gcc/ipa-comdats.cc

   1 /* Localize comdats.
   2    Copyright (C) 2014-2024 Free Software Foundation, Inc.
   3
   4 This file is part of GCC.
   5
   6 GCC is free software; you can redistribute it and/or modify it under
   7 the terms of the GNU General Public License as published by the Free
   8 Software Foundation; either version 3, or (at your option) any later
   9 version.
  10
  11 GCC is distributed in the hope that it will be useful, but WITHOUT ANY
  12 WARRANTY; without even the implied warranty of MERCHANTABILITY or
  13 FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  14 for more details.
  15
  16 You should have received a copy of the GNU General Public License
  17 along with GCC; see the file COPYING3.  If not see
  18 <http://www.gnu.org/licenses/>.  */
  19
  20 /* This is very simple pass that looks for static symbols that are used
  21    exclusively by symbol within one comdat group.  In this case it makes
  22    sense to bring the symbol itself into the group to avoid dead code
  23    that would arrise when the comdat group from current unit is replaced
  24    by a different copy.  Consider for example:
  25
  26     static int q(void)
  27     {
  28       ....
  29     }
  30     inline int t(void)
  31     {
  32       return q();
  33     }
  34
  35    if Q is used only by T, it makes sense to put Q into T's comdat group.
  36
  37    The pass solve simple dataflow across the callgraph trying to prove what
  38    symbols are used exclusively from a given comdat group.
  39
  40    The implementation maintains a queue linked by AUX pointer terminated by
  41    pointer value 1. Lattice values are NULL for TOP, actual comdat group, or
  42    ERROR_MARK_NODE for bottom.
  43
  44    TODO: When symbol is used only by comdat symbols, but from different groups,
  45    it would make sense to produce a new comdat group for it with anonymous name.
  46
  47    TODO2: We can't mix variables and functions within one group.  Currently
  48    we just give up on references of symbols of different types.  We also should
  49    handle this by anonymous comdat group section.  */
  50
  51 #include "config.h"
  52 #include "system.h"
  53 #include "coretypes.h"
  54 #include "tm.h"
  55 #include "tree.h"
  56 #include "tree-pass.h"
  57 #include "cgraph.h"
  58
  59 /* Main dataflow loop propagating comdat groups across
  60    the symbol table.  All references to SYMBOL are examined
  61    and NEWGROUP is updated accordingly. MAP holds current lattice
  62    values for individual symbols.  */
  63
  64 tree
  65 propagate_comdat_group (struct symtab_node *symbol,
  66                         tree newgroup, hash_map<symtab_node *, tree> &map)
  67 {
  68   int i;
  69   struct ipa_ref *ref;
  70
  71   /* Walk all references to SYMBOL, recursively dive into aliases.  */
  72
  73   for (i = 0;
  74        symbol->iterate_referring (i, ref)
  75        && newgroup != error_mark_node; i++)
  76     {
  77       struct symtab_node *symbol2 = ref->referring;
  78
  79       if (ref->use == IPA_REF_ALIAS)
  80         {
  81           newgroup = propagate_comdat_group (symbol2, newgroup, map);
  82           continue;
  83         }
  84
  85       /* One COMDAT group cannot hold both variables and functions at
  86          a same time.  For now we just go to BOTTOM, in future we may
  87          invent special comdat groups for this case.  */
  88
  89       if (symbol->type != symbol2->type)
  90         {
  91           newgroup = error_mark_node;
  92           break;
  93         }
  94
  95       /* If we see inline clone, its comdat group actually
  96          corresponds to the comdat group of the function it is inlined
  97          to.  */
  98
  99       if (cgraph_node * cn = dyn_cast <cgraph_node *> (symbol2))
 100         {
 101           if (cn->inlined_to)
