This is the mail archive of the
mailing list for the GCC project.
Re: [patch] Externalize data-ref analysis - part 2
- From: Daniel Berlin <dberlin at dberlin dot org>
- To: Ira Rosen <IRAR at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Sebastian Pop <sebastian dot pop at cri dot ensmp dot fr>, Diego Novillo <dnovillo at redhat dot com>
- Date: Wed, 06 Jul 2005 11:35:31 -0400
- Subject: Re: [patch] Externalize data-ref analysis - part 2
- References: <OF10AA5E2D.1631164F-ONC225702E.0031E59E-C225702E.0040FDF2@il.ibm.com>
On Tue, 2005-06-28 at 14:49 +0300, Ira Rosen wrote:
> This is the second part - of generalization/externalization data-ref
> analysis from the vectorizer.
> In data-refs bases comparison we switch from calling get_alias_set and
> alias_set_conflict to check type memory tags of symbols directly as
> suggested by Diego.
> Bootstrapped and tested on ppc-darwin.
> O.K. for mainline?
Why do you keep passing around both the base and the data reference
object, when the base is in the data reference object.
I would have expected that you could simply pass the data reference
object and do
tree base_a = DR_OBJECT_BASE (dra)
tree base_b = DR_OBJECT_BASE (drb)
at the top of the functions where you need it?