On Tue, Jun 11, 2019 at 09:22:15PM -0700, Richard Henderson wrote:
> On 6/11/19 6:52 PM, David Gibson wrote:
> > On Sun, Jun 02, 2019 at 12:08:52PM +0100, Mark Cave-Ayland wrote:
> >> Rather than perform the VSR register decoding within the helper itself,
> >> introduce a new VSX_CMP macro which performs the decode based upon xT, xA
> >> and xB at translation time.
> >>
> >> Subsequent commits will make the same changes for other instructions however
> >> the xvcmp* instructions are different in that they return a set of flags to be
> >> optionally written back to the crf[6] register. Move this logic from the
> >> helper function to the generator function, along with the
> >> float_status update.
> > 
> > What's the advantage of this.  Since we still have a helper, don't we
> > suffer the cost of the helper call *plus* the now-generated
> > instructions?
> 
> Not as such.
> 
> The generated instructions are all of the form
> 
> 	lea	offset+x*16(env), reg
> 
> which is cheaper than
> 
> 	mov	$x, reg
> 
> and then
> 
> 	shl	$4, reg
> 	lea	offset(env, reg), reg
> 
> within the helper.

Fair enough.  Richard, I was kind of hoping for an A-b or R-b on the
series from you, but didn't see one..?

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson