|Green Compiler ? firstname.lastname@example.org (Abid) (2012-12-20)|
|Re: Green Compiler ? email@example.com (glen herrmannsfeldt) (2012-12-23)|
|Re: Green Compiler ? firstname.lastname@example.org (Nils M Holm) (2012-12-23)|
|Re: Green Compiler ? DrDiettrich1@aol.com (Hans-Peter Diettrich) (2012-12-24)|
|Re: Green Compiler ? email@example.com (Peter Dassow) (2012-12-26)|
|Re: Green Compiler ? firstname.lastname@example.org (2012-12-27)|
|Re: Green Compiler ? email@example.com (glen herrmannsfeldt) (2012-12-28)|
|Re: Green Compiler ? Pidgeot18@verizon.invalid (Joshua Cranmer) (2012-12-27)|
|Re: Green Compiler ? firstname.lastname@example.org (Nils M Holm) (2012-12-28)|
|Re: Green Compiler ? DrDiettrich1@aol.com (Hans-Peter Diettrich) (2012-12-28)|
|[14 later articles]|
|From:||Hans-Peter Diettrich <DrDiettrich1@aol.com>|
|Date:||Mon, 24 Dec 2012 05:16:56 +0100|
|Posted-Date:||24 Dec 2012 00:14:11 EST|
Nils M Holm schrieb:
> Abid <email@example.com> wrote:
>> Do we need to change this model and make it
>> three dimensional by adding power axis in the search space?
> In principle, I would say that higher execution speed equals more
> grenn-ness. Less time spent dissipating heat means less energy
Not really. You mean more efficient code, I suppose?
> So the question actually is the same as ever: how to we make code
> run fast?
> - How do we squeeze more meaning into fewer/faster instructions?
Okay, but the available machine instructions are limited. In detail on
> - How do we make code so small that it fits in caches?
It's not only the code that has to fit into the caches. Instruction
reordering and branch prediction are one thing, but the placement of the
data (variables) is quite another story.
> - How do we organize code execution in such a way that cache
> stalls are minimized? (Locality)
Such optimization requires compiler hints, so that the compiler can not
only arrange the code, but also the data. This requires knowledge about
the *frequency* of subroutine use, branches taken, and variables used in
these critical pathes.
But who should provide such hints? IMO only a profiling tool can provide
that information, when a program is run with typical (real life) data.
Return to the
Search the comp.compilers archives again.