From 4d76079fdfa3d36ebd95ac8b489519945e8ee88f Mon Sep 17 00:00:00 2001 From: Tamar Christina Date: Wed, 4 Nov 2020 22:33:11 +0000 Subject: [PATCH] middle-end: Move load/store-lanes check till late. This moves the code that checks for load/store lanes further in the pipeline and places it after slp_optimize. This would allow us to perform optimizations on the SLP tree and only bail out if we really have a permute. With this change it allows us to handle permutes such as {1,1,1,1} which should be handled by a load and replicate. This change however makes it all or nothing. Either all instances can be handled or none at all. This is why some of the test cases have been adjusted. gcc/ChangeLog: * tree-vect-slp.c (vect_analyze_slp_instance): Moved load/store lanes check to ... * tree-vect-loop.c (vect_analyze_loop_2): ..Here gcc/testsuite/ChangeLog: * gcc.dg/vect/slp-11b.c: Update output scan. * gcc.dg/vect/slp-perm-6.c: Likewise. --- gcc/testsuite/gcc.dg/vect/slp-11b.c | 2 +- gcc/testsuite/gcc.dg/vect/slp-perm-6.c | 8 ++-- gcc/tree-vect-loop.c | 72 ++++++++++++++++++++++++++++++++++ gcc/tree-vect-slp.c | 48 ----------------------- 4 files changed, 77 insertions(+), 53 deletions(-) diff --git a/gcc/testsuite/gcc.dg/vect/slp-11b.c b/gcc/testsuite/gcc.dg/vect/slp-11b.c index 0cc2377..3f16c9c 100644 --- a/gcc/testsuite/gcc.dg/vect/slp-11b.c +++ b/gcc/testsuite/gcc.dg/vect/slp-11b.c @@ -45,4 +45,4 @@ int main (void) /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { target { vect_strided4 && vect_int_mult } } } } */ /* { dg-final { scan-tree-dump-times "vectorized 0 loops" 1 "vect" { target { ! { vect_strided4 && vect_int_mult } } } } } */ -/* { dg-final { scan-tree-dump-times "vectorizing stmts using SLP" 0 "vect" } } */ +/* { dg-final { scan-tree-dump-times "vectorizing stmts using SLP" 1 "vect" } } */ diff --git a/gcc/testsuite/gcc.dg/vect/slp-perm-6.c b/gcc/testsuite/gcc.dg/vect/slp-perm-6.c index 3848929..816486a 100644 --- a/gcc/testsuite/gcc.dg/vect/slp-perm-6.c +++ b/gcc/testsuite/gcc.dg/vect/slp-perm-6.c @@ -106,7 +106,7 @@ int main (int argc, const char* argv[]) /* { dg-final { scan-tree-dump-times "vectorizing stmts using SLP" 2 "vect" { target { vect_perm3_int && { {! vect_load_lanes } && {! vect_partial_vectors_usage_1 } } } } } } */ /* The epilogues are vectorized using partial vectors. */ /* { dg-final { scan-tree-dump-times "vectorizing stmts using SLP" 4 "vect" { target { vect_perm3_int && { {! vect_load_lanes } && vect_partial_vectors_usage_1 } } } } } */ -/* { dg-final { scan-tree-dump-times "vectorizing stmts using SLP" 1 "vect" { target vect_load_lanes } } } */ -/* { dg-final { scan-tree-dump "Built SLP cancelled: can use load/store-lanes" "vect" { target { vect_perm3_int && vect_load_lanes } } } } */ -/* { dg-final { scan-tree-dump "LOAD_LANES" "vect" { target vect_load_lanes } } } */ -/* { dg-final { scan-tree-dump "STORE_LANES" "vect" { target vect_load_lanes } } } */ +/* { dg-final { scan-tree-dump-times "vectorizing stmts using SLP" 2 "vect" { target vect_load_lanes } } } */ +/* { dg-final { scan-tree-dump "Built SLP cancelled: can use load/store-lanes" "vect" { xfail { vect_perm3_int && vect_load_lanes } } } } */ +/* { dg-final { scan-tree-dump "LOAD_LANES" "vect" { xfail vect_load_lanes } } } */ +/* { dg-final { scan-tree-dump "STORE_LANES" "vect" { xfail vect_load_lanes } } } */ diff --git a/gcc/tree-vect-loop.c b/gcc/tree-vect-loop.c index 5e7188a..547894c 100644 --- a/gcc/tree-vect-loop.c +++ b/gcc/tree-vect-loop.c @@ -2365,6 +2365,78 @@ start_over: "unsupported SLP instances\n"); goto again; } + + /* Check whether any load in ALL SLP instances is possibly permuted. */ + slp_tree load_node, slp_root; + unsigned i, x; + slp_instance