 102             symbol2 = cn->inlined_to;
 103         }
 104
 105       /* The actual merge operation.  */
 106
 107       tree *val2 = map.get (symbol2);
 108
 109       if (val2 && *val2 != newgroup)
 110         {
 111           if (!newgroup)
 112             newgroup = *val2;
 113           else
 114             newgroup = error_mark_node;
 115         }
 116     }
 117
 118   /* If we analyze function, walk also callers.  */
 119
 120   cgraph_node *cnode = dyn_cast <cgraph_node *> (symbol);
 121
 122   if (cnode)
 123     for (struct cgraph_edge * edge = cnode->callers;
 124          edge && newgroup != error_mark_node; edge = edge->next_caller)
 125       {
 126         struct symtab_node *symbol2 = edge->caller;
 127
 128         if (cgraph_node * cn = dyn_cast <cgraph_node *> (symbol2))
 129           {
 130             /* Thunks cannot call across section boundary.  */
 131             if (cn->thunk)
 132               newgroup = propagate_comdat_group (symbol2, newgroup, map);
 133             /* If we see inline clone, its comdat group actually
 134                corresponds to the comdat group of the function it
 135                is inlined to.  */
 136             if (cn->inlined_to)
 137               symbol2 = cn->inlined_to;
 138           }
 139
 140         /* The actual merge operation.  */
 141
 142         tree *val2 = map.get (symbol2);
 143
 144         if (val2 && *val2 != newgroup)
 145           {
 146             if (!newgroup)
 147               newgroup = *val2;
 148             else
 149               newgroup = error_mark_node;
 150           }
 151       }
 152   return newgroup;
 153 }
 154
 155
 156 /* Add all references of SYMBOL that are defined into queue started by FIRST
 157    and linked by AUX pointer (unless they are already enqueued).
 158    Walk recursively inlined functions.  */
 159
 160 void
 161 enqueue_references (symtab_node **first,
 162                     symtab_node *symbol)
 163 {
 164   int i;
 165   struct ipa_ref *ref = NULL;
 166
 167   for (i = 0; symbol->iterate_reference (i, ref); i++)
 168     {
 169       symtab_node *node = ref->referred->ultimate_alias_target ();
 170
 171       /* Always keep thunks in same sections as target function.  */
 172       if (is_a <cgraph_node *>(node))
 173         node = dyn_cast <cgraph_node *> (node)->function_symbol ();
 174       if (!node->aux && node->definition)
 175         {
 176            node->aux = *first;
 177            *first = node;
 178         }
 179     }
 180
 181   if (cgraph_node *cnode = dyn_cast <cgraph_node *> (symbol))
 182     {
 183       struct cgraph_edge *edge;
 184
 185       for (edge = cnode->callees; edge; edge = edge->next_callee)
 186         if (!edge->inline_failed)
 187           enqueue_references (first, edge->callee);
 188         else
 189           {
 190             symtab_node *node = edge->callee->ultimate_alias_target ();
 191
 192             /* Always keep thunks in same sections as target function.  */
 193             if (is_a <cgraph_node *>(node))
 194               node = dyn_cast <cgraph_node *> (node)->function_symbol ();
 195             if (!node->aux && node->definition)
 196               {
 197                  node->aux = *first;
 198                  *first = node;
 199               }
 200           }
 201     }
 202 }
 203
 204 /* Set comdat group of SYMBOL to GROUP.
 205    Callback for for_node_and_aliases.  */
 206
 207 bool
 208 set_comdat_group (symtab_node *symbol,
 209                   void *head_p)
 210 {
 211   symtab_node *head = (symtab_node *)head_p;
 212
 213   gcc_assert (!symbol->get_comdat_group ());
 214   if (symbol->real_symbol_p ())
 215     {
 216       symbol->set_comdat_group (head->get_comdat_group ());
 217       symbol->add_to_same_comdat_group (head);
 218     }
 219   return false;
 220 }
 221
 222 /* Set comdat group of SYMBOL to GROUP.
