----- Original Message -----
> From: "Chandler Carruth" <[email protected]>
> To: "Hal Finkel" <[email protected]>
> Cc: "Nadav Rotem" <[email protected]>, "Alexey Bataev" 
> <[email protected]>, "llvm cfe" <[email protected]>
> Sent: Tuesday, April 22, 2014 2:40:49 PM
> Subject: Re: [PATCH] #pragma vectorize
> 
> 
> 
> FWIW, only joining in the naming side of the discussion ...
> 
> 
> On Tue, Apr 22, 2014 at 11:39 AM, Hal Finkel < [email protected] >
> wrote:
> 
> 
> 
> > I also like ‘interleave’.
> > 
> 
> I like this as well.
> I very much dislike the term 'interleave'. We had a great deal of
> trouble with this term in the C++ committee. There are execution
> models which want this information but do not guarantee
> "interleaved" execution, and this is observable.

In this particular case, I think this objection is misplaced. The particular 
transformation that we're discussing is, literally, one that provides 
interleaving of loop iterations. We could also call it unsequenced (as I 
mentioned in some earlier e-mail), but in some sense, this transformation is 
more specific than that.

> 
> 
> Interleaved also seems to be promising much more than it can if this
> pragma does not assert safety.

Agreed; but we should keep the interleave (or whatever we call it) and the 
vectorize pragma consistent in this regard.

> 
> 
> If this is just a cost model hint, I like "widen" quite a bit better,
> and maybe there is a way to work "hint" or "cost" into the name?

In some sense it is a cost model hint, but I'm not sure a user would see it as 
such.

Using hint for the non-safety-asserting variants seems like a good idea. We 
should be clear with the users whether they are providing only a hint, or 
asserting something more.

 -Hal

-- 
Hal Finkel
Assistant Computational Scientist
Leadership Computing Facility
Argonne National Laboratory

_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to