/* Expansion pass for OMP directives. Outlines regions of certain OMP directives to separate functions, converts others into explicit calls to the runtime library (libgomp) and so forth Copyright (C) 2005-2017 Free Software Foundation, Inc. This file is part of GCC. GCC is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 3, or (at your option) any later version. GCC is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with GCC; see the file COPYING3. If not see . */ #include "config.h" #include "system.h" #include "coretypes.h" #include "memmodel.h" #include "backend.h" #include "target.h" #include "rtl.h" #include "tree.h" #include "gimple.h" #include "cfghooks.h" #include "tree-pass.h" #include "ssa.h" #include "optabs.h" #include "cgraph.h" #include "pretty-print.h" #include "diagnostic-core.h" #include "fold-const.h" #include "stor-layout.h" #include "cfganal.h" #include "internal-fn.h" #include "gimplify.h" #include "gimple-iterator.h" #include "gimplify-me.h" #include "gimple-walk.h" #include "tree-cfg.h" #include "tree-into-ssa.h" #include "tree-ssa.h" #include "splay-tree.h" #include "cfgloop.h" #include "omp-general.h" #include "omp-offload.h" #include "tree-cfgcleanup.h" #include "symbol-summary.h" #include "cilk.h" #include "gomp-constants.h" #include "gimple-pretty-print.h" #include "hsa-common.h" #include "debug.h" /* OMP region information. Every parallel and workshare directive is enclosed between two markers, the OMP_* directive and a corresponding GIMPLE_OMP_RETURN statement. */ struct omp_region { /* The enclosing region. */ struct omp_region *outer; /* First child region. */ struct omp_region *inner; /* Next peer region. */ struct omp_region *next; /* Block containing the omp directive as its last stmt. */ basic_block entry; /* Block containing the GIMPLE_OMP_RETURN as its last stmt. */ basic_block exit; /* Block containing the GIMPLE_OMP_CONTINUE as its last stmt. */ basic_block cont; /* If this is a combined parallel+workshare region, this is a list of additional arguments needed by the combined parallel+workshare library call. */ vec *ws_args; /* The code for the omp directive of this region. */ enum gimple_code type; /* Schedule kind, only used for GIMPLE_OMP_FOR type regions. */ enum omp_clause_schedule_kind sched_kind; /* Schedule modifiers. */ unsigned char sched_modifiers; /* True if this is a combined parallel+workshare region. */ bool is_combined_parallel; /* The ordered stmt if type is GIMPLE_OMP_ORDERED and it has a depend clause. */ gomp_ordered *ord_stmt; }; static struct omp_region *root_omp_region; static bool omp_any_child_fn_dumped; static void expand_omp_build_assign (gimple_stmt_iterator *, tree, tree, bool = false); static gphi *find_phi_with_arg_on_edge (tree, edge); static void expand_omp (struct omp_region *region); /* Return true if REGION is a combined parallel+workshare region. */ static inline bool is_combined_parallel (struct omp_region *region) { return region->is_combined_parallel; } /* Given two blocks PAR_ENTRY_BB and WS_ENTRY_BB such that WS_ENTRY_BB is the immediate dominator of PAR_ENTRY_BB, return true if there are no data dependencies that would prevent expanding the parallel directive at PAR_ENTRY_BB as a combined parallel+workshare region. When expanding a combined parallel+workshare region, the call to the child function may need additional arguments in the case of GIMPLE_OMP_FOR regions. In some cases, these arguments are computed out of variables passed in from the parent to the child via 'struct .omp_data_s'. For instance: #pragma omp parallel for schedule (guided, i * 4) for (j ...) Is lowered into: # BLOCK 2 (PAR_ENTRY_BB) .omp_data_o.i = i; #pragma omp parallel [child fn: bar.omp_fn.0 ( ..., D.1598) # BLOCK 3 (WS_ENTRY_BB) .omp_data_i = &.omp_data_o; D.1667 = .omp_data_i->i; D.1598 = D.1667 * 4; #pragma omp for schedule (guided, D.1598) When we outline the parallel region, the call to the child function 'bar.omp_fn.0' will need the value D.1598 in its argument list, but that value is computed *after* the call site. So, in principle we cannot do the transformation. To see whether the code in WS_ENTRY_BB blocks the combined parallel+workshare call, we collect all the variables used in the GIMPLE_OMP_FOR header check whether they appear on the LHS of any statement in WS_ENTRY_BB. If so, then we cannot emit the combined call. FIXME. If we had the SSA form built at this point, we could merely hoist the code in block 3 into block 2 and be done with it. But at this point we don't have dataflow information and though we could hack something up here, it is really not worth the aggravation. */ static bool workshare_safe_to_combine_p (basic_block ws_entry_bb) { struct omp_for_data fd; gimple *ws_stmt = last_stmt (ws_entry_bb); if (gimple_code (ws_stmt) == GIMPLE_OMP_SECTIONS) return true; gcc_assert (gimple_code (ws_stmt) == GIMPLE_OMP_FOR); omp_extract_for_data (as_a (ws_stmt), &fd, NULL); if (fd.collapse > 1 && TREE_CODE (fd.loop.n2) != INTEGER_CST) return false; if (fd.iter_type != long_integer_type_node) return false; /* FIXME. We give up too easily here. If any of these arguments are not constants, they will likely involve variables that have been mapped into fields of .omp_data_s for sharing with the child function. With appropriate data flow, it would be possible to see through this. */ if (!is_gimple_min_invariant (fd.loop.n1) || !is_gimple_min_invariant (fd.loop.n2) || !is_gimple_min_invariant (fd.loop.step) || (fd.chunk_size && !is_gimple_min_invariant (fd.chunk_size))) return false; return true; } /* Adjust CHUNK_SIZE from SCHEDULE clause, depending on simd modifier presence (SIMD_SCHEDULE). */ static tree omp_adjust_chunk_size (tree chunk_size, bool simd_schedule) { if (!simd_schedule) return chunk_size; int vf = omp_max_vf (); if (vf == 1) return chunk_size; tree type = TREE_TYPE (chunk_size); chunk_size = fold_build2 (PLUS_EXPR, type, chunk_size, build_int_cst (type, vf - 1)); return fold_build2 (BIT_AND_EXPR, type, chunk_size, build_int_cst (type, -vf)); } /* Collect additional arguments needed to emit a combined parallel+workshare call. WS_STMT is the workshare directive being expanded. */ static vec * get_ws_args_for (gimple *par_stmt, gimple *ws_stmt) { tree t; location_t loc = gimple_location (ws_stmt); vec *ws_args; if (gomp_for *for_stmt = dyn_cast (ws_stmt)) { struct omp_for_data fd; tree n1, n2; omp_extract_for_data (for_stmt, &fd, NULL); n1 = fd.loop.n1; n2 = fd.loop.n2; if (gimple_omp_for_combined_into_p (for_stmt)) { tree innerc = omp_find_clause (gimple_omp_parallel_clauses (par_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n1 = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n2 = OMP_CLAUSE_DECL (innerc); } vec_alloc (ws_args, 3 + (fd.chunk_size != 0)); t = fold_convert_loc (loc, long_integer_type_node, n1); ws_args->quick_push (t); t = fold_convert_loc (loc, long_integer_type_node, n2); ws_args->quick_push (t); t = fold_convert_loc (loc, long_integer_type_node, fd.loop.step); ws_args->quick_push (t); if (fd.chunk_size) { t = fold_convert_loc (loc, long_integer_type_node, fd.chunk_size); t = omp_adjust_chunk_size (t, fd.simd_schedule); ws_args->quick_push (t); } return ws_args; } else if (gimple_code (ws_stmt) == GIMPLE_OMP_SECTIONS) { /* Number of sections is equal to the number of edges from the GIMPLE_OMP_SECTIONS_SWITCH statement, except for the one to the exit of the sections region. */ basic_block bb = single_succ (gimple_bb (ws_stmt)); t = build_int_cst (unsigned_type_node, EDGE_COUNT (bb->succs) - 1); vec_alloc (ws_args, 1); ws_args->quick_push (t); return ws_args; } gcc_unreachable (); } /* Discover whether REGION is a combined parallel+workshare region. */ static void determine_parallel_type (struct omp_region *region) { basic_block par_entry_bb, par_exit_bb; basic_block ws_entry_bb, ws_exit_bb; if (region == NULL || region->inner == NULL || region->exit == NULL || region->inner->exit == NULL || region->inner->cont == NULL) return; /* We only support parallel+for and parallel+sections. */ if (region->type != GIMPLE_OMP_PARALLEL || (region->inner->type != GIMPLE_OMP_FOR && region->inner->type != GIMPLE_OMP_SECTIONS)) return; /* Check for perfect nesting PAR_ENTRY_BB -> WS_ENTRY_BB and WS_EXIT_BB -> PAR_EXIT_BB. */ par_entry_bb = region->entry; par_exit_bb = region->exit; ws_entry_bb = region->inner->entry; ws_exit_bb = region->inner->exit; if (single_succ (par_entry_bb) == ws_entry_bb && single_succ (ws_exit_bb) == par_exit_bb && workshare_safe_to_combine_p (ws_entry_bb) && (gimple_omp_parallel_combined_p (last_stmt (par_entry_bb)) || (last_and_only_stmt (ws_entry_bb) && last_and_only_stmt (par_exit_bb)))) { gimple *par_stmt = last_stmt (par_entry_bb); gimple *ws_stmt = last_stmt (ws_entry_bb); if (region->inner->type == GIMPLE_OMP_FOR) { /* If this is a combined parallel loop, we need to determine whether or not to use the combined library calls. There are two cases where we do not apply the transformation: static loops and any kind of ordered loop. In the first case, we already open code the loop so there is no need to do anything else. In the latter case, the combined parallel loop call would still need extra synchronization to implement ordered semantics, so there would not be any gain in using the combined call. */ tree clauses = gimple_omp_for_clauses (ws_stmt); tree c = omp_find_clause (clauses, OMP_CLAUSE_SCHEDULE); if (c == NULL || ((OMP_CLAUSE_SCHEDULE_KIND (c) & OMP_CLAUSE_SCHEDULE_MASK) == OMP_CLAUSE_SCHEDULE_STATIC) || omp_find_clause (clauses, OMP_CLAUSE_ORDERED)) { region->is_combined_parallel = false; region->inner->is_combined_parallel = false; return; } } region->is_combined_parallel = true; region->inner->is_combined_parallel = true; region->ws_args = get_ws_args_for (par_stmt, ws_stmt); } } /* Debugging dumps for parallel regions. */ void dump_omp_region (FILE *, struct omp_region *, int); void debug_omp_region (struct omp_region *); void debug_all_omp_regions (void); /* Dump the parallel region tree rooted at REGION. */ void dump_omp_region (FILE *file, struct omp_region *region, int indent) { fprintf (file, "%*sbb %d: %s\n", indent, "", region->entry->index, gimple_code_name[region->type]); if (region->inner) dump_omp_region (file, region->inner, indent + 4); if (region->cont) { fprintf (file, "%*sbb %d: GIMPLE_OMP_CONTINUE\n", indent, "", region->cont->index); } if (region->exit) fprintf (file, "%*sbb %d: GIMPLE_OMP_RETURN\n", indent, "", region->exit->index); else fprintf (file, "%*s[no exit marker]\n", indent, ""); if (region->next) dump_omp_region (file, region->next, indent); } DEBUG_FUNCTION void debug_omp_region (struct omp_region *region) { dump_omp_region (stderr, region, 0); } DEBUG_FUNCTION void debug_all_omp_regions (void) { dump_omp_region (stderr, root_omp_region, 0); } /* Create a new parallel region starting at STMT inside region PARENT. */ static struct omp_region * new_omp_region (basic_block bb, enum gimple_code type, struct omp_region *parent) { struct omp_region *region = XCNEW (struct omp_region); region->outer = parent; region->entry = bb; region->type = type; if (parent) { /* This is a nested region. Add it to the list of inner regions in PARENT. */ region->next = parent->inner; parent->inner = region; } else { /* This is a toplevel region. Add it to the list of toplevel regions in ROOT_OMP_REGION. */ region->next = root_omp_region; root_omp_region = region; } return region; } /* Release the memory associated with the region tree rooted at REGION. */ static void free_omp_region_1 (struct omp_region *region) { struct omp_region *i, *n; for (i = region->inner; i ; i = n) { n = i->next; free_omp_region_1 (i); } free (region); } /* Release the memory for the entire omp region tree. */ void omp_free_regions (void) { struct omp_region *r, *n; for (r = root_omp_region; r ; r = n) { n = r->next; free_omp_region_1 (r); } root_omp_region = NULL; } /* A convenience function to build an empty GIMPLE_COND with just the condition. */ static gcond * gimple_build_cond_empty (tree cond) { enum tree_code pred_code; tree lhs, rhs; gimple_cond_get_ops_from_tree (cond, &pred_code, &lhs, &rhs); return gimple_build_cond (pred_code, lhs, rhs, NULL_TREE, NULL_TREE); } /* Return true if a parallel REGION is within a declare target function or within a target region and is not a part of a gridified target. */ static bool parallel_needs_hsa_kernel_p (struct omp_region *region) { bool indirect = false; for (region = region->outer; region; region = region->outer) { if (region->type == GIMPLE_OMP_PARALLEL) indirect = true; else if (region->type == GIMPLE_OMP_TARGET) { gomp_target *tgt_stmt = as_a (last_stmt (region->entry)); if (omp_find_clause (gimple_omp_target_clauses (tgt_stmt), OMP_CLAUSE__GRIDDIM_)) return indirect; else return true; } } if (lookup_attribute ("omp declare target", DECL_ATTRIBUTES (current_function_decl))) return true; return false; } /* Build the function calls to GOMP_parallel_start etc to actually generate the parallel operation. REGION is the parallel region being expanded. BB is the block where to insert the code. WS_ARGS will be set if this is a call to a combined parallel+workshare construct, it contains the list of additional arguments needed by the workshare construct. */ static void expand_parallel_call (struct omp_region *region, basic_block bb, gomp_parallel *entry_stmt, vec *ws_args) { tree t, t1, t2, val, cond, c, clauses, flags; gimple_stmt_iterator gsi; gimple *stmt; enum built_in_function start_ix; int start_ix2; location_t clause_loc; vec *args; clauses = gimple_omp_parallel_clauses (entry_stmt); /* Determine what flavor of GOMP_parallel we will be emitting. */ start_ix = BUILT_IN_GOMP_PARALLEL; if (is_combined_parallel (region)) { switch (region->inner->type) { case GIMPLE_OMP_FOR: gcc_assert (region->inner->sched_kind != OMP_CLAUSE_SCHEDULE_AUTO); switch (region->inner->sched_kind) { case OMP_CLAUSE_SCHEDULE_RUNTIME: start_ix2 = 3; break; case OMP_CLAUSE_SCHEDULE_DYNAMIC: case OMP_CLAUSE_SCHEDULE_GUIDED: if (region->inner->sched_modifiers & OMP_CLAUSE_SCHEDULE_NONMONOTONIC) { start_ix2 = 3 + region->inner->sched_kind; break; } /* FALLTHRU */ default: start_ix2 = region->inner->sched_kind; break; } start_ix2 += (int) BUILT_IN_GOMP_PARALLEL_LOOP_STATIC; start_ix = (enum built_in_function) start_ix2; break; case GIMPLE_OMP_SECTIONS: start_ix = BUILT_IN_GOMP_PARALLEL_SECTIONS; break; default: gcc_unreachable (); } } /* By default, the value of NUM_THREADS is zero (selected at run time) and there is no conditional. */ cond = NULL_TREE; val = build_int_cst (unsigned_type_node, 0); flags = build_int_cst (unsigned_type_node, 0); c = omp_find_clause (clauses, OMP_CLAUSE_IF); if (c) cond = OMP_CLAUSE_IF_EXPR (c); c = omp_find_clause (clauses, OMP_CLAUSE_NUM_THREADS); if (c) { val = OMP_CLAUSE_NUM_THREADS_EXPR (c); clause_loc = OMP_CLAUSE_LOCATION (c); } else clause_loc = gimple_location (entry_stmt); c = omp_find_clause (clauses, OMP_CLAUSE_PROC_BIND); if (c) flags = build_int_cst (unsigned_type_node, OMP_CLAUSE_PROC_BIND_KIND (c)); /* Ensure 'val' is of the correct type. */ val = fold_convert_loc (clause_loc, unsigned_type_node, val); /* If we found the clause 'if (cond)', build either (cond != 0) or (cond ? val : 1u). */ if (cond) { cond = gimple_boolify (cond); if (integer_zerop (val)) val = fold_build2_loc (clause_loc, EQ_EXPR, unsigned_type_node, cond, build_int_cst (TREE_TYPE (cond), 0)); else { basic_block cond_bb, then_bb, else_bb; edge e, e_then, e_else; tree tmp_then, tmp_else, tmp_join, tmp_var; tmp_var = create_tmp_var (TREE_TYPE (val)); if (gimple_in_ssa_p (cfun)) { tmp_then = make_ssa_name (tmp_var); tmp_else = make_ssa_name (tmp_var); tmp_join = make_ssa_name (tmp_var); } else { tmp_then = tmp_var; tmp_else = tmp_var; tmp_join = tmp_var; } e = split_block_after_labels (bb); cond_bb = e->src; bb = e->dest; remove_edge (e); then_bb = create_empty_bb (cond_bb); else_bb = create_empty_bb (then_bb); set_immediate_dominator (CDI_DOMINATORS, then_bb, cond_bb); set_immediate_dominator (CDI_DOMINATORS, else_bb, cond_bb); stmt = gimple_build_cond_empty (cond); gsi = gsi_start_bb (cond_bb); gsi_insert_after (&gsi, stmt, GSI_CONTINUE_LINKING); gsi = gsi_start_bb (then_bb); expand_omp_build_assign (&gsi, tmp_then, val, true); gsi = gsi_start_bb (else_bb); expand_omp_build_assign (&gsi, tmp_else, build_int_cst (unsigned_type_node, 1), true); make_edge (cond_bb, then_bb, EDGE_TRUE_VALUE); make_edge (cond_bb, else_bb, EDGE_FALSE_VALUE); add_bb_to_loop (then_bb, cond_bb->loop_father); add_bb_to_loop (else_bb, cond_bb->loop_father); e_then = make_edge (then_bb, bb, EDGE_FALLTHRU); e_else = make_edge (else_bb, bb, EDGE_FALLTHRU); if (gimple_in_ssa_p (cfun)) { gphi *phi = create_phi_node (tmp_join, bb); add_phi_arg (phi, tmp_then, e_then, UNKNOWN_LOCATION); add_phi_arg (phi, tmp_else, e_else, UNKNOWN_LOCATION); } val = tmp_join; } gsi = gsi_start_bb (bb); val = force_gimple_operand_gsi (&gsi, val, true, NULL_TREE, false, GSI_CONTINUE_LINKING); } gsi = gsi_last_bb (bb); t = gimple_omp_parallel_data_arg (entry_stmt); if (t == NULL) t1 = null_pointer_node; else t1 = build_fold_addr_expr (t); tree child_fndecl = gimple_omp_parallel_child_fn (entry_stmt); t2 = build_fold_addr_expr (child_fndecl); vec_alloc (args, 4 + vec_safe_length (ws_args)); args->quick_push (t2); args->quick_push (t1); args->quick_push (val); if (ws_args) args->splice (*ws_args); args->quick_push (flags); t = build_call_expr_loc_vec (UNKNOWN_LOCATION, builtin_decl_explicit (start_ix), args); force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); if (hsa_gen_requested_p () && parallel_needs_hsa_kernel_p (region)) { cgraph_node *child_cnode = cgraph_node::get (child_fndecl); hsa_register_kernel (child_cnode); } } /* Insert a function call whose name is FUNC_NAME with the information from ENTRY_STMT into the basic_block BB. */ static void expand_cilk_for_call (basic_block bb, gomp_parallel *entry_stmt, vec *ws_args) { tree t, t1, t2; gimple_stmt_iterator gsi; vec *args; gcc_assert (vec_safe_length (ws_args) == 2); tree func_name = (*ws_args)[0]; tree grain = (*ws_args)[1]; tree clauses = gimple_omp_parallel_clauses (entry_stmt); tree count = omp_find_clause (clauses, OMP_CLAUSE__CILK_FOR_COUNT_); gcc_assert (count != NULL_TREE); count = OMP_CLAUSE_OPERAND (count, 0); gsi = gsi_last_bb (bb); t = gimple_omp_parallel_data_arg (entry_stmt); if (t == NULL) t1 = null_pointer_node; else t1 = build_fold_addr_expr (t); t2 = build_fold_addr_expr (gimple_omp_parallel_child_fn (entry_stmt)); vec_alloc (args, 4); args->quick_push (t2); args->quick_push (t1); args->quick_push (count); args->quick_push (grain); t = build_call_expr_loc_vec (UNKNOWN_LOCATION, func_name, args); force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); } /* Build the function call to GOMP_task to actually generate the task operation. BB is the block where to insert the code. */ static void expand_task_call (struct omp_region *region, basic_block bb, gomp_task *entry_stmt) { tree t1, t2, t3; gimple_stmt_iterator gsi; location_t loc = gimple_location (entry_stmt); tree clauses = gimple_omp_task_clauses (entry_stmt); tree ifc = omp_find_clause (clauses, OMP_CLAUSE_IF); tree untied = omp_find_clause (clauses, OMP_CLAUSE_UNTIED); tree mergeable = omp_find_clause (clauses, OMP_CLAUSE_MERGEABLE); tree depend = omp_find_clause (clauses, OMP_CLAUSE_DEPEND); tree finalc = omp_find_clause (clauses, OMP_CLAUSE_FINAL); tree priority = omp_find_clause (clauses, OMP_CLAUSE_PRIORITY); unsigned int iflags = (untied ? GOMP_TASK_FLAG_UNTIED : 0) | (mergeable ? GOMP_TASK_FLAG_MERGEABLE : 0) | (depend ? GOMP_TASK_FLAG_DEPEND : 0); bool taskloop_p = gimple_omp_task_taskloop_p (entry_stmt); tree startvar = NULL_TREE, endvar = NULL_TREE, step = NULL_TREE; tree num_tasks = NULL_TREE; bool ull = false; if (taskloop_p) { gimple *g = last_stmt (region->outer->entry); gcc_assert (gimple_code (g) == GIMPLE_OMP_FOR && gimple_omp_for_kind (g) == GF_OMP_FOR_KIND_TASKLOOP); struct omp_for_data fd; omp_extract_for_data (as_a (g), &fd, NULL); startvar = omp_find_clause (clauses, OMP_CLAUSE__LOOPTEMP_); endvar = omp_find_clause (OMP_CLAUSE_CHAIN (startvar), OMP_CLAUSE__LOOPTEMP_); startvar = OMP_CLAUSE_DECL (startvar); endvar = OMP_CLAUSE_DECL (endvar); step = fold_convert_loc (loc, fd.iter_type, fd.loop.step); if (fd.loop.cond_code == LT_EXPR) iflags |= GOMP_TASK_FLAG_UP; tree tclauses = gimple_omp_for_clauses (g); num_tasks = omp_find_clause (tclauses, OMP_CLAUSE_NUM_TASKS); if (num_tasks) num_tasks = OMP_CLAUSE_NUM_TASKS_EXPR (num_tasks); else { num_tasks = omp_find_clause (tclauses, OMP_CLAUSE_GRAINSIZE); if (num_tasks) { iflags |= GOMP_TASK_FLAG_GRAINSIZE; num_tasks = OMP_CLAUSE_GRAINSIZE_EXPR (num_tasks); } else num_tasks = integer_zero_node; } num_tasks = fold_convert_loc (loc, long_integer_type_node, num_tasks); if (ifc == NULL_TREE) iflags |= GOMP_TASK_FLAG_IF; if (omp_find_clause (tclauses, OMP_CLAUSE_NOGROUP)) iflags |= GOMP_TASK_FLAG_NOGROUP; ull = fd.iter_type == long_long_unsigned_type_node; } else if (priority) iflags |= GOMP_TASK_FLAG_PRIORITY; tree flags = build_int_cst (unsigned_type_node, iflags); tree cond = boolean_true_node; if (ifc) { if (taskloop_p) { tree t = gimple_boolify (OMP_CLAUSE_IF_EXPR (ifc)); t = fold_build3_loc (loc, COND_EXPR, unsigned_type_node, t, build_int_cst (unsigned_type_node, GOMP_TASK_FLAG_IF), build_int_cst (unsigned_type_node, 0)); flags = fold_build2_loc (loc, PLUS_EXPR, unsigned_type_node, flags, t); } else cond = gimple_boolify (OMP_CLAUSE_IF_EXPR (ifc)); } if (finalc) { tree t = gimple_boolify (OMP_CLAUSE_FINAL_EXPR (finalc)); t = fold_build3_loc (loc, COND_EXPR, unsigned_type_node, t, build_int_cst (unsigned_type_node, GOMP_TASK_FLAG_FINAL), build_int_cst (unsigned_type_node, 0)); flags = fold_build2_loc (loc, PLUS_EXPR, unsigned_type_node, flags, t); } if (depend) depend = OMP_CLAUSE_DECL (depend); else depend = build_int_cst (ptr_type_node, 0); if (priority) priority = fold_convert (integer_type_node, OMP_CLAUSE_PRIORITY_EXPR (priority)); else priority = integer_zero_node; gsi = gsi_last_bb (bb); tree t = gimple_omp_task_data_arg (entry_stmt); if (t == NULL) t2 = null_pointer_node; else t2 = build_fold_addr_expr_loc (loc, t); t1 = build_fold_addr_expr_loc (loc, gimple_omp_task_child_fn (entry_stmt)); t = gimple_omp_task_copy_fn (entry_stmt); if (t == NULL) t3 = null_pointer_node; else t3 = build_fold_addr_expr_loc (loc, t); if (taskloop_p) t = build_call_expr (ull ? builtin_decl_explicit (BUILT_IN_GOMP_TASKLOOP_ULL) : builtin_decl_explicit (BUILT_IN_GOMP_TASKLOOP), 11, t1, t2, t3, gimple_omp_task_arg_size (entry_stmt), gimple_omp_task_arg_align (entry_stmt), flags, num_tasks, priority, startvar, endvar, step); else t = build_call_expr (builtin_decl_explicit (BUILT_IN_GOMP_TASK), 9, t1, t2, t3, gimple_omp_task_arg_size (entry_stmt), gimple_omp_task_arg_align (entry_stmt), cond, flags, depend, priority); force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); } /* Chain all the DECLs in LIST by their TREE_CHAIN fields. */ static tree vec2chain (vec *v) { tree chain = NULL_TREE, t; unsigned ix; FOR_EACH_VEC_SAFE_ELT_REVERSE (v, ix, t) { DECL_CHAIN (t) = chain; chain = t; } return chain; } /* Remove barriers in REGION->EXIT's block. Note that this is only valid for GIMPLE_OMP_PARALLEL regions. Since the end of a parallel region is an implicit barrier, any workshare inside the GIMPLE_OMP_PARALLEL that left a barrier at the end of the GIMPLE_OMP_PARALLEL region can now be removed. */ static void remove_exit_barrier (struct omp_region *region) { gimple_stmt_iterator gsi; basic_block exit_bb; edge_iterator ei; edge e; gimple *stmt; int any_addressable_vars = -1; exit_bb = region->exit; /* If the parallel region doesn't return, we don't have REGION->EXIT block at all. */ if (! exit_bb) return; /* The last insn in the block will be the parallel's GIMPLE_OMP_RETURN. The workshare's GIMPLE_OMP_RETURN will be in a preceding block. The kinds of statements that can appear in between are extremely limited -- no memory operations at all. Here, we allow nothing at all, so the only thing we allow to precede this GIMPLE_OMP_RETURN is a label. */ gsi = gsi_last_bb (exit_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_RETURN); gsi_prev (&gsi); if (!gsi_end_p (gsi) && gimple_code (gsi_stmt (gsi)) != GIMPLE_LABEL) return; FOR_EACH_EDGE (e, ei, exit_bb->preds) { gsi = gsi_last_bb (e->src); if (gsi_end_p (gsi)) continue; stmt = gsi_stmt (gsi); if (gimple_code (stmt) == GIMPLE_OMP_RETURN && !gimple_omp_return_nowait_p (stmt)) { /* OpenMP 3.0 tasks unfortunately prevent this optimization in many cases. If there could be tasks queued, the barrier might be needed to let the tasks run before some local variable of the parallel that the task uses as shared runs out of scope. The task can be spawned either from within current function (this would be easy to check) or from some function it calls and gets passed an address of such a variable. */ if (any_addressable_vars < 0) { gomp_parallel *parallel_stmt = as_a (last_stmt (region->entry)); tree child_fun = gimple_omp_parallel_child_fn (parallel_stmt); tree local_decls, block, decl; unsigned ix; any_addressable_vars = 0; FOR_EACH_LOCAL_DECL (DECL_STRUCT_FUNCTION (child_fun), ix, decl) if (TREE_ADDRESSABLE (decl)) { any_addressable_vars = 1; break; } for (block = gimple_block (stmt); !any_addressable_vars && block && TREE_CODE (block) == BLOCK; block = BLOCK_SUPERCONTEXT (block)) { for (local_decls = BLOCK_VARS (block); local_decls; local_decls = DECL_CHAIN (local_decls)) if (TREE_ADDRESSABLE (local_decls)) { any_addressable_vars = 1; break; } if (block == gimple_block (parallel_stmt)) break; } } if (!any_addressable_vars) gimple_omp_return_set_nowait (stmt); } } } static void remove_exit_barriers (struct omp_region *region) { if (region->type == GIMPLE_OMP_PARALLEL) remove_exit_barrier (region); if (region->inner) { region = region->inner; remove_exit_barriers (region); while (region->next) { region = region->next; remove_exit_barriers (region); } } } /* Optimize omp_get_thread_num () and omp_get_num_threads () calls. These can't be declared as const functions, but within one parallel body they are constant, so they can be transformed there into __builtin_omp_get_{thread_num,num_threads} () which are declared const. Similarly for task body, except that in untied task omp_get_thread_num () can change at any task scheduling point. */ static void optimize_omp_library_calls (gimple *entry_stmt) { basic_block bb; gimple_stmt_iterator gsi; tree thr_num_tree = builtin_decl_explicit (BUILT_IN_OMP_GET_THREAD_NUM); tree thr_num_id = DECL_ASSEMBLER_NAME (thr_num_tree); tree num_thr_tree = builtin_decl_explicit (BUILT_IN_OMP_GET_NUM_THREADS); tree num_thr_id = DECL_ASSEMBLER_NAME (num_thr_tree); bool untied_task = (gimple_code (entry_stmt) == GIMPLE_OMP_TASK && omp_find_clause (gimple_omp_task_clauses (entry_stmt), OMP_CLAUSE_UNTIED) != NULL); FOR_EACH_BB_FN (bb, cfun) for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi)) { gimple *call = gsi_stmt (gsi); tree decl; if (is_gimple_call (call) && (decl = gimple_call_fndecl (call)) && DECL_EXTERNAL (decl) && TREE_PUBLIC (decl) && DECL_INITIAL (decl) == NULL) { tree built_in; if (DECL_NAME (decl) == thr_num_id) { /* In #pragma omp task untied omp_get_thread_num () can change during the execution of the task region. */ if (untied_task) continue; built_in = builtin_decl_explicit (BUILT_IN_OMP_GET_THREAD_NUM); } else if (DECL_NAME (decl) == num_thr_id) built_in = builtin_decl_explicit (BUILT_IN_OMP_GET_NUM_THREADS); else continue; if (DECL_ASSEMBLER_NAME (decl) != DECL_ASSEMBLER_NAME (built_in) || gimple_call_num_args (call) != 0) continue; if (flag_exceptions && !TREE_NOTHROW (decl)) continue; if (TREE_CODE (TREE_TYPE (decl)) != FUNCTION_TYPE || !types_compatible_p (TREE_TYPE (TREE_TYPE (decl)), TREE_TYPE (TREE_TYPE (built_in)))) continue; gimple_call_set_fndecl (call, built_in); } } } /* Callback for expand_omp_build_assign. Return non-NULL if *tp needs to be regimplified. */ static tree expand_omp_regimplify_p (tree *tp, int *walk_subtrees, void *) { tree t = *tp; /* Any variable with DECL_VALUE_EXPR needs to be regimplified. */ if (VAR_P (t) && DECL_HAS_VALUE_EXPR_P (t)) return t; if (TREE_CODE (t) == ADDR_EXPR) recompute_tree_invariant_for_addr_expr (t); *walk_subtrees = !TYPE_P (t) && !DECL_P (t); return NULL_TREE; } /* Prepend or append TO = FROM assignment before or after *GSI_P. */ static void expand_omp_build_assign (gimple_stmt_iterator *gsi_p, tree to, tree from, bool after) { bool simple_p = DECL_P (to) && TREE_ADDRESSABLE (to); from = force_gimple_operand_gsi (gsi_p, from, simple_p, NULL_TREE, !after, after ? GSI_CONTINUE_LINKING : GSI_SAME_STMT); gimple *stmt = gimple_build_assign (to, from); if (after) gsi_insert_after (gsi_p, stmt, GSI_CONTINUE_LINKING); else gsi_insert_before (gsi_p, stmt, GSI_SAME_STMT); if (walk_tree (&from, expand_omp_regimplify_p, NULL, NULL) || walk_tree (&to, expand_omp_regimplify_p, NULL, NULL)) { gimple_stmt_iterator gsi = gsi_for_stmt (stmt); gimple_regimplify_operands (stmt, &gsi); } } /* Expand the OpenMP parallel or task directive starting at REGION. */ static void expand_omp_taskreg (struct omp_region *region) { basic_block entry_bb, exit_bb, new_bb; struct function *child_cfun; tree child_fn, block, t; gimple_stmt_iterator gsi; gimple *entry_stmt, *stmt; edge e; vec *ws_args; entry_stmt = last_stmt (region->entry); child_fn = gimple_omp_taskreg_child_fn (entry_stmt); child_cfun = DECL_STRUCT_FUNCTION (child_fn); entry_bb = region->entry; if (gimple_code (entry_stmt) == GIMPLE_OMP_TASK) exit_bb = region->cont; else