Function names in AOCL-DLP follow a systematic naming convention that encodes the data types:
Pattern: [input_A][input_B][accumulation]o[output]
Abbreviation |
Data Type |
Description |
|---|---|---|
|
float (32-bit) |
IEEE 754 single precision |
|
bfloat16 (16-bit) |
Brain floating point format |
|
uint8_t (8-bit) |
Unsigned 8-bit integer |
|
int8_t (8-bit) |
Signed 8-bit integer |
|
int4 (4-bit) |
Signed 4-bit integer (packed) |
|
int32_t (32-bit) |
Signed 32-bit integer |