 223    Callback for for_node_thunks_and_aliases.  */
 224
 225 bool
 226 set_comdat_group_1 (cgraph_node *symbol,
 227                     void *head_p)
 228 {
 229   return set_comdat_group (symbol, head_p);
 230 }
 231
 232 /* The actual pass with the main dataflow loop.  */
 233
 234 static unsigned int
 235 ipa_comdats (void)
 236 {
 237   hash_map<symtab_node *, tree> map (251);
 238   hash_map<tree, symtab_node *> comdat_head_map (251);
 239   symtab_node *symbol;
 240   bool comdat_group_seen = false;
 241   symtab_node *first = (symtab_node *) (void *) 1;
 242   tree group;
 243
 244   /* Start the dataflow by assigning comdat group to symbols that are in comdat
 245      groups already.  All other externally visible symbols must stay, we use
 246      ERROR_MARK_NODE as bottom for the propagation.  */
 247
 248   FOR_EACH_DEFINED_SYMBOL (symbol)
 249     if (!symbol->real_symbol_p ())
 250       ;
 251     else if ((group = symbol->get_comdat_group ()) != NULL)
 252       {
 253         map.put (symbol, group);
 254         comdat_head_map.put (group, symbol);
 255         comdat_group_seen = true;
 256
 257         /* Mark the symbol so we won't waste time visiting it for dataflow.  */
 258         symbol->aux = (symtab_node *) (void *) 1;
 259       }
 260     /* See symbols that cannot be privatized to comdats; that is externally
 261        visible symbols or otherwise used ones.  We also do not want to mangle
 262        user section names.  */
 263     else if (symbol->externally_visible
 264              || symbol->force_output
 265              || symbol->used_from_other_partition
 266              || TREE_THIS_VOLATILE (symbol->decl)
 267              || symbol->get_section ()
 268              || (TREE_CODE (symbol->decl) == FUNCTION_DECL
 269                  && (DECL_STATIC_CONSTRUCTOR (symbol->decl)
 270                      || DECL_STATIC_DESTRUCTOR (symbol->decl))))
 271       {
 272         symtab_node *target = symbol->ultimate_alias_target ();
 273
 274         /* Always keep thunks in same sections as target function.  */
 275         if (is_a <cgraph_node *>(target))
 276           target = dyn_cast <cgraph_node *> (target)->function_symbol ();
 277         map.put (target, error_mark_node);
 278
 279         /* Mark the symbol so we won't waste time visiting it for dataflow.  */
 280         symbol->aux = (symtab_node *) (void *) 1;
 281       }
 282     else
 283       {
 284         /* Enqueue symbol for dataflow.  */
 285         symbol->aux = first;
 286         first = symbol;
 287       }
 288
 289   if (!comdat_group_seen)
 290     {
 291       FOR_EACH_DEFINED_SYMBOL (symbol)
 292         symbol->aux = NULL;
 293       return 0;
 294     }
 295
 296   /* The actual dataflow.  */
 297
 298   while (first != (void *) 1)
 299     {
 300       tree group = NULL;
 301       tree newgroup, *val;
 302
 303       symbol = first;
 304       first = (symtab_node *)first->aux;
 305
 306       /* Get current lattice value of SYMBOL.  */
 307       val = map.get (symbol);
 308       if (val)
 309         group = *val;
 310
 311       /* If it is bottom, there is nothing to do; do not clear AUX
 312          so we won't re-queue the symbol.  */
 313       if (group == error_mark_node)
 314         continue;
 315
 316       newgroup = propagate_comdat_group (symbol, group, map);
 317
 318       /* If nothing changed, proceed to next symbol.  */
 319       if (newgroup == group)
 320         {
 321           symbol->aux = NULL;
 322           continue;
 323         }
 324
 325       /* Update lattice value and enqueue all references for re-visiting.  */
 326       gcc_assert (newgroup);
 327       if (val)
 328         *val = newgroup;
 329       else
 330         map.put (symbol, newgroup);