exit_bb = region->exit; bool is_cilk_for = (flag_cilkplus && gimple_code (entry_stmt) == GIMPLE_OMP_PARALLEL && omp_find_clause (gimple_omp_parallel_clauses (entry_stmt), OMP_CLAUSE__CILK_FOR_COUNT_) != NULL_TREE); if (is_cilk_for) /* If it is a _Cilk_for statement, it is modelled *like* a parallel for, and the inner statement contains the name of the built-in function and grain. */ ws_args = region->inner->ws_args; else if (is_combined_parallel (region)) ws_args = region->ws_args; else ws_args = NULL; if (child_cfun->cfg) { /* Due to inlining, it may happen that we have already outlined the region, in which case all we need to do is make the sub-graph unreachable and emit the parallel call. */ edge entry_succ_e, exit_succ_e; entry_succ_e = single_succ_edge (entry_bb); gsi = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_PARALLEL || gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_TASK); gsi_remove (&gsi, true); new_bb = entry_bb; if (exit_bb) { exit_succ_e = single_succ_edge (exit_bb); make_edge (new_bb, exit_succ_e->dest, EDGE_FALLTHRU); } remove_edge_and_dominated_blocks (entry_succ_e); } else { unsigned srcidx, dstidx, num; /* If the parallel region needs data sent from the parent function, then the very first statement (except possible tree profile counter updates) of the parallel body is a copy assignment .OMP_DATA_I = &.OMP_DATA_O. Since &.OMP_DATA_O is passed as an argument to the child function, we need to replace it with the argument as seen by the child function. In most cases, this will end up being the identity assignment .OMP_DATA_I = .OMP_DATA_I. However, if the parallel body had a function call that has been inlined, the original PARM_DECL .OMP_DATA_I may have been converted into a different local variable. In which case, we need to keep the assignment. */ if (gimple_omp_taskreg_data_arg (entry_stmt)) { basic_block entry_succ_bb = single_succ_p (entry_bb) ? single_succ (entry_bb) : FALLTHRU_EDGE (entry_bb)->dest; tree arg; gimple *parcopy_stmt = NULL; for (gsi = gsi_start_bb (entry_succ_bb); ; gsi_next (&gsi)) { gimple *stmt; gcc_assert (!gsi_end_p (gsi)); stmt = gsi_stmt (gsi); if (gimple_code (stmt) != GIMPLE_ASSIGN) continue; if (gimple_num_ops (stmt) == 2) { tree arg = gimple_assign_rhs1 (stmt); /* We're ignore the subcode because we're effectively doing a STRIP_NOPS. */ if (TREE_CODE (arg) == ADDR_EXPR && TREE_OPERAND (arg, 0) == gimple_omp_taskreg_data_arg (entry_stmt)) { parcopy_stmt = stmt; break; } } } gcc_assert (parcopy_stmt != NULL); arg = DECL_ARGUMENTS (child_fn); if (!gimple_in_ssa_p (cfun)) { if (gimple_assign_lhs (parcopy_stmt) == arg) gsi_remove (&gsi, true); else { /* ?? Is setting the subcode really necessary ?? */ gimple_omp_set_subcode (parcopy_stmt, TREE_CODE (arg)); gimple_assign_set_rhs1 (parcopy_stmt, arg); } } else { tree lhs = gimple_assign_lhs (parcopy_stmt); gcc_assert (SSA_NAME_VAR (lhs) == arg); /* We'd like to set the rhs to the default def in the child_fn, but it's too early to create ssa names in the child_fn. Instead, we set the rhs to the parm. In move_sese_region_to_fn, we introduce a default def for the parm, map the parm to it's default def, and once we encounter this stmt, replace the parm with the default def. */ gimple_assign_set_rhs1 (parcopy_stmt, arg); update_stmt (parcopy_stmt); } } /* Declare local variables needed in CHILD_CFUN. */ block = DECL_INITIAL (child_fn); BLOCK_VARS (block) = vec2chain (child_cfun->local_decls); /* The gimplifier could record temporaries in parallel/task block rather than in containing function's local_decls chain, which would mean cgraph missed finalizing them. Do it now. */ for (t = BLOCK_VARS (block); t; t = DECL_CHAIN (t)) if (VAR_P (t) && TREE_STATIC (t) && !DECL_EXTERNAL (t)) varpool_node::finalize_decl (t); DECL_SAVED_TREE (child_fn) = NULL; /* We'll create a CFG for child_fn, so no gimple body is needed. */ gimple_set_body (child_fn, NULL); TREE_USED (block) = 1; /* Reset DECL_CONTEXT on function arguments. */ for (t = DECL_ARGUMENTS (child_fn); t; t = DECL_CHAIN (t)) DECL_CONTEXT (t) = child_fn; /* Split ENTRY_BB at GIMPLE_OMP_PARALLEL or GIMPLE_OMP_TASK, so that it can be moved to the child function. */ gsi = gsi_last_bb (entry_bb); stmt = gsi_stmt (gsi); gcc_assert (stmt && (gimple_code (stmt) == GIMPLE_OMP_PARALLEL || gimple_code (stmt) == GIMPLE_OMP_TASK)); e = split_block (entry_bb, stmt); gsi_remove (&gsi, true); entry_bb = e->dest; edge e2 = NULL; if (gimple_code (entry_stmt) == GIMPLE_OMP_PARALLEL) single_succ_edge (entry_bb)->flags = EDGE_FALLTHRU; else { e2 = make_edge (e->src, BRANCH_EDGE (entry_bb)->dest, EDGE_ABNORMAL); gcc_assert (e2->dest == region->exit); remove_edge (BRANCH_EDGE (entry_bb)); set_immediate_dominator (CDI_DOMINATORS, e2->dest, e->src); gsi = gsi_last_bb (region->exit); gcc_assert (!gsi_end_p (gsi) && gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_RETURN); gsi_remove (&gsi, true); } /* Convert GIMPLE_OMP_{RETURN,CONTINUE} into a RETURN_EXPR. */ if (exit_bb) { gsi = gsi_last_bb (exit_bb); gcc_assert (!gsi_end_p (gsi) && (gimple_code (gsi_stmt (gsi)) == (e2 ? GIMPLE_OMP_CONTINUE : GIMPLE_OMP_RETURN))); stmt = gimple_build_return (NULL); gsi_insert_after (&gsi, stmt, GSI_SAME_STMT); gsi_remove (&gsi, true); } /* Move the parallel region into CHILD_CFUN. */ if (gimple_in_ssa_p (cfun)) { init_tree_ssa (child_cfun); init_ssa_operands (child_cfun); child_cfun->gimple_df->in_ssa_p = true; block = NULL_TREE; } else block = gimple_block (entry_stmt); /* Make sure to generate early debug for the function before outlining anything. */ if (! gimple_in_ssa_p (cfun)) (*debug_hooks->early_global_decl) (cfun->decl); new_bb = move_sese_region_to_fn (child_cfun, entry_bb, exit_bb, block); if (exit_bb) single_succ_edge (new_bb)->flags = EDGE_FALLTHRU; if (e2) { basic_block dest_bb = e2->dest; if (!exit_bb) make_edge (new_bb, dest_bb, EDGE_FALLTHRU); remove_edge (e2); set_immediate_dominator (CDI_DOMINATORS, dest_bb, new_bb); } /* When the OMP expansion process cannot guarantee an up-to-date loop tree arrange for the child function to fixup loops. */ if (loops_state_satisfies_p (LOOPS_NEED_FIXUP)) child_cfun->x_current_loops->state |= LOOPS_NEED_FIXUP; /* Remove non-local VAR_DECLs from child_cfun->local_decls list. */ num = vec_safe_length (child_cfun->local_decls); for (srcidx = 0, dstidx = 0; srcidx < num; srcidx++) { t = (*child_cfun->local_decls)[srcidx]; if (DECL_CONTEXT (t) == cfun->decl) continue; if (srcidx != dstidx) (*child_cfun->local_decls)[dstidx] = t; dstidx++; } if (dstidx != num) vec_safe_truncate (child_cfun->local_decls, dstidx); /* Inform the callgraph about the new function. */ child_cfun->curr_properties = cfun->curr_properties; child_cfun->has_simduid_loops |= cfun->has_simduid_loops; child_cfun->has_force_vectorize_loops |= cfun->has_force_vectorize_loops; cgraph_node *node = cgraph_node::get_create (child_fn); node->parallelized_function = 1; cgraph_node::add_new_function (child_fn, true); bool need_asm = DECL_ASSEMBLER_NAME_SET_P (current_function_decl) && !DECL_ASSEMBLER_NAME_SET_P (child_fn); /* Fix the callgraph edges for child_cfun. Those for cfun will be fixed in a following pass. */ push_cfun (child_cfun); if (need_asm) assign_assembler_name_if_needed (child_fn); if (optimize) optimize_omp_library_calls (entry_stmt); cgraph_edge::rebuild_edges (); /* Some EH regions might become dead, see PR34608. If pass_cleanup_cfg isn't the first pass to happen with the new child, these dead EH edges might cause problems. Clean them up now. */ if (flag_exceptions) { basic_block bb; bool changed = false; FOR_EACH_BB_FN (bb, cfun) changed |= gimple_purge_dead_eh_edges (bb); if (changed) cleanup_tree_cfg (); } if (gimple_in_ssa_p (cfun)) update_ssa (TODO_update_ssa); if (flag_checking && !loops_state_satisfies_p (LOOPS_NEED_FIXUP)) verify_loop_structure (); pop_cfun (); if (dump_file && !gimple_in_ssa_p (cfun)) { omp_any_child_fn_dumped = true; dump_function_header (dump_file, child_fn, dump_flags); dump_function_to_file (child_fn, dump_file, dump_flags); } } /* Emit a library call to launch the children threads. */ if (is_cilk_for) expand_cilk_for_call (new_bb, as_a (entry_stmt), ws_args); else if (gimple_code (entry_stmt) == GIMPLE_OMP_PARALLEL) expand_parallel_call (region, new_bb, as_a (entry_stmt), ws_args); else expand_task_call (region, new_bb, as_a (entry_stmt)); if (gimple_in_ssa_p (cfun)) update_ssa (TODO_update_ssa_only_virtuals); } /* Information about members of an OpenACC collapsed loop nest. */ struct oacc_collapse { tree base; /* Base value. */ tree iters; /* Number of steps. */ tree step; /* Step size. */ tree tile; /* Tile increment (if tiled). */ tree outer; /* Tile iterator var. */ }; /* Helper for expand_oacc_for. Determine collapsed loop information. Fill in COUNTS array. Emit any initialization code before GSI. Return the calculated outer loop bound of BOUND_TYPE. */ static tree expand_oacc_collapse_init (const struct omp_for_data *fd, gimple_stmt_iterator *gsi, oacc_collapse *counts, tree bound_type, location_t loc) { tree tiling = fd->tiling; tree total = build_int_cst (bound_type, 1); int ix; gcc_assert (integer_onep (fd->loop.step)); gcc_assert (integer_zerop (fd->loop.n1)); /* When tiling, the first operand of the tile clause applies to the innermost loop, and we work outwards from there. Seems backwards, but whatever. */ for (ix = fd->collapse; ix--;) { const omp_for_data_loop *loop = &fd->loops[ix]; tree iter_type = TREE_TYPE (loop->v); tree diff_type = iter_type; tree plus_type = iter_type; gcc_assert (loop->cond_code == fd->loop.cond_code); if (POINTER_TYPE_P (iter_type)) plus_type = sizetype; if (POINTER_TYPE_P (diff_type) || TYPE_UNSIGNED (diff_type)) diff_type = signed_type_for (diff_type); if (tiling) { tree num = build_int_cst (integer_type_node, fd->collapse); tree loop_no = build_int_cst (integer_type_node, ix); tree tile = TREE_VALUE (tiling); gcall *call = gimple_build_call_internal (IFN_GOACC_TILE, 5, num, loop_no, tile, /* gwv-outer=*/integer_zero_node, /* gwv-inner=*/integer_zero_node); counts[ix].outer = create_tmp_var (iter_type, ".outer"); counts[ix].tile = create_tmp_var (diff_type, ".tile"); gimple_call_set_lhs (call, counts[ix].tile); gimple_set_location (call, loc); gsi_insert_before (gsi, call, GSI_SAME_STMT); tiling = TREE_CHAIN (tiling); } else { counts[ix].tile = NULL; counts[ix].outer = loop->v; } tree b = loop->n1; tree e = loop->n2; tree s = loop->step; bool up = loop->cond_code == LT_EXPR; tree dir = build_int_cst (diff_type, up ? +1 : -1); bool negating; tree expr; b = force_gimple_operand_gsi (gsi, b, true, NULL_TREE, true, GSI_SAME_STMT); e = force_gimple_operand_gsi (gsi, e, true, NULL_TREE, true, GSI_SAME_STMT); /* Convert the step, avoiding possible unsigned->signed overflow. */ negating = !up && TYPE_UNSIGNED (TREE_TYPE (s)); if (negating) s = fold_build1 (NEGATE_EXPR, TREE_TYPE (s), s); s = fold_convert (diff_type, s); if (negating) s = fold_build1 (NEGATE_EXPR, diff_type, s); s = force_gimple_operand_gsi (gsi, s, true, NULL_TREE, true, GSI_SAME_STMT); /* Determine the range, avoiding possible unsigned->signed overflow. */ negating = !up && TYPE_UNSIGNED (iter_type); expr = fold_build2 (MINUS_EXPR, plus_type, fold_convert (plus_type, negating ? b : e), fold_convert (plus_type, negating ? e : b)); expr = fold_convert (diff_type, expr); if (negating) expr = fold_build1 (NEGATE_EXPR, diff_type, expr); tree range = force_gimple_operand_gsi (gsi, expr, true, NULL_TREE, true, GSI_SAME_STMT); /* Determine number of iterations. */ expr = fold_build2 (MINUS_EXPR, diff_type, range, dir); expr = fold_build2 (PLUS_EXPR, diff_type, expr, s); expr = fold_build2 (TRUNC_DIV_EXPR, diff_type, expr, s); tree iters = force_gimple_operand_gsi (gsi, expr, true, NULL_TREE, true, GSI_SAME_STMT); counts[ix].base = b; counts[ix].iters = iters; counts[ix].step = s; total = fold_build2 (MULT_EXPR, bound_type, total, fold_convert (bound_type, iters)); } return total; } /* Emit initializers for collapsed loop members. INNER is true if this is for the element loop of a TILE. IVAR is the outer loop iteration variable, from which collapsed loop iteration values are calculated. COUNTS array has been initialized by expand_oacc_collapse_inits. */ static void expand_oacc_collapse_vars (const struct omp_for_data *fd, bool inner, gimple_stmt_iterator *gsi, const oacc_collapse *counts, tree ivar) { tree ivar_type = TREE_TYPE (ivar); /* The most rapidly changing iteration variable is the innermost one. */ for (int ix = fd->collapse; ix--;) { const omp_for_data_loop *loop = &fd->loops[ix]; const oacc_collapse *collapse = &counts[ix]; tree v = inner ? loop->v : collapse->outer; tree iter_type = TREE_TYPE (v); tree diff_type = TREE_TYPE (collapse->step); tree plus_type = iter_type; enum tree_code plus_code = PLUS_EXPR; tree expr; if (POINTER_TYPE_P (iter_type)) { plus_code = POINTER_PLUS_EXPR; plus_type = sizetype; } expr = ivar; if (ix) { tree mod = fold_convert (ivar_type, collapse->iters); ivar = fold_build2 (TRUNC_DIV_EXPR, ivar_type, expr, mod); expr = fold_build2 (TRUNC_MOD_EXPR, ivar_type, expr, mod); ivar = force_gimple_operand_gsi (gsi, ivar, true, NULL_TREE, true, GSI_SAME_STMT); } expr = fold_build2 (MULT_EXPR, diff_type, fold_convert (diff_type, expr), collapse->step); expr = fold_build2 (plus_code, iter_type, inner ? collapse->outer : collapse->base, fold_convert (plus_type, expr)); expr = force_gimple_operand_gsi (gsi, expr, false, NULL_TREE, true, GSI_SAME_STMT); gassign *ass = gimple_build_assign (v, expr); gsi_insert_before (gsi, ass, GSI_SAME_STMT); } } /* Helper function for expand_omp_{for_*,simd}. If this is the outermost of the combined collapse > 1 loop constructs, generate code like: if (__builtin_expect (N32 cond3 N31, 0)) goto ZERO_ITER_BB; if (cond3 is <) adj = STEP3 - 1; else adj = STEP3 + 1; count3 = (adj + N32 - N31) / STEP3; if (__builtin_expect (N22 cond2 N21, 0)) goto ZERO_ITER_BB; if (cond2 is <) adj = STEP2 - 1; else adj = STEP2 + 1; count2 = (adj + N22 - N21) / STEP2; if (__builtin_expect (N12 cond1 N11, 0)) goto ZERO_ITER_BB; if (cond1 is <) adj = STEP1 - 1; else adj = STEP1 + 1; count1 = (adj + N12 - N11) / STEP1; count = count1 * count2 * count3; Furthermore, if ZERO_ITER_BB is NULL, create a BB which does: count = 0; and set ZERO_ITER_BB to that bb. If this isn't the outermost of the combined loop constructs, just initialize COUNTS array from the _looptemp_ clauses. */ /* NOTE: It *could* be better to moosh all of the BBs together, creating one larger BB with all the computation and the unexpected jump at the end. I.e. bool zero3, zero2, zero1, zero; zero3 = N32 c3 N31; count3 = (N32 - N31) /[cl] STEP3; zero2 = N22 c2 N21; count2 = (N22 - N21) /[cl] STEP2; zero1 = N12 c1 N11; count1 = (N12 - N11) /[cl] STEP1; zero = zero3 || zero2 || zero1; count = count1 * count2 * count3; if (__builtin_expect(zero, false)) goto zero_iter_bb; After all, we expect the zero=false, and thus we expect to have to evaluate all of the comparison expressions, so short-circuiting oughtn't be a win. Since the condition isn't protecting a denominator, we're not concerned about divide-by-zero, so we can fully evaluate count even if a numerator turned out to be wrong. It seems like putting this all together would create much better scheduling opportunities, and less pressure on the chip's branch predictor. */ static void expand_omp_for_init_counts (struct omp_for_data *fd, gimple_stmt_iterator *gsi, basic_block &entry_bb, tree *counts, basic_block &zero_iter1_bb, int &first_zero_iter1, basic_block &zero_iter2_bb, int &first_zero_iter2, basic_block &l2_dom_bb) { tree t, type = TREE_TYPE (fd->loop.v); edge e, ne; int i; /* Collapsed loops need work for expansion into SSA form. */ gcc_assert (!gimple_in_ssa_p (cfun)); if (gimple_omp_for_combined_into_p (fd->for_stmt) && TREE_CODE (fd->loop.n2) != INTEGER_CST) { gcc_assert (fd->ordered == 0); /* First two _looptemp_ clauses are for istart/iend, counts[0] isn't supposed to be handled, as the inner loop doesn't use it. */ tree innerc = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); for (i = 0; i < fd->collapse; i++) { innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); if (i) counts[i] = OMP_CLAUSE_DECL (innerc); else counts[0] = NULL_TREE; } return; } for (i = fd->collapse; i < fd->ordered; i++) { tree itype = TREE_TYPE (fd->loops[i].v); counts[i] = NULL_TREE; t = fold_binary (fd->loops[i].cond_code, boolean_type_node, fold_convert (itype, fd->loops[i].n1), fold_convert (itype, fd->loops[i].n2)); if (t && integer_zerop (t)) { for (i = fd->collapse; i < fd->ordered; i++) counts[i] = build_int_cst (type, 0); break; } } for (i = 0; i < (fd->ordered ? fd->ordered : fd->collapse); i++) { tree itype = TREE_TYPE (fd->loops[i].v); if (i >= fd->collapse && counts[i]) continue; if ((SSA_VAR_P (fd->loop.n2) || i >= fd->collapse) && ((t = fold_binary (fd->loops[i].cond_code, boolean_type_node, fold_convert (itype, fd->loops[i].n1), fold_convert (itype, fd->loops[i].n2))) == NULL_TREE || !integer_onep (t))) { gcond *cond_stmt; tree n1, n2; n1 = fold_convert (itype, unshare_expr (fd->loops[i].n1)); n1 = force_gimple_operand_gsi (gsi, n1, true, NULL_TREE, true, GSI_SAME_STMT); n2 = fold_convert (itype, unshare_expr (fd->loops[i].n2)); n2 = force_gimple_operand_gsi (gsi, n2, true, NULL_TREE, true, GSI_SAME_STMT); cond_stmt = gimple_build_cond (fd->loops[i].cond_code, n1, n2, NULL_TREE, NULL_TREE); gsi_insert_before (gsi, cond_stmt, GSI_SAME_STMT); if (walk_tree (gimple_cond_lhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL) || walk_tree (gimple_cond_rhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL)) { *gsi = gsi_for_stmt (cond_stmt); gimple_regimplify_operands (cond_stmt, gsi); } e = split_block (entry_bb, cond_stmt); basic_block &zero_iter_bb = i < fd->collapse ? zero_iter1_bb : zero_iter2_bb; int &first_zero_iter = i < fd->collapse ? first_zero_iter1 : first_zero_iter2; if (zero_iter_bb == NULL) { gassign *assign_stmt; first_zero_iter = i; zero_iter_bb = create_empty_bb (entry_bb); add_bb_to_loop (zero_iter_bb, entry_bb->loop_father); *gsi = gsi_after_labels (zero_iter_bb); if (i < fd->collapse) assign_stmt = gimple_build_assign (fd->loop.n2, build_zero_cst (type)); else { counts[i] = create_tmp_reg (type, ".count"); assign_stmt = gimple_build_assign (counts[i], build_zero_cst (type)); } gsi_insert_before (gsi, assign_stmt, GSI_SAME_STMT); set_immediate_dominator (CDI_DOMINATORS, zero_iter_bb, entry_bb); } ne = make_edge (entry_bb, zero_iter_bb, EDGE_FALSE_VALUE); ne->probability = REG_BR_PROB_BASE / 2000 - 1; e->flags = EDGE_TRUE_VALUE; e->probability = REG_BR_PROB_BASE - ne->probability; if (l2_dom_bb == NULL) l2_dom_bb = entry_bb; entry_bb = e->dest; *gsi = gsi_last_bb (entry_bb); } if (POINTER_TYPE_P (itype)) itype = signed_type_for (itype); t = build_int_cst (itype, (fd->loops[i].cond_code == LT_EXPR ? -1 : 1)); t = fold_build2 (PLUS_EXPR, itype, fold_convert (itype, fd->loops[i].step), t); t = fold_build2 (PLUS_EXPR, itype, t, fold_convert (itype, fd->loops[i].n2)); t = fold_build2 (MINUS_EXPR, itype, t, fold_convert (itype, fd->loops[i].n1)); /* ?? We could probably use CEIL_DIV_EXPR instead of TRUNC_DIV_EXPR and adjusting by hand. Unless we can't generate the same code in the end because generically we don't know that the values involved must be negative for GT?? */ if (TYPE_UNSIGNED (itype) && fd->loops[i].cond_code == GT_EXPR) t = fold_build2 (TRUNC_DIV_EXPR, itype, fold_build1 (NEGATE_EXPR, itype, t), fold_build1 (NEGATE_EXPR, itype, fold_convert (itype, fd->loops[i].step))); else t = fold_build2 (TRUNC_DIV_EXPR, itype, t, fold_convert (itype, fd->loops[i].step)); t = fold_convert (type, t); if (TREE_CODE (t) == INTEGER_CST) counts[i] = t; else { if (i < fd->collapse || i != first_zero_iter2) counts[i] = create_tmp_reg (type, ".count"); expand_omp_build_assign (gsi, counts[i], t); } if (SSA_VAR_P (fd->loop.n2) && i < fd->collapse) { if (i == 0) t = counts[0]; else t = fold_build2 (MULT_EXPR, type, fd->loop.n2, counts[i]); expand_omp_build_assign (gsi, fd->loop.n2, t); } } } /* Helper function for expand_omp_{for_*,simd}. Generate code like: T = V; V3 = N31 + (T % count3) * STEP3; T = T / count3; V2 = N21 + (T % count2) * STEP2; T = T / count2; V1 = N11 + T * STEP1; if this loop doesn't have an inner loop construct combined with it. If it does have an inner loop construct combined with it and the iteration count isn't known constant, store values from counts array into its _looptemp_ temporaries instead. */ static void expand_omp_for_init_vars (struct omp_for_data *fd, gimple_stmt_iterator *gsi, tree *counts, gimple *inner_stmt, tree startvar) { int i; if (gimple_omp_for_combined_p (fd->for_stmt)) { /* If fd->loop.n2 is constant, then no propagation of the counts is needed, they are constant. */ if (TREE_CODE (fd->loop.n2) == INTEGER_CST) return; tree clauses = gimple_code (inner_stmt) != GIMPLE_OMP_FOR ? gimple_omp_taskreg_clauses (inner_stmt) : gimple_omp_for_clauses (inner_stmt); /* First two _looptemp_ clauses are for istart/iend, counts[0] isn't supposed to be handled, as the inner loop doesn't use it. */ tree innerc = omp_find_clause (clauses, OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); for (i = 0; i < fd->collapse; i++) { innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); if (i) { tree tem = OMP_CLAUSE_DECL (innerc); tree t = fold_convert (TREE_TYPE (tem), counts[i]); t = force_gimple_operand_gsi (gsi, t, false, NULL_TREE, false, GSI_CONTINUE_LINKING); gassign *stmt = gimple_build_assign (tem, t); gsi_insert_after (gsi, stmt, GSI_CONTINUE_LINKING); } } return; } tree type = TREE_TYPE (fd->loop.v); tree tem = create_tmp_reg (type, ".tem"); gassign *stmt = gimple_build_assign (tem, startvar); gsi_insert_after (gsi, stmt, GSI_CONTINUE_LINKING); for (i = fd->collapse - 1; i >= 0; i--) { tree vtype = TREE_TYPE (fd->loops[i].v), itype, t; itype = vtype; if (POINTER_TYPE_P (vtype)) itype = signed_type_for (vtype); if (i != 0) t = fold_build2 (TRUNC_MOD_EXPR, type, tem, counts[i]); else t = tem; t = fold_convert (itype, t); t = fold_build2 (MULT_EXPR, itype, t, fold_convert (itype, fd->loops[i].step)); if (POINTER_TYPE_P (vtype)) t = fold_build_pointer_plus (fd->loops[i].n1, t); else t = fold_build2 (PLUS_EXPR, itype, fd->loops[i].n1, t); t = force_gimple_operand_gsi (gsi, t, DECL_P (fd->loops[i].v) && TREE_ADDRESSABLE (fd->loops[i].v), NULL_TREE, false, GSI_CONTINUE_LINKING); stmt = gimple_build_assign (fd->loops[i].v, t); gsi_insert_after (gsi, stmt, GSI_CONTINUE_LINKING); if (i != 0) { t = fold_build2 (TRUNC_DIV_EXPR, type, tem, counts[i]); t = force_gimple_operand_gsi (gsi, t, false, NULL_TREE, false, GSI_CONTINUE_LINKING); stmt = gimple_build_assign (tem, t); gsi_insert_after (gsi, stmt, GSI_CONTINUE_LINKING); } } } /* Helper function for expand_omp_for_*. Generate code like: L10: V3 += STEP3; if (V3 cond3 N32) goto BODY_BB; else goto L11; L11: V3 = N31; V2 += STEP2; if (V2 cond2 N22) goto BODY_BB; else goto L12; L12: V2 = N21; V1 += STEP1; goto BODY_BB; */ static basic_block extract_omp_for_update_vars (struct omp_for_data *fd, basic_block cont_bb, basic_block body_bb) { basic_block last_bb, bb, collapse_bb = NULL; int i; gimple_stmt_iterator gsi; edge e; tree t; gimple *stmt; last_bb = cont_bb; for (i = fd->collapse - 1; i >= 0; i--) { tree vtype = TREE_TYPE (fd->loops[i].v); bb = create_empty_bb (last_bb); add_bb_to_loop (bb, last_bb->loop_father); gsi = gsi_start_bb (bb); if (i < fd->collapse - 1) { e = make_edge (last_bb, bb, EDGE_FALSE_VALUE); e->probability = REG_BR_PROB_BASE / 8; t = fd->loops[i + 1].n1; t = force_gimple_operand_gsi (&gsi, t, DECL_P (fd->loops[i + 1].v) && TREE_ADDRESSABLE (fd->loops[i + 1].v), NULL_TREE, false, GSI_CONTINUE_LINKING); stmt = gimple_build_assign (fd->loops[i + 1].v, t); gsi_insert_after (&gsi, stmt, GSI_CONTINUE_LINKING); } else collapse_bb = bb; set_immediate_dominator (CDI_DOMINATORS, bb, last_bb); if (POINTER_TYPE_P (vtype)) t = fold_build_pointer_plus (fd->loops[i].v, fd->loops[i].step); else t = fold_build2 (PLUS_EXPR, vtype, fd->loops[i].v, fd->loops[i].step); t = force_gimple_operand_gsi (&gsi, t, DECL_P (fd->loops[i].v) && TREE_ADDRESSABLE (fd->loops[i].v), NULL_TREE, false, GSI_CONTINUE_LINKING); stmt = gimple_build_assign (fd->loops[i].v, t); gsi_insert_after (&gsi, stmt, GSI_CONTINUE_LINKING); if (i > 0) { t = fd->loops[i].n2; t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); tree v = fd->loops[i].v; if (DECL_P (v) && TREE_ADDRESSABLE (v)) v = force_gimple_operand_gsi (&gsi, v, true, NULL_TREE, false, GSI_CONTINUE_LINKING); t = fold_build2 (fd->loops[i].cond_code, boolean_type_node, v, t); stmt = gimple_build_cond_empty (t); gsi_insert_after (&gsi, stmt, GSI_CONTINUE_LINKING); e = make_edge (bb, body_bb, EDGE_TRUE_VALUE); e->probability = REG_BR_PROB_BASE * 7 / 8; } else make_edge (bb, body_bb, EDGE_FALLTHRU); last_bb = bb; } return collapse_bb; } /* Expand #pragma omp ordered depend(source). */ static void expand_omp_ordered_source (gimple_stmt_iterator *gsi, struct omp_for_data *fd, tree *counts, location_t loc) { enum built_in_function source_ix = fd->iter_type == long_integer_type_node ? BUILT_IN_GOMP_DOACROSS_POST : BUILT_IN_GOMP_DOACROSS_ULL_POST; gimple *g = gimple_build_call (builtin_decl_explicit (source_ix), 1, build_fold_addr_expr (counts[fd->ordered])); gimple_set_location (g, loc); gsi_insert_before (gsi, g, GSI_SAME_STMT); } /* Expand a single depend from #pragma omp ordered depend(sink:...). */ static void expand_omp_ordered_sink (gimple_stmt_iterator *gsi, struct omp_for_data *fd, tree *counts, tree c, location_t loc) { auto_vec args; enum built_in_function sink_ix = fd->iter_type == long_integer_type_node ? BUILT_IN_GOMP_DOACROSS_WAIT : BUILT_IN_GOMP_DOACROSS_ULL_WAIT; tree t, off, coff = NULL_TREE, deps = OMP_CLAUSE_DECL (c), cond = NULL_TREE; int i; gimple_stmt_iterator gsi2 = *gsi; bool warned_step = false; for (i = 0; i < fd->ordered; i++) { tree step = NULL_TREE; off = TREE_PURPOSE (deps); if (TREE_CODE (off) == TRUNC_DIV_EXPR) { step = TREE_OPERAND (off, 1); off = TREE_OPERAND (off, 0); } if (!integer_zerop (off)) { gcc_assert (fd->loops[i].cond_code == LT_EXPR || fd->loops[i].cond_code == GT_EXPR); bool forward = fd->loops[i].cond_code == LT_EXPR; if (step) { /* Non-simple Fortran DO loops. If step is variable, we don't know at compile even the direction, so can't warn. */ if (TREE_CODE (step) != INTEGER_CST) break; forward = tree_int_cst_sgn (step) != -1; } if (forward ^ OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) warning_at (loc, 0, "% clause waiting for " "lexically later iteration"); break; } deps = TREE_CHAIN (deps); } /* If all offsets corresponding to the collapsed loops are zero, this depend clause can be ignored. FIXME: but there is still a flush needed. We need to emit one __sync_synchronize () for it though (perhaps conditionally)? Solve this together with the conservative dependence folding optimization. if (i >= fd->collapse) return; */ deps = OMP_CLAUSE_DECL (c); gsi_prev (&gsi2); edge e1 = split_block (gsi_bb (gsi2), gsi_stmt (gsi2)); edge e2 = split_block_after_labels (e1->dest); gsi2 = gsi_after_labels (e1->dest); *gsi = gsi_last_bb (e1->src); for (i = 0; i < fd->ordered; i++) { tree itype = TREE_TYPE (fd->loops[i].v); tree step = NULL_TREE; tree orig_off = NULL_TREE; if (POINTER_TYPE_P (itype)) itype = sizetype; if (i) deps = TREE_CHAIN (deps); off = TREE_PURPOSE (deps); if (TREE_CODE (off) == TRUNC_DIV_EXPR) { step = TREE_OPERAND (off, 1); off = TREE_OPERAND (off, 0); gcc_assert (fd->loops[i].cond_code == LT_EXPR && integer_onep (fd->loops[i].step) && !POINTER_TYPE_P (TREE_TYPE (fd->loops[i].v))); } tree s = fold_convert_loc (loc, itype, step ? step : fd->loops[i].step); if (step) { off = fold_convert_loc (loc, itype, off); orig_off = off; off = fold_build2_loc (loc, TRUNC_DIV_EXPR, itype, off, s); } if (integer_zerop (off)) t = boolean_true_node; else { tree a; tree co = fold_convert_loc (loc, itype, off); if (POINTER_TYPE_P (TREE_TYPE (fd->loops[i].v))) { if (OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) co = fold_build1_loc (loc, NEGATE_EXPR, itype, co); a = fold_build2_loc (loc, POINTER_PLUS_EXPR, TREE_TYPE (fd->loops[i].v), fd->loops[i].v, co); } else if (OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) a = fold_build2_loc (loc, MINUS_EXPR, TREE_TYPE (fd->loops[i].v), fd->loops[i].v, co); else a = fold_build2_loc (loc, PLUS_EXPR, TREE_TYPE (fd->loops[i].v), fd->loops[i].v, co); if (step) { tree t1, t2; if (OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) t1 = fold_build2_loc (loc, GE_EXPR, boolean_type_node, a, fd->loops[i].n1); else t1 = fold_build2_loc (loc, LT_EXPR, boolean_type_node, a, fd->loops[i].n2); if (OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) t2 = fold_build2_loc (loc, LT_EXPR, boolean_type_node, a, fd->loops[i].n2); else t2 = fold_build2_loc (loc, GE_EXPR, boolean_type_node, a, fd->loops[i].n1); t = fold_build2_loc (loc, LT_EXPR, boolean_type_node, step, build_int_cst (TREE_TYPE (step), 0)); if (TREE_CODE (step) != INTEGER_CST) { t1 = unshare_expr (t1); t1 = force_gimple_operand_gsi (gsi, t1, true, NULL_TREE, false, GSI_CONTINUE_LINKING); t2 = unshare_expr (t2); t2 = force_gimple_operand_gsi (gsi, t2, true, NULL_TREE, false, GSI_CONTINUE_LINKING); } t = fold_build3_loc (loc, COND_EXPR, boolean_type_node, t, t2, t1); } else if (fd->loops[i].cond_code == LT_EXPR) { if (OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) t = fold_build2_loc (loc, GE_EXPR, boolean_type_node, a, fd->loops[i].n1); else t = fold_build2_loc (loc, LT_EXPR, boolean_type_node, a, fd->loops[i].n2); } else if (OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) t = fold_build2_loc (loc, GT_EXPR, boolean_type_node, a, fd->loops[i].n2); else t = fold_build2_loc (loc, LE_EXPR, boolean_type_node, a, fd->loops[i].n1); } if (cond) cond = fold_build2_loc (loc, BIT_AND_EXPR, boolean_type_node, cond, t); else cond = t; off = fold_convert_loc (loc, itype, off); if (step || (fd->loops[i].cond_code == LT_EXPR ? !integer_onep (fd->loops[i].step) : !integer_minus_onep (fd->loops[i].step))) { if (step == NULL_TREE && TYPE_UNSIGNED (itype) && fd->loops[i].cond_code == GT_EXPR) t = fold_build2_loc (loc, TRUNC_MOD_EXPR, itype, off, fold_build1_loc (loc, NEGATE_EXPR, itype, s)); else t = fold_build2_loc (loc, TRUNC_MOD_EXPR, itype, orig_off ? orig_off : off, s); t = fold_build2_loc (loc, EQ_EXPR, boolean_type_node, t, build_int_cst (itype, 0)); if (integer_zerop (t) && !warned_step) { warning_at (loc, 0, "% refers to iteration never " "in the iteration space"); warned_step = true; } cond = fold_build2_loc (loc, BIT_AND_EXPR, boolean_type_node, cond, t); } if (i <= fd->collapse - 1 && fd->collapse > 1) t = fd->loop.v; else if (counts[i]) t = counts[i]; else { t = fold_build2_loc (loc, MINUS_EXPR, TREE_TYPE (fd->loops[i].v), fd->loops[i].v, fd->loops[i].n1); t = fold_convert_loc (loc, fd->iter_type, t); } if (step) /* We have divided off by step already earlier. */; else if (TYPE_UNSIGNED (itype) && fd->loops[i].cond_code == GT_EXPR) off = fold_build2_loc (loc, TRUNC_DIV_EXPR, itype, off, fold_build1_loc (loc, NEGATE_EXPR, itype, s)); else off = fold_build2_loc (loc, TRUNC_DIV_EXPR, itype, off, s); if (OMP_CLAUSE_DEPEND_SINK_NEGATIVE (deps)) off = fold_build1_loc (loc, NEGATE_EXPR, itype, off); off = fold_convert_loc (loc, fd->iter_type, off); if (i <= fd->collapse - 1 && fd->collapse > 1) { if (i) off = fold_build2_loc (loc, PLUS_EXPR, fd->iter_type, coff, off); if (i < fd->collapse - 1) { coff = fold_build2_loc (loc, MULT_EXPR, fd->iter_type, off, counts[i]); continue; } } off = unshare_expr (off); t = fold_build2_loc (loc, PLUS_EXPR, fd->iter_type, t, off); t = force_gimple_operand_gsi (&gsi2, t, true, NULL_TREE, true, GSI_SAME_STMT); args.safe_push (t); } gimple *g = gimple_build_call_vec (builtin_decl_explicit (sink_ix), args); gimple_set_location (g, loc); gsi_insert_before (&gsi2, g, GSI_SAME_STMT); cond = unshare_expr (cond); cond = force_gimple_operand_gsi (gsi, cond, true, NULL_TREE, false, GSI_CONTINUE_LINKING); gsi_insert_after (gsi, gimple_build_cond_empty (cond), GSI_NEW_STMT); edge e3 = make_edge (e1->src, e2->dest, EDGE_FALSE_VALUE); e3->probability = REG_BR_PROB_BASE / 8; e1->probability = REG_BR_PROB_BASE - e3->probability; e1->flags = EDGE_TRUE_VALUE; set_immediate_dominator (CDI_DOMINATORS, e2->dest, e1->src); *gsi = gsi_after_labels (e2->dest); } /* Expand all #pragma omp ordered depend(source) and #pragma omp ordered depend(sink:...) constructs in the current #pragma omp for ordered(n) region. */ static void expand_omp_ordered_source_sink (struct omp_region *region, struct omp_for_data *fd, tree *counts, basic_block cont_bb) { struct omp_region *inner; int i; for (i = fd->collapse - 1; i < fd->ordered; i++) if (i == fd->collapse - 1 && fd->collapse > 1) counts[i] = NULL_TREE; else if (i >= fd->collapse && !cont_bb) counts[i] = build_zero_cst (fd->iter_type); else if (!POINTER_TYPE_P (TREE_TYPE (fd->loops[i].v)) && integer_onep (fd->loops[i].step)) counts[i] = NULL_TREE; else counts[i] = create_tmp_var (fd->iter_type, ".orditer"); tree atype = build_array_type_nelts (fd->iter_type, fd->ordered - fd->collapse + 1); counts[fd->ordered] = create_tmp_var (atype, ".orditera"); TREE_ADDRESSABLE (counts[fd->ordered]) = 1; for (inner = region->inner; inner; inner = inner->next) if (inner->type == GIMPLE_OMP_ORDERED) { gomp_ordered *ord_stmt = inner->ord_stmt; gimple_stmt_iterator gsi = gsi_for_stmt (ord_stmt); location_t loc = gimple_location (ord_stmt); tree c; for (c = gimple_omp_ordered_clauses (ord_stmt); c; c = OMP_CLAUSE_CHAIN (c)) if (OMP_CLAUSE_DEPEND_KIND (c) == OMP_CLAUSE_DEPEND_SOURCE) break; if (c) expand_omp_ordered_source (&gsi, fd, counts, loc); for (c = gimple_omp_ordered_clauses (ord_stmt); c; c = OMP_CLAUSE_CHAIN (c)) if (OMP_CLAUSE_DEPEND_KIND (c) == OMP_CLAUSE_DEPEND_SINK) expand_omp_ordered_sink (&gsi, fd, counts, c, loc); gsi_remove (&gsi, true); } } /* Wrap the body into fd->ordered - fd->collapse loops that aren't collapsed. */ static basic_block expand_omp_for_ordered_loops (struct omp_for_data *fd, tree *counts, basic_block cont_bb, basic_block body_bb, bool ordered_lastprivate) { if (fd->ordered == fd->collapse) return cont_bb; if (!cont_bb) { gimple_stmt_iterator gsi = gsi_after_labels (body_bb); for (int i = fd->collapse; i < fd->ordered; i++) { tree type = TREE_TYPE (fd->loops[i].v); tree n1 = fold_convert (type, fd->loops[i].n1); expand_omp_build_assign (&gsi, fd->loops[i].v, n1); tree aref = build4 (ARRAY_REF, fd->iter_type, counts[fd->ordered], size_int (i - fd->collapse + 1), NULL_TREE, NULL_TREE); expand_omp_build_assign (&gsi, aref, build_zero_cst (fd->iter_type)); } return NULL; } for (int i = fd->ordered - 1; i >= fd->collapse; i--) { tree t, type = TREE_TYPE (fd->loops[i].v); gimple_stmt_iterator gsi = gsi_after_labels (body_bb); expand_omp_build_assign (&gsi, fd->loops[i].v, fold_convert (type, fd->loops[i].n1)); if (counts[i]) expand_omp_build_assign (&gsi, counts[i], build_zero_cst (fd->iter_type)); tree aref = build4 (ARRAY_REF, fd->iter_type, counts[fd->ordered], size_int (i - fd->collapse + 1), NULL_TREE, NULL_TREE); expand_omp_build_assign (&gsi, aref, build_zero_cst (fd->iter_type)); if (!gsi_end_p (gsi)) gsi_prev (&gsi); else gsi = gsi_last_bb (body_bb); edge e1 = split_block (body_bb, gsi_stmt (gsi)); basic_block new_body = e1->dest; if (body_bb == cont_bb) cont_bb = new_body; edge e2 = NULL; basic_block new_header; if (EDGE_COUNT (cont_bb->preds) > 0) { gsi = gsi_last_bb (cont_bb); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (fd->loops[i].v, fold_convert (sizetype, fd->loops[i].step)); else t = fold_build2 (PLUS_EXPR, type, fd->loops[i].v, fold_convert (type, fd->loops[i].step)); expand_omp_build_assign (&gsi, fd->loops[i].v, t); if (counts[i]) { t = fold_build2 (PLUS_EXPR, fd->iter_type, counts[i], build_int_cst (fd->iter_type, 1)); expand_omp_build_assign (&gsi, counts[i], t); t = counts[i]; } else { t = fold_build2 (MINUS_EXPR, TREE_TYPE (fd->loops[i].v), fd->loops[i].v, fd->loops[i].n1); t = fold_convert (fd->iter_type, t); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); } aref = build4 (ARRAY_REF, fd->iter_type, counts[fd->ordered], size_int (i - fd->collapse + 1), NULL_TREE, NULL_TREE); expand_omp_build_assign (&gsi, aref, t); gsi_prev (&gsi); e2 = split_block (cont_bb, gsi_stmt (gsi)); new_header = e2->dest; } else new_header = cont_bb; gsi = gsi_after_labels (new_header); tree v = force_gimple_operand_gsi (&gsi, fd->loops[i].v, true, NULL_TREE, true, GSI_SAME_STMT); tree n2 = force_gimple_operand_gsi (&gsi, fold_convert (type, fd->loops[i].n2), true, NULL_TREE, true, GSI_SAME_STMT); t = build2 (fd->loops[i].cond_code, boolean_type_node, v, n2); gsi_insert_before (&gsi, gimple_build_cond_empty (t), GSI_NEW_STMT); edge e3 = split_block (new_header, gsi_stmt (gsi)); cont_bb = e3->dest; remove_edge (e1); make_edge (body_bb, new_header, EDGE_FALLTHRU); e3->flags = EDGE_FALSE_VALUE; e3->probability = REG_BR_PROB_BASE / 8; e1 = make_edge (new_header, new_body, EDGE_TRUE_VALUE); e1->probability = REG_BR_PROB_BASE - e3->probability; set_immediate_dominator (CDI_DOMINATORS, new_header, body_bb); set_immediate_dominator (CDI_DOMINATORS, new_body, new_header); if (e2) { struct loop *loop = alloc_loop (); loop->header = new_header; loop->latch = e2->src; add_loop (loop, body_bb->loop_father); } } /* If there are any lastprivate clauses and it is possible some loops might have zero iterations, ensure all the decls are initialized, otherwise we could crash evaluating C++ class iterators with lastprivate clauses. */ bool need_inits = false; for (int i = fd->collapse; ordered_lastprivate && i < fd->ordered; i++) if (need_inits) { tree type = TREE_TYPE (fd->loops[i].v); gimple_stmt_iterator gsi = gsi_after_labels (body_bb); expand_omp_build_assign (&gsi, fd->loops[i].v, fold_convert (type, fd->loops[i].n1)); } else { tree type = TREE_TYPE (fd->loops[i].v); tree this_cond = fold_build2 (fd->loops[i].cond_code, boolean_type_node, fold_convert (type, fd->loops[i].n1), fold_convert (type, fd->loops[i].n2)); if (!integer_onep (this_cond)) need_inits = true; } return cont_bb; } /* A subroutine of expand_omp_for. Generate code for a parallel loop with any schedule. Given parameters: for (V = N1; V cond N2; V += STEP) BODY; where COND is "<" or ">", we generate pseudocode more = GOMP_loop_foo_start (N1, N2, STEP, CHUNK, &istart0, &iend0); if (more) goto L0; else goto L3; L0: V = istart0; iend = iend0; L1: BODY; V += STEP; if (V cond iend) goto L1; else goto L2; L2: if (GOMP_loop_foo_next (&istart0, &iend0)) goto L0; else goto L3; L3: If this is a combined omp parallel loop, instead of the call to GOMP_loop_foo_start, we call GOMP_loop_foo_next. If this is gimple_omp_for_combined_p loop, then instead of assigning V and iend in L0 we assign the first two _looptemp_ clause decls of the inner GIMPLE_OMP_FOR and V += STEP; and if (V cond iend) goto L1; else goto L2; are removed. For collapsed loops, given parameters: collapse(3) for (V1 = N11; V1 cond1 N12; V1 += STEP1) for (V2 = N21; V2 cond2 N22; V2 += STEP2) for (V3 = N31; V3 cond3 N32; V3 += STEP3) BODY; we generate pseudocode if (__builtin_expect (N32 cond3 N31, 0)) goto Z0; if (cond3 is <) adj = STEP3 - 1; else adj = STEP3 + 1; count3 = (adj + N32 - N31) / STEP3; if (__builtin_expect (N22 cond2 N21, 0)) goto Z0; if (cond2 is <) adj = STEP2 - 1; else adj = STEP2 + 1; count2 = (adj + N22 - N21) / STEP2; if (__builtin_expect (N12 cond1 N11, 0)) goto Z0; if (cond1 is <) adj = STEP1 - 1; else adj = STEP1 + 1; count1 = (adj + N12 - N11) / STEP1; count = count1 * count2 * count3; goto Z1; Z0: count = 0; Z1: more = GOMP_loop_foo_start (0, count, 1, CHUNK, &istart0, &iend0); if (more) goto L0; else goto L3; L0: V = istart0; T = V; V3 = N31 + (T % count3) * STEP3; T = T / count3; V2 = N21 + (T % count2) * STEP2; T = T / count2; V1 = N11 + T * STEP1; iend = iend0; L1: BODY; V += 1; if (V < iend) goto L10; else goto L2; L10: V3 += STEP3; if (V3 cond3 N32) goto L1; else goto L11; L11: V3 = N31; V2 += STEP2; if (V2 cond2 N22) goto L1; else goto L12; L12: V2 = N21; V1 += STEP1; goto L1; L2: if (GOMP_loop_foo_next (&istart0, &iend0)) goto L0; else goto L3; L3: */ static void expand_omp_for_generic (struct omp_region *region, struct omp_for_data *fd, enum built_in_function start_fn, enum built_in_function next_fn, gimple *inner_stmt) { tree type, istart0, iend0, iend; tree t, vmain, vback, bias = NULL_TREE; basic_block entry_bb, cont_bb, exit_bb, l0_bb, l1_bb, collapse_bb; basic_block l2_bb = NULL, l3_bb = NULL; gimple_stmt_iterator gsi; gassign *assign_stmt; bool in_combined_parallel = is_combined_parallel (region); bool broken_loop = region->cont == NULL; edge e, ne; tree *counts = NULL; int i; bool ordered_lastprivate = false; gcc_assert (!broken_loop || !in_combined_parallel); gcc_assert (fd->iter_type == long_integer_type_node || !in_combined_parallel); entry_bb = region->entry; cont_bb = region->cont; collapse_bb = NULL; gcc_assert (EDGE_COUNT (entry_bb->succs) == 2); gcc_assert (broken_loop || BRANCH_EDGE (entry_bb)->dest == FALLTHRU_EDGE (cont_bb)->dest); l0_bb = split_edge (FALLTHRU_EDGE (entry_bb)); l1_bb = single_succ (l0_bb); if (!broken_loop) { l2_bb = create_empty_bb (cont_bb); gcc_assert (BRANCH_EDGE (cont_bb)->dest == l1_bb || (single_succ_edge (BRANCH_EDGE (cont_bb)->dest)->dest == l1_bb)); gcc_assert (EDGE_COUNT (cont_bb->succs) == 2); } else l2_bb = NULL; l3_bb = BRANCH_EDGE (entry_bb)->dest; exit_bb = region->exit; gsi = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); if (fd->ordered && omp_find_clause (gimple_omp_for_clauses (gsi_stmt (gsi)), OMP_CLAUSE_LASTPRIVATE)) ordered_lastprivate = false; if (fd->collapse > 1 || fd->ordered) { int first_zero_iter1 = -1, first_zero_iter2 = -1; basic_block zero_iter1_bb = NULL, zero_iter2_bb = NULL, l2_dom_bb = NULL; counts = XALLOCAVEC (tree, fd->ordered ? fd->ordered + 1 : fd->collapse); expand_omp_for_init_counts (fd, &gsi, entry_bb, counts, zero_iter1_bb, first_zero_iter1, zero_iter2_bb, first_zero_iter2, l2_dom_bb); if (zero_iter1_bb) { /* Some counts[i] vars might be uninitialized if some loop has zero iterations. But the body shouldn't be executed in that case, so just avoid uninit warnings. */ for (i = first_zero_iter1; i < (fd->ordered ? fd->ordered : fd->collapse); i++) if (SSA_VAR_P (counts[i])) TREE_NO_WARNING (counts[i]) = 1; gsi_prev (&gsi); e = split_block (entry_bb, gsi_stmt (gsi)); entry_bb = e->dest; make_edge (zero_iter1_bb, entry_bb, EDGE_FALLTHRU); gsi = gsi_last_bb (entry_bb); set_immediate_dominator (CDI_DOMINATORS, entry_bb, get_immediate_dominator (CDI_DOMINATORS, zero_iter1_bb)); } if (zero_iter2_bb) { /* Some counts[i] vars might be uninitialized if some loop has zero iterations. But the body shouldn't be executed in that case, so just avoid uninit warnings. */ for (i = first_zero_iter2; i < fd->ordered; i++) if (SSA_VAR_P (counts[i])) TREE_NO_WARNING (counts[i]) = 1; if (zero_iter1_bb) make_edge (zero_iter2_bb, entry_bb, EDGE_FALLTHRU); else { gsi_prev (&gsi); e = split_block (entry_bb, gsi_stmt (gsi)); entry_bb = e->dest; make_edge (zero_iter2_bb, entry_bb, EDGE_FALLTHRU); gsi = gsi_last_bb (entry_bb); set_immediate_dominator (CDI_DOMINATORS, entry_bb, get_immediate_dominator (CDI_DOMINATORS, zero_iter2_bb)); } } if (fd->collapse == 1) { counts[0] = fd->loop.n2; fd->loop = fd->loops[0]; } } type = TREE_TYPE (fd->loop.v); istart0 = create_tmp_var (fd->iter_type, ".istart0"); iend0 = create_tmp_var (fd->iter_type, ".iend0"); TREE_ADDRESSABLE (istart0) = 1; TREE_ADDRESSABLE (iend0) = 1; /* See if we need to bias by LLONG_MIN. */ if (fd->iter_type == long_long_unsigned_type_node && TREE_CODE (type) == INTEGER_TYPE && !TYPE_UNSIGNED (type) && fd->ordered == 0) { tree n1, n2; if (fd->loop.cond_code == LT_EXPR) { n1 = fd->loop.n1; n2 = fold_build2 (PLUS_EXPR, type, fd->loop.n2, fd->loop.step); } else { n1 = fold_build2 (MINUS_EXPR, type, fd->loop.n2, fd->loop.step); n2 = fd->loop.n1; } if (TREE_CODE (n1) != INTEGER_CST || TREE_CODE (n2) != INTEGER_CST || ((tree_int_cst_sgn (n1) < 0) ^ (tree_int_cst_sgn (n2) < 0))) bias = fold_convert (fd->iter_type, TYPE_MIN_VALUE (type)); } gimple_stmt_iterator gsif = gsi; gsi_prev (&gsif); tree arr = NULL_TREE; if (in_combined_parallel) { gcc_assert (fd->ordered == 0); /* In a combined parallel loop, emit a call to GOMP_loop_foo_next. */ t = build_call_expr (builtin_decl_explicit (next_fn), 2, build_fold_addr_expr (istart0), build_fold_addr_expr (iend0)); } else { tree t0, t1, t2, t3, t4; /* If this is not a combined parallel loop, emit a call to GOMP_loop_foo_start in ENTRY_BB. */ t4 = build_fold_addr_expr (iend0); t3 = build_fold_addr_expr (istart0); if (fd->ordered) { t0 = build_int_cst (unsigned_type_node, fd->ordered - fd->collapse + 1); arr = create_tmp_var (build_array_type_nelts (fd->iter_type, fd->ordered - fd->collapse + 1), ".omp_counts"); DECL_NAMELESS (arr) = 1; TREE_ADDRESSABLE (arr) = 1; TREE_STATIC (arr) = 1; vec *v; vec_alloc (v, fd->ordered - fd->collapse + 1); int idx; for (idx = 0; idx < fd->ordered - fd->collapse + 1; idx++) { tree c; if (idx == 0 && fd->collapse > 1) c = fd->loop.n2; else c = counts[idx + fd->collapse - 1]; tree purpose = size_int (idx); CONSTRUCTOR_APPEND_ELT (v, purpose, c); if (TREE_CODE (c) != INTEGER_CST) TREE_STATIC (arr) = 0; } DECL_INITIAL (arr) = build_constructor (TREE_TYPE (arr), v); if (!TREE_STATIC (arr)) force_gimple_operand_gsi (&gsi, build1 (DECL_EXPR, void_type_node, arr), true, NULL_TREE, true, GSI_SAME_STMT); t1 = build_fold_addr_expr (arr); t2 = NULL_TREE; } else { t2 = fold_convert (fd->iter_type, fd->loop.step); t1 = fd->loop.n2; t0 = fd->loop.n1; if (gimple_omp_for_combined_into_p (fd->for_stmt)) { tree innerc = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); t0 = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); t1 = OMP_CLAUSE_DECL (innerc); } if (POINTER_TYPE_P (TREE_TYPE (t0)) && TYPE_PRECISION (TREE_TYPE (t0)) != TYPE_PRECISION (fd->iter_type)) { /* Avoid casting pointers to integer of a different size. */ tree itype = signed_type_for (type); t1 = fold_convert (fd->iter_type, fold_convert (itype, t1)); t0 = fold_convert (fd->iter_type, fold_convert (itype, t0)); } else { t1 = fold_convert (fd->iter_type, t1); t0 = fold_convert (fd->iter_type, t0); } if (bias) { t1 = fold_build2 (PLUS_EXPR, fd->iter_type, t1, bias); t0 = fold_build2 (PLUS_EXPR, fd->iter_type, t0, bias); } } if (fd->iter_type == long_integer_type_node || fd->ordered) { if (fd->chunk_size) { t = fold_convert (fd->iter_type, fd->chunk_size); t = omp_adjust_chunk_size (t, fd->simd_schedule); if (fd->ordered) t = build_call_expr (builtin_decl_explicit (start_fn), 5, t0, t1, t, t3, t4); else t = build_call_expr (builtin_decl_explicit (start_fn), 6, t0, t1, t2, t, t3, t4); } else if (fd->ordered) t = build_call_expr (builtin_decl_explicit (start_fn), 4, t0, t1, t3, t4); else t = build_call_expr (builtin_decl_explicit (start_fn), 5, t0, t1, t2, t3, t4); } else { tree t5; tree c_bool_type; tree bfn_decl; /* The GOMP_loop_ull_*start functions have additional boolean argument, true for < loops and false for > loops. In Fortran, the C bool type can be different from boolean_type_node. */ bfn_decl = builtin_decl_explicit (start_fn); c_bool_type = TREE_TYPE (TREE_TYPE (bfn_decl)); t5 = build_int_cst (c_bool_type, fd->loop.cond_code == LT_EXPR ? 1 : 0); if (fd->chunk_size) { tree bfn_decl = builtin_decl_explicit (start_fn); t = fold_convert (fd->iter_type, fd->chunk_size); t = omp_adjust_chunk_size (t, fd->simd_schedule); t = build_call_expr (bfn_decl, 7, t5, t0, t1, t2, t, t3, t4); } else t = build_call_expr (builtin_decl_explicit (start_fn), 6, t5, t0, t1, t2, t3, t4); } } if (TREE_TYPE (t) != boolean_type_node) t = fold_build2 (NE_EXPR, boolean_type_node, t, build_int_cst (TREE_TYPE (t), 0)); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); if (arr && !TREE_STATIC (arr)) { tree clobber = build_constructor (TREE_TYPE (arr), NULL); TREE_THIS_VOLATILE (clobber) = 1; gsi_insert_before (&gsi, gimple_build_assign (arr, clobber), GSI_SAME_STMT); } gsi_insert_after (&gsi, gimple_build_cond_empty (t), GSI_SAME_STMT); /* Remove the GIMPLE_OMP_FOR statement. */ gsi_remove (&gsi, true); if (gsi_end_p (gsif)) gsif = gsi_after_labels (gsi_bb (gsif)); gsi_next (&gsif); /* Iteration setup for sequential loop goes in L0_BB. */ tree startvar = fd->loop.v; tree endvar = NULL_TREE; if (gimple_omp_for_combined_p (fd->for_stmt)) { gcc_assert (gimple_code (inner_stmt) == GIMPLE_OMP_FOR && gimple_omp_for_kind (inner_stmt) == GF_OMP_FOR_KIND_SIMD); tree innerc = omp_find_clause (gimple_omp_for_clauses (inner_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); startvar = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); endvar = OMP_CLAUSE_DECL (innerc); } gsi = gsi_start_bb (l0_bb); t = istart0; if (fd->ordered && fd->collapse == 1) t = fold_build2 (MULT_EXPR, fd->iter_type, t, fold_convert (fd->iter_type, fd->loop.step)); else if (bias) t = fold_build2 (MINUS_EXPR, fd->iter_type, t, bias); if (fd->ordered && fd->collapse == 1) { if (POINTER_TYPE_P (TREE_TYPE (startvar))) t = fold_build2 (POINTER_PLUS_EXPR, TREE_TYPE (startvar), fd->loop.n1, fold_convert (sizetype, t)); else { t = fold_convert (TREE_TYPE (startvar), t); t = fold_build2 (PLUS_EXPR, TREE_TYPE (startvar), fd->loop.n1, t); } } else { if (POINTER_TYPE_P (TREE_TYPE (startvar))) t = fold_convert (signed_type_for (TREE_TYPE (startvar)), t); t = fold_convert (TREE_TYPE (startvar), t); } t = force_gimple_operand_gsi (&gsi, t, DECL_P (startvar) && TREE_ADDRESSABLE (startvar), NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (startvar, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); t = iend0; if (fd->ordered && fd->collapse == 1) t = fold_build2 (MULT_EXPR, fd->iter_type, t, fold_convert (fd->iter_type, fd->loop.step)); else if (bias) t = fold_build2 (MINUS_EXPR, fd->iter_type, t, bias); if (fd->ordered && fd->collapse == 1) { if (POINTER_TYPE_P (TREE_TYPE (startvar))) t = fold_build2 (POINTER_PLUS_EXPR, TREE_TYPE (startvar), fd->loop.n1, fold_convert (sizetype, t)); else { t = fold_convert (TREE_TYPE (startvar), t); t = fold_build2 (PLUS_EXPR, TREE_TYPE (startvar), fd->loop.n1, t); } } else { if (POINTER_TYPE_P (TREE_TYPE (startvar))) t = fold_convert (signed_type_for (TREE_TYPE (startvar)), t); t = fold_convert (TREE_TYPE (startvar), t); } iend = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); if (endvar) { assign_stmt = gimple_build_assign (endvar, iend); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); if (useless_type_conversion_p (TREE_TYPE (fd->loop.v), TREE_TYPE (iend))) assign_stmt = gimple_build_assign (fd->loop.v, iend); else assign_stmt = gimple_build_assign (fd->loop.v, NOP_EXPR, iend); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } /* Handle linear clause adjustments. */ tree itercnt = NULL_TREE; if (gimple_omp_for_kind (fd->for_stmt) == GF_OMP_FOR_KIND_FOR) for (tree c = gimple_omp_for_clauses (fd->for_stmt); c; c = OMP_CLAUSE_CHAIN (c)) if (OMP_CLAUSE_CODE (c) == OMP_CLAUSE_LINEAR && !OMP_CLAUSE_LINEAR_NO_COPYIN (c)) { tree d = OMP_CLAUSE_DECL (c); bool is_ref = omp_is_reference (d); tree t = d, a, dest; if (is_ref) t = build_simple_mem_ref_loc (OMP_CLAUSE_LOCATION (c), t); tree type = TREE_TYPE (t); if (POINTER_TYPE_P (type)) type = sizetype; dest = unshare_expr (t); tree v = create_tmp_var (TREE_TYPE (t), NULL); expand_omp_build_assign (&gsif, v, t); if (itercnt == NULL_TREE) { itercnt = startvar; tree n1 = fd->loop.n1; if (POINTER_TYPE_P (TREE_TYPE (itercnt))) { itercnt = fold_convert (signed_type_for (TREE_TYPE (itercnt)), itercnt); n1 = fold_convert (TREE_TYPE (itercnt), n1); } itercnt = fold_build2 (MINUS_EXPR, TREE_TYPE (itercnt), itercnt, n1); itercnt = fold_build2 (EXACT_DIV_EXPR, TREE_TYPE (itercnt), itercnt, fd->loop.step); itercnt = force_gimple_operand_gsi (&gsi, itercnt, true, NULL_TREE, false, GSI_CONTINUE_LINKING); } a = fold_build2 (MULT_EXPR, type, fold_convert (type, itercnt), fold_convert (type, OMP_CLAUSE_LINEAR_STEP (c))); t = fold_build2 (type == TREE_TYPE (t) ? PLUS_EXPR : POINTER_PLUS_EXPR, TREE_TYPE (t), v, a); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (dest, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } if (fd->collapse > 1) expand_omp_for_init_vars (fd, &gsi, counts, inner_stmt, startvar); if (fd->ordered) { /* Until now, counts array contained number of iterations or variable containing it for ith loop. From now on, we need those counts only for collapsed loops, and only for the 2nd till the last collapsed one. Move those one element earlier, we'll use counts[fd->collapse - 1] for the first source/sink iteration counter and so on and counts[fd->ordered] as the array holding the current counter values for depend(source). */ if (fd->collapse > 1) memmove (counts, counts + 1, (fd->collapse - 1) * sizeof (counts[0])); if (broken_loop) { int i; for (i = fd->collapse; i < fd->ordered; i++) { tree type = TREE_TYPE (fd->loops[i].v); tree this_cond = fold_build2 (fd->loops[i].cond_code, boolean_type_node, fold_convert (type, fd->loops[i].n1), fold_convert (type, fd->loops[i].n2)); if (!integer_onep (this_cond)) break; } if (i < fd->ordered) { cont_bb = create_empty_bb (EXIT_BLOCK_PTR_FOR_FN (cfun)->prev_bb); add_bb_to_loop (cont_bb, l1_bb->loop_father); gimple_stmt_iterator gsi = gsi_after_labels (cont_bb); gimple *g = gimple_build_omp_continue (fd->loop.v, fd->loop.v); gsi_insert_before (&gsi, g, GSI_SAME_STMT); make_edge (cont_bb, l3_bb, EDGE_FALLTHRU); make_edge (cont_bb, l1_bb, 0); l2_bb = create_empty_bb (cont_bb); broken_loop = false; } } expand_omp_ordered_source_sink (region, fd, counts, cont_bb); cont_bb = expand_omp_for_ordered_loops (fd, counts, cont_bb, l1_bb, ordered_lastprivate); if (counts[fd->collapse - 1]) { gcc_assert (fd->collapse == 1); gsi = gsi_last_bb (l0_bb); expand_omp_build_assign (&gsi, counts[fd->collapse - 1], istart0, true); gsi = gsi_last_bb (cont_bb); t = fold_build2 (PLUS_EXPR, fd->iter_type, counts[fd->collapse - 1], build_int_cst (fd->iter_type, 1)); expand_omp_build_assign (&gsi, counts[fd->collapse - 1], t); tree aref = build4 (ARRAY_REF, fd->iter_type, counts[fd->ordered], size_zero_node, NULL_TREE, NULL_TREE); expand_omp_build_assign (&gsi, aref, counts[fd->collapse - 1]); t = counts[fd->collapse - 1]; } else if (fd->collapse > 1) t = fd->loop.v; else { t = fold_build2 (MINUS_EXPR, TREE_TYPE (fd->loops[0].v), fd->loops[0].v, fd->loops[0].n1); t = fold_convert (fd->iter_type, t); } gsi = gsi_last_bb (l0_bb); tree aref = build4 (ARRAY_REF, fd->iter_type, counts[fd->ordered], size_zero_node, NULL_TREE, NULL_TREE); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); expand_omp_build_assign (&gsi, aref, t, true); } if (!broken_loop) { /* Code to control the increment and predicate for the sequential loop goes in the CONT_BB. */ gsi = gsi_last_bb (cont_bb); gomp_continue *cont_stmt = as_a (gsi_stmt (gsi)); gcc_assert (gimple_code (cont_stmt) == GIMPLE_OMP_CONTINUE); vmain = gimple_omp_continue_control_use (cont_stmt); vback = gimple_omp_continue_control_def (cont_stmt); if (!gimple_omp_for_combined_p (fd->for_stmt)) { if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (vmain, fd->loop.step); else t = fold_build2 (PLUS_EXPR, type, vmain, fd->loop.step); t = force_gimple_operand_gsi (&gsi, t, DECL_P (vback) && TREE_ADDRESSABLE (vback), NULL_TREE, true, GSI_SAME_STMT); assign_stmt = gimple_build_assign (vback, t); gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT); if (fd->ordered && counts[fd->collapse - 1] == NULL_TREE) { if (fd->collapse > 1) t = fd->loop.v; else { t = fold_build2 (MINUS_EXPR, TREE_TYPE (fd->loops[0].v), fd->loops[0].v, fd->loops[0].n1); t = fold_convert (fd->iter_type, t); } tree aref = build4 (ARRAY_REF, fd->iter_type, counts[fd->ordered], size_zero_node, NULL_TREE, NULL_TREE); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); expand_omp_build_assign (&gsi, aref, t); } t = build2 (fd->loop.cond_code, boolean_type_node, DECL_P (vback) && TREE_ADDRESSABLE (vback) ? t : vback, iend); gcond *cond_stmt = gimple_build_cond_empty (t); gsi_insert_before (&gsi, cond_stmt, GSI_SAME_STMT); } /* Remove GIMPLE_OMP_CONTINUE. */ gsi_remove (&gsi, true); if (fd->collapse > 1 && !gimple_omp_for_combined_p (fd->for_stmt)) collapse_bb = extract_omp_for_update_vars (fd, cont_bb, l1_bb); /* Emit code to get the next parallel iteration in L2_BB. */ gsi = gsi_start_bb (l2_bb); t = build_call_expr (builtin_decl_explicit (next_fn), 2, build_fold_addr_expr (istart0), build_fold_addr_expr (iend0)); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); if (TREE_TYPE (t) != boolean_type_node) t = fold_build2 (NE_EXPR, boolean_type_node, t, build_int_cst (TREE_TYPE (t), 0)); gcond *cond_stmt = gimple_build_cond_empty (t); gsi_insert_after (&gsi, cond_stmt, GSI_CONTINUE_LINKING); } /* Add the loop cleanup function. */ gsi = gsi_last_bb (exit_bb); if (gimple_omp_return_nowait_p (gsi_stmt (gsi))) t = builtin_decl_explicit (BUILT_IN_GOMP_LOOP_END_NOWAIT); else if (gimple_omp_return_lhs (gsi_stmt (gsi))) t = builtin_decl_explicit (BUILT_IN_GOMP_LOOP_END_CANCEL); else t = builtin_decl_explicit (BUILT_IN_GOMP_LOOP_END); gcall *call_stmt = gimple_build_call (t, 0); if (gimple_omp_return_lhs (gsi_stmt (gsi))) gimple_call_set_lhs (call_stmt, gimple_omp_return_lhs (gsi_stmt (gsi))); gsi_insert_after (&gsi, call_stmt, GSI_SAME_STMT); if (fd->ordered) { tree arr = counts[fd->ordered]; tree clobber = build_constructor (TREE_TYPE (arr), NULL); TREE_THIS_VOLATILE (clobber) = 1; gsi_insert_after (&gsi, gimple_build_assign (arr, clobber), GSI_SAME_STMT); } gsi_remove (&gsi, true); /* Connect the new blocks. */ find_edge (entry_bb, l0_bb)->flags = EDGE_TRUE_VALUE; find_edge (entry_bb, l3_bb)->flags = EDGE_FALSE_VALUE; if (!broken_loop) { gimple_seq phis; e = find_edge (cont_bb, l3_bb); ne = make_edge (l2_bb, l3_bb, EDGE_FALSE_VALUE); phis = phi_nodes (l3_bb); for (gsi = gsi_start (phis); !gsi_end_p (gsi); gsi_next (&gsi)) { gimple *phi = gsi_stmt (gsi); SET_USE (PHI_ARG_DEF_PTR_FROM_EDGE (phi, ne), PHI_ARG_DEF_FROM_EDGE (phi, e)); } remove_edge (e); make_edge (cont_bb, l2_bb, EDGE_FALSE_VALUE); e = find_edge (cont_bb, l1_bb); if (e == NULL) { e = BRANCH_EDGE (cont_bb); gcc_assert (single_succ (e->dest) == l1_bb); } if (gimple_omp_for_combined_p (fd->for_stmt)) { remove_edge (e); e = NULL; } else if (fd->collapse > 1) { remove_edge (e); e = make_edge (cont_bb, collapse_bb, EDGE_TRUE_VALUE); } else e->flags = EDGE_TRUE_VALUE; if (e) { e->probability = REG_BR_PROB_BASE * 7 / 8; find_edge (cont_bb, l2_bb)->probability = REG_BR_PROB_BASE / 8; } else { e = find_edge (cont_bb, l2_bb); e->flags = EDGE_FALLTHRU; } make_edge (l2_bb, l0_bb, EDGE_TRUE_VALUE); if (gimple_in_ssa_p (cfun)) { /* Add phis to the outer loop that connect to the phis in the inner, original loop, and move the loop entry value of the inner phi to the loop entry value of the outer phi. */ gphi_iterator psi; for (psi = gsi_start_phis (l3_bb); !gsi_end_p (psi); gsi_next (&psi)) { source_location locus; gphi *nphi; gphi *exit_phi = psi.phi (); edge l2_to_l3 = find_edge (l2_bb, l3_bb); tree exit_res = PHI_ARG_DEF_FROM_EDGE (exit_phi, l2_to_l3); basic_block latch = BRANCH_EDGE (cont_bb)->dest; edge