Searched refs:MS_VMLAL_F16 (Results 1 – 2 of 2) sorted by relevance
| /third_party/mindspore/mindspore/ccsrc/backend/kernel_compiler/cpu/nnacl/fp16_grad/ |
| D | convolution_grad_filter.c | 69 sum_0 = MS_VMLAL_F16(vget_low_f16(x_0), vget_low_f16(dy_0), sum_0); in FilterGrad32Arm()
70 sum_1 = MS_VMLAL_F16(vget_high_f16(x_0), vget_high_f16(dy_0), sum_1); in FilterGrad32Arm()
74 sum_2 = MS_VMLAL_F16(vget_low_f16(x_1), vget_low_f16(dy_1), sum_2); in FilterGrad32Arm()
75 sum_3 = MS_VMLAL_F16(vget_high_f16(x_1), vget_high_f16(dy_1), sum_3); in FilterGrad32Arm()
79 sum_4 = MS_VMLAL_F16(vget_low_f16(x_2), vget_low_f16(dy_2), sum_4); in FilterGrad32Arm()
80 sum_5 = MS_VMLAL_F16(vget_high_f16(x_2), vget_high_f16(dy_2), sum_5); in FilterGrad32Arm()
84 sum_6 = MS_VMLAL_F16(vget_low_f16(x_3), vget_low_f16(dy_3), sum_6); in FilterGrad32Arm()
85 sum_7 = MS_VMLAL_F16(vget_high_f16(x_3), vget_high_f16(dy_3), sum_7); in FilterGrad32Arm()
142 sum_0 = MS_VMLAL_F16(vget_low_f16(x_0), vget_low_f16(dy_0), sum_0); in FilterGrad16Arm()
143 sum_1 = MS_VMLAL_F16(vget_high_f16(x_0), vget_high_f16(dy_0), sum_1); in FilterGrad16Arm()
[all …]
|
| /third_party/mindspore/mindspore/ccsrc/backend/kernel_compiler/cpu/nnacl/intrinsics/ |
| D | ms_simd_instructions_fp16.h | 152 static inline float32x4_t MS_VMLAL_F16(float16x4_t x, float16x4_t dy, float32x4_t sum) { in MS_VMLAL_F16() function
|