vmla

功能说明

• 当src0、src1和dst类型相同时（均为half或float），以block（32Byte）为单位完成以下计算，一次完成8个block的计算。

[dst] = [src0] * [src1] + [dst]

• 当src0和src1为half类型，dst为f32类型时，src0和src1各取低4个block，合计64个half数。高4个block被忽略，相乘后与dst的64个f32（8个block）相加。

[dst(64个f32)] = [src0(低4个block的64个half)] * [src1(低4个block的64个half)] + [dst(64个f32)]

函数原型

void vmla(__ubuf__ half *dst, __ubuf__ half *src0, __ubuf__ half *src1, uint8_t repeat, uint8_t dstBlockStride, uint8_t src0BlockStride, uint8_t src1BlockStride, uint8_t dstRepeatStride, uint8_t src0RepeatStride, uint8_t src1RepeatStride); 
 
void vmla(__ubuf__ float *dst, __ubuf__ float *src0, __ubuf__ float *src1, uint8_t repeat, uint8_t dstBlockStride, uint8_t src0BlockStride, uint8_t src1BlockStride, uint8_t dstRepeatStride, uint8_t src0RepeatStride, uint8_t src1RepeatStride); 
 
void vmla(__ubuf__ float *dst, __ubuf__ half *src0, __ubuf__ half *src1, uint8_t repeat, uint8_t dstBlockStride, uint8_t src0BlockStride, uint8_t src1BlockStride, uint8_t dstRepeatStride, uint8_t src0RepeatStride, uint8_t src1RepeatStride);

流水类型

PIPE_V