2.95, x86: severe performance problems with short arithmetic

[Richard Henderson]

On Wed, Aug 11, 1999 at 10:23:25AM -0700, Zack Weinberg wrote:
> For 2.96/the new_ia32_branch, I wonder if it would be possible to
> use the peephole2 framework you posted to widen register operations
> throughout a function.

Nope.  Such a pass needs a global view of the function.  
A peepholer, by nature, does not have that.

> I wonder if it wouldn't be profitable to do the extend at the same
> time as the fetch, like this:

Yes.


r~