This is the mail archive of the
mailing list for the GCC project.
Re: [1/6] Handle gphis in gimple_get_lhs
- From: Jeff Law <law at redhat dot com>
- To: gcc-patches at gcc dot gnu dot org, richard dot sandiford at arm dot com
- Date: Tue, 28 Aug 2018 12:22:29 -0600
- Subject: Re: [1/6] Handle gphis in gimple_get_lhs
- References: <firstname.lastname@example.org> <email@example.com>
On 08/28/2018 05:20 AM, Richard Sandiford wrote:
> Several callers of gimple_get_lhs deal with statements that might
> be phis. This patch makes gimple_get_lhs handle that case too,
> so that the callers don't have to.
> 2018-08-28 Richard Sandiford <firstname.lastname@example.org>
> * gimple.c (gimple_get_lhs): Handle gphis.
> * tree-ssa-phionlycprop.c (get_lhs_or_phi_result): Delete and...
> (propagate_rhs_into_lhs, eliminate_const_or_copy): ...use
> gimple_get_lhs instead.
> * tree-ssa-dom.c (eliminate_redundant_computations): Don't handle
> phis specially before calling gimple_get_lhs.
> * tree-ssa-scopedtables.c (avail_exprs_stack::lookup_avail_expr):
> * tree-vect-loop.c (vectorizable_live_operation): Likewise.
> * tree-vect-slp.c (vect_get_slp_vect_defs): Likewise.
> (vect_get_slp_defs): Likewise.
> * tree-vect-stmts.c (vect_get_vec_def_for_operand_1): Likewise.
> (vect_get_vec_def_for_stmt_copy, vect_transform_stmt): Likewise.
I pondered doing this repeatedly through the years. So I'm all for it.
Changes look fine to me.