The following fixes a mistake when applying the bias for peeling via
masking to the inital value of SLP inductions.
This resolves gcc.target/aarch64/sve/peel_ind_1.c (a scan-assembler
only unfortunately) when forcing single-lane SLP for it.
Bootstrap and regtest running on x86_64-unknown-linux-gnu.
PR tree-optimization/116610
* tree-vect-loop.cc (vectorizable_induction): Use MINUS_EXPR
to apply a mask peeling adjustment.
---
gcc/tree-vect-loop.cc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc
index 62cf9205059..c981ab656ae 100644
--- a/gcc/tree-vect-loop.cc
+++ b/gcc/tree-vect-loop.cc
@@ -10543,7 +10543,7 @@ vectorizable_induction (loop_vec_info loop_vinfo,
vec_steps.safe_push (vec_step);
tree step_mul = gimple_build_vector (&init_stmts, &mul_elts);
if (peel_mul)
- step_mul = gimple_build (&init_stmts, PLUS_EXPR, step_vectype,
+ step_mul = gimple_build (&init_stmts, MINUS_EXPR, step_vectype,
step_mul, peel_mul);
if (!init_node)
vec_init = gimple_build_vector (&init_stmts, &init_elts);
--
2.43.0
