使用 ARM NEON 内联函数将 float32 强制转换为 int16。

huangapple go评论108阅读模式
英文:

typecast float32 to int16 using arm neon intrinsics

问题

我是一个初学者,对于 ARM NEON Intrinsics 不太熟悉,我想要将一个 float32 数组按照一个标量(2^13 = 8192)进行缩放,并将其转换为 int16_t 数组。
我认为我需要执行以下步骤:

  1. 加载 float 缓冲数组
  2. 与标量相乘(2^13 = 8192)
  3. 将它们转换为 32 位整数
  4. 将 32 位转换为 16 位整数
  5. 存储到 16 位缓冲区中

请帮我检查并纠正以下代码:

    // 将 float 转换为 int16
    uint32_t blk_cnt;
    float32x4_t f32x4;
    int32x4_t i32x4;
    int16x4_t i16x4;
    float32_t scale = 8192.0;
    /* 一次计算 4 个复杂样本 */
    blk_cnt = sz >> 2U;
    while (blk_cnt > 0U) {
        f32x4 = vld1q_f32 ((float32_t *) inpout);
        f32x4 = vmulq_n_f32(f32x4, scale);
        i32x4 = vcvtq_s32_f32 (f32x4);
        i16x4 = vmovn_s32 (i32x4);
        vst1_s16 (out, i16x4);      
        /* 增加指针 */
        out += 4;
        inpout += 4;
        /* 减少循环计数器 */
        blk_cnt--;
    }

请注意,此代码中的注释已被翻译为中文。

英文:

I'm a newbie to arm neon intrinsics and I would like to scale the float32 array with a scalar (2^13 = 8192) and typecast to int16_t array.
I believe I need to perform the below steps:

  1. Load the float buffer array
  2. Multiply with the scalar (2^13 = 8192)
  3. Convert them to 32-bit integers
  4. Convert 32-bit to 16-bit integers
  5. Store them into 16-bit buffer

Could you please check and correct the below code:

    // convert float to int16
    uint32_t blk_cnt;
    float32x4_t f32x4;
    int32x4_t i32x4;
    int16x4_t i16x4;
    float32_t scale = 8192.0;
    /* Compute 4 complex samples at a time */
    blk_cnt = sz >> 2U;
    while (blk_cnt > 0U) {
        f32x4 = vld1q_f32 ((float32_t *) inpout);
        f32x4 = vmulq_n_f32(f32x4, scale);
        i32x4 = vcvtq_s32_f32 (f32x4);
        i16x4 = vmovn_s32 (i32x4);
        vst1_s16 (out, i16x4);      
        /* Increment pointers */
        out += 4;
        inpout += 4;
        /* Decrement the loop counter */
        blk_cnt--;
    }

答案1

得分: 1

你很可能正在处理q13(13个小数位)的定点值。
只需将浮点数转换为q13 int32(vcvt_n_s32_f32),然后通过vqmovn缩小为int16。

链接:https://developer.arm.com/documentation/dui0473/m/vfp-instructions/vcvt--between-floating-point-and-fixed-point-

英文:

You are most probably dealing with q13 (13 fraction bits) fixed point values.
You just need to convert the floats to q13 int32 (vcvt_n_s32_f32), then shrink them to int16 by vqmovn.

https://developer.arm.com/documentation/dui0473/m/vfp-instructions/vcvt--between-floating-point-and-fixed-point-

huangapple
  • 本文由 发表于 2023年8月10日 14:17:30
  • 转载请务必保留本文链接:https://go.coder-hub.com/76873053.html
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定