|Any references for efficient division code sequences? email@example.com (Shankar Unni) (2001-06-28)|
|Re: Any references for efficient division code sequences? firstname.lastname@example.org (2001-07-02)|
|Re: Any references for efficient division code sequences? email@example.com (Raja Venkateswaran) (2001-07-02)|
|Re: Any references for efficient division code sequences? David.Chase@naturalbridge.com (David Chase) (2001-07-03)|
|Re: Any references for efficient division code sequences? firstname.lastname@example.org (Suresh Kadiyala) (2001-07-17)|
|From:||Suresh Kadiyala <email@example.com>|
|Date:||17 Jul 2001 23:30:29 -0400|
|Organization:||Excite@Home - The Leader in Broadband http://home.com/faster|
|Posted-Date:||17 Jul 2001 23:30:29 EDT|
A while back I worked on exactly this problem and found a few references
that addressed multiplication by constants and related problems. I need to
dig through a few piles of papers to get to the references. I remember
ending up writing a program to generate efficient shift-add sequences for
implementing multiplication. I believe similar techniques could be used
for division and remainders as well.
Shankar Unni wrote:
> Specifically for dividing by small constants (< 16). We're working on a
> small MIPS-like network processor that has no MUL/DIV unit, and need to
> be able to do a *fast* mod by small numbers.
> Are there any references that discuss code sequences for dividing (or
> specifically in our case, remaindering) integers by small constants (for
> any processor that doesn't have a MUL/DIV unit)?
Return to the
Search the comp.compilers archives again.