X Index Register: Difference between revisions
(always 8 bits in emulation mode) |
(flowed LDX/STX into body) |
||
(4 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
The '''X Index Register''' on 65x processors often holds the current index when iterating over things. It can be incremented or decremented by one with [[INX]] or [[DEX]], but there is no instruction to add or subtract more than one. Although INX and DEX can affect the [[negative flag]], the indexed [[addressing mode]]s always interpret the bit pattern in the X index register as a non-negative integer. | The '''X Index Register''' on 65x processors often holds the current index when iterating over things. It can be incremented or decremented by one with [[INX]] or [[DEX]], but there is no instruction to add or subtract more than one. Although INX and DEX can affect the [[negative flag]], the indexed [[addressing mode]]s always interpret the bit pattern in the X index register as a non-negative integer. | ||
It can be loaded with [[LDX]] and stored to memory with [[STX]]. | |||
On the [[S-SMP]] it is always 8 bits wide. It is the divisor for division commands.<sup>[2]</sup> | On the [[S-SMP]] it is always 8 bits wide. It is the divisor for division commands.<sup>[2]</sup> | ||
On the [[65c816]], it may be 8 or 16 bits wide | On the [[65c816]], it may be 8 or 16 bits wide: | ||
* When the [[index register select]] flag is clear, X is 16 bit | |||
* When the index register select flag is set, X is 8 bit | |||
Indexing may cross bank boundaries.<sup>[1]</sup> In [[emulation mode]] it is always 8 bits wide. | |||
Unlike the [[Y index register]], the value of the [[stack pointer]] can be transferred to/from X with [[TXS]] and [[TSX]]. | Unlike the [[Y index register]], the value of the [[stack pointer]] can be transferred to/from X with [[TXS]] and [[TSX]]. | ||
Line 9: | Line 15: | ||
It is not specified to have any particular value after reset. | It is not specified to have any particular value after reset. | ||
It can be pushed to the [[stack]] with [[PHX]] and pulled with [[PLX]]. | |||
It can be transferred to the [[accumulator]] with [[TXA]] or to Y with [[TXY]]. Conversely, it can be made to reflect the accumulator with [[TAX]] or Y with [[TYX]]. | |||
[[CPX]] can compare it to something. | |||
=== References === | === References === |
Latest revision as of 23:29, 16 August 2024
The X Index Register on 65x processors often holds the current index when iterating over things. It can be incremented or decremented by one with INX or DEX, but there is no instruction to add or subtract more than one. Although INX and DEX can affect the negative flag, the indexed addressing modes always interpret the bit pattern in the X index register as a non-negative integer.
It can be loaded with LDX and stored to memory with STX.
On the S-SMP it is always 8 bits wide. It is the divisor for division commands.[2]
On the 65c816, it may be 8 or 16 bits wide:
- When the index register select flag is clear, X is 16 bit
- When the index register select flag is set, X is 8 bit
Indexing may cross bank boundaries.[1] In emulation mode it is always 8 bits wide.
Unlike the Y index register, the value of the stack pointer can be transferred to/from X with TXS and TSX.
It is not specified to have any particular value after reset.
It can be pushed to the stack with PHX and pulled with PLX.
It can be transferred to the accumulator with TXA or to Y with TXY. Conversely, it can be made to reflect the accumulator with TAX or Y with TYX.
CPX can compare it to something.
References
- https://wilsonminesco.com/816myths
- 8.1.2 of page 3-8-4 of Book I of the official Super Nintendo development manual