[Bug middle-end/20283] optimising muldiv() type operations

[ajrobb at bigfoot dot com]

------- Additional Comments From ajrobb at bigfoot dot com  2005-03-02 16:53 -------
Subject: Re:  optimising muldiv() type operations

Hi,
Thanks for getting back. Your code below still performs a separate 
multiply by 47 before 'dividing' by 40.  My enhancement request is to 
combine these two operations making the 2nd-4th (and 8th) lines below 
redundant.

Well done though for spotting it should have been unsigned (signed 
division is horrible as it is not defined/portable).

pinskia at gcc dot gnu dot org wrote:

>------- Additional Comments From pinskia at gcc dot gnu dot org  2005-03-02 16:35 -------
>With unsigned instead of signed, I get the following code with the mainline:
>        movl    4(%esp), %eax
>        leal    (%eax,%eax,2), %edx
>        sall    $4, %edx
>        subl    %eax, %edx
>        movl    $-858993459, %eax
>        mull    %edx
>        shrl    $5, %edx
>        movl    %edx, %eax
>        ret
>
>Is this okay, If so then there is no bug here really.
>
>  
>



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20283