Re: kernel + gcc 4.1 = several problems

From: Denis Vlasenko <vda.linux@googlemail.com>
To: Linus Torvalds <torvalds@osdl.org>
Cc: Grzegorz Kulewski <kangur@polcom.net>,
	Alan <alan@lxorguk.ukuu.org.uk>,
	Mikael Pettersson <mikpe@it.uu.se>,
	s0348365@sms.ed.ac.uk, 76306.1226@compuserve.com, akpm@osdl.org,
	bunk@stusta.de, greg@kroah.com, linux-kernel@vger.kernel.org,
	yanmin_zhang@linux.intel.com
Subject: Re: kernel + gcc 4.1 = several problems
Date: Wed, 3 Jan 2007 20:47:02 +0100	[thread overview]
Message-ID: <200701032047.02941.vda.linux@googlemail.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0701030731080.4473@woody.osdl.org>

On Wednesday 03 January 2007 17:03, Linus Torvalds wrote:
> On Wed, 3 Jan 2007, Grzegorz Kulewski wrote:
> > Could you explain why CMOV is pointless now? Are there any benchmarks proving
> > that?
> 
> CMOV (and, more generically, any "predicated instruction") tends to 
> generally a bad idea on an aggressively out-of-order CPU. It doesn't 
> always have to be horrible, but in practice it is seldom very nice, and 
> (as usual) on the P4 it can be really quite bad.
> 
> On a P4, I think a cmov basically takes 10 cycles.
> 
> But even ignoring the usual P4 "I suck at things that aren't totally 
> normal", cmov is actually not a great idea. You can always replace it by
> 
> 		j<negated condition> forward
> 		mov ..., %reg
> 	forward:
...
...
> In contrast, if you use a predicated instruction, ALL of it is on the 
> critical path. Calculating the conditional is on the critical path. 
> Calculating the value that gets used is obviously ALSO on the critical 
> path, but so is the calculation for the value that DOESN'T get used too. 
> So the cmov - rather than speeding things up - actually slows things down, 
> because it makes more code be dependent on each other.

Why CPU people do not internally convert cmov into jmp,mov pair?
--
vda