Re: 2.6.13/14 x86 Makefile - Pentiums penalized ?

[Denis Vlasenko]

> > > It's documented as being suboptimal to use inc/dec due to it modifying all 
> > > of eflags resulting in dependency related stalls. add/sub only modifies 
> > > one bit of eflags so is more optimal. However there is a problem of 
> > 
> > ?! add/sub doesn't modify "only one bit in eflags", it modifies all.
> > In fact, it's dec/inc which does not modify all bits.
> > It doesn't touch 'carry' bit (IIRC).
> > 
> > If inc/dec is slower on P4, it must be just another P4 quirk.
> 
> You're right about the add and the number of modified bits. The documented 
> part is found in the P4 optimisation manual;
> 
> "The inc and dec instructions should always be avoided. Using add
>  and sub instructions instead avoids data dependence and improves
>  performance."

I read this as "get saner processor".

But frankly any CPU optimization manuals, not just Intel's,
go to semi-absurd suggestions like "align all data
and branch targets to $BIGNUM bytes", "do not use instruction X
(because we failed to make it work reasonably fast), use Y".

Of course next gen CPU will prefer Y over X
(for example, AMD's recomended way to do NOPs).
--
vda
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/