diff options
| author | Richard Biener <rguenther@suse.de> | 2024-11-25 13:32:15 +0100 |
|---|---|---|
| committer | Richard Biener <rguenth@gcc.gnu.org> | 2024-11-25 14:24:30 +0100 |
| commit | cd8db107b9bef73fd822ffb420f96ed2bc622a19 (patch) | |
| tree | a951d973e88f067189e6c460b419d58ed15baf2f | |
| parent | d9c908b75039653f2b7717b4b7cdffdc4f0fcc7d (diff) | |
| download | gcc-cd8db107b9bef73fd822ffb420f96ed2bc622a19.zip gcc-cd8db107b9bef73fd822ffb420f96ed2bc622a19.tar.gz gcc-cd8db107b9bef73fd822ffb420f96ed2bc622a19.tar.bz2 | |
target/116760 - 416.gamess slowdown with SLP
For the TWOTFF loop vectorization the backend scales constructor
and vector extract cost to make higher VFs less profitable. This
heuristic currently fails to consider VMAT_STRIDED_SLP which we
now get with single-lane SLP, causing a huge regression in SPEC 2k6
416.gamess for the respective loop nest.
The following fixes this, matching behavior to that of GCC 14 by
treating single-lane VMAT_STRIDED_SLP the same as VMAT_ELEMENTWISE.
PR target/116760
* config/i386/i386.cc (ix86_vector_costs::add_stmt_cost):
Scale vec_construct for single-lane VMAT_STRIDED_SLP the
same as VMAT_ELEMENTWISE.
* tree-vect-stmts.cc (vectorizable_store): Pass SLP node
down to costing for vec_to_scalar for VMAT_STRIDED_SLP.
| -rw-r--r-- | gcc/config/i386/i386.cc | 8 | ||||
| -rw-r--r-- | gcc/tree-vect-stmts.cc | 3 |
2 files changed, 7 insertions, 4 deletions
diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc index 8ab9120..fda2112 100644 --- a/gcc/config/i386/i386.cc +++ b/gcc/config/i386/i386.cc @@ -25340,9 +25340,11 @@ ix86_vector_costs::add_stmt_cost (int count, vect_cost_for_stmt kind, || (STMT_VINFO_MEMORY_ACCESS_TYPE (stmt_info) == VMAT_GATHER_SCATTER))) || (node - && ((SLP_TREE_MEMORY_ACCESS_TYPE (node) == VMAT_ELEMENTWISE - && (TREE_CODE (DR_STEP (STMT_VINFO_DATA_REF - (SLP_TREE_REPRESENTATIVE (node)))) + && (((SLP_TREE_MEMORY_ACCESS_TYPE (node) == VMAT_ELEMENTWISE + || (SLP_TREE_MEMORY_ACCESS_TYPE (node) == VMAT_STRIDED_SLP + && SLP_TREE_LANES (node) == 1)) + && (TREE_CODE (DR_STEP (STMT_VINFO_DATA_REF + (SLP_TREE_REPRESENTATIVE (node)))) != INTEGER_CST)) || (SLP_TREE_MEMORY_ACCESS_TYPE (node) == VMAT_GATHER_SCATTER))))) diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index 82cd389..c2d5818 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -8832,7 +8832,8 @@ vectorizable_store (vec_info *vinfo, if (nstores > 1) inside_cost += record_stmt_cost (cost_vec, 1, vec_to_scalar, - stmt_info, 0, vect_body); + stmt_info, slp_node, + 0, vect_body); /* Take a single lane vector type store as scalar store to avoid ICE like 110776. */ if (VECTOR_TYPE_P (ltype) |