gcc/gimple-ssa-isolate-paths.c

   1 /* Detect paths through the CFG which can never be executed in a conforming
   2    program and isolate them.
   3
   4    Copyright (C) 2013-2014 Free Software Foundation, Inc.
   5
   6 This file is part of GCC.
   7
   8 GCC is free software; you can redistribute it and/or modify
   9 it under the terms of the GNU General Public License as published by
  10 the Free Software Foundation; either version 3, or (at your option)
  11 any later version.
  12
  13 GCC is distributed in the hope that it will be useful,
  14 but WITHOUT ANY WARRANTY; without even the implied warranty of
  15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16 GNU General Public License for more details.
  17
  18 You should have received a copy of the GNU General Public License
  19 along with GCC; see the file COPYING3.  If not see
  20 <http://www.gnu.org/licenses/>.  */
  21
  22 #include "config.h"
  23 #include "system.h"
  24 #include "coretypes.h"
  25 #include "tree.h"
  26 #include "flags.h"
  27 #include "basic-block.h"
  28 #include "tree-ssa-alias.h"
  29 #include "internal-fn.h"
  30 #include "gimple-expr.h"
  31 #include "is-a.h"
  32 #include "gimple.h"
  33 #include "gimple-iterator.h"
  34 #include "gimple-walk.h"
  35 #include "tree-ssa.h"
  36 #include "stringpool.h"
  37 #include "tree-ssanames.h"
  38 #include "gimple-ssa.h"
  39 #include "tree-ssa-operands.h"
  40 #include "tree-phinodes.h"
  41 #include "ssa-iterators.h"
  42 #include "cfgloop.h"
  43 #include "tree-pass.h"
  44 #include "tree-cfg.h"
  45
  46
  47 static bool cfg_altered;
  48
  49 /* Callback for walk_stmt_load_store_ops.
  50
  51    Return TRUE if OP will dereference the tree stored in DATA, FALSE
  52    otherwise.
  53
  54    This routine only makes a superficial check for a dereference.  Thus,
  55    it must only be used if it is safe to return a false negative.  */
  56 static bool
  57 check_loadstore (gimple stmt, tree op, tree, void *data)
  58 {
  59   if ((TREE_CODE (op) == MEM_REF || TREE_CODE (op) == TARGET_MEM_REF)
  60       && operand_equal_p (TREE_OPERAND (op, 0), (tree)data, 0))
  61     {
  62       TREE_THIS_VOLATILE (op) = 1;
  63       TREE_SIDE_EFFECTS (op) = 1;
  64       update_stmt (stmt);
  65       return true;
  66     }
  67   return false;
  68 }
  69
  70 /* Insert a trap after SI and remove SI and all statements after the trap.  */
  71
  72 static void
  73 insert_trap_and_remove_trailing_statements (gimple_stmt_iterator *si_p, tree op)
  74 {
  75   /* We want the NULL pointer dereference to actually occur so that
  76      code that wishes to catch the signal can do so.