 331       enqueue_references (&first, symbol);
 332
 333       /* We may need to revisit the symbol unless it is BOTTOM.  */
 334       if (newgroup != error_mark_node)
 335         symbol->aux = NULL;
 336     }
 337
 338   /* Finally assign symbols to the sections.  */
 339
 340   FOR_EACH_DEFINED_SYMBOL (symbol)
 341     {
 342       struct cgraph_node *fun;
 343       symbol->aux = NULL;
 344       if (!symbol->get_comdat_group ()
 345           && !symbol->alias
 346           && (!(fun = dyn_cast <cgraph_node *> (symbol))
 347               || !fun->thunk)
 348           && symbol->real_symbol_p ())
 349         {
 350           tree *val = map.get (symbol);
 351
 352           /* A NULL here means that SYMBOL is unreachable in the definition
 353              of ipa-comdats. Either ipa-comdats is wrong about this or someone
 354              forgot to cleanup and remove unreachable functions earlier.  */
 355           gcc_assert (val);
 356
 357           tree group = *val;
 358
 359           if (group == error_mark_node)
 360             continue;
 361           if (dump_file)
 362             {
 363               fprintf (dump_file, "Localizing symbol\n");
 364               symbol->dump (dump_file);
 365               fprintf (dump_file, "To group: %s\n", IDENTIFIER_POINTER (group));
 366             }
 367           if (is_a <cgraph_node *> (symbol))
 368            dyn_cast <cgraph_node *>(symbol)->call_for_symbol_thunks_and_aliases
 369                   (set_comdat_group_1,
 370                    *comdat_head_map.get (group),
 371                    true);
 372           else
 373            symbol->call_for_symbol_and_aliases
 374                   (set_comdat_group,
 375                    *comdat_head_map.get (group),
 376                    true);
 377         }
 378     }
 379
 380 #if 0
 381   /* Recompute calls comdat local flag.  This need to be done after all changes
 382      are made.  */
 383   cgraph_node *function;
 384   FOR_EACH_DEFINED_FUNCTION (function)
 385     if (function->get_comdat_group ())
 386       function->calls_comdat_local = function->check_calls_comdat_local_p ();
 387 #endif
 388   return 0;
 389 }
 390
 391 namespace {
 392
 393 const pass_data pass_data_ipa_comdats =
 394 {
 395   IPA_PASS, /* type */
 396   "comdats", /* name */
 397   OPTGROUP_NONE, /* optinfo_flags */
 398   TV_IPA_COMDATS, /* tv_id */
 399   0, /* properties_required */
 400   0, /* properties_provided */
 401   0, /* properties_destroyed */
 402   0, /* todo_flags_start */
 403   0, /* todo_flags_finish */
 404 };
 405
 406 class pass_ipa_comdats : public ipa_opt_pass_d
 407 {
 408 public:
 409   pass_ipa_comdats (gcc::context *ctxt)
 410     : ipa_opt_pass_d (pass_data_ipa_comdats, ctxt,
 411                       NULL, /* generate_summary */
 412                       NULL, /* write_summary */
 413                       NULL, /* read_summary */
 414                       NULL, /* write_optimization_summary */
 415                       NULL, /* read_optimization_summary */
 416                       NULL, /* stmt_fixup */
 417                       0, /* function_transform_todo_flags_start */
 418                       NULL, /* function_transform */
 419                       NULL) /* variable_transform */
 420   {}
 421
 422   /* opt_pass methods: */
 423   bool gate (function *) final override;
 424   unsigned int execute (function *) final override { return ipa_comdats (); }
 425
 426 }; // class pass_ipa_comdats
 427
 428 bool
 429 pass_ipa_comdats::gate (function *)
 430 {
 431   return HAVE_COMDAT_GROUP;
 432 }
 433
 434 } // anon namespace
 435
 436 ipa_opt_pass_d *
 437 make_pass_ipa_comdats (gcc::context *ctxt)
 438 {
 439   return new pass_ipa_comdats (ctxt);
 440 }