Program Statements - Nintendo Ultra64 Programmer's Manual

Revision 1.0
Vector Register Element Syntax
In some circumstances, a scalar element of a vector register may be specified.
These circumstances include the target register of most vector
computational instructions and the source/destination register of all vector
loads, stores, and moves.
For vector computational instructions, a vector register element syntax is
one of:
• an integer (or integer expression) in the range 0...7, enclosed by
square brackets ([]), representing the ordinal index of one of the 8
16-bit vector elements of the register.
• an integer (or integer expression) in the range 0...3, followed by
the letter 'h', enclosed by square brackets ([]), representing the
ordinal index of one of the 4 16-bit vector elements of the register
halves.
• an integer (or integer expression) in the range 0...1, followed by
the letter 'q', enclosed by square brackets ([]), representing the
ordinal index of one of the 2 16-bit vector elements of the register
quarters.
For vector loads, stores, and moves, the vector register element syntax is as
follows:
• an integer (or integer expression) in the range 0...15, enclosed
by square brackets ([]), representing which of the 16 bytes of the
register to use as a source or destination.
In any case where a vector element may optionally be specified, but is not, a
0 is assumed.
Additional usage of vector register element syntax is explained along with
the instructions that use them in a later chapter.

Program Statements

A program statement consists of an optional label, an operator keyword, and
the operand(s). The operator may be a scalar instruction or a vector
instruction.
Syntax
113