vdupq_n_bf16ADD TO COMPARE ADDED TO COMPARE
Arm 64-bit (64 bits)/ NEON
View official documentation
Purpose:
Duplicates a scalar into every element of the destination vector.
Result:
bfloat16x8_t
Example:
#include <stdio.h>
#include <arm_neon.h>
int main() {
bfloat16_t value = 42.0f;
bfloat16x8_t result = vdupq_n_bf16(value);
bfloat16_t output[8];
vst1q_bf16(output, result);
printf("Input value: %.1f\n", (float)value);
printf("Result vector: {
");
for(int i = 0; i < 7; i++) {
printf("%.1f, ", (float)output[i]);
}
printf("%.1f
}
\n", (float)output[7]);
return 0;
}Prototypes
Assembly Instruction:
DUP
Usage:
bfloat16x8_t result =
vdupq_n_bf16(
bfloat16_t a
)
SIMD Intrinsics Summary
| SIMD Engines: | 6 |
| C Intrinsics: | 10444 |
| NEON: | 4353 |
| AVX2: | 405 |
| AVX512: | 4717 |
| SSE4.2: | 598 |
| VSX: | 192 |
| IBM-Z: | 179 |