This is the mail archive of the
mailing list for the GCC project.
Re: Handling aliasing memory accesses in gimple/ssa
- From: Kyrill Tkachov <kyrylo dot tkachov at foss dot arm dot com>
- To: Richard Biener <richard dot guenther at gmail dot com>
- Cc: "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>
- Date: Tue, 10 May 2016 14:51:49 +0100
- Subject: Re: Handling aliasing memory accesses in gimple/ssa
- Authentication-results: sourceware.org; auth=none
- References: <5731E517 dot 40104 at foss dot arm dot com> <CAFiYyc1xUNiAXOZHMQQSVrzC7diRnrirDw9EQXGwSgfJ3GvnDw at mail dot gmail dot com>
On 10/05/16 14:46, Richard Biener wrote:
On Tue, May 10, 2016 at 3:41 PM, Kyrill Tkachov
I'm taking a stab at fixing PR 22141 by merging adjacent stores into wider
stores in a late gimple pass.
My current plan is to go through all the assignments in a basic block and
keep track of
LHS expressions that are COMPONENT_REF, BIT_FIELD_REF, ARRAY_REF or
we encounter an expression that affects the recorded stores or a mergeable
store that aliases with
any of the preceeding loads. My question is how to express that question in
Is taking get_alias_set of the two expressions that I want to compare and
checking alias_sets_conflict_p enough?
That's too conservative. You want to use stmt_may_clobber_ref_p /
and likely feed both with the "base" you want to perform the merging with.
I think that's exactly what I need.