latch_to_l1 = find_edge (latch, l1_bb); gphi *inner_phi = find_phi_with_arg_on_edge (exit_res, latch_to_l1); tree t = gimple_phi_result (exit_phi); tree new_res = copy_ssa_name (t, NULL); nphi = create_phi_node (new_res, l0_bb); edge l0_to_l1 = find_edge (l0_bb, l1_bb); t = PHI_ARG_DEF_FROM_EDGE (inner_phi, l0_to_l1); locus = gimple_phi_arg_location_from_edge (inner_phi, l0_to_l1); edge entry_to_l0 = find_edge (entry_bb, l0_bb); add_phi_arg (nphi, t, entry_to_l0, locus); edge l2_to_l0 = find_edge (l2_bb, l0_bb); add_phi_arg (nphi, exit_res, l2_to_l0, UNKNOWN_LOCATION); add_phi_arg (inner_phi, new_res, l0_to_l1, UNKNOWN_LOCATION); }; } set_immediate_dominator (CDI_DOMINATORS, l2_bb, recompute_dominator (CDI_DOMINATORS, l2_bb)); set_immediate_dominator (CDI_DOMINATORS, l3_bb, recompute_dominator (CDI_DOMINATORS, l3_bb)); set_immediate_dominator (CDI_DOMINATORS, l0_bb, recompute_dominator (CDI_DOMINATORS, l0_bb)); set_immediate_dominator (CDI_DOMINATORS, l1_bb, recompute_dominator (CDI_DOMINATORS, l1_bb)); /* We enter expand_omp_for_generic with a loop. This original loop may have its own loop struct, or it may be part of an outer loop struct (which may be the fake loop). */ struct loop *outer_loop = entry_bb->loop_father; bool orig_loop_has_loop_struct = l1_bb->loop_father != outer_loop; add_bb_to_loop (l2_bb, outer_loop); /* We've added a new loop around the original loop. Allocate the corresponding loop struct. */ struct loop *new_loop = alloc_loop (); new_loop->header = l0_bb; new_loop->latch = l2_bb; add_loop (new_loop, outer_loop); /* Allocate a loop structure for the original loop unless we already had one. */ if (!orig_loop_has_loop_struct && !gimple_omp_for_combined_p (fd->for_stmt)) { struct loop *orig_loop = alloc_loop (); orig_loop->header = l1_bb; /* The loop may have multiple latches. */ add_loop (orig_loop, new_loop); } } } /* A subroutine of expand_omp_for. Generate code for a parallel loop with static schedule and no specified chunk size. Given parameters: for (V = N1; V cond N2; V += STEP) BODY; where COND is "<" or ">", we generate pseudocode if ((__typeof (V)) -1 > 0 && N2 cond N1) goto L2; if (cond is <) adj = STEP - 1; else adj = STEP + 1; if ((__typeof (V)) -1 > 0 && cond is >) n = -(adj + N2 - N1) / -STEP; else n = (adj + N2 - N1) / STEP; q = n / nthreads; tt = n % nthreads; if (threadid < tt) goto L3; else goto L4; L3: tt = 0; q = q + 1; L4: s0 = q * threadid + tt; e0 = s0 + q; V = s0 * STEP + N1; if (s0 >= e0) goto L2; else goto L0; L0: e = e0 * STEP + N1; L1: BODY; V += STEP; if (V cond e) goto L1; L2: */ static void expand_omp_for_static_nochunk (struct omp_region *region, struct omp_for_data *fd, gimple *inner_stmt) { tree n, q, s0, e0, e, t, tt, nthreads, threadid; tree type, itype, vmain, vback; basic_block entry_bb, second_bb, third_bb, exit_bb, seq_start_bb; basic_block body_bb, cont_bb, collapse_bb = NULL; basic_block fin_bb; gimple_stmt_iterator gsi; edge ep; bool broken_loop = region->cont == NULL; tree *counts = NULL; tree n1, n2, step; itype = type = TREE_TYPE (fd->loop.v); if (POINTER_TYPE_P (type)) itype = signed_type_for (type); entry_bb = region->entry; cont_bb = region->cont; gcc_assert (EDGE_COUNT (entry_bb->succs) == 2); fin_bb = BRANCH_EDGE (entry_bb)->dest; gcc_assert (broken_loop || (fin_bb == FALLTHRU_EDGE (cont_bb)->dest)); seq_start_bb = split_edge (FALLTHRU_EDGE (entry_bb)); body_bb = single_succ (seq_start_bb); if (!broken_loop) { gcc_assert (BRANCH_EDGE (cont_bb)->dest == body_bb || single_succ (BRANCH_EDGE (cont_bb)->dest) == body_bb); gcc_assert (EDGE_COUNT (cont_bb->succs) == 2); } exit_bb = region->exit; /* Iteration space partitioning goes in ENTRY_BB. */ gsi = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); if (fd->collapse > 1) { int first_zero_iter = -1, dummy = -1; basic_block l2_dom_bb = NULL, dummy_bb = NULL; counts = XALLOCAVEC (tree, fd->collapse); expand_omp_for_init_counts (fd, &gsi, entry_bb, counts, fin_bb, first_zero_iter, dummy_bb, dummy, l2_dom_bb); t = NULL_TREE; } else if (gimple_omp_for_combined_into_p (fd->for_stmt)) t = integer_one_node; else t = fold_binary (fd->loop.cond_code, boolean_type_node, fold_convert (type, fd->loop.n1), fold_convert (type, fd->loop.n2)); if (fd->collapse == 1 && TYPE_UNSIGNED (type) && (t == NULL_TREE || !integer_onep (t))) { n1 = fold_convert (type, unshare_expr (fd->loop.n1)); n1 = force_gimple_operand_gsi (&gsi, n1, true, NULL_TREE, true, GSI_SAME_STMT); n2 = fold_convert (type, unshare_expr (fd->loop.n2)); n2 = force_gimple_operand_gsi (&gsi, n2, true, NULL_TREE, true, GSI_SAME_STMT); gcond *cond_stmt = gimple_build_cond (fd->loop.cond_code, n1, n2, NULL_TREE, NULL_TREE); gsi_insert_before (&gsi, cond_stmt, GSI_SAME_STMT); if (walk_tree (gimple_cond_lhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL) || walk_tree (gimple_cond_rhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL)) { gsi = gsi_for_stmt (cond_stmt); gimple_regimplify_operands (cond_stmt, &gsi); } ep = split_block (entry_bb, cond_stmt); ep->flags = EDGE_TRUE_VALUE; entry_bb = ep->dest; ep->probability = REG_BR_PROB_BASE - (REG_BR_PROB_BASE / 2000 - 1); ep = make_edge (ep->src, fin_bb, EDGE_FALSE_VALUE); ep->probability = REG_BR_PROB_BASE / 2000 - 1; if (gimple_in_ssa_p (cfun)) { int dest_idx = find_edge (entry_bb, fin_bb)->dest_idx; for (gphi_iterator gpi = gsi_start_phis (fin_bb); !gsi_end_p (gpi); gsi_next (&gpi)) { gphi *phi = gpi.phi (); add_phi_arg (phi, gimple_phi_arg_def (phi, dest_idx), ep, UNKNOWN_LOCATION); } } gsi = gsi_last_bb (entry_bb); } switch (gimple_omp_for_kind (fd->for_stmt)) { case GF_OMP_FOR_KIND_FOR: nthreads = builtin_decl_explicit (BUILT_IN_OMP_GET_NUM_THREADS); threadid = builtin_decl_explicit (BUILT_IN_OMP_GET_THREAD_NUM); break; case GF_OMP_FOR_KIND_DISTRIBUTE: nthreads = builtin_decl_explicit (BUILT_IN_OMP_GET_NUM_TEAMS); threadid = builtin_decl_explicit (BUILT_IN_OMP_GET_TEAM_NUM); break; default: gcc_unreachable (); } nthreads = build_call_expr (nthreads, 0); nthreads = fold_convert (itype, nthreads); nthreads = force_gimple_operand_gsi (&gsi, nthreads, true, NULL_TREE, true, GSI_SAME_STMT); threadid = build_call_expr (threadid, 0); threadid = fold_convert (itype, threadid); threadid = force_gimple_operand_gsi (&gsi, threadid, true, NULL_TREE, true, GSI_SAME_STMT); n1 = fd->loop.n1; n2 = fd->loop.n2; step = fd->loop.step; if (gimple_omp_for_combined_into_p (fd->for_stmt)) { tree innerc = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n1 = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n2 = OMP_CLAUSE_DECL (innerc); } n1 = force_gimple_operand_gsi (&gsi, fold_convert (type, n1), true, NULL_TREE, true, GSI_SAME_STMT); n2 = force_gimple_operand_gsi (&gsi, fold_convert (itype, n2), true, NULL_TREE, true, GSI_SAME_STMT); step = force_gimple_operand_gsi (&gsi, fold_convert (itype, step), true, NULL_TREE, true, GSI_SAME_STMT); t = build_int_cst (itype, (fd->loop.cond_code == LT_EXPR ? -1 : 1)); t = fold_build2 (PLUS_EXPR, itype, step, t); t = fold_build2 (PLUS_EXPR, itype, t, n2); t = fold_build2 (MINUS_EXPR, itype, t, fold_convert (itype, n1)); if (TYPE_UNSIGNED (itype) && fd->loop.cond_code == GT_EXPR) t = fold_build2 (TRUNC_DIV_EXPR, itype, fold_build1 (NEGATE_EXPR, itype, t), fold_build1 (NEGATE_EXPR, itype, step)); else t = fold_build2 (TRUNC_DIV_EXPR, itype, t, step); t = fold_convert (itype, t); n = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); q = create_tmp_reg (itype, "q"); t = fold_build2 (TRUNC_DIV_EXPR, itype, n, nthreads); t = force_gimple_operand_gsi (&gsi, t, false, NULL_TREE, true, GSI_SAME_STMT); gsi_insert_before (&gsi, gimple_build_assign (q, t), GSI_SAME_STMT); tt = create_tmp_reg (itype, "tt"); t = fold_build2 (TRUNC_MOD_EXPR, itype, n, nthreads); t = force_gimple_operand_gsi (&gsi, t, false, NULL_TREE, true, GSI_SAME_STMT); gsi_insert_before (&gsi, gimple_build_assign (tt, t), GSI_SAME_STMT); t = build2 (LT_EXPR, boolean_type_node, threadid, tt); gcond *cond_stmt = gimple_build_cond_empty (t); gsi_insert_before (&gsi, cond_stmt, GSI_SAME_STMT); second_bb = split_block (entry_bb, cond_stmt)->dest; gsi = gsi_last_bb (second_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); gsi_insert_before (&gsi, gimple_build_assign (tt, build_int_cst (itype, 0)), GSI_SAME_STMT); gassign *assign_stmt = gimple_build_assign (q, PLUS_EXPR, q, build_int_cst (itype, 1)); gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT); third_bb = split_block (second_bb, assign_stmt)->dest; gsi = gsi_last_bb (third_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); t = build2 (MULT_EXPR, itype, q, threadid); t = build2 (PLUS_EXPR, itype, t, tt); s0 = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); t = fold_build2 (PLUS_EXPR, itype, s0, q); e0 = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); t = build2 (GE_EXPR, boolean_type_node, s0, e0); gsi_insert_before (&gsi, gimple_build_cond_empty (t), GSI_SAME_STMT); /* Remove the GIMPLE_OMP_FOR statement. */ gsi_remove (&gsi, true); /* Setup code for sequential iteration goes in SEQ_START_BB. */ gsi = gsi_start_bb (seq_start_bb); tree startvar = fd->loop.v; tree endvar = NULL_TREE; if (gimple_omp_for_combined_p (fd->for_stmt)) { tree clauses = gimple_code (inner_stmt) == GIMPLE_OMP_PARALLEL ? gimple_omp_parallel_clauses (inner_stmt) : gimple_omp_for_clauses (inner_stmt); tree innerc = omp_find_clause (clauses, OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); startvar = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); endvar = OMP_CLAUSE_DECL (innerc); if (fd->collapse > 1 && TREE_CODE (fd->loop.n2) != INTEGER_CST && gimple_omp_for_kind (fd->for_stmt) == GF_OMP_FOR_KIND_DISTRIBUTE) { int i; for (i = 1; i < fd->collapse; i++) { innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); } innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); if (innerc) { /* If needed (distribute parallel for with lastprivate), propagate down the total number of iterations. */ tree t = fold_convert (TREE_TYPE (OMP_CLAUSE_DECL (innerc)), fd->loop.n2); t = force_gimple_operand_gsi (&gsi, t, false, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (OMP_CLAUSE_DECL (innerc), t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } } } t = fold_convert (itype, s0); t = fold_build2 (MULT_EXPR, itype, t, step); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (n1, t); else t = fold_build2 (PLUS_EXPR, type, t, n1); t = fold_convert (TREE_TYPE (startvar), t); t = force_gimple_operand_gsi (&gsi, t, DECL_P (startvar) && TREE_ADDRESSABLE (startvar), NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (startvar, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); t = fold_convert (itype, e0); t = fold_build2 (MULT_EXPR, itype, t, step); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (n1, t); else t = fold_build2 (PLUS_EXPR, type, t, n1); t = fold_convert (TREE_TYPE (startvar), t); e = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); if (endvar) { assign_stmt = gimple_build_assign (endvar, e); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); if (useless_type_conversion_p (TREE_TYPE (fd->loop.v), TREE_TYPE (e))) assign_stmt = gimple_build_assign (fd->loop.v, e); else assign_stmt = gimple_build_assign (fd->loop.v, NOP_EXPR, e); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } /* Handle linear clause adjustments. */ tree itercnt = NULL_TREE; if (gimple_omp_for_kind (fd->for_stmt) == GF_OMP_FOR_KIND_FOR) for (tree c = gimple_omp_for_clauses (fd->for_stmt); c; c = OMP_CLAUSE_CHAIN (c)) if (OMP_CLAUSE_CODE (c) == OMP_CLAUSE_LINEAR && !OMP_CLAUSE_LINEAR_NO_COPYIN (c)) { tree d = OMP_CLAUSE_DECL (c); bool is_ref = omp_is_reference (d); tree t = d, a, dest; if (is_ref) t = build_simple_mem_ref_loc (OMP_CLAUSE_LOCATION (c), t); if (itercnt == NULL_TREE) { if (gimple_omp_for_combined_into_p (fd->for_stmt)) { itercnt = fold_build2 (MINUS_EXPR, itype, fold_convert (itype, n1), fold_convert (itype, fd->loop.n1)); itercnt = fold_build2 (EXACT_DIV_EXPR, itype, itercnt, step); itercnt = fold_build2 (PLUS_EXPR, itype, itercnt, s0); itercnt = force_gimple_operand_gsi (&gsi, itercnt, true, NULL_TREE, false, GSI_CONTINUE_LINKING); } else itercnt = s0; } tree type = TREE_TYPE (t); if (POINTER_TYPE_P (type)) type = sizetype; a = fold_build2 (MULT_EXPR, type, fold_convert (type, itercnt), fold_convert (type, OMP_CLAUSE_LINEAR_STEP (c))); dest = unshare_expr (t); t = fold_build2 (type == TREE_TYPE (t) ? PLUS_EXPR : POINTER_PLUS_EXPR, TREE_TYPE (t), t, a); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (dest, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } if (fd->collapse > 1) expand_omp_for_init_vars (fd, &gsi, counts, inner_stmt, startvar); if (!broken_loop) { /* The code controlling the sequential loop replaces the GIMPLE_OMP_CONTINUE. */ gsi = gsi_last_bb (cont_bb); gomp_continue *cont_stmt = as_a (gsi_stmt (gsi)); gcc_assert (gimple_code (cont_stmt) == GIMPLE_OMP_CONTINUE); vmain = gimple_omp_continue_control_use (cont_stmt); vback = gimple_omp_continue_control_def (cont_stmt); if (!gimple_omp_for_combined_p (fd->for_stmt)) { if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (vmain, step); else t = fold_build2 (PLUS_EXPR, type, vmain, step); t = force_gimple_operand_gsi (&gsi, t, DECL_P (vback) && TREE_ADDRESSABLE (vback), NULL_TREE, true, GSI_SAME_STMT); assign_stmt = gimple_build_assign (vback, t); gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT); t = build2 (fd->loop.cond_code, boolean_type_node, DECL_P (vback) && TREE_ADDRESSABLE (vback) ? t : vback, e); gsi_insert_before (&gsi, gimple_build_cond_empty (t), GSI_SAME_STMT); } /* Remove the GIMPLE_OMP_CONTINUE statement. */ gsi_remove (&gsi, true); if (fd->collapse > 1 && !gimple_omp_for_combined_p (fd->for_stmt)) collapse_bb = extract_omp_for_update_vars (fd, cont_bb, body_bb); } /* Replace the GIMPLE_OMP_RETURN with a barrier, or nothing. */ gsi = gsi_last_bb (exit_bb); if (!gimple_omp_return_nowait_p (gsi_stmt (gsi))) { t = gimple_omp_return_lhs (gsi_stmt (gsi)); gsi_insert_after (&gsi, omp_build_barrier (t), GSI_SAME_STMT); } gsi_remove (&gsi, true); /* Connect all the blocks. */ ep = make_edge (entry_bb, third_bb, EDGE_FALSE_VALUE); ep->probability = REG_BR_PROB_BASE / 4 * 3; ep = find_edge (entry_bb, second_bb); ep->flags = EDGE_TRUE_VALUE; ep->probability = REG_BR_PROB_BASE / 4; find_edge (third_bb, seq_start_bb)->flags = EDGE_FALSE_VALUE; find_edge (third_bb, fin_bb)->flags = EDGE_TRUE_VALUE; if (!broken_loop) { ep = find_edge (cont_bb, body_bb); if (ep == NULL) { ep = BRANCH_EDGE (cont_bb); gcc_assert (single_succ (ep->dest) == body_bb); } if (gimple_omp_for_combined_p (fd->for_stmt)) { remove_edge (ep); ep = NULL; } else if (fd->collapse > 1) { remove_edge (ep); ep = make_edge (cont_bb, collapse_bb, EDGE_TRUE_VALUE); } else ep->flags = EDGE_TRUE_VALUE; find_edge (cont_bb, fin_bb)->flags = ep ? EDGE_FALSE_VALUE : EDGE_FALLTHRU; } set_immediate_dominator (CDI_DOMINATORS, second_bb, entry_bb); set_immediate_dominator (CDI_DOMINATORS, third_bb, entry_bb); set_immediate_dominator (CDI_DOMINATORS, seq_start_bb, third_bb); set_immediate_dominator (CDI_DOMINATORS, body_bb, recompute_dominator (CDI_DOMINATORS, body_bb)); set_immediate_dominator (CDI_DOMINATORS, fin_bb, recompute_dominator (CDI_DOMINATORS, fin_bb)); struct loop *loop = body_bb->loop_father; if (loop != entry_bb->loop_father) { gcc_assert (broken_loop || loop->header == body_bb); gcc_assert (broken_loop || loop->latch == region->cont || single_pred (loop->latch) == region->cont); return; } if (!broken_loop && !gimple_omp_for_combined_p (fd->for_stmt)) { loop = alloc_loop (); loop->header = body_bb; if (collapse_bb == NULL) loop->latch = cont_bb; add_loop (loop, body_bb->loop_father); } } /* Return phi in E->DEST with ARG on edge E. */ static gphi * find_phi_with_arg_on_edge (tree arg, edge e) { basic_block bb = e->dest; for (gphi_iterator gpi = gsi_start_phis (bb); !gsi_end_p (gpi); gsi_next (&gpi)) { gphi *phi = gpi.phi (); if (PHI_ARG_DEF_FROM_EDGE (phi, e) == arg) return phi; } return NULL; } /* A subroutine of expand_omp_for. Generate code for a parallel loop with static schedule and a specified chunk size. Given parameters: for (V = N1; V cond N2; V += STEP) BODY; where COND is "<" or ">", we generate pseudocode if ((__typeof (V)) -1 > 0 && N2 cond N1) goto L2; if (cond is <) adj = STEP - 1; else adj = STEP + 1; if ((__typeof (V)) -1 > 0 && cond is >) n = -(adj + N2 - N1) / -STEP; else n = (adj + N2 - N1) / STEP; trip = 0; V = threadid * CHUNK * STEP + N1; -- this extra definition of V is here so that V is defined if the loop is not entered L0: s0 = (trip * nthreads + threadid) * CHUNK; e0 = min (s0 + CHUNK, n); if (s0 < n) goto L1; else goto L4; L1: V = s0 * STEP + N1; e = e0 * STEP + N1; L2: BODY; V += STEP; if (V cond e) goto L2; else goto L3; L3: trip += 1; goto L0; L4: */ static void expand_omp_for_static_chunk (struct omp_region *region, struct omp_for_data *fd, gimple *inner_stmt) { tree n, s0, e0, e, t; tree trip_var, trip_init, trip_main, trip_back, nthreads, threadid; tree type, itype, vmain, vback, vextra; basic_block entry_bb, exit_bb, body_bb, seq_start_bb, iter_part_bb; basic_block trip_update_bb = NULL, cont_bb, collapse_bb = NULL, fin_bb; gimple_stmt_iterator gsi; edge se; bool broken_loop = region->cont == NULL; tree *counts = NULL; tree n1, n2, step; itype = type = TREE_TYPE (fd->loop.v); if (POINTER_TYPE_P (type)) itype = signed_type_for (type); entry_bb = region->entry; se = split_block (entry_bb, last_stmt (entry_bb)); entry_bb = se->src; iter_part_bb = se->dest; cont_bb = region->cont; gcc_assert (EDGE_COUNT (iter_part_bb->succs) == 2); fin_bb = BRANCH_EDGE (iter_part_bb)->dest; gcc_assert (broken_loop || fin_bb == FALLTHRU_EDGE (cont_bb)->dest); seq_start_bb = split_edge (FALLTHRU_EDGE (iter_part_bb)); body_bb = single_succ (seq_start_bb); if (!broken_loop) { gcc_assert (BRANCH_EDGE (cont_bb)->dest == body_bb || single_succ (BRANCH_EDGE (cont_bb)->dest) == body_bb); gcc_assert (EDGE_COUNT (cont_bb->succs) == 2); trip_update_bb = split_edge (FALLTHRU_EDGE (cont_bb)); } exit_bb = region->exit; /* Trip and adjustment setup goes in ENTRY_BB. */ gsi = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); if (fd->collapse > 1) { int first_zero_iter = -1, dummy = -1; basic_block l2_dom_bb = NULL, dummy_bb = NULL; counts = XALLOCAVEC (tree, fd->collapse); expand_omp_for_init_counts (fd, &gsi, entry_bb, counts, fin_bb, first_zero_iter, dummy_bb, dummy, l2_dom_bb); t = NULL_TREE; } else if (gimple_omp_for_combined_into_p (fd->for_stmt)) t = integer_one_node; else t = fold_binary (fd->loop.cond_code, boolean_type_node, fold_convert (type, fd->loop.n1), fold_convert (type, fd->loop.n2)); if (fd->collapse == 1 && TYPE_UNSIGNED (type) && (t == NULL_TREE || !integer_onep (t))) { n1 = fold_convert (type, unshare_expr (fd->loop.n1)); n1 = force_gimple_operand_gsi (&gsi, n1, true, NULL_TREE, true, GSI_SAME_STMT); n2 = fold_convert (type, unshare_expr (fd->loop.n2)); n2 = force_gimple_operand_gsi (&gsi, n2, true, NULL_TREE, true, GSI_SAME_STMT); gcond *cond_stmt = gimple_build_cond (fd->loop.cond_code, n1, n2, NULL_TREE, NULL_TREE); gsi_insert_before (&gsi, cond_stmt, GSI_SAME_STMT); if (walk_tree (gimple_cond_lhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL) || walk_tree (gimple_cond_rhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL)) { gsi = gsi_for_stmt (cond_stmt); gimple_regimplify_operands (cond_stmt, &gsi); } se = split_block (entry_bb, cond_stmt); se->flags = EDGE_TRUE_VALUE; entry_bb = se->dest; se->probability = REG_BR_PROB_BASE - (REG_BR_PROB_BASE / 2000 - 1); se = make_edge (se->src, fin_bb, EDGE_FALSE_VALUE); se->probability = REG_BR_PROB_BASE / 2000 - 1; if (gimple_in_ssa_p (cfun)) { int dest_idx = find_edge (iter_part_bb, fin_bb)->dest_idx; for (gphi_iterator gpi = gsi_start_phis (fin_bb); !gsi_end_p (gpi); gsi_next (&gpi)) { gphi *phi = gpi.phi (); add_phi_arg (phi, gimple_phi_arg_def (phi, dest_idx), se, UNKNOWN_LOCATION); } } gsi = gsi_last_bb (entry_bb); } switch (gimple_omp_for_kind (fd->for_stmt)) { case GF_OMP_FOR_KIND_FOR: nthreads = builtin_decl_explicit (BUILT_IN_OMP_GET_NUM_THREADS); threadid = builtin_decl_explicit (BUILT_IN_OMP_GET_THREAD_NUM); break; case GF_OMP_FOR_KIND_DISTRIBUTE: nthreads = builtin_decl_explicit (BUILT_IN_OMP_GET_NUM_TEAMS); threadid = builtin_decl_explicit (BUILT_IN_OMP_GET_TEAM_NUM); break; default: gcc_unreachable (); } nthreads = build_call_expr (nthreads, 0); nthreads = fold_convert (itype, nthreads); nthreads = force_gimple_operand_gsi (&gsi, nthreads, true, NULL_TREE, true, GSI_SAME_STMT); threadid = build_call_expr (threadid, 0); threadid = fold_convert (itype, threadid); threadid = force_gimple_operand_gsi (&gsi, threadid, true, NULL_TREE, true, GSI_SAME_STMT); n1 = fd->loop.n1; n2 = fd->loop.n2; step = fd->loop.step; if (gimple_omp_for_combined_into_p (fd->for_stmt)) { tree innerc = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n1 = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n2 = OMP_CLAUSE_DECL (innerc); } n1 = force_gimple_operand_gsi (&gsi, fold_convert (type, n1), true, NULL_TREE, true, GSI_SAME_STMT); n2 = force_gimple_operand_gsi (&gsi, fold_convert (itype, n2), true, NULL_TREE, true, GSI_SAME_STMT); step = force_gimple_operand_gsi (&gsi, fold_convert (itype, step), true, NULL_TREE, true, GSI_SAME_STMT); tree chunk_size = fold_convert (itype, fd->chunk_size); chunk_size = omp_adjust_chunk_size (chunk_size, fd->simd_schedule); chunk_size = force_gimple_operand_gsi (&gsi, chunk_size, true, NULL_TREE, true, GSI_SAME_STMT); t = build_int_cst (itype, (fd->loop.cond_code == LT_EXPR ? -1 : 1)); t = fold_build2 (PLUS_EXPR, itype, step, t); t = fold_build2 (PLUS_EXPR, itype, t, n2); t = fold_build2 (MINUS_EXPR, itype, t, fold_convert (itype, n1)); if (TYPE_UNSIGNED (itype) && fd->loop.cond_code == GT_EXPR) t = fold_build2 (TRUNC_DIV_EXPR, itype, fold_build1 (NEGATE_EXPR, itype, t), fold_build1 (NEGATE_EXPR, itype, step)); else t = fold_build2 (TRUNC_DIV_EXPR, itype, t, step); t = fold_convert (itype, t); n = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); trip_var = create_tmp_reg (itype, ".trip"); if (gimple_in_ssa_p (cfun)) { trip_init = make_ssa_name (trip_var); trip_main = make_ssa_name (trip_var); trip_back = make_ssa_name (trip_var); } else { trip_init = trip_var; trip_main = trip_var; trip_back = trip_var; } gassign *assign_stmt = gimple_build_assign (trip_init, build_int_cst (itype, 0)); gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT); t = fold_build2 (MULT_EXPR, itype, threadid, chunk_size); t = fold_build2 (MULT_EXPR, itype, t, step); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (n1, t); else t = fold_build2 (PLUS_EXPR, type, t, n1); vextra = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); /* Remove the GIMPLE_OMP_FOR. */ gsi_remove (&gsi, true); gimple_stmt_iterator gsif = gsi; /* Iteration space partitioning goes in ITER_PART_BB. */ gsi = gsi_last_bb (iter_part_bb); t = fold_build2 (MULT_EXPR, itype, trip_main, nthreads); t = fold_build2 (PLUS_EXPR, itype, t, threadid); t = fold_build2 (MULT_EXPR, itype, t, chunk_size); s0 = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); t = fold_build2 (PLUS_EXPR, itype, s0, chunk_size); t = fold_build2 (MIN_EXPR, itype, t, n); e0 = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); t = build2 (LT_EXPR, boolean_type_node, s0, n); gsi_insert_after (&gsi, gimple_build_cond_empty (t), GSI_CONTINUE_LINKING); /* Setup code for sequential iteration goes in SEQ_START_BB. */ gsi = gsi_start_bb (seq_start_bb); tree startvar = fd->loop.v; tree endvar = NULL_TREE; if (gimple_omp_for_combined_p (fd->for_stmt)) { tree clauses = gimple_code (inner_stmt) == GIMPLE_OMP_PARALLEL ? gimple_omp_parallel_clauses (inner_stmt) : gimple_omp_for_clauses (inner_stmt); tree innerc = omp_find_clause (clauses, OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); startvar = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); endvar = OMP_CLAUSE_DECL (innerc); if (fd->collapse > 1 && TREE_CODE (fd->loop.n2) != INTEGER_CST && gimple_omp_for_kind (fd->for_stmt) == GF_OMP_FOR_KIND_DISTRIBUTE) { int i; for (i = 1; i < fd->collapse; i++) { innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); } innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); if (innerc) { /* If needed (distribute parallel for with lastprivate), propagate down the total number of iterations. */ tree t = fold_convert (TREE_TYPE (OMP_CLAUSE_DECL (innerc)), fd->loop.n2); t = force_gimple_operand_gsi (&gsi, t, false, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (OMP_CLAUSE_DECL (innerc), t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } } } t = fold_convert (itype, s0); t = fold_build2 (MULT_EXPR, itype, t, step); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (n1, t); else t = fold_build2 (PLUS_EXPR, type, t, n1); t = fold_convert (TREE_TYPE (startvar), t); t = force_gimple_operand_gsi (&gsi, t, DECL_P (startvar) && TREE_ADDRESSABLE (startvar), NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (startvar, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); t = fold_convert (itype, e0); t = fold_build2 (MULT_EXPR, itype, t, step); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (n1, t); else t = fold_build2 (PLUS_EXPR, type, t, n1); t = fold_convert (TREE_TYPE (startvar), t); e = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); if (endvar) { assign_stmt = gimple_build_assign (endvar, e); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); if (useless_type_conversion_p (TREE_TYPE (fd->loop.v), TREE_TYPE (e))) assign_stmt = gimple_build_assign (fd->loop.v, e); else assign_stmt = gimple_build_assign (fd->loop.v, NOP_EXPR, e); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } /* Handle linear clause adjustments. */ tree itercnt = NULL_TREE, itercntbias = NULL_TREE; if (gimple_omp_for_kind (fd->for_stmt) == GF_OMP_FOR_KIND_FOR) for (tree c = gimple_omp_for_clauses (fd->for_stmt); c; c = OMP_CLAUSE_CHAIN (c)) if (OMP_CLAUSE_CODE (c) == OMP_CLAUSE_LINEAR && !OMP_CLAUSE_LINEAR_NO_COPYIN (c)) { tree d = OMP_CLAUSE_DECL (c); bool is_ref = omp_is_reference (d); tree t = d, a, dest; if (is_ref) t = build_simple_mem_ref_loc (OMP_CLAUSE_LOCATION (c), t); tree type = TREE_TYPE (t); if (POINTER_TYPE_P (type)) type = sizetype; dest = unshare_expr (t); tree v = create_tmp_var (TREE_TYPE (t), NULL); expand_omp_build_assign (&gsif, v, t); if (itercnt == NULL_TREE) { if (gimple_omp_for_combined_into_p (fd->for_stmt)) { itercntbias = fold_build2 (MINUS_EXPR, itype, fold_convert (itype, n1), fold_convert (itype, fd->loop.n1)); itercntbias = fold_build2 (EXACT_DIV_EXPR, itype, itercntbias, step); itercntbias = force_gimple_operand_gsi (&gsif, itercntbias, true, NULL_TREE, true, GSI_SAME_STMT); itercnt = fold_build2 (PLUS_EXPR, itype, itercntbias, s0); itercnt = force_gimple_operand_gsi (&gsi, itercnt, true, NULL_TREE, false, GSI_CONTINUE_LINKING); } else itercnt = s0; } a = fold_build2 (MULT_EXPR, type, fold_convert (type, itercnt), fold_convert (type, OMP_CLAUSE_LINEAR_STEP (c))); t = fold_build2 (type == TREE_TYPE (t) ? PLUS_EXPR : POINTER_PLUS_EXPR, TREE_TYPE (t), v, a); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (dest, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } if (fd->collapse > 1) expand_omp_for_init_vars (fd, &gsi, counts, inner_stmt, startvar); if (!broken_loop) { /* The code controlling the sequential loop goes in CONT_BB, replacing the GIMPLE_OMP_CONTINUE. */ gsi = gsi_last_bb (cont_bb); gomp_continue *cont_stmt = as_a (gsi_stmt (gsi)); vmain = gimple_omp_continue_control_use (cont_stmt); vback = gimple_omp_continue_control_def (cont_stmt); if (!gimple_omp_for_combined_p (fd->for_stmt)) { if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (vmain, step); else t = fold_build2 (PLUS_EXPR, type, vmain, step); if (DECL_P (vback) && TREE_ADDRESSABLE (vback)) t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, true, GSI_SAME_STMT); assign_stmt = gimple_build_assign (vback, t); gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT); if (tree_int_cst_equal (fd->chunk_size, integer_one_node)) t = build2 (EQ_EXPR, boolean_type_node, build_int_cst (itype, 0), build_int_cst (itype, 1)); else t = build2 (fd->loop.cond_code, boolean_type_node, DECL_P (vback) && TREE_ADDRESSABLE (vback) ? t : vback, e); gsi_insert_before (&gsi, gimple_build_cond_empty (t), GSI_SAME_STMT); } /* Remove GIMPLE_OMP_CONTINUE. */ gsi_remove (&gsi, true); if (fd->collapse > 1 && !gimple_omp_for_combined_p (fd->for_stmt)) collapse_bb = extract_omp_for_update_vars (fd, cont_bb, body_bb); /* Trip update code goes into TRIP_UPDATE_BB. */ gsi = gsi_start_bb (trip_update_bb); t = build_int_cst (itype, 1); t = build2 (PLUS_EXPR, itype, trip_main, t); assign_stmt = gimple_build_assign (trip_back, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } /* Replace the GIMPLE_OMP_RETURN with a barrier, or nothing. */ gsi = gsi_last_bb (exit_bb); if (!gimple_omp_return_nowait_p (gsi_stmt (gsi))) { t = gimple_omp_return_lhs (gsi_stmt (gsi)); gsi_insert_after (&gsi, omp_build_barrier (t), GSI_SAME_STMT); } gsi_remove (&gsi, true); /* Connect the new blocks. */ find_edge (iter_part_bb, seq_start_bb)->flags = EDGE_TRUE_VALUE; find_edge (iter_part_bb, fin_bb)->flags = EDGE_FALSE_VALUE; if (!broken_loop) { se = find_edge (cont_bb, body_bb); if (se == NULL) { se = BRANCH_EDGE (cont_bb); gcc_assert (single_succ (se->dest) == body_bb); } if (gimple_omp_for_combined_p (fd->for_stmt)) { remove_edge (se); se = NULL; } else if (fd->collapse > 1) { remove_edge (se); se = make_edge (cont_bb, collapse_bb, EDGE_TRUE_VALUE); } else se->flags = EDGE_TRUE_VALUE; find_edge (cont_bb, trip_update_bb)->flags = se ? EDGE_FALSE_VALUE : EDGE_FALLTHRU; redirect_edge_and_branch (single_succ_edge (trip_update_bb), iter_part_bb); } if (gimple_in_ssa_p (cfun)) { gphi_iterator psi; gphi *phi; edge re, ene; edge_var_map *vm; size_t i; gcc_assert (fd->collapse == 1 && !broken_loop); /* When we redirect the edge from trip_update_bb to iter_part_bb, we remove arguments of the phi nodes in fin_bb. We need to create appropriate phi nodes in iter_part_bb instead. */ se = find_edge (iter_part_bb, fin_bb); re = single_succ_edge (trip_update_bb); vec *head = redirect_edge_var_map_vector (re); ene = single_succ_edge (entry_bb); psi = gsi_start_phis (fin_bb); for (i = 0; !gsi_end_p (psi) && head->iterate (i, &vm); gsi_next (&psi), ++i) { gphi *nphi; source_location locus; phi = psi.phi (); t = gimple_phi_result (phi); gcc_assert (t == redirect_edge_var_map_result (vm)); if (!single_pred_p (fin_bb)) t = copy_ssa_name (t, phi); nphi = create_phi_node (t, iter_part_bb); t = PHI_ARG_DEF_FROM_EDGE (phi, se); locus = gimple_phi_arg_location_from_edge (phi, se); /* A special case -- fd->loop.v is not yet computed in iter_part_bb, we need to use vextra instead. */ if (t == fd->loop.v) t = vextra; add_phi_arg (nphi, t, ene, locus); locus = redirect_edge_var_map_location (vm); tree back_arg = redirect_edge_var_map_def (vm); add_phi_arg (nphi, back_arg, re, locus); edge ce = find_edge (cont_bb, body_bb); if (ce == NULL) { ce = BRANCH_EDGE (cont_bb); gcc_assert (single_succ (ce->dest) == body_bb); ce = single_succ_edge (ce->dest); } gphi *inner_loop_phi = find_phi_with_arg_on_edge (back_arg, ce); gcc_assert (inner_loop_phi != NULL); add_phi_arg (inner_loop_phi, gimple_phi_result (nphi), find_edge (seq_start_bb, body_bb), locus); if (!single_pred_p (fin_bb)) add_phi_arg (phi, gimple_phi_result (nphi), se, locus); } gcc_assert (gsi_end_p (psi) && (head == NULL || i == head->length ())); redirect_edge_var_map_clear (re); if (single_pred_p (fin_bb)) while (1) { psi = gsi_start_phis (fin_bb); if (gsi_end_p (psi)) break; remove_phi_node (&psi, false); } /* Make phi node for trip. */ phi = create_phi_node (trip_main, iter_part_bb); add_phi_arg (phi, trip_back, single_succ_edge (trip_update_bb), UNKNOWN_LOCATION); add_phi_arg (phi, trip_init, single_succ_edge (entry_bb), UNKNOWN_LOCATION); } if (!broken_loop) set_immediate_dominator (CDI_DOMINATORS, trip_update_bb, cont_bb); set_immediate_dominator (CDI_DOMINATORS, iter_part_bb, recompute_dominator (CDI_DOMINATORS, iter_part_bb)); set_immediate_dominator (CDI_DOMINATORS, fin_bb, recompute_dominator (CDI_DOMINATORS, fin_bb)); set_immediate_dominator (CDI_DOMINATORS, seq_start_bb, recompute_dominator (CDI_DOMINATORS, seq_start_bb)); set_immediate_dominator (CDI_DOMINATORS, body_bb, recompute_dominator (CDI_DOMINATORS, body_bb)); if (!broken_loop) { struct loop *loop = body_bb->loop_father; struct loop *trip_loop = alloc_loop (); trip_loop->header = iter_part_bb; trip_loop->latch = trip_update_bb; add_loop (trip_loop, iter_part_bb->loop_father); if (loop != entry_bb->loop_father) { gcc_assert (loop->header == body_bb); gcc_assert (loop->latch == region->cont || single_pred (loop->latch) == region->cont); trip_loop->inner = loop; return; } if (!gimple_omp_for_combined_p (fd->for_stmt)) { loop = alloc_loop (); loop->header = body_bb; if (collapse_bb == NULL) loop->latch = cont_bb; add_loop (loop, trip_loop); } } } /* A subroutine of expand_omp_for. Generate code for _Cilk_for loop. Given parameters: for (V = N1; V cond N2; V += STEP) BODY; where COND is "<" or ">" or "!