Re: [PATCH] tcp_cubic: faster cube root

2007-03-07 Thread David Miller
From: Andi Kleen <[EMAIL PROTECTED]> Date: Wed, 7 Mar 2007 13:12:46 +0100 > Well that still needs the ugly div64_64 function. At least my goal was to > eliminate that, not make it faster (I don't see any evidence this function > is particularly performance critical). You prefer to keep div64_64?

Re: [PATCH] tcp_cubic: faster cube root

2007-03-07 Thread Andi Kleen
On Tue, Mar 06, 2007 at 08:20:52PM -0800, David Miller wrote: > From: Stephen Hemminger <[EMAIL PROTECTED]> > Date: Tue, 6 Mar 2007 14:47:06 -0800 > > > The Newton-Raphson method is quadratically convergent so > > only a small fixed number of steps are necessary. > > Therefore it is faster to unro

Re: [PATCH] tcp_cubic: faster cube root

2007-03-06 Thread David Miller
From: Stephen Hemminger <[EMAIL PROTECTED]> Date: Tue, 6 Mar 2007 14:47:06 -0800 > The Newton-Raphson method is quadratically convergent so > only a small fixed number of steps are necessary. > Therefore it is faster to unroll the loop. Since div64_64 is no longer > inline it won't cause code expl

[PATCH] tcp_cubic: faster cube root

2007-03-06 Thread Stephen Hemminger
The Newton-Raphson method is quadratically convergent so only a small fixed number of steps are necessary. Therefore it is faster to unroll the loop. Since div64_64 is no longer inline it won't cause code explosion. Also fixes a bug that can occur if x^2 was bigger than 32 bits. Signed-off-by: St