Re: [ft-devel] FT_MulFix assembly

[Werner LEMBERG]

> Werner:  Miles' version is shorter, is only wrong by one ulp and only
>          when the product overflows and is negative.  My variation,
>          called another() above, fixes that slight difference.
>          Which would you prefer, if anything?

I tend to prefer the faster one $(Q#|(B IIRC, we never rely on product
overflow of 32bit entities.

>          I can write a patch which uses it only when it works (64
>          bit sizeof(long)).

Please do so.


    Werner