Re: [PATCH v2] Use long for mantissa for generic mp code

[Richard Henderson <rth at twiddle dot net>]

On 03/08/2013 04:38 AM, Siddhesh Poyarekar wrote:
> +typedef long mantissa_t;
> +typedef int64_t mantissa_store_t;
> +
> +#define TWOPOW(i) (1L << i)
> +
> +#define  RADIX     TWOPOW (24)		/* 2^24    */
> +
> +/* Divide D by RADIX and put the remainder in R.  D must be a non-negative
> +   integral value.  */
> +#define DIV_RADIX(d, r) \
> +  ({									      \
> +    r = ((uint64_t) d) % RADIX;							      \
> +    d /= RADIX;								      \
> +  })

One unsigned divide and one signed divide is even worse than two signed
divides, surely.  Perhaps it would be easier to just write

    r = d & (RADIX - 1);
    d >>= LG2_RADIX;

or whatever?  Or just give up on being able to avoid the signed divide?

> +
> +/* Put the integer component of a double X in R and retain the fraction in
> +   X.  */
> +#define INTEGER_OF(x, i) \
> +  ({									      \
> +    i = (mantissa_t) x;							      \
> +    x -= i;								      \
> +  })

We need a better comment here, since we're no longer talking about doubles.
Otherwise the patch looks good to me.


r~