=", we generate pseudocode for (ind_var = low; ind_var < high; ind_var++) { V = n1 + (ind_var * STEP) } In the above pseudocode, low and high are function parameters of the child function. In the function below, we are inserting a temp. variable that will be making a call to two OMP functions that will not be found in the body of _Cilk_for (since OMP_FOR cannot be mixed with _Cilk_for). These functions are replaced with low and high by the function that handles taskreg. */ static void expand_cilk_for (struct omp_region *region, struct omp_for_data *fd) { bool broken_loop = region->cont == NULL; basic_block entry_bb = region->entry; basic_block cont_bb = region->cont; gcc_assert (EDGE_COUNT (entry_bb->succs) == 2); gcc_assert (broken_loop || BRANCH_EDGE (entry_bb)->dest == FALLTHRU_EDGE (cont_bb)->dest); basic_block l0_bb = FALLTHRU_EDGE (entry_bb)->dest; basic_block l1_bb, l2_bb; if (!broken_loop) { gcc_assert (BRANCH_EDGE (cont_bb)->dest == l0_bb); gcc_assert (EDGE_COUNT (cont_bb->succs) == 2); l1_bb = split_block (cont_bb, last_stmt (cont_bb))->dest; l2_bb = BRANCH_EDGE (entry_bb)->dest; } else { BRANCH_EDGE (entry_bb)->flags &= ~EDGE_ABNORMAL; l1_bb = split_edge (BRANCH_EDGE (entry_bb)); l2_bb = single_succ (l1_bb); } basic_block exit_bb = region->exit; basic_block l2_dom_bb = NULL; gimple_stmt_iterator gsi = gsi_last_bb (entry_bb); /* Below statements until the "tree high_val = ..." are pseudo statements used to pass information to be used by expand_omp_taskreg. low_val and high_val will be replaced by the __low and __high parameter from the child function. The call_exprs part is a place-holder, it is mainly used to distinctly identify to the top-level part that this is where we should put low and high (reasoning given in header comment). */ gomp_parallel *par_stmt = as_a (last_stmt (region->outer->entry)); tree child_fndecl = gimple_omp_parallel_child_fn (par_stmt); tree t, low_val = NULL_TREE, high_val = NULL_TREE; for (t = DECL_ARGUMENTS (child_fndecl); t; t = TREE_CHAIN (t)) { if (!strcmp (IDENTIFIER_POINTER (DECL_NAME (t)), "__high")) high_val = t; else if (!strcmp (IDENTIFIER_POINTER (DECL_NAME (t)), "__low")) low_val = t; } gcc_assert (low_val && high_val); tree type = TREE_TYPE (low_val); tree ind_var = create_tmp_reg (type, "__cilk_ind_var"); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); /* Not needed in SSA form right now. */ gcc_assert (!gimple_in_ssa_p (cfun)); if (l2_dom_bb == NULL) l2_dom_bb = l1_bb; tree n1 = low_val; tree n2 = high_val; gimple *stmt = gimple_build_assign (ind_var, n1); /* Replace the GIMPLE_OMP_FOR statement. */ gsi_replace (&gsi, stmt, true); if (!broken_loop) { /* Code to control the increment goes in the CONT_BB. */ gsi = gsi_last_bb (cont_bb); stmt = gsi_stmt (gsi); gcc_assert (gimple_code (stmt) == GIMPLE_OMP_CONTINUE); stmt = gimple_build_assign (ind_var, PLUS_EXPR, ind_var, build_one_cst (type)); /* Replace GIMPLE_OMP_CONTINUE. */ gsi_replace (&gsi, stmt, true); } /* Emit the condition in L1_BB. */ gsi = gsi_after_labels (l1_bb); t = fold_build2 (MULT_EXPR, TREE_TYPE (fd->loop.step), fold_convert (TREE_TYPE (fd->loop.step), ind_var), fd->loop.step); if (POINTER_TYPE_P (TREE_TYPE (fd->loop.n1))) t = fold_build2 (POINTER_PLUS_EXPR, TREE_TYPE (fd->loop.n1), fd->loop.n1, fold_convert (sizetype, t)); else t = fold_build2 (PLUS_EXPR, TREE_TYPE (fd->loop.n1), fd->loop.n1, fold_convert (TREE_TYPE (fd->loop.n1), t)); t = fold_convert (TREE_TYPE (fd->loop.v), t); expand_omp_build_assign (&gsi, fd->loop.v, t); /* The condition is always '<' since the runtime will fill in the low and high values. */ stmt = gimple_build_cond (LT_EXPR, ind_var, n2, NULL_TREE, NULL_TREE); gsi_insert_before (&gsi, stmt, GSI_SAME_STMT); /* Remove GIMPLE_OMP_RETURN. */ gsi = gsi_last_bb (exit_bb); gsi_remove (&gsi, true); /* Connect the new blocks. */ remove_edge (FALLTHRU_EDGE (entry_bb)); edge e, ne; if (!broken_loop) { remove_edge (BRANCH_EDGE (entry_bb)); make_edge (entry_bb, l1_bb, EDGE_FALLTHRU); e = BRANCH_EDGE (l1_bb); ne = FALLTHRU_EDGE (l1_bb); e->flags = EDGE_TRUE_VALUE; } else { single_succ_edge (entry_bb)->flags = EDGE_FALLTHRU; ne = single_succ_edge (l1_bb); e = make_edge (l1_bb, l0_bb, EDGE_TRUE_VALUE); } ne->flags = EDGE_FALSE_VALUE; e->probability = REG_BR_PROB_BASE * 7 / 8; ne->probability = REG_BR_PROB_BASE / 8; set_immediate_dominator (CDI_DOMINATORS, l1_bb, entry_bb); set_immediate_dominator (CDI_DOMINATORS, l2_bb, l2_dom_bb); set_immediate_dominator (CDI_DOMINATORS, l0_bb, l1_bb); if (!broken_loop) { struct loop *loop = alloc_loop (); loop->header = l1_bb; loop->latch = cont_bb; add_loop (loop, l1_bb->loop_father); loop->safelen = INT_MAX; } /* Pick the correct library function based on the precision of the induction variable type. */ tree lib_fun = NULL_TREE; if (TYPE_PRECISION (type) == 32) lib_fun = cilk_for_32_fndecl; else if (TYPE_PRECISION (type) == 64) lib_fun = cilk_for_64_fndecl; else gcc_unreachable (); gcc_assert (fd->sched_kind == OMP_CLAUSE_SCHEDULE_CILKFOR); /* WS_ARGS contains the library function flavor to call: __libcilkrts_cilk_for_64 or __libcilkrts_cilk_for_32), and the user-defined grain value. If the user does not define one, then zero is passed in by the parser. */ vec_alloc (region->ws_args, 2); region->ws_args->quick_push (lib_fun); region->ws_args->quick_push (fd->chunk_size); } /* A subroutine of expand_omp_for. Generate code for a simd non-worksharing loop. Given parameters: for (V = N1; V cond N2; V += STEP) BODY; where COND is "<" or ">", we generate pseudocode V = N1; goto L1; L0: BODY; V += STEP; L1: if (V cond N2) goto L0; else goto L2; L2: For collapsed loops, given parameters: collapse(3) for (V1 = N11; V1 cond1 N12; V1 += STEP1) for (V2 = N21; V2 cond2 N22; V2 += STEP2) for (V3 = N31; V3 cond3 N32; V3 += STEP3) BODY; we generate pseudocode if (cond3 is <) adj = STEP3 - 1; else adj = STEP3 + 1; count3 = (adj + N32 - N31) / STEP3; if (cond2 is <) adj = STEP2 - 1; else adj = STEP2 + 1; count2 = (adj + N22 - N21) / STEP2; if (cond1 is <) adj = STEP1 - 1; else adj = STEP1 + 1; count1 = (adj + N12 - N11) / STEP1; count = count1 * count2 * count3; V = 0; V1 = N11; V2 = N21; V3 = N31; goto L1; L0: BODY; V += 1; V3 += STEP3; V2 += (V3 cond3 N32) ? 0 : STEP2; V3 = (V3 cond3 N32) ? V3 : N31; V1 += (V2 cond2 N22) ? 0 : STEP1; V2 = (V2 cond2 N22) ? V2 : N21; L1: if (V < count) goto L0; else goto L2; L2: */ static void expand_omp_simd (struct omp_region *region, struct omp_for_data *fd) { tree type, t; basic_block entry_bb, cont_bb, exit_bb, l0_bb, l1_bb, l2_bb, l2_dom_bb; gimple_stmt_iterator gsi; gimple *stmt; gcond *cond_stmt; bool broken_loop = region->cont == NULL; edge e, ne; tree *counts = NULL; int i; int safelen_int = INT_MAX; tree safelen = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE_SAFELEN); tree simduid = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__SIMDUID_); tree n1, n2; if (safelen) { safelen = OMP_CLAUSE_SAFELEN_EXPR (safelen); if (TREE_CODE (safelen) != INTEGER_CST) safelen_int = 0; else if (tree_fits_uhwi_p (safelen) && tree_to_uhwi (safelen) < INT_MAX) safelen_int = tree_to_uhwi (safelen); if (safelen_int == 1) safelen_int = 0; } type = TREE_TYPE (fd->loop.v); entry_bb = region->entry; cont_bb = region->cont; gcc_assert (EDGE_COUNT (entry_bb->succs) == 2); gcc_assert (broken_loop || BRANCH_EDGE (entry_bb)->dest == FALLTHRU_EDGE (cont_bb)->dest); l0_bb = FALLTHRU_EDGE (entry_bb)->dest; if (!broken_loop) { gcc_assert (BRANCH_EDGE (cont_bb)->dest == l0_bb); gcc_assert (EDGE_COUNT (cont_bb->succs) == 2); l1_bb = split_block (cont_bb, last_stmt (cont_bb))->dest; l2_bb = BRANCH_EDGE (entry_bb)->dest; } else { BRANCH_EDGE (entry_bb)->flags &= ~EDGE_ABNORMAL; l1_bb = split_edge (BRANCH_EDGE (entry_bb)); l2_bb = single_succ (l1_bb); } exit_bb = region->exit; l2_dom_bb = NULL; gsi = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); /* Not needed in SSA form right now. */ gcc_assert (!gimple_in_ssa_p (cfun)); if (fd->collapse > 1) { int first_zero_iter = -1, dummy = -1; basic_block zero_iter_bb = l2_bb, dummy_bb = NULL; counts = XALLOCAVEC (tree, fd->collapse); expand_omp_for_init_counts (fd, &gsi, entry_bb, counts, zero_iter_bb, first_zero_iter, dummy_bb, dummy, l2_dom_bb); } if (l2_dom_bb == NULL) l2_dom_bb = l1_bb; n1 = fd->loop.n1; n2 = fd->loop.n2; if (gimple_omp_for_combined_into_p (fd->for_stmt)) { tree innerc = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n1 = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n2 = OMP_CLAUSE_DECL (innerc); } tree step = fd->loop.step; bool is_simt = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__SIMT_); if (is_simt) { cfun->curr_properties &= ~PROP_gimple_lomp_dev; is_simt = safelen_int > 1; } tree simt_lane = NULL_TREE, simt_maxlane = NULL_TREE; if (is_simt) { simt_lane = create_tmp_var (unsigned_type_node); gimple *g = gimple_build_call_internal (IFN_GOMP_SIMT_LANE, 0); gimple_call_set_lhs (g, simt_lane); gsi_insert_before (&gsi, g, GSI_SAME_STMT); tree offset = fold_build2 (MULT_EXPR, TREE_TYPE (step), step, fold_convert (TREE_TYPE (step), simt_lane)); n1 = fold_convert (type, n1); if (POINTER_TYPE_P (type)) n1 = fold_build_pointer_plus (n1, offset); else n1 = fold_build2 (PLUS_EXPR, type, n1, fold_convert (type, offset)); /* Collapsed loops not handled for SIMT yet: limit to one lane only. */ if (fd->collapse > 1) simt_maxlane = build_one_cst (unsigned_type_node); else if (safelen_int < omp_max_simt_vf ()) simt_maxlane = build_int_cst (unsigned_type_node, safelen_int); tree vf = build_call_expr_internal_loc (UNKNOWN_LOCATION, IFN_GOMP_SIMT_VF, unsigned_type_node, 0); if (simt_maxlane) vf = fold_build2 (MIN_EXPR, unsigned_type_node, vf, simt_maxlane); vf = fold_convert (TREE_TYPE (step), vf); step = fold_build2 (MULT_EXPR, TREE_TYPE (step), step, vf); } expand_omp_build_assign (&gsi, fd->loop.v, fold_convert (type, n1)); if (fd->collapse > 1) { if (gimple_omp_for_combined_into_p (fd->for_stmt)) { gsi_prev (&gsi); expand_omp_for_init_vars (fd, &gsi, counts, NULL, n1); gsi_next (&gsi); } else for (i = 0; i < fd->collapse; i++) { tree itype = TREE_TYPE (fd->loops[i].v); if (POINTER_TYPE_P (itype)) itype = signed_type_for (itype); t = fold_convert (TREE_TYPE (fd->loops[i].v), fd->loops[i].n1); expand_omp_build_assign (&gsi, fd->loops[i].v, t); } } /* Remove the GIMPLE_OMP_FOR statement. */ gsi_remove (&gsi, true); if (!broken_loop) { /* Code to control the increment goes in the CONT_BB. */ gsi = gsi_last_bb (cont_bb); stmt = gsi_stmt (gsi); gcc_assert (gimple_code (stmt) == GIMPLE_OMP_CONTINUE); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (fd->loop.v, step); else t = fold_build2 (PLUS_EXPR, type, fd->loop.v, step); expand_omp_build_assign (&gsi, fd->loop.v, t); if (fd->collapse > 1) { i = fd->collapse - 1; if (POINTER_TYPE_P (TREE_TYPE (fd->loops[i].v))) { t = fold_convert (sizetype, fd->loops[i].step); t = fold_build_pointer_plus (fd->loops[i].v, t); } else { t = fold_convert (TREE_TYPE (fd->loops[i].v), fd->loops[i].step); t = fold_build2 (PLUS_EXPR, TREE_TYPE (fd->loops[i].v), fd->loops[i].v, t); } expand_omp_build_assign (&gsi, fd->loops[i].v, t); for (i = fd->collapse - 1; i > 0; i--) { tree itype = TREE_TYPE (fd->loops[i].v); tree itype2 = TREE_TYPE (fd->loops[i - 1].v); if (POINTER_TYPE_P (itype2)) itype2 = signed_type_for (itype2); t = build3 (COND_EXPR, itype2, build2 (fd->loops[i].cond_code, boolean_type_node, fd->loops[i].v, fold_convert (itype, fd->loops[i].n2)), build_int_cst (itype2, 0), fold_convert (itype2, fd->loops[i - 1].step)); if (POINTER_TYPE_P (TREE_TYPE (fd->loops[i - 1].v))) t = fold_build_pointer_plus (fd->loops[i - 1].v, t); else t = fold_build2 (PLUS_EXPR, itype2, fd->loops[i - 1].v, t); expand_omp_build_assign (&gsi, fd->loops[i - 1].v, t); t = build3 (COND_EXPR, itype, build2 (fd->loops[i].cond_code, boolean_type_node, fd->loops[i].v, fold_convert (itype, fd->loops[i].n2)), fd->loops[i].v, fold_convert (itype, fd->loops[i].n1)); expand_omp_build_assign (&gsi, fd->loops[i].v, t); } } /* Remove GIMPLE_OMP_CONTINUE. */ gsi_remove (&gsi, true); } /* Emit the condition in L1_BB. */ gsi = gsi_start_bb (l1_bb); t = fold_convert (type, n2); t = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); tree v = fd->loop.v; if (DECL_P (v) && TREE_ADDRESSABLE (v)) v = force_gimple_operand_gsi (&gsi, v, true, NULL_TREE, false, GSI_CONTINUE_LINKING); t = build2 (fd->loop.cond_code, boolean_type_node, v, t); cond_stmt = gimple_build_cond_empty (t); gsi_insert_after (&gsi, cond_stmt, GSI_CONTINUE_LINKING); if (walk_tree (gimple_cond_lhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL) || walk_tree (gimple_cond_rhs_ptr (cond_stmt), expand_omp_regimplify_p, NULL, NULL)) { gsi = gsi_for_stmt (cond_stmt); gimple_regimplify_operands (cond_stmt, &gsi); } /* Add 'V -= STEP * (SIMT_VF - 1)' after the loop. */ if (is_simt) { gsi = gsi_start_bb (l2_bb); step = fold_build2 (MINUS_EXPR, TREE_TYPE (step), fd->loop.step, step); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (fd->loop.v, step); else t = fold_build2 (PLUS_EXPR, type, fd->loop.v, step); expand_omp_build_assign (&gsi, fd->loop.v, t); } /* Remove GIMPLE_OMP_RETURN. */ gsi = gsi_last_bb (exit_bb); gsi_remove (&gsi, true); /* Connect the new blocks. */ remove_edge (FALLTHRU_EDGE (entry_bb)); if (!broken_loop) { remove_edge (BRANCH_EDGE (entry_bb)); make_edge (entry_bb, l1_bb, EDGE_FALLTHRU); e = BRANCH_EDGE (l1_bb); ne = FALLTHRU_EDGE (l1_bb); e->flags = EDGE_TRUE_VALUE; } else { single_succ_edge (entry_bb)->flags = EDGE_FALLTHRU; ne = single_succ_edge (l1_bb); e = make_edge (l1_bb, l0_bb, EDGE_TRUE_VALUE); } ne->flags = EDGE_FALSE_VALUE; e->probability = REG_BR_PROB_BASE * 7 / 8; ne->probability = REG_BR_PROB_BASE / 8; set_immediate_dominator (CDI_DOMINATORS, l1_bb, entry_bb); set_immediate_dominator (CDI_DOMINATORS, l0_bb, l1_bb); if (simt_maxlane) { cond_stmt = gimple_build_cond (LT_EXPR, simt_lane, simt_maxlane, NULL_TREE, NULL_TREE); gsi = gsi_last_bb (entry_bb); gsi_insert_after (&gsi, cond_stmt, GSI_NEW_STMT); make_edge (entry_bb, l2_bb, EDGE_FALSE_VALUE); FALLTHRU_EDGE (entry_bb)->flags = EDGE_TRUE_VALUE; FALLTHRU_EDGE (entry_bb)->probability = REG_BR_PROB_BASE * 7 / 8; BRANCH_EDGE (entry_bb)->probability = REG_BR_PROB_BASE / 8; l2_dom_bb = entry_bb; } set_immediate_dominator (CDI_DOMINATORS, l2_bb, l2_dom_bb); if (!broken_loop) { struct loop *loop = alloc_loop (); loop->header = l1_bb; loop->latch = cont_bb; add_loop (loop, l1_bb->loop_father); loop->safelen = safelen_int; if (simduid) { loop->simduid = OMP_CLAUSE__SIMDUID__DECL (simduid); cfun->has_simduid_loops = true; } /* If not -fno-tree-loop-vectorize, hint that we want to vectorize the loop. */ if ((flag_tree_loop_vectorize || (!global_options_set.x_flag_tree_loop_vectorize && !global_options_set.x_flag_tree_vectorize)) && flag_tree_loop_optimize && loop->safelen > 1) { loop->force_vectorize = true; cfun->has_force_vectorize_loops = true; } } else if (simduid) cfun->has_simduid_loops = true; } /* Taskloop construct is represented after gimplification with two GIMPLE_OMP_FOR constructs with GIMPLE_OMP_TASK sandwiched in between them. This routine expands the outer GIMPLE_OMP_FOR, which should just compute all the needed loop temporaries for GIMPLE_OMP_TASK. */ static void expand_omp_taskloop_for_outer (struct omp_region *region, struct omp_for_data *fd, gimple *inner_stmt) { tree type, bias = NULL_TREE; basic_block entry_bb, cont_bb, exit_bb; gimple_stmt_iterator gsi; gassign *assign_stmt; tree *counts = NULL; int i; gcc_assert (inner_stmt); gcc_assert (region->cont); gcc_assert (gimple_code (inner_stmt) == GIMPLE_OMP_TASK && gimple_omp_task_taskloop_p (inner_stmt)); type = TREE_TYPE (fd->loop.v); /* See if we need to bias by LLONG_MIN. */ if (fd->iter_type == long_long_unsigned_type_node && TREE_CODE (type) == INTEGER_TYPE && !TYPE_UNSIGNED (type)) { tree n1, n2; if (fd->loop.cond_code == LT_EXPR) { n1 = fd->loop.n1; n2 = fold_build2 (PLUS_EXPR, type, fd->loop.n2, fd->loop.step); } else { n1 = fold_build2 (MINUS_EXPR, type, fd->loop.n2, fd->loop.step); n2 = fd->loop.n1; } if (TREE_CODE (n1) != INTEGER_CST || TREE_CODE (n2) != INTEGER_CST || ((tree_int_cst_sgn (n1) < 0) ^ (tree_int_cst_sgn (n2) < 0))) bias = fold_convert (fd->iter_type, TYPE_MIN_VALUE (type)); } entry_bb = region->entry; cont_bb = region->cont; gcc_assert (EDGE_COUNT (entry_bb->succs) == 2); gcc_assert (BRANCH_EDGE (entry_bb)->dest == FALLTHRU_EDGE (cont_bb)->dest); exit_bb = region->exit; gsi = gsi_last_bb (entry_bb); gimple *for_stmt = gsi_stmt (gsi); gcc_assert (gimple_code (for_stmt) == GIMPLE_OMP_FOR); if (fd->collapse > 1) { int first_zero_iter = -1, dummy = -1; basic_block zero_iter_bb = NULL, dummy_bb = NULL, l2_dom_bb = NULL; counts = XALLOCAVEC (tree, fd->collapse); expand_omp_for_init_counts (fd, &gsi, entry_bb, counts, zero_iter_bb, first_zero_iter, dummy_bb, dummy, l2_dom_bb); if (zero_iter_bb) { /* Some counts[i] vars might be uninitialized if some loop has zero iterations. But the body shouldn't be executed in that case, so just avoid uninit warnings. */ for (i = first_zero_iter; i < fd->collapse; i++) if (SSA_VAR_P (counts[i])) TREE_NO_WARNING (counts[i]) = 1; gsi_prev (&gsi); edge e = split_block (entry_bb, gsi_stmt (gsi)); entry_bb = e->dest; make_edge (zero_iter_bb, entry_bb, EDGE_FALLTHRU); gsi = gsi_last_bb (entry_bb); set_immediate_dominator (CDI_DOMINATORS, entry_bb, get_immediate_dominator (CDI_DOMINATORS, zero_iter_bb)); } } tree t0, t1; t1 = fd->loop.n2; t0 = fd->loop.n1; if (POINTER_TYPE_P (TREE_TYPE (t0)) && TYPE_PRECISION (TREE_TYPE (t0)) != TYPE_PRECISION (fd->iter_type)) { /* Avoid casting pointers to integer of a different size. */ tree itype = signed_type_for (type); t1 = fold_convert (fd->iter_type, fold_convert (itype, t1)); t0 = fold_convert (fd->iter_type, fold_convert (itype, t0)); } else { t1 = fold_convert (fd->iter_type, t1); t0 = fold_convert (fd->iter_type, t0); } if (bias) { t1 = fold_build2 (PLUS_EXPR, fd->iter_type, t1, bias); t0 = fold_build2 (PLUS_EXPR, fd->iter_type, t0, bias); } tree innerc = omp_find_clause (gimple_omp_task_clauses (inner_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); tree startvar = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); tree endvar = OMP_CLAUSE_DECL (innerc); if (fd->collapse > 1 && TREE_CODE (fd->loop.n2) != INTEGER_CST) { gcc_assert (innerc); for (i = 1; i < fd->collapse; i++) { innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); } innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); if (innerc) { /* If needed (inner taskloop has lastprivate clause), propagate down the total number of iterations. */ tree t = force_gimple_operand_gsi (&gsi, fd->loop.n2, false, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (OMP_CLAUSE_DECL (innerc), t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } } t0 = force_gimple_operand_gsi (&gsi, t0, false, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (startvar, t0); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); t1 = force_gimple_operand_gsi (&gsi, t1, false, NULL_TREE, false, GSI_CONTINUE_LINKING); assign_stmt = gimple_build_assign (endvar, t1); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); if (fd->collapse > 1) expand_omp_for_init_vars (fd, &gsi, counts, inner_stmt, startvar); /* Remove the GIMPLE_OMP_FOR statement. */ gsi = gsi_for_stmt (for_stmt); gsi_remove (&gsi, true); gsi = gsi_last_bb (cont_bb); gsi_remove (&gsi, true); gsi = gsi_last_bb (exit_bb); gsi_remove (&gsi, true); FALLTHRU_EDGE (entry_bb)->probability = REG_BR_PROB_BASE; remove_edge (BRANCH_EDGE (entry_bb)); FALLTHRU_EDGE (cont_bb)->probability = REG_BR_PROB_BASE; remove_edge (BRANCH_EDGE (cont_bb)); set_immediate_dominator (CDI_DOMINATORS, exit_bb, cont_bb); set_immediate_dominator (CDI_DOMINATORS, region->entry, recompute_dominator (CDI_DOMINATORS, region->entry)); } /* Taskloop construct is represented after gimplification with two GIMPLE_OMP_FOR constructs with GIMPLE_OMP_TASK sandwiched in between them. This routine expands the inner GIMPLE_OMP_FOR. GOMP_taskloop{,_ull} function arranges for each task to be given just a single range of iterations. */ static void expand_omp_taskloop_for_inner (struct omp_region *region, struct omp_for_data *fd, gimple *inner_stmt) { tree e, t, type, itype, vmain, vback, bias = NULL_TREE; basic_block entry_bb, exit_bb, body_bb, cont_bb, collapse_bb = NULL; basic_block fin_bb; gimple_stmt_iterator gsi; edge ep; bool broken_loop = region->cont == NULL; tree *counts = NULL; tree n1, n2, step; itype = type = TREE_TYPE (fd->loop.v); if (POINTER_TYPE_P (type)) itype = signed_type_for (type); /* See if we need to bias by LLONG_MIN. */ if (fd->iter_type == long_long_unsigned_type_node && TREE_CODE (type) == INTEGER_TYPE && !TYPE_UNSIGNED (type)) { tree n1, n2; if (fd->loop.cond_code == LT_EXPR) { n1 = fd->loop.n1; n2 = fold_build2 (PLUS_EXPR, type, fd->loop.n2, fd->loop.step); } else { n1 = fold_build2 (MINUS_EXPR, type, fd->loop.n2, fd->loop.step); n2 = fd->loop.n1; } if (TREE_CODE (n1) != INTEGER_CST || TREE_CODE (n2) != INTEGER_CST || ((tree_int_cst_sgn (n1) < 0) ^ (tree_int_cst_sgn (n2) < 0))) bias = fold_convert (fd->iter_type, TYPE_MIN_VALUE (type)); } entry_bb = region->entry; cont_bb = region->cont; gcc_assert (EDGE_COUNT (entry_bb->succs) == 2); fin_bb = BRANCH_EDGE (entry_bb)->dest; gcc_assert (broken_loop || (fin_bb == FALLTHRU_EDGE (cont_bb)->dest)); body_bb = FALLTHRU_EDGE (entry_bb)->dest; if (!broken_loop) { gcc_assert (BRANCH_EDGE (cont_bb)->dest == body_bb); gcc_assert (EDGE_COUNT (cont_bb->succs) == 2); } exit_bb = region->exit; /* Iteration space partitioning goes in ENTRY_BB. */ gsi = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_FOR); if (fd->collapse > 1) { int first_zero_iter = -1, dummy = -1; basic_block l2_dom_bb = NULL, dummy_bb = NULL; counts = XALLOCAVEC (tree, fd->collapse); expand_omp_for_init_counts (fd, &gsi, entry_bb, counts, fin_bb, first_zero_iter, dummy_bb, dummy, l2_dom_bb); t = NULL_TREE; } else t = integer_one_node; step = fd->loop.step; tree innerc = omp_find_clause (gimple_omp_for_clauses (fd->for_stmt), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n1 = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); n2 = OMP_CLAUSE_DECL (innerc); if (bias) { n1 = fold_build2 (PLUS_EXPR, fd->iter_type, n1, bias); n2 = fold_build2 (PLUS_EXPR, fd->iter_type, n2, bias); } n1 = force_gimple_operand_gsi (&gsi, fold_convert (type, n1), true, NULL_TREE, true, GSI_SAME_STMT); n2 = force_gimple_operand_gsi (&gsi, fold_convert (itype, n2), true, NULL_TREE, true, GSI_SAME_STMT); step = force_gimple_operand_gsi (&gsi, fold_convert (itype, step), true, NULL_TREE, true, GSI_SAME_STMT); tree startvar = fd->loop.v; tree endvar = NULL_TREE; if (gimple_omp_for_combined_p (fd->for_stmt)) { tree clauses = gimple_omp_for_clauses (inner_stmt); tree innerc = omp_find_clause (clauses, OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); startvar = OMP_CLAUSE_DECL (innerc); innerc = omp_find_clause (OMP_CLAUSE_CHAIN (innerc), OMP_CLAUSE__LOOPTEMP_); gcc_assert (innerc); endvar = OMP_CLAUSE_DECL (innerc); } t = fold_convert (TREE_TYPE (startvar), n1); t = force_gimple_operand_gsi (&gsi, t, DECL_P (startvar) && TREE_ADDRESSABLE (startvar), NULL_TREE, false, GSI_CONTINUE_LINKING); gimple *assign_stmt = gimple_build_assign (startvar, t); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); t = fold_convert (TREE_TYPE (startvar), n2); e = force_gimple_operand_gsi (&gsi, t, true, NULL_TREE, false, GSI_CONTINUE_LINKING); if (endvar) { assign_stmt = gimple_build_assign (endvar, e); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); if (useless_type_conversion_p (TREE_TYPE (fd->loop.v), TREE_TYPE (e))) assign_stmt = gimple_build_assign (fd->loop.v, e); else assign_stmt = gimple_build_assign (fd->loop.v, NOP_EXPR, e); gsi_insert_after (&gsi, assign_stmt, GSI_CONTINUE_LINKING); } if (fd->collapse > 1) expand_omp_for_init_vars (fd, &gsi, counts, inner_stmt, startvar); if (!broken_loop) { /* The code controlling the sequential loop replaces the GIMPLE_OMP_CONTINUE. */ gsi = gsi_last_bb (cont_bb); gomp_continue *cont_stmt = as_a (gsi_stmt (gsi)); gcc_assert (gimple_code (cont_stmt) == GIMPLE_OMP_CONTINUE); vmain = gimple_omp_continue_control_use (cont_stmt); vback = gimple_omp_continue_control_def (cont_stmt); if (!gimple_omp_for_combined_p (fd->for_stmt)) { if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (vmain, step); else t = fold_build2 (PLUS_EXPR, type, vmain, step); t = force_gimple_operand_gsi (&gsi, t, DECL_P (vback) && TREE_ADDRESSABLE (vback), NULL_TREE, true, GSI_SAME_STMT); assign_stmt = gimple_build_assign (vback, t); gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT); t = build2 (fd->loop.cond_code, boolean_type_node, DECL_P (vback) && TREE_ADDRESSABLE (vback) ? t : vback, e); gsi_insert_before (&gsi, gimple_build_cond_empty (t), GSI_SAME_STMT); } /* Remove the GIMPLE_OMP_CONTINUE statement. */ gsi_remove (&gsi, true); if (fd->collapse > 1 && !gimple_omp_for_combined_p (fd->for_stmt)) collapse_bb = extract_omp_for_update_vars (fd, cont_bb, body_bb); } /* Remove the GIMPLE_OMP_FOR statement. */ gsi = gsi_for_stmt (fd->for_stmt); gsi_remove (&gsi, true); /* Remove the GIMPLE_OMP_RETURN statement. */ gsi = gsi_last_bb (exit_bb); gsi_remove (&gsi, true); FALLTHRU_EDGE (entry_bb)->probability = REG_BR_PROB_BASE; if (!broken_loop) remove_edge (BRANCH_EDGE (entry_bb)); else { remove_edge_and_dominated_blocks (BRANCH_EDGE (entry_bb)); region->outer->cont = NULL; } /* Connect all the blocks. */ if (!broken_loop) { ep = find_edge (cont_bb, body_bb); if (gimple_omp_for_combined_p (fd->for_stmt)) { remove_edge (ep); ep = NULL; } else if (fd->collapse > 1) { remove_edge (ep); ep = make_edge (cont_bb, collapse_bb, EDGE_TRUE_VALUE); } else ep->flags = EDGE_TRUE_VALUE; find_edge (cont_bb, fin_bb)->flags = ep ? EDGE_FALSE_VALUE : EDGE_FALLTHRU; } set_immediate_dominator (CDI_DOMINATORS, body_bb, recompute_dominator (CDI_DOMINATORS, body_bb)); if (!broken_loop) set_immediate_dominator (CDI_DOMINATORS, fin_bb, recompute_dominator (CDI_DOMINATORS, fin_bb)); if (!broken_loop && !gimple_omp_for_combined_p (fd->for_stmt)) { struct loop *loop = alloc_loop (); loop->header = body_bb; if (collapse_bb == NULL) loop->latch = cont_bb; add_loop (loop, body_bb->loop_father); } } /* A subroutine of expand_omp_for. Generate code for an OpenACC partitioned loop. The lowering here is abstracted, in that the loop parameters are passed through internal functions, which are further lowered by oacc_device_lower, once we get to the target compiler. The loop is of the form: for (V = B; V LTGT E; V += S) {BODY} where LTGT is < or >. We may have a specified chunking size, CHUNKING (constant 0 for no chunking) and we will have a GWV partitioning mask, specifying dimensions over which the loop is to be partitioned (see note below). We generate code that looks like (this ignores tiling): [incoming FALL->body, BRANCH->exit] typedef signedintify (typeof (V)) T; // underlying signed integral type T range = E - B; T chunk_no = 0; T DIR = LTGT == '<' ? +1 : -1; T chunk_max = GOACC_LOOP_CHUNK (dir, range, S, CHUNK_SIZE, GWV); T step = GOACC_LOOP_STEP (dir, range, S, CHUNK_SIZE, GWV); [created by splitting end of entry_bb] T offset = GOACC_LOOP_OFFSET (dir, range, S, CHUNK_SIZE, GWV, chunk_no); T bound = GOACC_LOOP_BOUND (dir, range, S, CHUNK_SIZE, GWV, offset); if (!