Re: [PATCH 6/6] target/tricore: Add shuffle insn

[Richard Henderson]

On 6/10/23 03:55, Bastian Koppelmann wrote:
> +/*
> + * table from
> + * https://graphics.stanford.edu/~seander/bithacks.html#BitReverseTable
> + */
> +static const unsigned char BitReverseTable256[256] = {
> +#   define R2(n) n, n + 2 * 64, n + 1 * 64, n + 3 * 64
> +#   define R4(n) R2(n), R2(n + 2 * 16), R2(n + 1 * 16), R2(n + 3 * 16)
> +#   define R6(n) R4(n), R4(n + 2 * 4 ), R4(n + 1 * 4 ), R4(n + 3 * 4 )
> +    R6(0), R6(2), R6(1), R6(3)
> +};

This is revbit8() from qemu/host-utils.h.

> +uint32_t helper_shuffle(uint32_t arg0, uint32_t arg1)
> +{
> +    uint8_t buf[4];
> +    uint8_t resbuf[4];
> +    uint32_t byte_select;
> +    uint32_t res = 0;
> +
> +    stl_le_p(buf, arg0);

While storing to a buffer works, it's just as easy to use shifts.

> +    byte_select = arg1 & 0x3;
> +    resbuf[0] = buf[byte_select];

  resb = extract32(arg0, byte_select * 8, 8);
  res |= resb << 0;

> +    resbuf[1] = buf[byte_select];

  res |= resb << 8;

etc.

> +    if (arg1 & 0x100) {
> +        resbuf[3] = BitReverseTable256[resbuf[3]];
> +    }

The bit-reversal is controlled by one bit for all bytes.  It can be done for all bytes in 
parallel.  Use the shifts from bitrev8, applied to the entire uint32_t result.


r~