The speed of the compiler, was: Re: Combine four insns

[Andi Kleen]

On Tue, Aug 10, 2010 at 01:51:46PM -0600, Tom Tromey wrote:
> >>>>> "Andi" == Andi Kleen <andi@firstfloor.org> writes:
> 
> Richard> I'm sure there is a way to open-code this using integer math.
> 
> Andi> I don't think so. Take a look at what PCMPESTRI does.  There's no easy 
> Andi> replacement, even if you use all the Hacker's Delight tricks 
> Andi> (it's really a cool instruction, but also very complicated :-)
> 
> This is still pending:
> 
> http://gcc.gnu.org/ml/gcc-patches/2010-03/msg00526.html

I see. I suspect my version is faster though :-) (if you have a suitable
CPU)

Also I think David got the "buffer near end of page before hole" case wrong
(although I'm not fully sure that really happens in gcc) 

> 
> I think any sort of hackery in this area is fine, if it speeds up the
> compiler.  All that is needed is someone with the time and motivation to
> do the testing.

My version passes testing, except for a few PCH test cases. I actually
ran it in parallel with the original cleanline and it gave the same
results. Any ideas why it could break PCH?

-Andi
-- 
ak@linux.intel.com -- Speaking for myself only.