(offset LTGT bound)) goto bottom_bb; [incoming] V = B + offset; {BODY} [incoming, may == body_bb FALL->exit_bb, BRANCH->body_bb] offset += step; if (offset LTGT bound) goto body_bb; [*] [created by splitting start of exit_bb] insert BRANCH->head_bb chunk_no++; if (chunk < chunk_max) goto head_bb; [incoming] V = B + ((range -/+ 1) / S +/- 1) * S [*] [*] Needed if V live at end of loop. */ static void expand_oacc_for (struct omp_region *region, struct omp_for_data *fd) { tree v = fd->loop.v; enum tree_code cond_code = fd->loop.cond_code; enum tree_code plus_code = PLUS_EXPR; tree chunk_size = integer_minus_one_node; tree gwv = integer_zero_node; tree iter_type = TREE_TYPE (v); tree diff_type = iter_type; tree plus_type = iter_type; struct oacc_collapse *counts = NULL; gcc_checking_assert (gimple_omp_for_kind (fd->for_stmt) == GF_OMP_FOR_KIND_OACC_LOOP); gcc_assert (!gimple_omp_for_combined_into_p (fd->for_stmt)); gcc_assert (cond_code == LT_EXPR || cond_code == GT_EXPR); if (POINTER_TYPE_P (iter_type)) { plus_code = POINTER_PLUS_EXPR; plus_type = sizetype; } if (POINTER_TYPE_P (diff_type) || TYPE_UNSIGNED (diff_type)) diff_type = signed_type_for (diff_type); basic_block entry_bb = region->entry; /* BB ending in OMP_FOR */ basic_block exit_bb = region->exit; /* BB ending in OMP_RETURN */ basic_block cont_bb = region->cont; /* BB ending in OMP_CONTINUE */ basic_block bottom_bb = NULL; /* entry_bb has two sucessors; the branch edge is to the exit block, fallthrough edge to body. */ gcc_assert (EDGE_COUNT (entry_bb->succs) == 2 && BRANCH_EDGE (entry_bb)->dest == exit_bb); /* If cont_bb non-NULL, it has 2 successors. The branch successor is body_bb, or to a block whose only successor is the body_bb. Its fallthrough successor is the final block (same as the branch successor of the entry_bb). */ if (cont_bb) { basic_block body_bb = FALLTHRU_EDGE (entry_bb)->dest; basic_block bed = BRANCH_EDGE (cont_bb)->dest; gcc_assert (FALLTHRU_EDGE (cont_bb)->dest == exit_bb); gcc_assert (bed == body_bb || single_succ_edge (bed)->dest == body_bb); } else gcc_assert (!gimple_in_ssa_p (cfun)); /* The exit block only has entry_bb and cont_bb as predecessors. */ gcc_assert (EDGE_COUNT (exit_bb->preds) == 1 + (cont_bb != NULL)); tree chunk_no; tree chunk_max = NULL_TREE; tree bound, offset; tree step = create_tmp_var (diff_type, ".step"); bool up = cond_code == LT_EXPR; tree dir = build_int_cst (diff_type, up ? +1 : -1); bool chunking = !gimple_in_ssa_p (cfun); bool negating; /* Tiling vars. */ tree tile_size = NULL_TREE; tree element_s = NULL_TREE; tree e_bound = NULL_TREE, e_offset = NULL_TREE, e_step = NULL_TREE; basic_block elem_body_bb = NULL; basic_block elem_cont_bb = NULL; /* SSA instances. */ tree offset_incr = NULL_TREE; tree offset_init = NULL_TREE; gimple_stmt_iterator gsi; gassign *ass; gcall *call; gimple *stmt; tree expr; location_t loc; edge split, be, fte; /* Split the end of entry_bb to create head_bb. */ split = split_block (entry_bb, last_stmt (entry_bb)); basic_block head_bb = split->dest; entry_bb = split->src; /* Chunk setup goes at end of entry_bb, replacing the omp_for. */ gsi = gsi_last_bb (entry_bb); gomp_for *for_stmt = as_a (gsi_stmt (gsi)); loc = gimple_location (for_stmt); if (gimple_in_ssa_p (cfun)) { offset_init = gimple_omp_for_index (for_stmt, 0); gcc_assert (integer_zerop (fd->loop.n1)); /* The SSA parallelizer does gang parallelism. */ gwv = build_int_cst (integer_type_node, GOMP_DIM_MASK (GOMP_DIM_GANG)); } if (fd->collapse > 1 || fd->tiling) { gcc_assert (!gimple_in_ssa_p (cfun) && up); counts = XALLOCAVEC (struct oacc_collapse, fd->collapse); tree total = expand_oacc_collapse_init (fd, &gsi, counts, TREE_TYPE (fd->loop.n2), loc); if (SSA_VAR_P (fd->loop.n2)) { total = force_gimple_operand_gsi (&gsi, total, false, NULL_TREE, true, GSI_SAME_STMT); ass = gimple_build_assign (fd->loop.n2, total); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); } } tree b = fd->loop.n1; tree e = fd->loop.n2; tree s = fd->loop.step; b = force_gimple_operand_gsi (&gsi, b, true, NULL_TREE, true, GSI_SAME_STMT); e = force_gimple_operand_gsi (&gsi, e, true, NULL_TREE, true, GSI_SAME_STMT); /* Convert the step, avoiding possible unsigned->signed overflow. */ negating = !up && TYPE_UNSIGNED (TREE_TYPE (s)); if (negating) s = fold_build1 (NEGATE_EXPR, TREE_TYPE (s), s); s = fold_convert (diff_type, s); if (negating) s = fold_build1 (NEGATE_EXPR, diff_type, s); s = force_gimple_operand_gsi (&gsi, s, true, NULL_TREE, true, GSI_SAME_STMT); if (!chunking) chunk_size = integer_zero_node; expr = fold_convert (diff_type, chunk_size); chunk_size = force_gimple_operand_gsi (&gsi, expr, true, NULL_TREE, true, GSI_SAME_STMT); if (fd->tiling) { /* Determine the tile size and element step, modify the outer loop step size. */ tile_size = create_tmp_var (diff_type, ".tile_size"); expr = build_int_cst (diff_type, 1); for (int ix = 0; ix < fd->collapse; ix++) expr = fold_build2 (MULT_EXPR, diff_type, counts[ix].tile, expr); expr = force_gimple_operand_gsi (&gsi, expr, true, NULL_TREE, true, GSI_SAME_STMT); ass = gimple_build_assign (tile_size, expr); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); element_s = create_tmp_var (diff_type, ".element_s"); ass = gimple_build_assign (element_s, s); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); expr = fold_build2 (MULT_EXPR, diff_type, s, tile_size); s = force_gimple_operand_gsi (&gsi, expr, true, NULL_TREE, true, GSI_SAME_STMT); } /* Determine the range, avoiding possible unsigned->signed overflow. */ negating = !up && TYPE_UNSIGNED (iter_type); expr = fold_build2 (MINUS_EXPR, plus_type, fold_convert (plus_type, negating ? b : e), fold_convert (plus_type, negating ? e : b)); expr = fold_convert (diff_type, expr); if (negating) expr = fold_build1 (NEGATE_EXPR, diff_type, expr); tree range = force_gimple_operand_gsi (&gsi, expr, true, NULL_TREE, true, GSI_SAME_STMT); chunk_no = build_int_cst (diff_type, 0); if (chunking) { gcc_assert (!gimple_in_ssa_p (cfun)); expr = chunk_no; chunk_max = create_tmp_var (diff_type, ".chunk_max"); chunk_no = create_tmp_var (diff_type, ".chunk_no"); ass = gimple_build_assign (chunk_no, expr); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); call = gimple_build_call_internal (IFN_GOACC_LOOP, 6, build_int_cst (integer_type_node, IFN_GOACC_LOOP_CHUNKS), dir, range, s, chunk_size, gwv); gimple_call_set_lhs (call, chunk_max); gimple_set_location (call, loc); gsi_insert_before (&gsi, call, GSI_SAME_STMT); } else chunk_size = chunk_no; call = gimple_build_call_internal (IFN_GOACC_LOOP, 6, build_int_cst (integer_type_node, IFN_GOACC_LOOP_STEP), dir, range, s, chunk_size, gwv); gimple_call_set_lhs (call, step); gimple_set_location (call, loc); gsi_insert_before (&gsi, call, GSI_SAME_STMT); /* Remove the GIMPLE_OMP_FOR. */ gsi_remove (&gsi, true); /* Fixup edges from head_bb. */ be = BRANCH_EDGE (head_bb); fte = FALLTHRU_EDGE (head_bb); be->flags |= EDGE_FALSE_VALUE; fte->flags ^= EDGE_FALLTHRU | EDGE_TRUE_VALUE; basic_block body_bb = fte->dest; if (gimple_in_ssa_p (cfun)) { gsi = gsi_last_bb (cont_bb); gomp_continue *cont_stmt = as_a (gsi_stmt (gsi)); offset = gimple_omp_continue_control_use (cont_stmt); offset_incr = gimple_omp_continue_control_def (cont_stmt); } else { offset = create_tmp_var (diff_type, ".offset"); offset_init = offset_incr = offset; } bound = create_tmp_var (TREE_TYPE (offset), ".bound"); /* Loop offset & bound go into head_bb. */ gsi = gsi_start_bb (head_bb); call = gimple_build_call_internal (IFN_GOACC_LOOP, 7, build_int_cst (integer_type_node, IFN_GOACC_LOOP_OFFSET), dir, range, s, chunk_size, gwv, chunk_no); gimple_call_set_lhs (call, offset_init); gimple_set_location (call, loc); gsi_insert_after (&gsi, call, GSI_CONTINUE_LINKING); call = gimple_build_call_internal (IFN_GOACC_LOOP, 7, build_int_cst (integer_type_node, IFN_GOACC_LOOP_BOUND), dir, range, s, chunk_size, gwv, offset_init); gimple_call_set_lhs (call, bound); gimple_set_location (call, loc); gsi_insert_after (&gsi, call, GSI_CONTINUE_LINKING); expr = build2 (cond_code, boolean_type_node, offset_init, bound); gsi_insert_after (&gsi, gimple_build_cond_empty (expr), GSI_CONTINUE_LINKING); /* V assignment goes into body_bb. */ if (!gimple_in_ssa_p (cfun)) { gsi = gsi_start_bb (body_bb); expr = build2 (plus_code, iter_type, b, fold_convert (plus_type, offset)); expr = force_gimple_operand_gsi (&gsi, expr, false, NULL_TREE, true, GSI_SAME_STMT); ass = gimple_build_assign (v, expr); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); if (fd->collapse > 1 || fd->tiling) expand_oacc_collapse_vars (fd, false, &gsi, counts, v); if (fd->tiling) { /* Determine the range of the element loop -- usually simply the tile_size, but could be smaller if the final iteration of the outer loop is a partial tile. */ tree e_range = create_tmp_var (diff_type, ".e_range"); expr = build2 (MIN_EXPR, diff_type, build2 (MINUS_EXPR, diff_type, bound, offset), build2 (MULT_EXPR, diff_type, tile_size, element_s)); expr = force_gimple_operand_gsi (&gsi, expr, false, NULL_TREE, true, GSI_SAME_STMT); ass = gimple_build_assign (e_range, expr); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); /* Determine bound, offset & step of inner loop. */ e_bound = create_tmp_var (diff_type, ".e_bound"); e_offset = create_tmp_var (diff_type, ".e_offset"); e_step = create_tmp_var (diff_type, ".e_step"); /* Mark these as element loops. */ tree t, e_gwv = integer_minus_one_node; tree chunk = build_int_cst (diff_type, 0); /* Never chunked. */ t = build_int_cst (integer_type_node, IFN_GOACC_LOOP_OFFSET); call = gimple_build_call_internal (IFN_GOACC_LOOP, 7, t, dir, e_range, element_s, chunk, e_gwv, chunk); gimple_call_set_lhs (call, e_offset); gimple_set_location (call, loc); gsi_insert_before (&gsi, call, GSI_SAME_STMT); t = build_int_cst (integer_type_node, IFN_GOACC_LOOP_BOUND); call = gimple_build_call_internal (IFN_GOACC_LOOP, 7, t, dir, e_range, element_s, chunk, e_gwv, e_offset); gimple_call_set_lhs (call, e_bound); gimple_set_location (call, loc); gsi_insert_before (&gsi, call, GSI_SAME_STMT); t = build_int_cst (integer_type_node, IFN_GOACC_LOOP_STEP); call = gimple_build_call_internal (IFN_GOACC_LOOP, 6, t, dir, e_range, element_s, chunk, e_gwv); gimple_call_set_lhs (call, e_step); gimple_set_location (call, loc); gsi_insert_before (&gsi, call, GSI_SAME_STMT); /* Add test and split block. */ expr = build2 (cond_code, boolean_type_node, e_offset, e_bound); stmt = gimple_build_cond_empty (expr); gsi_insert_before (&gsi, stmt, GSI_SAME_STMT); split = split_block (body_bb, stmt); elem_body_bb = split->dest; if (cont_bb == body_bb) cont_bb = elem_body_bb; body_bb = split->src; split->flags ^= EDGE_FALLTHRU | EDGE_TRUE_VALUE; /* Initialize the user's loop vars. */ gsi = gsi_start_bb (elem_body_bb); expand_oacc_collapse_vars (fd, true, &gsi, counts, e_offset); } } /* Loop increment goes into cont_bb. If this is not a loop, we will have spawned threads as if it was, and each one will execute one iteration. The specification is not explicit about whether such constructs are ill-formed or not, and they can occur, especially when noreturn routines are involved. */ if (cont_bb) { gsi = gsi_last_bb (cont_bb); gomp_continue *cont_stmt = as_a (gsi_stmt (gsi)); loc = gimple_location (cont_stmt); if (fd->tiling) { /* Insert element loop increment and test. */ expr = build2 (PLUS_EXPR, diff_type, e_offset, e_step); expr = force_gimple_operand_gsi (&gsi, expr, false, NULL_TREE, true, GSI_SAME_STMT); ass = gimple_build_assign (e_offset, expr); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); expr = build2 (cond_code, boolean_type_node, e_offset, e_bound); stmt = gimple_build_cond_empty (expr); gsi_insert_before (&gsi, stmt, GSI_SAME_STMT); split = split_block (cont_bb, stmt); elem_cont_bb = split->src; cont_bb = split->dest; split->flags ^= EDGE_FALLTHRU | EDGE_FALSE_VALUE; make_edge (elem_cont_bb, elem_body_bb, EDGE_TRUE_VALUE); make_edge (body_bb, cont_bb, EDGE_FALSE_VALUE); gsi = gsi_for_stmt (cont_stmt); } /* Increment offset. */ if (gimple_in_ssa_p (cfun)) expr = build2 (plus_code, iter_type, offset, fold_convert (plus_type, step)); else expr = build2 (PLUS_EXPR, diff_type, offset, step); expr = force_gimple_operand_gsi (&gsi, expr, false, NULL_TREE, true, GSI_SAME_STMT); ass = gimple_build_assign (offset_incr, expr); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); expr = build2 (cond_code, boolean_type_node, offset_incr, bound); gsi_insert_before (&gsi, gimple_build_cond_empty (expr), GSI_SAME_STMT); /* Remove the GIMPLE_OMP_CONTINUE. */ gsi_remove (&gsi, true); /* Fixup edges from cont_bb. */ be = BRANCH_EDGE (cont_bb); fte = FALLTHRU_EDGE (cont_bb); be->flags |= EDGE_TRUE_VALUE; fte->flags ^= EDGE_FALLTHRU | EDGE_FALSE_VALUE; if (chunking) { /* Split the beginning of exit_bb to make bottom_bb. We need to insert a nop at the start, because splitting is after a stmt, not before. */ gsi = gsi_start_bb (exit_bb); stmt = gimple_build_nop (); gsi_insert_before (&gsi, stmt, GSI_SAME_STMT); split = split_block (exit_bb, stmt); bottom_bb = split->src; exit_bb = split->dest; gsi = gsi_last_bb (bottom_bb); /* Chunk increment and test goes into bottom_bb. */ expr = build2 (PLUS_EXPR, diff_type, chunk_no, build_int_cst (diff_type, 1)); ass = gimple_build_assign (chunk_no, expr); gsi_insert_after (&gsi, ass, GSI_CONTINUE_LINKING); /* Chunk test at end of bottom_bb. */ expr = build2 (LT_EXPR, boolean_type_node, chunk_no, chunk_max); gsi_insert_after (&gsi, gimple_build_cond_empty (expr), GSI_CONTINUE_LINKING); /* Fixup edges from bottom_bb. */ split->flags ^= EDGE_FALLTHRU | EDGE_FALSE_VALUE; make_edge (bottom_bb, head_bb, EDGE_TRUE_VALUE); } } gsi = gsi_last_bb (exit_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_RETURN); loc = gimple_location (gsi_stmt (gsi)); if (!gimple_in_ssa_p (cfun)) { /* Insert the final value of V, in case it is live. This is the value for the only thread that survives past the join. */ expr = fold_build2 (MINUS_EXPR, diff_type, range, dir); expr = fold_build2 (PLUS_EXPR, diff_type, expr, s); expr = fold_build2 (TRUNC_DIV_EXPR, diff_type, expr, s); expr = fold_build2 (MULT_EXPR, diff_type, expr, s); expr = build2 (plus_code, iter_type, b, fold_convert (plus_type, expr)); expr = force_gimple_operand_gsi (&gsi, expr, false, NULL_TREE, true, GSI_SAME_STMT); ass = gimple_build_assign (v, expr); gsi_insert_before (&gsi, ass, GSI_SAME_STMT); } /* Remove the OMP_RETURN. */ gsi_remove (&gsi, true); if (cont_bb) { /* We now have one, two or three nested loops. Update the loop structures. */ struct loop *parent = entry_bb->loop_father; struct loop *body = body_bb->loop_father; if (chunking) { struct loop *chunk_loop = alloc_loop (); chunk_loop->header = head_bb; chunk_loop->latch = bottom_bb; add_loop (chunk_loop, parent); parent = chunk_loop; } else if (parent != body) { gcc_assert (body->header == body_bb); gcc_assert (body->latch == cont_bb || single_pred (body->latch) == cont_bb); parent = NULL; } if (parent) { struct loop *body_loop = alloc_loop (); body_loop->header = body_bb; body_loop->latch = cont_bb; add_loop (body_loop, parent); if (fd->tiling) { /* Insert tiling's element loop. */ struct loop *inner_loop = alloc_loop (); inner_loop->header = elem_body_bb; inner_loop->latch = elem_cont_bb; add_loop (inner_loop, body_loop); } } } } /* Expand the OMP loop defined by REGION. */ static void expand_omp_for (struct omp_region *region, gimple *inner_stmt) { struct omp_for_data fd; struct omp_for_data_loop *loops; loops = (struct omp_for_data_loop *) alloca (gimple_omp_for_collapse (last_stmt (region->entry)) * sizeof (struct omp_for_data_loop)); omp_extract_for_data (as_a (last_stmt (region->entry)), &fd, loops); region->sched_kind = fd.sched_kind; region->sched_modifiers = fd.sched_modifiers; gcc_assert (EDGE_COUNT (region->entry->succs) == 2); BRANCH_EDGE (region->entry)->flags &= ~EDGE_ABNORMAL; FALLTHRU_EDGE (region->entry)->flags &= ~EDGE_ABNORMAL; if (region->cont) { gcc_assert (EDGE_COUNT (region->cont->succs) == 2); BRANCH_EDGE (region->cont)->flags &= ~EDGE_ABNORMAL; FALLTHRU_EDGE (region->cont)->flags &= ~EDGE_ABNORMAL; } else /* If there isn't a continue then this is a degerate case where the introduction of abnormal edges during lowering will prevent original loops from being detected. Fix that up. */ loops_state_set (LOOPS_NEED_FIXUP); if (gimple_omp_for_kind (fd.for_stmt) & GF_OMP_FOR_SIMD) expand_omp_simd (region, &fd); else if (gimple_omp_for_kind (fd.for_stmt) == GF_OMP_FOR_KIND_CILKFOR) expand_cilk_for (region, &fd); else if (gimple_omp_for_kind (fd.for_stmt) == GF_OMP_FOR_KIND_OACC_LOOP) { gcc_assert (!inner_stmt); expand_oacc_for (region, &fd); } else if (gimple_omp_for_kind (fd.for_stmt) == GF_OMP_FOR_KIND_TASKLOOP) { if (gimple_omp_for_combined_into_p (fd.for_stmt)) expand_omp_taskloop_for_inner (region, &fd, inner_stmt); else expand_omp_taskloop_for_outer (region, &fd, inner_stmt); } else if (fd.sched_kind == OMP_CLAUSE_SCHEDULE_STATIC && !fd.have_ordered) { if (fd.chunk_size == NULL) expand_omp_for_static_nochunk (region, &fd, inner_stmt); else expand_omp_for_static_chunk (region, &fd, inner_stmt); } else { int fn_index, start_ix, next_ix; gcc_assert (gimple_omp_for_kind (fd.for_stmt) == GF_OMP_FOR_KIND_FOR); if (fd.chunk_size == NULL && fd.sched_kind == OMP_CLAUSE_SCHEDULE_STATIC) fd.chunk_size = integer_zero_node; gcc_assert (fd.sched_kind != OMP_CLAUSE_SCHEDULE_AUTO); switch (fd.sched_kind) { case OMP_CLAUSE_SCHEDULE_RUNTIME: fn_index = 3; break; case OMP_CLAUSE_SCHEDULE_DYNAMIC: case OMP_CLAUSE_SCHEDULE_GUIDED: if ((fd.sched_modifiers & OMP_CLAUSE_SCHEDULE_NONMONOTONIC) && !fd.ordered && !fd.have_ordered) { fn_index = 3 + fd.sched_kind; break; } /* FALLTHRU */ default: fn_index = fd.sched_kind; break; } if (!fd.ordered) fn_index += fd.have_ordered * 6; if (fd.ordered) start_ix = ((int)BUILT_IN_GOMP_LOOP_DOACROSS_STATIC_START) + fn_index; else start_ix = ((int)BUILT_IN_GOMP_LOOP_STATIC_START) + fn_index; next_ix = ((int)BUILT_IN_GOMP_LOOP_STATIC_NEXT) + fn_index; if (fd.iter_type == long_long_unsigned_type_node) { start_ix += ((int)BUILT_IN_GOMP_LOOP_ULL_STATIC_START - (int)BUILT_IN_GOMP_LOOP_STATIC_START); next_ix += ((int)BUILT_IN_GOMP_LOOP_ULL_STATIC_NEXT - (int)BUILT_IN_GOMP_LOOP_STATIC_NEXT); } expand_omp_for_generic (region, &fd, (enum built_in_function) start_ix, (enum built_in_function) next_ix, inner_stmt); } if (gimple_in_ssa_p (cfun)) update_ssa (TODO_update_ssa_only_virtuals); } /* Expand code for an OpenMP sections directive. In pseudo code, we generate v = GOMP_sections_start (n); L0: switch (v) { case 0: goto L2; case 1: section 1; goto L1; case 2: ... case n: ... default: abort (); } L1: v = GOMP_sections_next (); goto L0; L2: reduction; If this is a combined parallel sections, replace the call to GOMP_sections_start with call to GOMP_sections_next. */ static void expand_omp_sections (struct omp_region *region) { tree t, u, vin = NULL, vmain, vnext, l2; unsigned len; basic_block entry_bb, l0_bb, l1_bb, l2_bb, default_bb; gimple_stmt_iterator si, switch_si; gomp_sections *sections_stmt; gimple *stmt; gomp_continue *cont; edge_iterator ei; edge e; struct omp_region *inner; unsigned i, casei; bool exit_reachable = region->cont != NULL; gcc_assert (region->exit != NULL); entry_bb = region->entry; l0_bb = single_succ (entry_bb); l1_bb = region->cont; l2_bb = region->exit; if (single_pred_p (l2_bb) && single_pred (l2_bb) == l0_bb) l2 = gimple_block_label (l2_bb); else { /* This can happen if there are reductions. */ len = EDGE_COUNT (l0_bb->succs); gcc_assert (len > 0); e = EDGE_SUCC (l0_bb, len - 1); si = gsi_last_bb (e->dest); l2 = NULL_TREE; if (gsi_end_p (si) || gimple_code (gsi_stmt (si)) != GIMPLE_OMP_SECTION) l2 = gimple_block_label (e->dest); else FOR_EACH_EDGE (e, ei, l0_bb->succs) { si = gsi_last_bb (e->dest); if (gsi_end_p (si) || gimple_code (gsi_stmt (si)) != GIMPLE_OMP_SECTION) { l2 = gimple_block_label (e->dest); break; } } } if (exit_reachable) default_bb = create_empty_bb (l1_bb->prev_bb); else default_bb = create_empty_bb (l0_bb); /* We will build a switch() with enough cases for all the GIMPLE_OMP_SECTION regions, a '0' case to handle the end of more work and a default case to abort if something goes wrong. */ len = EDGE_COUNT (l0_bb->succs); /* Use vec::quick_push on label_vec throughout, since we know the size in advance. */ auto_vec label_vec (len); /* The call to GOMP_sections_start goes in ENTRY_BB, replacing the GIMPLE_OMP_SECTIONS statement. */ si = gsi_last_bb (entry_bb); sections_stmt = as_a (gsi_stmt (si)); gcc_assert (gimple_code (sections_stmt) == GIMPLE_OMP_SECTIONS); vin = gimple_omp_sections_control (sections_stmt); if (!is_combined_parallel (region)) { /* If we are not inside a combined parallel+sections region, call GOMP_sections_start. */ t = build_int_cst (unsigned_type_node, len - 1); u = builtin_decl_explicit (BUILT_IN_GOMP_SECTIONS_START); stmt = gimple_build_call (u, 1, t); } else { /* Otherwise, call GOMP_sections_next. */ u = builtin_decl_explicit (BUILT_IN_GOMP_SECTIONS_NEXT); stmt = gimple_build_call (u, 0); } gimple_call_set_lhs (stmt, vin); gsi_insert_after (&si, stmt, GSI_SAME_STMT); gsi_remove (&si, true); /* The switch() statement replacing GIMPLE_OMP_SECTIONS_SWITCH goes in L0_BB. */ switch_si = gsi_last_bb (l0_bb); gcc_assert (gimple_code (gsi_stmt (switch_si)) == GIMPLE_OMP_SECTIONS_SWITCH); if (exit_reachable) { cont = as_a (last_stmt (l1_bb)); gcc_assert (gimple_code (cont) == GIMPLE_OMP_CONTINUE); vmain = gimple_omp_continue_control_use (cont); vnext = gimple_omp_continue_control_def (cont); } else { vmain = vin; vnext = NULL_TREE; } t = build_case_label (build_int_cst (unsigned_type_node, 0), NULL, l2); label_vec.quick_push (t); i = 1; /* Convert each GIMPLE_OMP_SECTION into a CASE_LABEL_EXPR. */ for (inner = region->inner, casei = 1; inner; inner = inner->next, i++, casei++) { basic_block s_entry_bb, s_exit_bb; /* Skip optional reduction region. */ if (inner->type == GIMPLE_OMP_ATOMIC_LOAD) { --i; --casei; continue; } s_entry_bb = inner->entry; s_exit_bb = inner->exit; t = gimple_block_label (s_entry_bb); u = build_int_cst (unsigned_type_node, casei); u = build_case_label (u, NULL, t); label_vec.quick_push (u); si = gsi_last_bb (s_entry_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_SECTION); gcc_assert (i < len || gimple_omp_section_last_p (gsi_stmt (si))); gsi_remove (&si, true); single_succ_edge (s_entry_bb)->flags = EDGE_FALLTHRU; if (s_exit_bb == NULL) continue; si = gsi_last_bb (s_exit_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_RETURN); gsi_remove (&si, true); single_succ_edge (s_exit_bb)->flags = EDGE_FALLTHRU; } /* Error handling code goes in DEFAULT_BB. */ t = gimple_block_label (default_bb); u = build_case_label (NULL, NULL, t); make_edge (l0_bb, default_bb, 0); add_bb_to_loop (default_bb, current_loops->tree_root); stmt = gimple_build_switch (vmain, u, label_vec); gsi_insert_after (&switch_si, stmt, GSI_SAME_STMT); gsi_remove (&switch_si, true); si = gsi_start_bb (default_bb); stmt = gimple_build_call (builtin_decl_explicit (BUILT_IN_TRAP), 0); gsi_insert_after (&si, stmt, GSI_CONTINUE_LINKING); if (exit_reachable) { tree bfn_decl; /* Code to get the next section goes in L1_BB. */ si = gsi_last_bb (l1_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_CONTINUE); bfn_decl = builtin_decl_explicit (BUILT_IN_GOMP_SECTIONS_NEXT); stmt = gimple_build_call (bfn_decl, 0); gimple_call_set_lhs (stmt, vnext); gsi_insert_after (&si, stmt, GSI_SAME_STMT); gsi_remove (&si, true); single_succ_edge (l1_bb)->flags = EDGE_FALLTHRU; } /* Cleanup function replaces GIMPLE_OMP_RETURN in EXIT_BB. */ si = gsi_last_bb (l2_bb); if (gimple_omp_return_nowait_p (gsi_stmt (si))) t = builtin_decl_explicit (BUILT_IN_GOMP_SECTIONS_END_NOWAIT); else if (gimple_omp_return_lhs (gsi_stmt (si))) t = builtin_decl_explicit (BUILT_IN_GOMP_SECTIONS_END_CANCEL); else t = builtin_decl_explicit (BUILT_IN_GOMP_SECTIONS_END); stmt = gimple_build_call (t, 0); if (gimple_omp_return_lhs (gsi_stmt (si))) gimple_call_set_lhs (stmt, gimple_omp_return_lhs (gsi_stmt (si))); gsi_insert_after (&si, stmt, GSI_SAME_STMT); gsi_remove (&si, true); set_immediate_dominator (CDI_DOMINATORS, default_bb, l0_bb); } /* Expand code for an OpenMP single directive. We've already expanded much of the code, here we simply place the GOMP_barrier call. */ static void expand_omp_single (struct omp_region *region) { basic_block entry_bb, exit_bb; gimple_stmt_iterator si; entry_bb = region->entry; exit_bb = region->exit; si = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_SINGLE); gsi_remove (&si, true); single_succ_edge (entry_bb)->flags = EDGE_FALLTHRU; si = gsi_last_bb (exit_bb); if (!gimple_omp_return_nowait_p (gsi_stmt (si))) { tree t = gimple_omp_return_lhs (gsi_stmt (si)); gsi_insert_after (&si, omp_build_barrier (t), GSI_SAME_STMT); } gsi_remove (&si, true); single_succ_edge (exit_bb)->flags = EDGE_FALLTHRU; } /* Generic expansion for OpenMP synchronization directives: master, ordered and critical. All we need to do here is remove the entry and exit markers for REGION. */ static void expand_omp_synch (struct omp_region *region) { basic_block entry_bb, exit_bb; gimple_stmt_iterator si; entry_bb = region->entry; exit_bb = region->exit; si = gsi_last_bb (entry_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_SINGLE || gimple_code (gsi_stmt (si)) == GIMPLE_OMP_MASTER || gimple_code (gsi_stmt (si)) == GIMPLE_OMP_TASKGROUP || gimple_code (gsi_stmt (si)) == GIMPLE_OMP_ORDERED || gimple_code (gsi_stmt (si)) == GIMPLE_OMP_CRITICAL || gimple_code (gsi_stmt (si)) == GIMPLE_OMP_TEAMS); gsi_remove (&si, true); single_succ_edge (entry_bb)->flags = EDGE_FALLTHRU; if (exit_bb) { si = gsi_last_bb (exit_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_RETURN); gsi_remove (&si, true); single_succ_edge (exit_bb)->flags = EDGE_FALLTHRU; } } /* A subroutine of expand_omp_atomic. Attempt to implement the atomic operation as a normal volatile load. */ static bool expand_omp_atomic_load (basic_block load_bb, tree addr, tree loaded_val, int index) { enum built_in_function tmpbase; gimple_stmt_iterator gsi; basic_block store_bb; location_t loc; gimple *stmt; tree decl, call, type, itype; gsi = gsi_last_bb (load_bb); stmt = gsi_stmt (gsi); gcc_assert (gimple_code (stmt) == GIMPLE_OMP_ATOMIC_LOAD); loc = gimple_location (stmt); /* ??? If the target does not implement atomic_load_optab[mode], and mode is smaller than word size, then expand_atomic_load assumes that the load is atomic. We could avoid the builtin entirely in this case. */ tmpbase = (enum built_in_function) (BUILT_IN_ATOMIC_LOAD_N + index + 1); decl = builtin_decl_explicit (tmpbase); if (decl == NULL_TREE) return false; type = TREE_TYPE (loaded_val); itype = TREE_TYPE (TREE_TYPE (decl)); call = build_call_expr_loc (loc, decl, 2, addr, build_int_cst (NULL, gimple_omp_atomic_seq_cst_p (stmt) ? MEMMODEL_SEQ_CST : MEMMODEL_RELAXED)); if (!useless_type_conversion_p (type, itype)) call = fold_build1_loc (loc, VIEW_CONVERT_EXPR, type, call); call = build2_loc (loc, MODIFY_EXPR, void_type_node, loaded_val, call); force_gimple_operand_gsi (&gsi, call, true, NULL_TREE, true, GSI_SAME_STMT); gsi_remove (&gsi, true); store_bb = single_succ (load_bb); gsi = gsi_last_bb (store_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_ATOMIC_STORE); gsi_remove (&gsi, true); if (gimple_in_ssa_p (cfun)) update_ssa (TODO_update_ssa_no_phi); return true; } /* A subroutine of expand_omp_atomic. Attempt to implement the atomic operation as a normal volatile store. */ static bool expand_omp_atomic_store (basic_block load_bb, tree addr, tree loaded_val, tree stored_val, int index) { enum built_in_function tmpbase; gimple_stmt_iterator gsi; basic_block store_bb = single_succ (load_bb); location_t loc; gimple *stmt; tree decl, call, type, itype; machine_mode imode; bool exchange; gsi = gsi_last_bb (load_bb); stmt = gsi_stmt (gsi); gcc_assert (gimple_code (stmt) == GIMPLE_OMP_ATOMIC_LOAD); /* If the load value is needed, then this isn't a store but an exchange. */ exchange = gimple_omp_atomic_need_value_p (stmt); gsi = gsi_last_bb (store_bb); stmt = gsi_stmt (gsi); gcc_assert (gimple_code (stmt) == GIMPLE_OMP_ATOMIC_STORE); loc = gimple_location (stmt); /* ??? If the target does not implement atomic_store_optab[mode], and mode is smaller than word size, then expand_atomic_store assumes that the store is