Moving some VRP bits from DOM into tree-vrp

Roger Sayle roger@eyesopen.com
Tue Jun 14 14:18:00 GMT 2005


On Tue, 14 Jun 2005, Roger Sayle wrote:
> Agreed.  Whilst I can see there may be some value in a function like
> local_fold in the tree-ssa optimizers, it certainly isn't merited for
> this particular change.  Perhaps Jeff has accidentally introduced it
> prior to a legitimate use in a follow-up patch?

Digging a bit deeper, it looks like Jeff is just introducing a duplicate
copy of local_fold in tree-vrp.c based upon the identical function in
tree-ssa-dom.c.  I suspect that Jeff just didn't notice that the use
of local_fold wasn't appropriate in the original code, and simply moved
it to its new location blindly.

Roger
--



More information about the Gcc-patches mailing list