Re: [PATCH 2/3] netfilter : 3 patches to boost ip_tables performance

[Eric Dumazet <dada1@cosmosbay.com>]

Harald Welte a écrit :
> On Wed, Sep 21, 2005 at 11:32:24PM +0200, Eric Dumazet wrote:
>
> > Patch 2/3 (please apply after Patch 1/3)
> >
> > 2) Loop unrolling
>
>
> First of all, thanks for your performance analysis and patches.  I'm
> very inclined of merging them.
>
>
> > It seems that with current compilers and CFLAGS, the code from
> > ip_packet_match() is very bad, using lot of mispredicted conditional branches I made some patches 
> > and generated code on i386 and x86_64
> > is much better.
>
>
> This only describes your "compare_if_lstrings()" changes, and I'm happy
> to merge them.
>
> However, you also removed the use of the FWINV macro and replaced it by
> explicit code (including the bool1/bool2 variables, which are not really
> named intuitively).  Why was this neccessary?

It was necessary to get the best code with gcc-3.4.4 on i386 and gcc-4.0.1 on 
x86_64

For example :

bool1 = FWINV(ret != 0, IPT_INV_VIA_OUT);
if (bool1) {

gives a better code than :

if (FWINV(ret != 0, IPT_INV_VIA_OUT)) {

(one less conditional branch)

Dont ask me why, it is shocking but true :(

Eric

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/