ICachePreLoad (ISASI)

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product/Atlas A3 inference product	√
Atlas A2 training product/Atlas A2 inference product	√
Atlas 200I/500 A2 inference product	x
Atlas inference product AI Core	√
Atlas inference product Vector Core	x
Atlas training product	x

Preloads instructions to the iCache from the DDR address where the instructions are located.

__aicore__ inline void ICachePreLoad(const int64_t preFetchLen)

**Table 1** Parameters
Parameter	Input/Output	Description
preFetchLen	Input	Prefetch length. For the Atlas A2 training product/Atlas A2 inference product, the unit of preFetchLen is 2 KB. The value must be smaller than the iCache size/2 KB. The iCache sizes of the AIC and AIV are 32 KB and 16 KB, respectively. For the Atlas A3 training product/Atlas A3 inference product, the unit of preFetchLen is 2 KB. The value must be smaller than the iCache size/2 KB. The iCache sizes of the AIC and AIV are 32 KB and 16 KB, respectively. For the Atlas inference product AI Core, this parameter is invalid. The prefetch length is 128 bytes. For the Atlas 350 Accelerator Card, the unit of preFetchLen is 2 KB. The value must be smaller than the iCache size/2 KB. The iCache sizes of the AIC and AIV are 32 KB and 16 KB, respectively.

None

None

int64_t preFetchLen = 2; // Prefetch length
AscendC::ICachePreLoad(preFetchLen);

Parent topic: Cache Control