  77
  78      If the dereference is a load, then there's nothing to do as the
  79      LHS will be a throw-away SSA_NAME and the RHS is the NULL dereference.
  80
  81      If the dereference is a store and we can easily transform the RHS,
  82      then simplify the RHS to enable more DCE.   Note that we require the
  83      statement to be a GIMPLE_ASSIGN which filters out calls on the RHS.  */
  84   gimple stmt = gsi_stmt (*si_p);
  85   if (walk_stmt_load_store_ops (stmt, (void *)op, NULL, check_loadstore)
  86       && is_gimple_assign (stmt)
  87       && INTEGRAL_TYPE_P (TREE_TYPE (gimple_assign_lhs (stmt))))
  88     {
  89       /* We just need to turn the RHS into zero converted to the proper
  90          type.  */
  91       tree type = TREE_TYPE (gimple_assign_lhs (stmt));
  92       gimple_assign_set_rhs_code (stmt, INTEGER_CST);
  93       gimple_assign_set_rhs1 (stmt, fold_convert (type, integer_zero_node));
  94       update_stmt (stmt);
  95     }
  96
  97   gimple new_stmt
  98     = gimple_build_call (builtin_decl_explicit (BUILT_IN_TRAP), 0);
  99   gimple_seq seq = NULL;
 100   gimple_seq_add_stmt (&seq, new_stmt);
 101
 102   /* If we had a NULL pointer dereference, then we want to insert the
 103      __builtin_trap after the statement, for the other cases we want
 104      to insert before the statement.  */
 105   if (walk_stmt_load_store_ops (stmt, (void *)op,
 106                                 check_loadstore,
 107                                 check_loadstore))
 108     gsi_insert_after (si_p, seq, GSI_NEW_STMT);
 109   else
 110     gsi_insert_before (si_p, seq, GSI_NEW_STMT);
 111
 112   /* We must remove statements from the end of the block so that we
 113      never reference a released SSA_NAME.  */
 114   basic_block bb = gimple_bb (gsi_stmt (*si_p));
 115   for (gimple_stmt_iterator si = gsi_last_bb (bb);
 116        gsi_stmt (si) != gsi_stmt (*si_p);
 117        si = gsi_last_bb (bb))
 118     {
 119       stmt = gsi_stmt (si);
 120       unlink_stmt_vdef (stmt);
 121       gsi_remove (&si, true);
 122       release_defs (stmt);
 123     }
 124 }
 125
 126 /* BB when reached via incoming edge E will exhibit undefined behaviour
 127    at STMT.  Isolate and optimize the path which exhibits undefined
 128    behaviour.
 129
 130    Isolation is simple.  Duplicate BB and redirect E to BB'.
 131
 132    Optimization is simple as well.  Replace STMT in BB' with an
 133    unconditional trap and remove all outgoing edges from BB'.
 134
 135    DUPLICATE is a pre-existing duplicate, use it as BB' if it exists.
 136
 137    Return BB'.  */
 138
 139 basic_block
 140 isolate_path (basic_block bb, basic_block duplicate,
 141               edge e, gimple stmt, tree op)
 142 {
 143   gimple_stmt_iterator si, si2;
 144   edge_iterator ei;
 145   edge e2;
 146
 147   /* First duplicate BB if we have not done so already and remove all
 148      the duplicate's outgoing edges as duplicate is going to unconditionally
 149      trap.  Removing the outgoing edges is both an optimization and ensures
 150      we don't need to do any PHI node updates.  */
 151   if (!duplicate)
 152     {
 153       duplicate = duplicate_block (bb, NULL, NULL);
 154       for (ei = ei_start (duplicate->succs); (e2 = ei_safe_edge (ei)); )
 155         remove_edge (e2);
 156     }
 157
 158   /* Complete the isolation step by redirecting E to reach DUPLICATE.  */
 159   e2 = redirect_edge_and_branch (e, duplicate);
 160   if (e2)
 161     flush_pending_stmts (e2);
 162
 163
 164   /* There may be more than one statement in DUPLICATE which exhibits
 165      undefined behaviour.  Ultimately we want the first such statement in
 166      DUPLCIATE so that we're able to delete as much code as possible.
 167
 168      So each time we discover undefined behaviour in DUPLICATE, search for
 169      the statement which triggers undefined behaviour.  If found, then
 170      transform the statement into a trap and delete everything after the
 171      statement.  If not found, then this particular instance was subsumed by
 172      an earlier instance of undefined behaviour and there's nothing to do.
 173
 174      This is made more complicated by the fact that we have STMT, which is in
 175      BB rather than in DUPLICATE.  So we set up two iterators, one for each
 176      block and walk forward looking for STMT in BB, advancing each iterator at
 177      each step.
 178
 179      When we find STMT the second iterator should point to STMT's equivalent in
 180      duplicate.  If DUPLICATE ends before STMT is found in BB, then there's
 181      nothing to do.