instance; + bool can_use_lanes = true; + FOR_EACH_VEC_ELT (LOOP_VINFO_SLP_INSTANCES (loop_vinfo), x, instance) + { + slp_root = SLP_INSTANCE_TREE (instance); + int group_size = SLP_TREE_LANES (slp_root); + tree vectype = SLP_TREE_VECTYPE (slp_root); + bool loads_permuted = false; + FOR_EACH_VEC_ELT (SLP_INSTANCE_LOADS (instance), i, load_node) + { + if (!SLP_TREE_LOAD_PERMUTATION (load_node).exists ()) + continue; + unsigned j; + stmt_vec_info load_info; + FOR_EACH_VEC_ELT (SLP_TREE_SCALAR_STMTS (load_node), j, load_info) + if (SLP_TREE_LOAD_PERMUTATION (load_node)[j] != j) + { + loads_permuted = true; + break; + } + } + + /* If the loads and stores can be handled with load/store-lane + instructions record it and move on to the next instance. */ + if (loads_permuted + && vect_store_lanes_supported (vectype, group_size, false)) + { + FOR_EACH_VEC_ELT (SLP_INSTANCE_LOADS (instance), i, load_node) + { + stmt_vec_info stmt_vinfo = DR_GROUP_FIRST_ELEMENT + (SLP_TREE_SCALAR_STMTS (load_node)[0]); + /* Use SLP for strided accesses (or if we can't + load-lanes). */ + if (STMT_VINFO_STRIDED_P (stmt_vinfo) + || ! vect_load_lanes_supported + (STMT_VINFO_VECTYPE (stmt_vinfo), + DR_GROUP_SIZE (stmt_vinfo), false)) + break; + } + + can_use_lanes + = can_use_lanes && i == SLP_INSTANCE_LOADS (instance).length (); + + if (can_use_lanes && dump_enabled_p ()) + dump_printf_loc (MSG_NOTE, vect_location, + "SLP instance %p can use load/store-lanes\n", + instance); + } + else + { + can_use_lanes = false; + break; + } + } + + /* If all SLP instances can use load/store-lanes abort SLP and try again + with SLP disabled. */ + if (can_use_lanes) + { + ok = opt_result::failure_at (vect_location, + "Built SLP cancelled: can use " + "load/store-lanes\n"); + if (dump_enabled_p ()) + dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, + "Built SLP cancelled: all SLP instances support " + "load/store-lanes\n"); + goto again; + } } /* Dissolve SLP-only groups. */ diff --git a/gcc/tree-vect-slp.c b/gcc/tree-vect-slp.c index 11fe685..d498cd4 100644 --- a/gcc/tree-vect-slp.c +++ b/gcc/tree-vect-slp.c @@ -2263,54 +2263,6 @@ vect_build_slp_instance (vec_info *vinfo, "SLP size %u vs. limit %u.\n", tree_size, max_tree_size); - /* Check whether any load is possibly permuted. */ - slp_tree load_node; - bool loads_permuted = false; - FOR_EACH_VEC_ELT (SLP_INSTANCE_LOADS (new_instance), i, load_node) - { - if (!SLP_TREE_LOAD_PERMUTATION (load_node).exists ()) - continue; - unsigned j; - stmt_vec_info load_info; - FOR_EACH_VEC_ELT (SLP_TREE_SCALAR_STMTS (load_node), j, load_info) - if (SLP_TREE_LOAD_PERMUTATION (load_node)[j] != j) - { - loads_permuted = true; - break; - } - } - - /* If the loads and stores can be handled with load/store-lane - instructions do not generate this SLP instance. */ - if (is_a (vinfo) - && loads_permuted - && kind == slp_inst_kind_store - && vect_store_lanes_supported - (STMT_VINFO_VECTYPE (scalar_stmts[0]), group_size, false)) - { - slp_tree load_node; - FOR_EACH_VEC_ELT (SLP_INSTANCE_LOADS (new_instance), i, load_node) - { - stmt_vec_info stmt_vinfo = DR_GROUP_FIRST_ELEMENT - (SLP_TREE_SCALAR_STMTS (load_node)[0]); - /* Use SLP for strided accesses (or if we can't load-lanes). */ - if (STMT_VINFO_STRIDED_P (stmt_vinfo) - || ! vect_load_lanes_supported - (STMT_VINFO_VECTYPE (stmt_vinfo), - DR_GROUP_SIZE (stmt_vinfo), false)) - break; - } - if (i == SLP_INSTANCE_LOADS (new_instance).length ()) - { - if (dump_enabled_p ()) - dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, - "Built SLP cancelled: can use " - "load/store-lanes\n"); - vect_free_slp_instance (new_instance); - return false; - } - } - /* Fixup SLP reduction chains. */ if (kind == slp_inst_kind_reduc_chain) { -- 2.7.4