Load Vector Indexed

Hardware: G4


	Log In \| Not a Member?	Support

Load Vector Indexed

`vec_ld(arg1, arg2) vec_ldl(arg1, arg2)`

Each operation performs a 16-byte load from a 16-byte aligned address. arg1 is an integer value, while arg2 is a pointer. arg1 is added to arg2 and the sum is truncated, if necessary, to produce 16-byte aligned result. The arg2 type may also be a pointer to a const-qualified type.
vec_ldl marks the data as Least Recently Used (LRU), while vec_ld does not; otherwise the instructions are identical.

See AltiVec Alignment for a discussion of how to load data that may not be 16-byte aligned.

Result	arg1	arg2	vec_ld	vec_ldl
Result	arg1	arg2	Maps To
vector unsigned char	any integral type	vector unsigned char *	lvx	lvxl
vector unsigned char	any integral type	unsigned char *	lvx	lvxl
vector signed char	any integral type	vector signed char *	lvx	lvxl
vector signed char	any integral type	signed char *	lvx	lvxl
vector bool char	any integral type	vector bool char *	lvx	lvxl
vector unsigned short	any integral type	vector unsigned short *	lvx	lvxl
vector unsigned short	any integral type	unsigned short *	lvx	lvxl
vector signed short	any integral type	vector signed short *	lvx	lvxl
vector signed short	any integral type	short *	lvx	lvxl
vector bool short	any integral type	vector bool short *	lvx	lvxl
vector pixel	any integral type	vector pixel *	lvx	lvxl
vector unsigned long	any integral type	vector unsigned long *	lvx	lvxl
vector unsigned long	any integral type	unsigned int *	lvx	lvxl
vector unsigned long	any integral type	unsigned long *	lvx	lvxl
vector signed long	any integral type	vector signed long *	lvx	lvxl
vector signed long	any integral type	int *	lvx	lvxl
vector signed long	any integral type	long *	lvx	lvxl
vector bool long	any integral type	vector bool long *	lvx	lvxl
vector float	any integral type	vector float *	lvx	lvxl
vector float	any integral type	float *	lvx	lvxl