 182
 183      Ignore labels and debug statements.  */
 184   si = gsi_start_nondebug_after_labels_bb (bb);
 185   si2 = gsi_start_nondebug_after_labels_bb (duplicate);
 186   while (!gsi_end_p (si) && !gsi_end_p (si2) && gsi_stmt (si) != stmt)
 187     {
 188       gsi_next_nondebug (&si);
 189       gsi_next_nondebug (&si2);
 190     }
 191
 192   /* This would be an indicator that we never found STMT in BB, which should
 193      never happen.  */
 194   gcc_assert (!gsi_end_p (si));
 195
 196   /* If we did not run to the end of DUPLICATE, then SI points to STMT and
 197      SI2 points to the duplicate of STMT in DUPLICATE.  Insert a trap
 198      before SI2 and remove SI2 and all trailing statements.  */
 199   if (!gsi_end_p (si2))
 200     insert_trap_and_remove_trailing_statements (&si2, op);
 201
 202   return duplicate;
 203 }
 204
 205 /* Look for PHI nodes which feed statements in the same block where
 206    the value of the PHI node implies the statement is erroneous.
 207
 208    For example, a NULL PHI arg value which then feeds a pointer
 209    dereference.
 210
 211    When found isolate and optimize the path associated with the PHI
 212    argument feeding the erroneous statement.  */
 213 static void
 214 find_implicit_erroneous_behaviour (void)
 215 {
 216   basic_block bb;
 217
 218   FOR_EACH_BB_FN (bb, cfun)
 219     {
 220       gimple_stmt_iterator si;
 221
 222       /* Out of an abundance of caution, do not isolate paths to a
 223          block where the block has any abnormal outgoing edges.
 224
 225          We might be able to relax this in the future.  We have to detect
 226          when we have to split the block with the NULL dereference and
 227          the trap we insert.  We have to preserve abnormal edges out
 228          of the isolated block which in turn means updating PHIs at
 229          the targets of those abnormal outgoing edges.  */
 230       if (has_abnormal_or_eh_outgoing_edge_p (bb))
 231         continue;
 232
 233       /* First look for a PHI which sets a pointer to NULL and which
 234          is then dereferenced within BB.  This is somewhat overly
 235          conservative, but probably catches most of the interesting
 236          cases.   */
 237       for (si = gsi_start_phis (bb); !gsi_end_p (si); gsi_next (&si))
 238         {
 239           gimple phi = gsi_stmt (si);
 240           tree lhs = gimple_phi_result (phi);
 241
 242           /* If the result is not a pointer, then there is no need to
 243              examine the arguments.  */
 244           if (!POINTER_TYPE_P (TREE_TYPE (lhs)))
 245             continue;
 246
 247           /* PHI produces a pointer result.  See if any of the PHI's
 248              arguments are NULL.
 249
 250              When we remove an edge, we want to reprocess the current
 251              index, hence the ugly way we update I for each iteration.  */
 252           basic_block duplicate = NULL;
 253           for (unsigned i = 0, next_i = 0;
 254                i < gimple_phi_num_args (phi);
 255                i = next_i)
 256             {
 257               tree op = gimple_phi_arg_def (phi, i);
 258
 259               next_i = i + 1;
 260
 261               if (!integer_zerop (op))
 262                 continue;
 263
 264               edge e = gimple_phi_arg_edge (phi, i);
 265               imm_use_iterator iter;
 266               gimple use_stmt;
 267
 268               /* We've got a NULL PHI argument.  Now see if the
 269                  PHI's result is dereferenced within BB.  */
 270               FOR_EACH_IMM_USE_STMT (use_stmt, iter, lhs)
 271                 {
 272                   /* We only care about uses in BB.  Catching cases in
 273                      in other blocks would require more complex path
 274                      isolation code.   */
 275                   if (gimple_bb (use_stmt) != bb)
 276                     continue;
 277
 278                   if (infer_nonnull_range (use_stmt, lhs,
 279                                            flag_isolate_erroneous_paths_dereference,
 280                                            flag_isolate_erroneous_paths_attribute))
 281
 282                     {
 283                       duplicate = isolate_path (bb, duplicate,
 284                                                 e, use_stmt, lhs);
 285
 286                       /* When we remove an incoming edge, we need to
 287                          reprocess the Ith element.  */
 288                       next_i = i;
 289                       cfg_altered = true;
 290                     }
 291                 }
 292             }
 293         }
 294     }
 295 }
 296
 297 /* Look for statements which exhibit erroneous behaviour.  For example
 298    a NULL pointer dereference.
