[PATCH 2/3] ARM cost model: Address computation in vector mem ops not free

[Renato Golin]

On 7 February 2013 05:55, Nadav Rotem <nrotem at apple.com> wrote:

> I am not sure that its worth modeling this because it only affects the
> latency and not the throughput of the machine.
>

It seems like penalizing the insertion into a D-subregister is getting out
of hand. This is likely to occur when dealing with relative numbers (like
instruction costs), so I'm not particularly excited about this patch,
either. I'm also not unhappy, so if there is a tangible benefit, by all
means...

Some specific questions:

+      if (Stride > 0)
+        return Cost;

Can the stride ever be zero?

-    return 2;
+    return 3;

This seems arbitrary (I know it's not but), would be good to have them
exposed as function calls, for instance, getAddressComputationCost() +
whatever.

cheers,
--renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130207/a029de5e/attachment.html>