atomic. We could avoid the builtin entirely in this case. */ tmpbase = (exchange ? BUILT_IN_ATOMIC_EXCHANGE_N : BUILT_IN_ATOMIC_STORE_N); tmpbase = (enum built_in_function) ((int) tmpbase + index + 1); decl = builtin_decl_explicit (tmpbase); if (decl == NULL_TREE) return false; type = TREE_TYPE (stored_val); /* Dig out the type of the function's second argument. */ itype = TREE_TYPE (decl); itype = TYPE_ARG_TYPES (itype); itype = TREE_CHAIN (itype); itype = TREE_VALUE (itype); imode = TYPE_MODE (itype); if (exchange && !can_atomic_exchange_p (imode, true)) return false; if (!useless_type_conversion_p (itype, type)) stored_val = fold_build1_loc (loc, VIEW_CONVERT_EXPR, itype, stored_val); call = build_call_expr_loc (loc, decl, 3, addr, stored_val, build_int_cst (NULL, gimple_omp_atomic_seq_cst_p (stmt) ? MEMMODEL_SEQ_CST : MEMMODEL_RELAXED)); if (exchange) { if (!useless_type_conversion_p (type, itype)) call = build1_loc (loc, VIEW_CONVERT_EXPR, type, call); call = build2_loc (loc, MODIFY_EXPR, void_type_node, loaded_val, call); } force_gimple_operand_gsi (&gsi, call, true, NULL_TREE, true, GSI_SAME_STMT); gsi_remove (&gsi, true); /* Remove the GIMPLE_OMP_ATOMIC_LOAD that we verified above. */ gsi = gsi_last_bb (load_bb); gsi_remove (&gsi, true); if (gimple_in_ssa_p (cfun)) update_ssa (TODO_update_ssa_no_phi); return true; } /* A subroutine of expand_omp_atomic. Attempt to implement the atomic operation as a __atomic_fetch_op builtin. INDEX is log2 of the size of the data type, and thus usable to find the index of the builtin decl. Returns false if the expression is not of the proper form. */ static bool expand_omp_atomic_fetch_op (basic_block load_bb, tree addr, tree loaded_val, tree stored_val, int index) { enum built_in_function oldbase, newbase, tmpbase; tree decl, itype, call; tree lhs, rhs; basic_block store_bb = single_succ (load_bb); gimple_stmt_iterator gsi; gimple *stmt; location_t loc; enum tree_code code; bool need_old, need_new; machine_mode imode; bool seq_cst; /* We expect to find the following sequences: load_bb: GIMPLE_OMP_ATOMIC_LOAD (tmp, mem) store_bb: val = tmp OP something; (or: something OP tmp) GIMPLE_OMP_STORE (val) ???FIXME: Allow a more flexible sequence. Perhaps use data flow to pick the statements. */ gsi = gsi_after_labels (store_bb); stmt = gsi_stmt (gsi); loc = gimple_location (stmt); if (!is_gimple_assign (stmt)) return false; gsi_next (&gsi); if (gimple_code (gsi_stmt (gsi)) != GIMPLE_OMP_ATOMIC_STORE) return false; need_new = gimple_omp_atomic_need_value_p (gsi_stmt (gsi)); need_old = gimple_omp_atomic_need_value_p (last_stmt (load_bb)); seq_cst = gimple_omp_atomic_seq_cst_p (last_stmt (load_bb)); gcc_checking_assert (!need_old || !need_new); if (!operand_equal_p (gimple_assign_lhs (stmt), stored_val, 0)) return false; /* Check for one of the supported fetch-op operations. */ code = gimple_assign_rhs_code (stmt); switch (code) { case PLUS_EXPR: case POINTER_PLUS_EXPR: oldbase = BUILT_IN_ATOMIC_FETCH_ADD_N; newbase = BUILT_IN_ATOMIC_ADD_FETCH_N; break; case MINUS_EXPR: oldbase = BUILT_IN_ATOMIC_FETCH_SUB_N; newbase = BUILT_IN_ATOMIC_SUB_FETCH_N; break; case BIT_AND_EXPR: oldbase = BUILT_IN_ATOMIC_FETCH_AND_N; newbase = BUILT_IN_ATOMIC_AND_FETCH_N; break; case BIT_IOR_EXPR: oldbase = BUILT_IN_ATOMIC_FETCH_OR_N; newbase = BUILT_IN_ATOMIC_OR_FETCH_N; break; case BIT_XOR_EXPR: oldbase = BUILT_IN_ATOMIC_FETCH_XOR_N; newbase = BUILT_IN_ATOMIC_XOR_FETCH_N; break; default: return false; } /* Make sure the expression is of the proper form. */ if (operand_equal_p (gimple_assign_rhs1 (stmt), loaded_val, 0)) rhs = gimple_assign_rhs2 (stmt); else if (commutative_tree_code (gimple_assign_rhs_code (stmt)) && operand_equal_p (gimple_assign_rhs2 (stmt), loaded_val, 0)) rhs = gimple_assign_rhs1 (stmt); else return false; tmpbase = ((enum built_in_function) ((need_new ? newbase : oldbase) + index + 1)); decl = builtin_decl_explicit (tmpbase); if (decl == NULL_TREE) return false; itype = TREE_TYPE (TREE_TYPE (decl)); imode = TYPE_MODE (itype); /* We could test all of the various optabs involved, but the fact of the matter is that (with the exception of i486 vs i586 and xadd) all targets that support any atomic operaton optab also implements compare-and-swap. Let optabs.c take care of expanding any compare-and-swap loop. */ if (!can_compare_and_swap_p (imode, true) || !can_atomic_load_p (imode)) return false; gsi = gsi_last_bb (load_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_ATOMIC_LOAD); /* OpenMP does not imply any barrier-like semantics on its atomic ops. It only requires that the operation happen atomically. Thus we can use the RELAXED memory model. */ call = build_call_expr_loc (loc, decl, 3, addr, fold_convert_loc (loc, itype, rhs), build_int_cst (NULL, seq_cst ? MEMMODEL_SEQ_CST : MEMMODEL_RELAXED)); if (need_old || need_new) { lhs = need_old ? loaded_val : stored_val; call = fold_convert_loc (loc, TREE_TYPE (lhs), call); call = build2_loc (loc, MODIFY_EXPR, void_type_node, lhs, call); } else call = fold_convert_loc (loc, void_type_node, call); force_gimple_operand_gsi (&gsi, call, true, NULL_TREE, true, GSI_SAME_STMT); gsi_remove (&gsi, true); gsi = gsi_last_bb (store_bb); gcc_assert (gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_ATOMIC_STORE); gsi_remove (&gsi, true); gsi = gsi_last_bb (store_bb); stmt = gsi_stmt (gsi); gsi_remove (&gsi, true); if (gimple_in_ssa_p (cfun)) { release_defs (stmt); update_ssa (TODO_update_ssa_no_phi); } return true; } /* A subroutine of expand_omp_atomic. Implement the atomic operation as: oldval = *addr; repeat: newval = rhs; // with oldval replacing *addr in rhs oldval = __sync_val_compare_and_swap (addr, oldval, newval); if (oldval != newval) goto repeat; INDEX is log2 of the size of the data type, and thus usable to find the index of the builtin decl. */ static bool expand_omp_atomic_pipeline (basic_block load_bb, basic_block store_bb, tree addr, tree loaded_val, tree stored_val, int index) { tree loadedi, storedi, initial, new_storedi, old_vali; tree type, itype, cmpxchg, iaddr; gimple_stmt_iterator si; basic_block loop_header = single_succ (load_bb); gimple *phi, *stmt; edge e; enum built_in_function fncode; /* ??? We need a non-pointer interface to __atomic_compare_exchange in order to use the RELAXED memory model effectively. */ fncode = (enum built_in_function)((int)BUILT_IN_SYNC_VAL_COMPARE_AND_SWAP_N + index + 1); cmpxchg = builtin_decl_explicit (fncode); if (cmpxchg == NULL_TREE) return false; type = TYPE_MAIN_VARIANT (TREE_TYPE (TREE_TYPE (addr))); itype = TREE_TYPE (TREE_TYPE (cmpxchg)); if (!can_compare_and_swap_p (TYPE_MODE (itype), true) || !can_atomic_load_p (TYPE_MODE (itype))) return false; /* Load the initial value, replacing the GIMPLE_OMP_ATOMIC_LOAD. */ si = gsi_last_bb (load_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_ATOMIC_LOAD); /* For floating-point values, we'll need to view-convert them to integers so that we can perform the atomic compare and swap. Simplify the following code by always setting up the "i"ntegral variables. */ if (!INTEGRAL_TYPE_P (type) && !POINTER_TYPE_P (type)) { tree iaddr_val; iaddr = create_tmp_reg (build_pointer_type_for_mode (itype, ptr_mode, true)); iaddr_val = force_gimple_operand_gsi (&si, fold_convert (TREE_TYPE (iaddr), addr), false, NULL_TREE, true, GSI_SAME_STMT); stmt = gimple_build_assign (iaddr, iaddr_val); gsi_insert_before (&si, stmt, GSI_SAME_STMT); loadedi = create_tmp_var (itype); if (gimple_in_ssa_p (cfun)) loadedi = make_ssa_name (loadedi); } else { iaddr = addr; loadedi = loaded_val; } fncode = (enum built_in_function) (BUILT_IN_ATOMIC_LOAD_N + index + 1); tree loaddecl = builtin_decl_explicit (fncode); if (loaddecl) initial = fold_convert (TREE_TYPE (TREE_TYPE (iaddr)), build_call_expr (loaddecl, 2, iaddr, build_int_cst (NULL_TREE, MEMMODEL_RELAXED))); else initial = build2 (MEM_REF, TREE_TYPE (TREE_TYPE (iaddr)), iaddr, build_int_cst (TREE_TYPE (iaddr), 0)); initial = force_gimple_operand_gsi (&si, initial, true, NULL_TREE, true, GSI_SAME_STMT); /* Move the value to the LOADEDI temporary. */ if (gimple_in_ssa_p (cfun)) { gcc_assert (gimple_seq_empty_p (phi_nodes (loop_header))); phi = create_phi_node (loadedi, loop_header); SET_USE (PHI_ARG_DEF_PTR_FROM_EDGE (phi, single_succ_edge (load_bb)), initial); } else gsi_insert_before (&si, gimple_build_assign (loadedi, initial), GSI_SAME_STMT); if (loadedi != loaded_val) { gimple_stmt_iterator gsi2; tree x; x = build1 (VIEW_CONVERT_EXPR, type, loadedi); gsi2 = gsi_start_bb (loop_header); if (gimple_in_ssa_p (cfun)) { gassign *stmt; x = force_gimple_operand_gsi (&gsi2, x, true, NULL_TREE, true, GSI_SAME_STMT); stmt = gimple_build_assign (loaded_val, x); gsi_insert_before (&gsi2, stmt, GSI_SAME_STMT); } else { x = build2 (MODIFY_EXPR, TREE_TYPE (loaded_val), loaded_val, x); force_gimple_operand_gsi (&gsi2, x, true, NULL_TREE, true, GSI_SAME_STMT); } } gsi_remove (&si, true); si = gsi_last_bb (store_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_ATOMIC_STORE); if (iaddr == addr) storedi = stored_val; else storedi = force_gimple_operand_gsi (&si, build1 (VIEW_CONVERT_EXPR, itype, stored_val), true, NULL_TREE, true, GSI_SAME_STMT); /* Build the compare&swap statement. */ new_storedi = build_call_expr (cmpxchg, 3, iaddr, loadedi, storedi); new_storedi = force_gimple_operand_gsi (&si, fold_convert (TREE_TYPE (loadedi), new_storedi), true, NULL_TREE, true, GSI_SAME_STMT); if (gimple_in_ssa_p (cfun)) old_vali = loadedi; else { old_vali = create_tmp_var (TREE_TYPE (loadedi)); stmt = gimple_build_assign (old_vali, loadedi); gsi_insert_before (&si, stmt, GSI_SAME_STMT); stmt = gimple_build_assign (loadedi, new_storedi); gsi_insert_before (&si, stmt, GSI_SAME_STMT); } /* Note that we always perform the comparison as an integer, even for floating point. This allows the atomic operation to properly succeed even with NaNs and -0.0. */ tree ne = build2 (NE_EXPR, boolean_type_node, new_storedi, old_vali); stmt = gimple_build_cond_empty (ne); gsi_insert_before (&si, stmt, GSI_SAME_STMT); /* Update cfg. */ e = single_succ_edge (store_bb); e->flags &= ~EDGE_FALLTHRU; e->flags |= EDGE_FALSE_VALUE; e = make_edge (store_bb, loop_header, EDGE_TRUE_VALUE); /* Copy the new value to loadedi (we already did that before the condition if we are not in SSA). */ if (gimple_in_ssa_p (cfun)) { phi = gimple_seq_first_stmt (phi_nodes (loop_header)); SET_USE (PHI_ARG_DEF_PTR_FROM_EDGE (phi, e), new_storedi); } /* Remove GIMPLE_OMP_ATOMIC_STORE. */ gsi_remove (&si, true); struct loop *loop = alloc_loop (); loop->header = loop_header; loop->latch = store_bb; add_loop (loop, loop_header->loop_father); if (gimple_in_ssa_p (cfun)) update_ssa (TODO_update_ssa_no_phi); return true; } /* A subroutine of expand_omp_atomic. Implement the atomic operation as: GOMP_atomic_start (); *addr = rhs; GOMP_atomic_end (); The result is not globally atomic, but works so long as all parallel references are within #pragma omp atomic directives. According to responses received from omp@openmp.org, appears to be within spec. Which makes sense, since that's how several other compilers handle this situation as well. LOADED_VAL and ADDR are the operands of GIMPLE_OMP_ATOMIC_LOAD we're expanding. STORED_VAL is the operand of the matching GIMPLE_OMP_ATOMIC_STORE. We replace GIMPLE_OMP_ATOMIC_LOAD (loaded_val, addr) with loaded_val = *addr; and replace GIMPLE_OMP_ATOMIC_STORE (stored_val) with *addr = stored_val; */ static bool expand_omp_atomic_mutex (basic_block load_bb, basic_block store_bb, tree addr, tree loaded_val, tree stored_val) { gimple_stmt_iterator si; gassign *stmt; tree t; si = gsi_last_bb (load_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_ATOMIC_LOAD); t = builtin_decl_explicit (BUILT_IN_GOMP_ATOMIC_START); t = build_call_expr (t, 0); force_gimple_operand_gsi (&si, t, true, NULL_TREE, true, GSI_SAME_STMT); stmt = gimple_build_assign (loaded_val, build_simple_mem_ref (addr)); gsi_insert_before (&si, stmt, GSI_SAME_STMT); gsi_remove (&si, true); si = gsi_last_bb (store_bb); gcc_assert (gimple_code (gsi_stmt (si)) == GIMPLE_OMP_ATOMIC_STORE); stmt = gimple_build_assign (build_simple_mem_ref (unshare_expr (addr)), stored_val); gsi_insert_before (&si, stmt, GSI_SAME_STMT); t = builtin_decl_explicit (BUILT_IN_GOMP_ATOMIC_END); t = build_call_expr (t, 0); force_gimple_operand_gsi (&si, t, true, NULL_TREE, true, GSI_SAME_STMT); gsi_remove (&si, true); if (gimple_in_ssa_p (cfun)) update_ssa (TODO_update_ssa_no_phi); return true; } /* Expand an GIMPLE_OMP_ATOMIC statement. We try to expand using expand_omp_atomic_fetch_op. If it failed, we try to call expand_omp_atomic_pipeline, and if it fails too, the ultimate fallback is wrapping the operation in a mutex (expand_omp_atomic_mutex). REGION is the atomic region built by build_omp_regions_1(). */ static void expand_omp_atomic (struct omp_region *region) { basic_block load_bb = region->entry, store_bb = region->exit; gomp_atomic_load *load = as_a (last_stmt (load_bb)); gomp_atomic_store *store = as_a (last_stmt (store_bb)); tree loaded_val = gimple_omp_atomic_load_lhs (load); tree addr = gimple_omp_atomic_load_rhs (load); tree stored_val = gimple_omp_atomic_store_val (store); tree type = TYPE_MAIN_VARIANT (TREE_TYPE (TREE_TYPE (addr))); HOST_WIDE_INT index; /* Make sure the type is one of the supported sizes. */ index = tree_to_uhwi (TYPE_SIZE_UNIT (type)); index = exact_log2 (index); if (index >= 0 && index <= 4) { unsigned int align = TYPE_ALIGN_UNIT (type); /* __sync builtins require strict data alignment. */ if (exact_log2 (align) >= index) { /* Atomic load. */ if (loaded_val == stored_val && (GET_MODE_CLASS (TYPE_MODE (type)) == MODE_INT || GET_MODE_CLASS (TYPE_MODE (type)) == MODE_FLOAT) && GET_MODE_BITSIZE (TYPE_MODE (type)) <= BITS_PER_WORD && expand_omp_atomic_load (load_bb, addr, loaded_val, index)) return; /* Atomic store. */ if ((GET_MODE_CLASS (TYPE_MODE (type)) == MODE_INT || GET_MODE_CLASS (TYPE_MODE (type)) == MODE_FLOAT) && GET_MODE_BITSIZE (TYPE_MODE (type)) <= BITS_PER_WORD && store_bb == single_succ (load_bb) && first_stmt (store_bb) == store && expand_omp_atomic_store (load_bb, addr, loaded_val, stored_val, index)) return; /* When possible, use specialized atomic update functions. */ if ((INTEGRAL_TYPE_P (type) || POINTER_TYPE_P (type)) && store_bb == single_succ (load_bb) && expand_omp_atomic_fetch_op (load_bb, addr, loaded_val, stored_val, index)) return; /* If we don't have specialized __sync builtins, try and implement as a compare and swap loop. */ if (expand_omp_atomic_pipeline (load_bb, store_bb, addr, loaded_val, stored_val, index)) return; } } /* The ultimate fallback is wrapping the operation in a mutex. */ expand_omp_atomic_mutex (load_bb, store_bb, addr, loaded_val, stored_val); } /* Mark the loops inside the kernels region starting at REGION_ENTRY and ending at REGION_EXIT. */ static void mark_loops_in_oacc_kernels_region (basic_block region_entry, basic_block region_exit) { struct loop *outer = region_entry->loop_father; gcc_assert (region_exit == NULL || outer == region_exit->loop_father); /* Don't parallelize the kernels region if it contains more than one outer loop. */ unsigned int nr_outer_loops = 0; struct loop *single_outer = NULL; for (struct loop *loop = outer->inner; loop != NULL; loop = loop->next) { gcc_assert (loop_outer (loop) == outer); if (!dominated_by_p (CDI_DOMINATORS, loop->header, region_entry)) continue; if (region_exit != NULL && dominated_by_p (CDI_DOMINATORS, loop->header, region_exit)) continue; nr_outer_loops++; single_outer = loop; } if (nr_outer_loops != 1) return; for (struct loop *loop = single_outer->inner; loop != NULL; loop = loop->inner) if (loop->next) return; /* Mark the loops in the region. */ for (struct loop *loop = single_outer; loop != NULL; loop = loop->inner) loop->in_oacc_kernels_region = true; } /* Types used to pass grid and wortkgroup sizes to kernel invocation. */ struct GTY(()) grid_launch_attributes_trees { tree kernel_dim_array_type; tree kernel_lattrs_dimnum_decl; tree kernel_lattrs_grid_decl; tree kernel_lattrs_group_decl; tree kernel_launch_attributes_type; }; static GTY(()) struct grid_launch_attributes_trees *grid_attr_trees; /* Create types used to pass kernel launch attributes to target. */ static void grid_create_kernel_launch_attr_types (void) { if (grid_attr_trees) return; grid_attr_trees = ggc_alloc (); tree dim_arr_index_type = build_index_type (build_int_cst (integer_type_node, 2)); grid_attr_trees->kernel_dim_array_type = build_array_type (uint32_type_node, dim_arr_index_type); grid_attr_trees->kernel_launch_attributes_type = make_node (RECORD_TYPE); grid_attr_trees->kernel_lattrs_dimnum_decl = build_decl (BUILTINS_LOCATION, FIELD_DECL, get_identifier ("ndim"), uint32_type_node); DECL_CHAIN (grid_attr_trees->kernel_lattrs_dimnum_decl) = NULL_TREE; grid_attr_trees->kernel_lattrs_grid_decl = build_decl (BUILTINS_LOCATION, FIELD_DECL, get_identifier ("grid_size"), grid_attr_trees->kernel_dim_array_type); DECL_CHAIN (grid_attr_trees->kernel_lattrs_grid_decl) = grid_attr_trees->kernel_lattrs_dimnum_decl; grid_attr_trees->kernel_lattrs_group_decl = build_decl (BUILTINS_LOCATION, FIELD_DECL, get_identifier ("group_size"), grid_attr_trees->kernel_dim_array_type); DECL_CHAIN (grid_attr_trees->kernel_lattrs_group_decl) = grid_attr_trees->kernel_lattrs_grid_decl; finish_builtin_struct (grid_attr_trees->kernel_launch_attributes_type, "__gomp_kernel_launch_attributes", grid_attr_trees->kernel_lattrs_group_decl, NULL_TREE); } /* Insert before the current statement in GSI a store of VALUE to INDEX of array (of type kernel_dim_array_type) FLD_DECL of RANGE_VAR. VALUE must be of type uint32_type_node. */ static void grid_insert_store_range_dim (gimple_stmt_iterator *gsi, tree range_var, tree fld_decl, int index, tree value) { tree ref = build4 (ARRAY_REF, uint32_type_node, build3 (COMPONENT_REF, grid_attr_trees->kernel_dim_array_type, range_var, fld_decl, NULL_TREE), build_int_cst (integer_type_node, index), NULL_TREE, NULL_TREE); gsi_insert_before (gsi, gimple_build_assign (ref, value), GSI_SAME_STMT); } /* Return a tree representation of a pointer to a structure with grid and work-group size information. Statements filling that information will be inserted before GSI, TGT_STMT is the target statement which has the necessary information in it. */ static tree grid_get_kernel_launch_attributes (gimple_stmt_iterator *gsi, gomp_target *tgt_stmt) { grid_create_kernel_launch_attr_types (); tree lattrs = create_tmp_var (grid_attr_trees->kernel_launch_attributes_type, "__kernel_launch_attrs"); unsigned max_dim = 0; for (tree clause = gimple_omp_target_clauses (tgt_stmt); clause; clause = OMP_CLAUSE_CHAIN (clause)) { if (OMP_CLAUSE_CODE (clause) != OMP_CLAUSE__GRIDDIM_) continue; unsigned dim = OMP_CLAUSE__GRIDDIM__DIMENSION (clause); max_dim = MAX (dim, max_dim); grid_insert_store_range_dim (gsi, lattrs, grid_attr_trees->kernel_lattrs_grid_decl, dim, OMP_CLAUSE__GRIDDIM__SIZE (clause)); grid_insert_store_range_dim (gsi, lattrs, grid_attr_trees->kernel_lattrs_group_decl, dim, OMP_CLAUSE__GRIDDIM__GROUP (clause)); } tree dimref = build3 (COMPONENT_REF, uint32_type_node, lattrs, grid_attr_trees->kernel_lattrs_dimnum_decl, NULL_TREE); gcc_checking_assert (max_dim <= 2); tree dimensions = build_int_cstu (uint32_type_node, max_dim + 1); gsi_insert_before (gsi, gimple_build_assign (dimref, dimensions), GSI_SAME_STMT); TREE_ADDRESSABLE (lattrs) = 1; return build_fold_addr_expr (lattrs); } /* Build target argument identifier from the DEVICE identifier, value identifier ID and whether the element also has a SUBSEQUENT_PARAM. */ static tree get_target_argument_identifier_1 (int device, bool subseqent_param, int id) { tree t = build_int_cst (integer_type_node, device); if (subseqent_param) t = fold_build2 (BIT_IOR_EXPR, integer_type_node, t, build_int_cst (integer_type_node, GOMP_TARGET_ARG_SUBSEQUENT_PARAM)); t = fold_build2 (BIT_IOR_EXPR, integer_type_node, t, build_int_cst (integer_type_node, id)); return t; } /* Like above but return it in type that can be directly stored as an element of the argument array. */ static tree get_target_argument_identifier (int device, bool subseqent_param, int id) { tree t = get_target_argument_identifier_1 (device, subseqent_param, id); return fold_convert (ptr_type_node, t); } /* Return a target argument consisting of DEVICE identifier, value identifier ID, and the actual VALUE. */ static tree get_target_argument_value (gimple_stmt_iterator *gsi, int device, int id, tree value) { tree t = fold_build2 (LSHIFT_EXPR, integer_type_node, fold_convert (integer_type_node, value), build_int_cst (unsigned_type_node, GOMP_TARGET_ARG_VALUE_SHIFT)); t = fold_build2 (BIT_IOR_EXPR, integer_type_node, t, get_target_argument_identifier_1 (device, false, id)); t = fold_convert (ptr_type_node, t); return force_gimple_operand_gsi (gsi, t, true, NULL, true, GSI_SAME_STMT); } /* If VALUE is an integer constant greater than -2^15 and smaller than 2^15, push one argument to ARGS with both the DEVICE, ID and VALUE embedded in it, otherwise push an identifier (with DEVICE and ID) and the VALUE in two arguments. */ static void push_target_argument_according_to_value (gimple_stmt_iterator *gsi, int device, int id, tree value, vec *args) { if (tree_fits_shwi_p (value) && tree_to_shwi (value) > -(1 << 15) && tree_to_shwi (value) < (1 << 15)) args->quick_push (get_target_argument_value (gsi, device, id, value)); else { args->quick_push (get_target_argument_identifier (device, true, id)); value = fold_convert (ptr_type_node, value); value = force_gimple_operand_gsi (gsi, value, true, NULL, true, GSI_SAME_STMT); args->quick_push (value); } } /* Create an array of arguments that is then passed to GOMP_target. */ static tree get_target_arguments (gimple_stmt_iterator *gsi, gomp_target *tgt_stmt) { auto_vec args; tree clauses = gimple_omp_target_clauses (tgt_stmt); tree t, c = omp_find_clause (clauses, OMP_CLAUSE_NUM_TEAMS); if (c) t = OMP_CLAUSE_NUM_TEAMS_EXPR (c); else t = integer_minus_one_node; push_target_argument_according_to_value (gsi, GOMP_TARGET_ARG_DEVICE_ALL, GOMP_TARGET_ARG_NUM_TEAMS, t, &args); c = omp_find_clause (clauses, OMP_CLAUSE_THREAD_LIMIT); if (c) t = OMP_CLAUSE_THREAD_LIMIT_EXPR (c); else t = integer_minus_one_node; push_target_argument_according_to_value (gsi, GOMP_TARGET_ARG_DEVICE_ALL, GOMP_TARGET_ARG_THREAD_LIMIT, t, &args); /* Add HSA-specific grid sizes, if available. */ if (omp_find_clause (gimple_omp_target_clauses (tgt_stmt), OMP_CLAUSE__GRIDDIM_)) { int id = GOMP_TARGET_ARG_HSA_KERNEL_ATTRIBUTES; t = get_target_argument_identifier (GOMP_DEVICE_HSA, true, id); args.quick_push (t); args.quick_push (grid_get_kernel_launch_attributes (gsi, tgt_stmt)); } /* Produce more, perhaps device specific, arguments here. */ tree argarray = create_tmp_var (build_array_type_nelts (ptr_type_node, args.length () + 1), ".omp_target_args"); for (unsigned i = 0; i < args.length (); i++) { tree ref = build4 (ARRAY_REF, ptr_type_node, argarray, build_int_cst (integer_type_node, i), NULL_TREE, NULL_TREE); gsi_insert_before (gsi, gimple_build_assign (ref, args[i]), GSI_SAME_STMT); } tree ref = build4 (ARRAY_REF, ptr_type_node, argarray, build_int_cst (integer_type_node, args.length ()), NULL_TREE, NULL_TREE); gsi_insert_before (gsi, gimple_build_assign (ref, null_pointer_node), GSI_SAME_STMT); TREE_ADDRESSABLE (argarray) = 1; return build_fold_addr_expr (argarray); } /* Expand the GIMPLE_OMP_TARGET starting at REGION. */ static void expand_omp_target (struct omp_region *region) { basic_block entry_bb, exit_bb, new_bb; struct function *child_cfun; tree child_fn, block, t; gimple_stmt_iterator gsi; gomp_target *entry_stmt; gimple *stmt; edge e; bool offloaded, data_region; entry_stmt = as_a (last_stmt (region->entry)); new_bb = region->entry; offloaded = is_gimple_omp_offloaded (entry_stmt); switch (gimple_omp_target_kind (entry_stmt)) { case GF_OMP_TARGET_KIND_REGION: case GF_OMP_TARGET_KIND_UPDATE: case GF_OMP_TARGET_KIND_ENTER_DATA: case GF_OMP_TARGET_KIND_EXIT_DATA: case GF_OMP_TARGET_KIND_OACC_PARALLEL: case GF_OMP_TARGET_KIND_OACC_KERNELS: case GF_OMP_TARGET_KIND_OACC_UPDATE: case GF_OMP_TARGET_KIND_OACC_ENTER_EXIT_DATA: case GF_OMP_TARGET_KIND_OACC_DECLARE: data_region = false; break; case GF_OMP_TARGET_KIND_DATA: case GF_OMP_TARGET_KIND_OACC_DATA: case GF_OMP_TARGET_KIND_OACC_HOST_DATA: data_region = true; break; default: gcc_unreachable (); } child_fn = NULL_TREE; child_cfun = NULL; if (offloaded) { child_fn = gimple_omp_target_child_fn (entry_stmt); child_cfun = DECL_STRUCT_FUNCTION (child_fn); } /* Supported by expand_omp_taskreg, but not here. */ if (child_cfun != NULL) gcc_checking_assert (!child_cfun->cfg); gcc_checking_assert (!gimple_in_ssa_p (cfun)); entry_bb = region->entry; exit_bb = region->exit; if (gimple_omp_target_kind (entry_stmt) == GF_OMP_TARGET_KIND_OACC_KERNELS) mark_loops_in_oacc_kernels_region (region->entry, region->exit); if (offloaded) { unsigned srcidx, dstidx, num; /* If the offloading region needs data sent from the parent function, then the very first statement (except possible tree profile counter updates) of the offloading body is a copy assignment .OMP_DATA_I = &.OMP_DATA_O. Since &.OMP_DATA_O is passed as an argument to the child function, we need to replace it with the argument as seen by the child function. In most cases, this will end up being the identity assignment .OMP_DATA_I = .OMP_DATA_I. However, if the offloading body had a function call that has been inlined, the original PARM_DECL .OMP_DATA_I may have been converted into a different local variable. In which case, we need to keep the assignment. */ tree data_arg = gimple_omp_target_data_arg (entry_stmt); if (data_arg) { basic_block entry_succ_bb = single_succ (entry_bb); gimple_stmt_iterator gsi; tree arg; gimple *tgtcopy_stmt = NULL; tree sender = TREE_VEC_ELT (data_arg, 0); for (gsi = gsi_start_bb (entry_succ_bb); ; gsi_next (&gsi)) { gcc_assert (!gsi_end_p (gsi)); stmt = gsi_stmt (gsi); if (gimple_code (stmt) != GIMPLE_ASSIGN) continue; if (gimple_num_ops (stmt) == 2) { tree arg = gimple_assign_rhs1 (stmt); /* We're ignoring the subcode because we're effectively doing a STRIP_NOPS. */ if (TREE_CODE (arg) == ADDR_EXPR && TREE_OPERAND (arg, 0) == sender) { tgtcopy_stmt = stmt; break; } } } gcc_assert (tgtcopy_stmt != NULL); arg = DECL_ARGUMENTS (child_fn); gcc_assert (gimple_assign_lhs (tgtcopy_stmt) == arg); gsi_remove (&gsi, true); } /* Declare local variables needed in CHILD_CFUN. */ block = DECL_INITIAL (child_fn); BLOCK_VARS (block) = vec2chain (child_cfun->local_decls); /* The gimplifier could record temporaries in the offloading block rather than in containing function's local_decls chain, which would mean cgraph missed finalizing them. Do it now. */ for (t = BLOCK_VARS (block); t; t = DECL_CHAIN (t)) if (VAR_P (t) && TREE_STATIC (t) && !DECL_EXTERNAL (t)) varpool_node::finalize_decl (t); DECL_SAVED_TREE (child_fn) = NULL; /* We'll create a CFG for child_fn, so no gimple body is needed. */ gimple_set_body (child_fn, NULL); TREE_USED (block) = 1; /* Reset DECL_CONTEXT on function arguments. */ for (t = DECL_ARGUMENTS (child_fn); t; t = DECL_CHAIN (t)) DECL_CONTEXT (t) = child_fn; /* Split ENTRY_BB at GIMPLE_*, so that it can be moved to the child function. */ gsi = gsi_last_bb (entry_bb); stmt = gsi_stmt (gsi); gcc_assert (stmt && gimple_code (stmt) == gimple_code (entry_stmt)); e = split_block (entry_bb, stmt); gsi_remove (&gsi, true); entry_bb = e->dest; single_succ_edge (entry_bb)->flags = EDGE_FALLTHRU; /* Convert GIMPLE_OMP_RETURN into a RETURN_EXPR. */ if (exit_bb) { gsi = gsi_last_bb (exit_bb); gcc_assert (!gsi_end_p (gsi) && gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_RETURN); stmt = gimple_build_return (NULL); gsi_insert_after (&gsi, stmt, GSI_SAME_STMT); gsi_remove (&gsi, true); } /* Make sure to generate early debug for the function before outlining anything. */ if (! gimple_in_ssa_p (cfun)) (*debug_hooks->early_global_decl) (cfun->decl); /* Move the offloading region into CHILD_CFUN. */ block = gimple_block (entry_stmt); new_bb = move_sese_region_to_fn (child_cfun, entry_bb, exit_bb, block); if (exit_bb) single_succ_edge (new_bb)->flags = EDGE_FALLTHRU; /* When the OMP expansion process cannot guarantee an up-to-date loop tree arrange for the child function to fixup loops. */ if (loops_state_satisfies_p (LOOPS_NEED_FIXUP)) child_cfun->x_current_loops->state |= LOOPS_NEED_FIXUP; /* Remove non-local VAR_DECLs from child_cfun->local_decls list. */ num = vec_safe_length (child_cfun->local_decls); for (srcidx = 0, dstidx = 0; srcidx < num; srcidx++) { t = (*child_cfun->local_decls)[srcidx]; if (DECL_CONTEXT (t) == cfun->decl) continue; if (srcidx != dstidx) (*child_cfun->local_decls)[dstidx] = t; dstidx++; } if (dstidx != num) vec_safe_truncate (child_cfun->local_decls, dstidx); /* Inform the callgraph about the new function. */ child_cfun->curr_properties = cfun->curr_properties; child_cfun->has_simduid_loops |= cfun->has_simduid_loops; child_cfun->has_force_vectorize_loops |= cfun->has_force_vectorize_loops; cgraph_node *node = cgraph_node::get_create (child_fn); node->parallelized_function = 1; cgraph_node::add_new_function (child_fn, true); /* Add the new function to the offload table. */ if (ENABLE_OFFLOADING) vec_safe_push (offload_funcs, child_fn); bool need_asm = DECL_ASSEMBLER_NAME_SET_P (current_function_decl) && !DECL_ASSEMBLER_NAME_SET_P (child_fn); /* Fix the callgraph edges for child_cfun. Those for cfun will be fixed in a following pass. */ push_cfun (child_cfun); if (need_asm) assign_assembler_name_if_needed (child_fn); cgraph_edge::rebuild_edges (); /* Some EH regions might become dead, see PR34608. If pass_cleanup_cfg isn't the first pass to happen with the new child, these dead EH edges might cause problems. Clean them up now. */ if (flag_exceptions) { basic_block bb; bool changed = false; FOR_EACH_BB_FN (bb, cfun) changed |= gimple_purge_dead_eh_edges (bb); if (changed) cleanup_tree_cfg (); } if (flag_checking && !loops_state_satisfies_p (LOOPS_NEED_FIXUP)) verify_loop_structure (); pop_cfun (); if (dump_file && !gimple_in_ssa_p (cfun)) { omp_any_child_fn_dumped = true; dump_function_header (dump_file, child_fn, dump_flags); dump_function_to_file (child_fn, dump_file, dump_flags); } } /* Emit a library call to launch the offloading region, or do data transfers. */ tree t1, t2, t3, t4, device, cond, depend, c, clauses; enum built_in_function start_ix; location_t clause_loc; unsigned int flags_i = 0; bool oacc_kernels_p = false; switch (gimple_omp_target_kind (entry_stmt)) { case GF_OMP_TARGET_KIND_REGION: start_ix = BUILT_IN_GOMP_TARGET; break; case GF_OMP_TARGET_KIND_DATA: start_ix = BUILT_IN_GOMP_TARGET_DATA; break; case GF_OMP_TARGET_KIND_UPDATE: start_ix = BUILT_IN_GOMP_TARGET_UPDATE; break; case GF_OMP_TARGET_KIND_ENTER_DATA: start_ix = BUILT_IN_GOMP_TARGET_ENTER_EXIT_DATA; break; case GF_OMP_TARGET_KIND_EXIT_DATA: start_ix = BUILT_IN_GOMP_TARGET_ENTER_EXIT_DATA; flags_i |= GOMP_TARGET_FLAG_EXIT_DATA; break; case GF_OMP_TARGET_KIND_OACC_KERNELS: oacc_kernels_p = true; /* FALLTHROUGH */ case GF_OMP_TARGET_KIND_OACC_PARALLEL: start_ix = BUILT_IN_GOACC_PARALLEL; break; case GF_OMP_TARGET_KIND_OACC_DATA: case GF_OMP_TARGET_KIND_OACC_HOST_DATA: start_ix = BUILT_IN_GOACC_DATA_START; break; case GF_OMP_TARGET_KIND_OACC_UPDATE: start_ix = BUILT_IN_GOACC_UPDATE; break; case GF_OMP_TARGET_KIND_OACC_ENTER_EXIT_DATA: start_ix = BUILT_IN_GOACC_ENTER_EXIT_DATA; break; case GF_OMP_TARGET_KIND_OACC_DECLARE: start_ix = BUILT_IN_GOACC_DECLARE; break; default: gcc_unreachable (); } clauses = gimple_omp_target_clauses (entry_stmt); /* By default, the value of DEVICE is GOMP_DEVICE_ICV (let runtime library choose) and there is no conditional. */ cond = NULL_TREE; device = build_int_cst (integer_type_node, GOMP_DEVICE_ICV); c = omp_find_clause (clauses, OMP_CLAUSE_IF); if (c) cond = OMP_CLAUSE_IF_EXPR (c); c = omp_find_clause (clauses, OMP_CLAUSE_DEVICE); if (c) { /* Even if we pass it to all library function calls, it is currently only defined/used for the OpenMP target ones. */ gcc_checking_assert (start_ix == BUILT_IN_GOMP_TARGET || start_ix == BUILT_IN_GOMP_TARGET_DATA || start_ix == BUILT_IN_GOMP_TARGET_UPDATE || start_ix == BUILT_IN_GOMP_TARGET_ENTER_EXIT_DATA); device = OMP_CLAUSE_DEVICE_ID (c); clause_loc = OMP_CLAUSE_LOCATION (c); } else clause_loc = gimple_location (entry_stmt); c = omp_find_clause (clauses, OMP_CLAUSE_NOWAIT); if (c) flags_i |= GOMP_TARGET_FLAG_NOWAIT; /* Ensure 'device' is of the correct type. */ device = fold_convert_loc (clause_loc, integer_type_node, device); /* If we found the clause 'if (cond)', build (cond ? device : GOMP_DEVICE_HOST_FALLBACK). */ if (cond) { cond = gimple_boolify (cond); basic_block cond_bb, then_bb, else_bb; edge e; tree tmp_var; tmp_var = create_tmp_var (TREE_TYPE (device)); if (offloaded) e = split_block_after_labels (new_bb); else { gsi = gsi_last_bb (new_bb); gsi_prev (&gsi); e = split_block (new_bb, gsi_stmt (gsi)); } cond_bb = e->src; new_bb = e->dest; remove_edge (e); then_bb = create_empty_bb (cond_bb); else_bb = create_empty_bb (then_bb); set_immediate_dominator (CDI_DOMINATORS, then_bb, cond_bb); set_immediate_dominator (CDI_DOMINATORS, else_bb, cond_bb); stmt = gimple_build_cond_empty (cond); gsi = gsi_last_bb (cond_bb); gsi_insert_after (&gsi, stmt, GSI_CONTINUE_LINKING); gsi = gsi_start_bb (then_bb); stmt = gimple_build_assign (tmp_var, device); gsi_insert_after (&gsi, stmt, GSI_CONTINUE_LINKING); gsi = gsi_start_bb (else_bb); stmt = gimple_build_assign (tmp_var, build_int_cst (integer_type_node, GOMP_DEVICE_HOST_FALLBACK)); gsi_insert_after (&gsi, stmt, GSI_CONTINUE_LINKING); make_edge (cond_bb, then_bb, EDGE_TRUE_VALUE); make_edge (cond_bb, else_bb, EDGE_FALSE_VALUE); add_bb_to_loop (then_bb, cond_bb->loop_father); add_bb_to_loop (else_bb, cond_bb->loop_father); make_edge (then_bb, new_bb, EDGE_FALLTHRU); make_edge (else_bb, new_bb, EDGE_FALLTHRU); device = tmp_var; gsi = gsi_last_bb (new_bb); } else { gsi = gsi_last_bb (new_bb); device = force_gimple_operand_gsi (&gsi, device, true, NULL_TREE, true, GSI_SAME_STMT); } t = gimple_omp_target_data_arg (entry_stmt); if (t == NULL) { t1 = size_zero_node; t2 = build_zero_cst (ptr_type_node); t3 = t2; t4 = t2; } else { t1 = TYPE_MAX_VALUE (TYPE_DOMAIN (TREE_TYPE (TREE_VEC_ELT (t, 1)))); t1 = size_binop (PLUS_EXPR, t1, size_int (1)); t2 = build_fold_addr_expr (TREE_VEC_ELT (t, 0)); t3 = build_fold_addr_expr (TREE_VEC_ELT (t, 1)); t4 = build_fold_addr_expr (TREE_VEC_ELT (t, 2)); } gimple *g; bool tagging = false; /* The maximum number used by any start_ix, without varargs. */ auto_vec args; args.quick_push (device); if (offloaded) args.quick_push (build_fold_addr_expr (child_fn)); args.quick_push (t1); args.quick_push (t2); args.quick_push (t3); args.quick_push (t4); switch (start_ix) { case BUILT_IN_GOACC_DATA_START: case BUILT_IN_GOACC_DECLARE: case BUILT_IN_GOMP_TARGET_DATA: break; case BUILT_IN_GOMP_TARGET: case BUILT_IN_GOMP_TARGET_UPDATE: case BUILT_IN_GOMP_TARGET_ENTER_EXIT_DATA: args.quick_push (build_int_cst (unsigned_type_node, flags_i)); c = omp_find_clause (clauses, OMP_CLAUSE_DEPEND); if (c) depend = OMP_CLAUSE_DECL (c); else depend = build_int_cst (ptr_type_node, 0); args.quick_push (depend); if (start_ix == BUILT_IN_GOMP_TARGET) args.quick_push (get_target_arguments (&gsi, entry_stmt)); break; case BUILT_IN_GOACC_PARALLEL: { oacc_set_fn_attrib (child_fn, clauses, oacc_kernels_p, &args); tagging = true; } /* FALLTHRU */ case BUILT_IN_GOACC_ENTER_EXIT_DATA: case BUILT_IN_GOACC_UPDATE: { tree t_async = NULL_TREE; /* If present, use the value specified by the respective clause, making sure that is of the correct type. */ c = omp_find_clause (clauses, OMP_CLAUSE_ASYNC); if (c) t_async = fold_convert_loc (OMP_CLAUSE_LOCATION (c), integer_type_node, OMP_CLAUSE_ASYNC_EXPR (c)); else if (!tagging) /* Default values for t_async. */ t_async = fold_convert_loc (gimple_location (entry_stmt), integer_type_node, build_int_cst (integer_type_node, GOMP_ASYNC_SYNC)); if (tagging && t_async) { unsigned HOST_WIDE_INT i_async = GOMP_LAUNCH_OP_MAX; if (TREE_CODE (t_async) == INTEGER_CST) { /* See if we can pack the async arg in to the tag's operand. */ i_async = TREE_INT_CST_LOW (t_async); if (i_async < GOMP_LAUNCH_OP_MAX) t_async = NULL_TREE; else i_async = GOMP_LAUNCH_OP_MAX; } args.safe_push (oacc_launch_pack (GOMP_LAUNCH_ASYNC, NULL_TREE, i_async)); } if (t_async) args.safe_push (t_async); /* Save the argument index, and ... */ unsigned t_wait_idx = args.length (); unsigned num_waits = 0; c = omp_find_clause (clauses, OMP_CLAUSE_WAIT); if (!tagging || c) /* ... push a placeholder. */ args.safe_push (integer_zero_node); for (; c; c = OMP_CLAUSE_CHAIN (c)) if (OMP_CLAUSE_CODE (c) == OMP_CLAUSE_WAIT) { args.safe_push (fold_convert_loc (OMP_CLAUSE_LOCATION (c), integer_type_node, OMP_CLAUSE_WAIT_EXPR (c))); num_waits++; } if (!tagging || num_waits) { tree len; /* Now that we know the number, update the placeholder. */ if (tagging) len = oacc_launch_pack (GOMP_LAUNCH_WAIT, NULL_TREE, num_waits); else len = build_int_cst (integer_type_node, num_waits); len = fold_convert_loc (gimple_location (entry_stmt), unsigned_type_node, len); args[t_wait_idx] = len; } } break; default: gcc_unreachable (); } if (tagging) /* Push terminal marker - zero. */ args.safe_push (oacc_launch_pack (0, NULL_TREE, 0)); g = gimple_build_call_vec (builtin_decl_explicit (start_ix), args); gimple_set_location (g, gimple_location (entry_stmt)); gsi_insert_before (&gsi, g, GSI_SAME_STMT); if (!offloaded) { g = gsi_stmt (gsi); gcc_assert (g && gimple_code (g) == GIMPLE_OMP_TARGET); gsi_remove (&gsi, true); } if (data_region && region->exit) { gsi = gsi_last_bb (region->exit); g = gsi_stmt (gsi); gcc_assert (g && gimple_code (g) == GIMPLE_OMP_RETURN); gsi_remove (&gsi, true); } } /* Expand KFOR loop as a HSA grifidied kernel, i.e. as a body only with iteration variable derived from the thread number. INTRA_GROUP means this is an expansion of a loop iterating over work-items within a separate iteration over groups. */ static void grid_expand_omp_for_loop (struct omp_region *kfor, bool intra_group) { gimple_stmt_iterator gsi; gomp_for *for_stmt = as_a (last_stmt (kfor->entry)); gcc_checking_assert (gimple_omp_for_kind (for_stmt) == GF_OMP_FOR_KIND_GRID_LOOP); size_t collapse = gimple_omp_for_collapse (for_stmt); struct omp_for_data_loop *loops = XALLOCAVEC (struct omp_for_data_loop, gimple_omp_for_collapse (for_stmt)); struct omp_for_data fd; remove_edge (BRANCH_EDGE (kfor->entry)); basic_block body_bb = FALLTHRU_EDGE (kfor->entry)->dest; gcc_assert (kfor->cont); omp_extract_for_data (for_stmt, &fd, loops); gsi = gsi_start_bb (body_bb); for (size_t dim = 0; dim < collapse; dim++) { tree type, itype; itype = type = TREE_TYPE (fd.loops[dim].v); if (POINTER_TYPE_P (type)) itype = signed_type_for (type); tree n1 = fd.loops[dim].n1; tree step = fd.loops[dim].step; n1 = force_gimple_operand_gsi (&gsi, fold_convert (type, n1), true, NULL_TREE, true, GSI_SAME_STMT); step = force_gimple_operand_gsi (&gsi, fold_convert (itype, step), true, NULL_TREE, true, GSI_SAME_STMT); tree threadid; if (gimple_omp_for_grid_group_iter (for_stmt)) { gcc_checking_assert (!intra_group); threadid = build_call_expr (builtin_decl_explicit (BUILT_IN_HSA_WORKGROUPID), 1, build_int_cstu (unsigned_type_node, dim)); } else if (intra_group) threadid = build_call_expr (builtin_decl_explicit (BUILT_IN_HSA_WORKITEMID), 1, build_int_cstu (unsigned_type_node, dim)); else threadid = build_call_expr (builtin_decl_explicit (BUILT_IN_HSA_WORKITEMABSID), 1, build_int_cstu (unsigned_type_node, dim)); threadid = fold_convert (itype, threadid); threadid = force_gimple_operand_gsi (&gsi, threadid, true, NULL_TREE, true, GSI_SAME_STMT); tree startvar = fd.loops[dim].v; tree t = fold_build2 (MULT_EXPR, itype, threadid, step); if (POINTER_TYPE_P (type)) t = fold_build_pointer_plus (n1, t); else t = fold_build2 (PLUS_EXPR, type, t, n1); t = fold_convert (type, t); t = force_gimple_operand_gsi (&gsi, t, DECL_P (startvar) && TREE_ADDRESSABLE (startvar), NULL_TREE, true, GSI_SAME_STMT); gassign *assign_stmt = gimple_build_assign (startvar, t); gsi_insert_before (&gsi, assign_stmt, GSI_SAME_STMT); } /* Remove the omp for statement. */ gsi = gsi_last_bb (kfor->entry); gsi_remove (&gsi, true); /* Remove the GIMPLE_OMP_CONTINUE statement. */ gsi = gsi_last_bb (kfor->cont); gcc_assert (!gsi_end_p (gsi) && gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_CONTINUE); gsi_remove (&gsi, true); /* Replace the GIMPLE_OMP_RETURN with a barrier, if necessary. */ gsi = gsi_last_bb (kfor->exit); gcc_assert (!gsi_end_p (gsi) && gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_RETURN); if (intra_group) gsi_insert_before (&gsi, omp_build_barrier (NULL_TREE), GSI_SAME_STMT); gsi_remove (&gsi, true); /* Fixup the much simpler CFG. */ remove_edge (find_edge (kfor->cont, body_bb)); if (kfor->cont != body_bb) set_immediate_dominator (CDI_DOMINATORS, kfor->cont, body_bb); set_immediate_dominator (CDI_DOMINATORS, kfor->exit, kfor->cont); } /* Structure passed to grid_remap_kernel_arg_accesses so that it can remap argument_decls. */ struct grid_arg_decl_map { tree old_arg; tree new_arg; }; /* Invoked through walk_gimple_op, will remap all PARM_DECLs to the ones pertaining to kernel function. */ static tree grid_remap_kernel_arg_accesses (tree *tp, int *walk_subtrees, void *data) { struct walk_stmt_info *wi = (struct walk_stmt_info *) data; struct grid_arg_decl_map *adm = (struct grid_arg_decl_map *) wi->info; tree t = *tp; if (t == adm->old_arg) *tp = adm->new_arg; *walk_subtrees = !TYPE_P (t) && !DECL_P (t); return NULL_TREE; } /* If TARGET region contains a kernel body for loop, remove its region from the TARGET and expand it in HSA gridified kernel fashion. */ static void grid_expand_target_grid_body (struct omp_region *target) { if (!hsa_gen_requested_p ()) return; gomp_target *tgt_stmt = as_a (last_stmt (target->entry)); struct omp_region **pp; for (pp = &target->inner; *pp; pp = &(*pp)->next) if ((*pp)->type == GIMPLE_OMP_GRID_BODY) break; struct omp_region *gpukernel = *pp; tree orig_child_fndecl = gimple_omp_target_child_fn (tgt_stmt); if (!gpukernel) { /* HSA cannot handle OACC stuff. */ if (gimple_omp_target_kind (tgt_stmt) != GF_OMP_TARGET_KIND_REGION) return; gcc_checking_assert (orig_child_fndecl); gcc_assert (!omp_find_clause (gimple_omp_target_clauses (tgt_stmt), OMP_CLAUSE__GRIDDIM_)); cgraph_node *n = cgraph_node::get (orig_child_fndecl); hsa_register_kernel (n); return; } gcc_assert (omp_find_clause (gimple_omp_target_clauses (tgt_stmt), OMP_CLAUSE__GRIDDIM_)); tree inside_block = gimple_block (first_stmt (single_succ (gpukernel->entry))); *pp = gpukernel->next; for (pp = &gpukernel->inner; *pp; pp = &(*pp)->next) if ((*pp)->type == GIMPLE_OMP_FOR) break; struct omp_region *kfor = *pp; gcc_assert (kfor); gomp_for *for_stmt = as_a (last_stmt (kfor->entry)); gcc_assert (gimple_omp_for_kind (for_stmt) == GF_OMP_FOR_KIND_GRID_LOOP); *pp = kfor->next; if (kfor->inner) { if (gimple_omp_for_grid_group_iter (for_stmt)) { struct omp_region **next_pp; for (pp = &kfor->inner; *pp; pp = next_pp) { next_pp = &(*pp)->next; if ((*pp)->type != GIMPLE_OMP_FOR) continue; gomp_for *inner = as_a (last_stmt ((*pp)->entry)); gcc_assert (gimple_omp_for_kind (inner) == GF_OMP_FOR_KIND_GRID_LOOP); grid_expand_omp_for_loop (*pp, true); *pp = (*pp)->next; next_pp = pp; } } expand_omp (kfor->inner); } if (gpukernel->inner) expand_omp (gpukernel->inner); tree kern_fndecl = copy_node (orig_child_fndecl); DECL_NAME (kern_fndecl) = clone_function_name (kern_fndecl, "kernel"); SET_DECL_ASSEMBLER_NAME (kern_fndecl, DECL_NAME (kern_fndecl)); tree tgtblock = gimple_block (tgt_stmt); tree fniniblock = make_node (BLOCK); BLOCK_ABSTRACT_ORIGIN (fniniblock) = tgtblock; BLOCK_SOURCE_LOCATION (fniniblock) = BLOCK_SOURCE_LOCATION (tgtblock); BLOCK_SOURCE_END_LOCATION (fniniblock) = BLOCK_SOURCE_END_LOCATION (tgtblock); BLOCK_SUPERCONTEXT (fniniblock) = kern_fndecl; DECL_INITIAL (kern_fndecl) = fniniblock; push_struct_function (kern_fndecl); cfun->function_end_locus = gimple_location (tgt_stmt); init_tree_ssa (cfun); pop_cfun (); /* Make sure to generate early debug for the function before outlining anything. */ if (! gimple_in_ssa_p (cfun)) (*debug_hooks->early_global_decl) (cfun->decl); tree old_parm_decl = DECL_ARGUMENTS (kern_fndecl); gcc_assert (!DECL_CHAIN (old_parm_decl)); tree new_parm_decl = copy_node (DECL_ARGUMENTS (kern_fndecl)); DECL_CONTEXT (new_parm_decl) = kern_fndecl; DECL_ARGUMENTS (kern_fndecl) = new_parm_decl; gcc_assert (VOID_TYPE_P (TREE_TYPE (DECL_RESULT (kern_fndecl)))); DECL_RESULT (kern_fndecl) = copy_node (DECL_RESULT (kern_fndecl)); DECL_CONTEXT (DECL_RESULT (kern_fndecl)) = kern_fndecl; struct function *kern_cfun = DECL_STRUCT_FUNCTION (kern_fndecl); kern_cfun->curr_properties = cfun->curr_properties; grid_expand_omp_for_loop (kfor, false); /* Remove the omp for statement. */ gimple_stmt_iterator gsi = gsi_last_bb (gpukernel->entry); gsi_remove (&gsi, true); /* Replace the GIMPLE_OMP_RETURN at the end of the kernel region with a real return. */ gsi = gsi_last_bb (gpukernel->exit); gcc_assert (!gsi_end_p (gsi) && gimple_code (gsi_stmt (gsi)) == GIMPLE_OMP_RETURN); gimple *ret_stmt = gimple_build_return (NULL); gsi_insert_after (&gsi, ret_stmt, GSI_SAME_STMT); gsi_remove (&gsi, true); /* Statements in the first BB in the target construct have been produced by target lowering and must be copied inside the GPUKERNEL, with the two exceptions of the first OMP statement and the OMP_DATA assignment statement. */ gsi = gsi_start_bb (single_succ (gpukernel->entry)); tree data_arg = gimple_omp_target_data_arg (tgt_stmt); tree sender = data_arg ? TREE_VEC_ELT (data_arg, 0) : NULL; for (gimple_stmt_iterator tsi = gsi_start_bb (single_succ (target->entry)); !gsi_end_p (tsi); gsi_next (&tsi)) { gimple *stmt = gsi_stmt (tsi); if (is_gimple_omp (stmt)) break; if (sender && is_gimple_assign (stmt) && TREE_CODE (gimple_assign_rhs1 (stmt)) == ADDR_EXPR && TREE_OPERAND (gimple_assign_rhs1 (stmt), 0) == sender) continue; gimple *copy = gimple_copy (stmt); gsi_insert_before (&gsi, copy, GSI_SAME_STMT); gimple_set_block (copy, fniniblock); } move_sese_region_to_fn (kern_cfun, single_succ (gpukernel->entry), gpukernel->exit, inside_block); cgraph_node *kcn = cgraph_node::get_create (kern_fndecl); kcn->mark_force_output (); cgraph_node *orig_child = cgraph_node::get (orig_child_fndecl); hsa_register_kernel (kcn, orig_child); cgraph_node::add_new_function (kern_fndecl, true); push_cfun (kern_cfun); cgraph_edge::rebuild_edges (); /* Re-map any mention of the PARM_DECL of the original function to the PARM_DECL of the new one. TODO: It would be great if lowering produced references into the GPU kernel decl straight away and we did not have to do this. */ struct grid_arg_decl_map adm; adm.old_arg = old_parm_decl; adm.new_arg = new_parm_decl; basic_block bb; FOR_EACH_BB_FN (bb, kern_cfun) { for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi)) { gimple *stmt = gsi_stmt (gsi); struct walk_stmt_info wi; memset (&wi, 0, sizeof (wi)); wi.info = &adm; walk_gimple_op (stmt, grid_remap_kernel_arg_accesses, &wi); } } pop_cfun (); return; } /* Expand the parallel region tree rooted at REGION. Expansion proceeds in depth-first order. Innermost regions are expanded first. This way, parallel regions that require a new function to be created (e.g., GIMPLE_OMP_PARALLEL) can be expanded without having any internal dependencies in their body. */ static void expand_omp (struct omp_region *region) { omp_any_child_fn_dumped = false; while (region) { location_t saved_location; gimple *inner_stmt = NULL; /* First, determine whether this is a combined parallel+workshare region. */ if (region->type == GIMPLE_OMP_PARALLEL) determine_parallel_type (region); else if (region->type == GIMPLE_OMP_TARGET) grid_expand_target_grid_body (region); if (region->type == GIMPLE_OMP_FOR && gimple_omp_for_combined_p (last_stmt (region->entry))) inner_stmt = last_stmt (region->inner->entry); if (region->inner) expand_omp (region->inner); saved_location = input_location; if (gimple_has_location (last_stmt (region->entry))) input_location = gimple_location (last_stmt (region->entry)); switch (region->type) { case GIMPLE_OMP_PARALLEL: case GIMPLE_OMP_TASK: expand_omp_taskreg (region); break; case GIMPLE_OMP_FOR: expand_omp_for (region, inner_stmt); break; case GIMPLE_OMP_SECTIONS: expand_omp_sections (region); break; case GIMPLE_OMP_SECTION: /* Individual omp sections are handled together with their parent GIMPLE_OMP_SECTIONS region. */ break; case GIMPLE_OMP_SINGLE: expand_omp_single (region); break; case GIMPLE_OMP_ORDERED: { gomp_ordered *ord_stmt = as_a (last_stmt (region->entry)); if (omp_find_clause (gimple_omp_ordered_clauses (ord_stmt), OMP_CLAUSE_DEPEND)) { /* We'll expand these when expanding corresponding worksharing region with ordered(n) clause. */ gcc_assert (region->outer && region->outer->type == GIMPLE_OMP_FOR); region->ord_stmt = ord_stmt; break; } } /* FALLTHRU */ case GIMPLE_OMP_MASTER: case GIMPLE_OMP_TASKGROUP: case GIMPLE_OMP_CRITICAL: case GIMPLE_OMP_TEAMS: expand_omp_synch (region); break; case GIMPLE_OMP_ATOMIC_LOAD: expand_omp_atomic (region); break; case GIMPLE_OMP_TARGET: expand_omp_target (region); break; default: gcc_unreachable (); } input_location = saved_location; region = region->next; } if (omp_any_child_fn_dumped) { if (dump_file) dump_function_header (dump_file, current_function_decl, dump_flags); omp_any_child_fn_dumped = false; } } /* Helper for build_omp_regions. Scan the dominator tree starting at block BB. PARENT is the region that contains BB. If SINGLE_TREE is true, the function ends once a single tree is built (otherwise, whole forest of OMP constructs may be built). */ static void build_omp_regions_1 (basic_block bb, struct omp_region *parent, bool single_tree) { gimple_stmt_iterator gsi; gimple *stmt; basic_block son; gsi = gsi_last_bb (bb); if (!gsi_end_p (gsi) && is_gimple_omp (gsi_stmt (gsi))) { struct omp_region *region; enum gimple_code code; stmt = gsi_stmt (gsi); code = gimple_code (stmt); if (code == GIMPLE_OMP_RETURN) { /* STMT is the return point out of region PARENT. Mark it as the exit point and make PARENT the immediately enclosing region. */ gcc_assert (parent); region = parent; region->exit = bb; parent = parent->outer; } else if (code == GIMPLE_OMP_ATOMIC_STORE) { /* GIMPLE_OMP_ATOMIC_STORE is analogous to GIMPLE_OMP_RETURN, but matches with GIMPLE_OMP_ATOMIC_LOAD. */ gcc_assert (parent); gcc_assert (parent->type == GIMPLE_OMP_ATOMIC_LOAD); region = parent; region->exit = bb; parent = parent->outer; } else if (code == GIMPLE_OMP_CONTINUE) { gcc_assert (parent); parent->cont = bb; } else if (code == GIMPLE_OMP_SECTIONS_SWITCH) { /* GIMPLE_OMP_SECTIONS_SWITCH is part of GIMPLE_OMP_SECTIONS, and we do nothing for it. */ } else { region = new_omp_region (bb, code, parent); /* Otherwise... */ if (code == GIMPLE_OMP_TARGET) { switch (gimple_omp_target_kind (stmt)) { case GF_OMP_TARGET_KIND_REGION: case GF_OMP_TARGET_KIND_DATA: case GF_OMP_TARGET_KIND_OACC_PARALLEL: case GF_OMP_TARGET_KIND_OACC_KERNELS: case GF_OMP_TARGET_KIND_OACC_DATA: case GF_OMP_TARGET_KIND_OACC_HOST_DATA: break; case GF_OMP_TARGET_KIND_UPDATE: case GF_OMP_TARGET_KIND_ENTER_DATA: case GF_OMP_TARGET_KIND_EXIT_DATA: case GF_OMP_TARGET_KIND_OACC_UPDATE: case GF_OMP_TARGET_KIND_OACC_ENTER_EXIT_DATA: case GF_OMP_TARGET_KIND_OACC_DECLARE: /* ..., other than for those stand-alone directives... */ region = NULL; break; default: gcc_unreachable (); } } else if (code == GIMPLE_OMP_ORDERED && omp_find_clause (gimple_omp_ordered_clauses (as_a (stmt)), OMP_CLAUSE_DEPEND)) /* #pragma omp ordered depend is also just a stand-alone directive. */ region = NULL; /* ..., this directive becomes the parent for a new region. */ if (region) parent = region; } } if (single_tree && !parent) return; for (son = first_dom_son (CDI_DOMINATORS, bb); son; son = next_dom_son (CDI_DOMINATORS, son)) build_omp_regions_1 (son, parent, single_tree); } /* Builds the tree of OMP regions rooted at ROOT, storing it to root_omp_region. */ static void build_omp_regions_root (basic_block root) { gcc_assert (root_omp_region == NULL); build_omp_regions_1 (root, NULL, true); gcc_assert (root_omp_region != NULL); } /* Expands omp construct (and its subconstructs) starting in HEAD. */ void omp_expand_local (basic_block head) { build_omp_regions_root (head); if (dump_file && (dump_flags & TDF_DETAILS)) { fprintf (dump_file, "\nOMP region tree\n\n"); dump_omp_region (dump_file, root_omp_region, 0); fprintf (dump_file, "\n"); } remove_exit_barriers (root_omp_region); expand_omp (root_omp_region); omp_free_regions (); } /* Scan the CFG and build a tree of OMP regions. Return the root of the OMP region tree. */ static void build_omp_regions (void) { gcc_assert (root_omp_region == NULL); calculate_dominance_info (CDI_DOMINATORS); build_omp_regions_1 (ENTRY_BLOCK_PTR_FOR_FN (cfun), NULL, false); } /* Main entry point for expanding OMP-GIMPLE into runtime calls. */ static unsigned int execute_expand_omp (void) { build_omp_regions (); if (!root_omp_region) return 0; if (dump_file) { fprintf (dump_file, "\nOMP region tree\n\n"); dump_omp_region (dump_file, root_omp_region, 0); fprintf (dump_file, "\n"); } remove_exit_barriers (root_omp_region); expand_omp (root_omp_region); if (flag_checking && !loops_state_satisfies_p (LOOPS_NEED_FIXUP)) verify_loop_structure (); cleanup_tree_cfg (); omp_free_regions (); return 0; } /* OMP expansion -- the default pass, run before creation of SSA form. */ namespace { const pass_data pass_data_expand_omp = { GIMPLE_PASS, /* type */ "ompexp", /* name */ OPTGROUP_OMP, /* optinfo_flags */ TV_NONE, /* tv_id */ PROP_gimple_any, /* properties_required */ PROP_gimple_eomp, /* properties_provided */ 0, /* properties_destroyed */ 0, /* todo_flags_start */ 0, /* todo_flags_finish */ }; class pass_expand_omp : public gimple_opt_pass { public: pass_expand_omp (gcc::context *ctxt) : gimple_opt_pass (pass_data_expand_omp, ctxt) {} /* opt_pass methods: */ virtual unsigned int execute (function *) { bool gate = ((flag_cilkplus != 0 || flag_openacc != 0 || flag_openmp != 0 || flag_openmp_simd != 0) && !seen_error ()); /* This pass always runs, to provide PROP_gimple_eomp. But often, there is nothing to do. */ if (!gate) return 0; return execute_expand_omp (); } }; // class pass_expand_omp } // anon namespace gimple_opt_pass * make_pass_expand_omp (gcc::context *ctxt) { return new pass_expand_omp (ctxt); } namespace { const pass_data pass_data_expand_omp_ssa = { GIMPLE_PASS, /* type */ "ompexpssa", /* name */ OPTGROUP_OMP, /* optinfo_flags */ TV_NONE, /* tv_id */ PROP_cfg | PROP_ssa, /* properties_required */ PROP_gimple_eomp, /* properties_provided */ 0, /* properties_destroyed */ 0, /* todo_flags_start */ TODO_cleanup_cfg | TODO_rebuild_alias, /* todo_flags_finish */ }; class pass_expand_omp_ssa : public gimple_opt_pass { public: pass_expand_omp_ssa (gcc::context *ctxt) : gimple_opt_pass (pass_data_expand_omp_ssa, ctxt) {} /* opt_pass methods: */ virtual bool gate (function *fun) { return !(fun->curr_properties & PROP_gimple_eomp); } virtual unsigned int execute (function *) { return execute_expand_omp (); } opt_pass * clone () { return new pass_expand_omp_ssa (m_ctxt); } }; // class pass_expand_omp_ssa } // anon namespace gimple_opt_pass * make_pass_expand_omp_ssa (gcc::context *ctxt) { return new pass_expand_omp_ssa (ctxt); } /* Called from tree-cfg.c::make_edges to create cfg edges for all relevant GIMPLE_* codes. */ bool omp_make_gimple_edges (basic_block bb, struct omp_region **region, int *region_idx) { gimple *last = last_stmt (bb); enum gimple_code code = gimple_code (last); struct omp_region *cur_region = *region; bool fallthru = false; switch (code) { case GIMPLE_OMP_PARALLEL: case GIMPLE_OMP_TASK: case GIMPLE_OMP_FOR: case GIMPLE_OMP_SINGLE: case GIMPLE_OMP_TEAMS: case GIMPLE_OMP_MASTER: case GIMPLE_OMP_TASKGROUP: case GIMPLE_OMP_CRITICAL: case GIMPLE_OMP_SECTION: case GIMPLE_OMP_GRID_BODY: cur_region = new_omp_region (bb, code, cur_region); fallthru = true; break; case GIMPLE_OMP_ORDERED: cur_region = new_omp_region (bb, code, cur_region); fallthru = true; if (omp_find_clause (gimple_omp_ordered_clauses (as_a (last)), OMP_CLAUSE_DEPEND)) cur_region = cur_region->outer; break; case GIMPLE_OMP_TARGET: cur_region = new_omp_region (bb, code, cur_region); fallthru = true; switch (gimple_omp_target_kind (last)) { case GF_OMP_TARGET_KIND_REGION: case GF_OMP_TARGET_KIND_DATA: case GF_OMP_TARGET_KIND_OACC_PARALLEL: case GF_OMP_TARGET_KIND_OACC_KERNELS: case GF_OMP_TARGET_KIND_OACC_DATA: case GF_OMP_TARGET_KIND_OACC_HOST_DATA: break; case GF_OMP_TARGET_KIND_UPDATE: case GF_OMP_TARGET_KIND_ENTER_DATA: case GF_OMP_TARGET_KIND_EXIT_DATA: case GF_OMP_TARGET_KIND_OACC_UPDATE: case GF_OMP_TARGET_KIND_OACC_ENTER_EXIT_DATA: case GF_OMP_TARGET_KIND_OACC_DECLARE: cur_region = cur_region->outer; break; default: gcc_unreachable (); } break; case GIMPLE_OMP_SECTIONS: cur_region = new_omp_region (bb, code, cur_region); fallthru = true; break; case GIMPLE_OMP_SECTIONS_SWITCH: fallthru = false; break; case GIMPLE_OMP_ATOMIC_LOAD: case GIMPLE_OMP_ATOMIC_STORE: fallthru = true; break; case GIMPLE_OMP_RETURN: /* In the case of a GIMPLE_OMP_SECTION, the edge will go somewhere other than the next block. This will be created later. */ cur_region->exit = bb; if (cur_region->type == GIMPLE_OMP_TASK) /* Add an edge corresponding to not scheduling the task immediately. */ make_edge (cur_region->entry, bb, EDGE_ABNORMAL); fallthru = cur_region->type != GIMPLE_OMP_SECTION; cur_region = cur_region->outer; break; case GIMPLE_OMP_CONTINUE: cur_region->cont = bb; switch (cur_region->type) { case GIMPLE_OMP_FOR: /* Mark all GIMPLE_OMP_FOR and GIMPLE_OMP_CONTINUE succs edges as abnormal to prevent splitting them. */ single_succ_edge (cur_region->entry)->flags |= EDGE_ABNORMAL; /* Make the loopback edge. */ make_edge (bb, single_succ (cur_region->entry), EDGE_ABNORMAL); /* Create an edge from GIMPLE_OMP_FOR to exit, which corresponds to the case that the body of the loop is not executed at all. */ make_edge (cur_region->entry, bb->next_bb, EDGE_ABNORMAL); make_edge (bb, bb->next_bb, EDGE_FALLTHRU | EDGE_ABNORMAL); fallthru = false; break; case GIMPLE_OMP_SECTIONS: /* Wire up the edges into and out of the nested sections. */ { basic_block switch_bb = single_succ (cur_region->entry); struct omp_region *i; for (i = cur_region->inner; i ; i = i->next) { gcc_assert (i->type == GIMPLE_OMP_SECTION); make_edge (switch_bb, i->entry, 0); make_edge (i->exit, bb, EDGE_FALLTHRU); } /* Make the loopback edge to the block with GIMPLE_OMP_SECTIONS_SWITCH. */ make_edge (bb, switch_bb, 0); /* Make the edge from the switch to exit. */ make_edge (switch_bb, bb->next_bb, 0); fallthru = false; } break; case GIMPLE_OMP_TASK: fallthru = true; break; default: gcc_unreachable (); } break; default: gcc_unreachable (); } if (*region != cur_region) { *region = cur_region; if (cur_region) *region_idx = cur_region->entry->index; else *region_idx = 0; } return fallthru; } #include "gt-omp-expand.h"