Richard Biener <rguent...@suse.de> writes:
> The final loop IV use after the loop has that not in LC SSA
> (and inserts not simplified _2 = _3 - 0 stmts). In particular
> since it splits the exit edge when there's a virtual PHI in the
> destination it breaks virtual LC SSA form (but likely also
> non-virtual).
>
> The following properly inserts LC PHIs instead.
>
> Bootstrap & regtest pending on x86_64-unknown-linux-gnu.
>
> OK if that succeeds?
LGTM, thanks.
Richard
>
> Thanks,
> Richard.
>
> 2022-07-04 Richard Biener <rguent...@suse.de>
>
> * tree-vect-loop-manip.cc (vect_set_loop_condition_normal):
> Maintain LC SSA.
> ---
> gcc/tree-vect-loop-manip.cc | 19 ++++++++++++++++---
> 1 file changed, 16 insertions(+), 3 deletions(-)
>
> diff --git a/gcc/tree-vect-loop-manip.cc b/gcc/tree-vect-loop-manip.cc
> index 895957472c9..eba71c50428 100644
> --- a/gcc/tree-vect-loop-manip.cc
> +++ b/gcc/tree-vect-loop-manip.cc
> @@ -920,9 +920,22 @@ vect_set_loop_condition_normal (class loop *loop, tree
> niters, tree step,
>
> if (final_iv)
> {
> - gassign *assign = gimple_build_assign (final_iv, MINUS_EXPR,
> - indx_after_incr, init);
> - gsi_insert_on_edge_immediate (single_exit (loop), assign);
> + gassign *assign;
> + edge exit = single_exit (loop);
> + gcc_assert (single_pred_p (exit->dest));
> + tree phi_dest
> + = integer_zerop (init) ? final_iv : copy_ssa_name (indx_after_incr);
> + /* Make sure to maintain LC SSA form here and elide the subtraction
> + if the value is zero. */
> + gphi *phi = create_phi_node (phi_dest, exit->dest);
> + add_phi_arg (phi, indx_after_incr, exit, UNKNOWN_LOCATION);
> + if (!integer_zerop (init))
> + {
> + assign = gimple_build_assign (final_iv, MINUS_EXPR,
> + phi_dest, init);
> + gimple_stmt_iterator gsi = gsi_after_labels (exit->dest);
> + gsi_insert_before (&gsi, assign, GSI_SAME_STMT);
> + }
> }
>
> return cond_stmt;
