pairs on subsequent processors, including the 8086, which was a full 16-bit extension of the 8080.}