AI Engine-ML Intrinsics User Guide  (v2023.2)
Load/Store Operations

The compiler supports pointer dereferencing and pointer arithmetic. No special intrinsics are needed to load or store vectors. For example: More...

Overview

The compiler supports pointer dereferencing and pointer arithmetic. No special intrinsics are needed to load or store vectors. For example:

v16int16 *input, *output;
// ... initialize input and output data pointers
v16int16 mydata = *input++;
...
*output = mydata;

The data pointers must be aligned to 256-bit boundary for vector loads/stores. Load/Store behaviour is undefined when stack-allocated vector variables are unaligned.

Linear pointer manipulation can be directly in the code with the use of the de-referencing, increment, decrement, etc, operators. For more complex pointer manipulation please see Addressing intrinsics .

Modules

 Addressing intrinsics
 
 Compressed Load Operations
 Compressed load operations load a compressed vector and expand it into an AIE-ML register.
 
 Compressed Sparse Load Operations
 Compressed sparse load operations load a compressed sparse vector and expand it into an AIE-ML register.
 
 Load 4x Operations
 Load 4x intrinsics load four 64-bit values to a vector register from data memory.
 
 Streams
 
v16int16
Definition: me_chess.h:491