 299
 300    When found, optimize the block containing the erroneous behaviour.  */
 301 static void
 302 find_explicit_erroneous_behaviour (void)
 303 {
 304   basic_block bb;
 305
 306   FOR_EACH_BB_FN (bb, cfun)
 307     {
 308       gimple_stmt_iterator si;
 309
 310       /* Out of an abundance of caution, do not isolate paths to a
 311          block where the block has any abnormal outgoing edges.
 312
 313          We might be able to relax this in the future.  We have to detect
 314          when we have to split the block with the NULL dereference and
 315          the trap we insert.  We have to preserve abnormal edges out
 316          of the isolated block which in turn means updating PHIs at
 317          the targets of those abnormal outgoing edges.  */
 318       if (has_abnormal_or_eh_outgoing_edge_p (bb))
 319         continue;
 320
 321       /* Now look at the statements in the block and see if any of
 322          them explicitly dereference a NULL pointer.  This happens
 323          because of jump threading and constant propagation.  */
 324       for (si = gsi_start_bb (bb); !gsi_end_p (si); gsi_next (&si))
 325         {
 326           gimple stmt = gsi_stmt (si);
 327
 328           /* By passing null_pointer_node, we can use infer_nonnull_range
 329              to detect explicit NULL pointer dereferences and other uses
 330              where a non-NULL value is required.  */
 331           if (infer_nonnull_range (stmt, null_pointer_node,
 332                                    flag_isolate_erroneous_paths_dereference,
 333                                    flag_isolate_erroneous_paths_attribute))
 334             {
 335               insert_trap_and_remove_trailing_statements (&si,
 336                                                           null_pointer_node);
 337
 338               /* And finally, remove all outgoing edges from BB.  */
 339               edge e;
 340               for (edge_iterator ei = ei_start (bb->succs);
 341                    (e = ei_safe_edge (ei)); )
 342                 remove_edge (e);
 343
 344               /* Ignore any more operands on this statement and
 345                  continue the statement iterator (which should
 346                  terminate its loop immediately.  */
 347               cfg_altered = true;
 348               break;
 349             }
 350         }
 351     }
 352 }
 353 /* Search the function for statements which, if executed, would cause
 354    the program to fault such as a dereference of a NULL pointer.
 355
 356    Such a program can't be valid if such a statement was to execute
 357    according to ISO standards.
 358
 359    We detect explicit NULL pointer dereferences as well as those implied
 360    by a PHI argument having a NULL value which unconditionally flows into
 361    a dereference in the same block as the PHI.
 362
 363    In the former case we replace the offending statement with an
 364    unconditional trap and eliminate the outgoing edges from the statement's
 365    basic block.  This may expose secondary optimization opportunities.
 366
 367    In the latter case, we isolate the path(s) with the NULL PHI
 368    feeding the dereference.  We can then replace the offending statement
 369    and eliminate the outgoing edges in the duplicate.  Again, this may
 370    expose secondary optimization opportunities.
 371
 372    A warning for both cases may be advisable as well.
 373
 374    Other statically detectable violations of the ISO standard could be
 375    handled in a similar way, such as out-of-bounds array indexing.  */
 376
 377 static unsigned int
 378 gimple_ssa_isolate_erroneous_paths (void)
 379 {
 380   initialize_original_copy_tables ();
 381
 382   /* Search all the blocks for edges which, if traversed, will
 383      result in undefined behaviour.  */
 384   cfg_altered = false;
 385
 386   /* First handle cases where traversal of a particular edge
 387      triggers undefined behaviour.  These cases require creating
 388      duplicate blocks and thus new SSA_NAMEs.
 389
 390      We want that process complete prior to the phase where we start
 391      removing edges from the CFG.  Edge removal may ultimately result in
 392      removal of PHI nodes and thus releasing SSA_NAMEs back to the
 393      name manager.
 394
 395      If the two processes run in parallel we could release an SSA_NAME
 396      back to the manager but we could still have dangling references
 397      to the released SSA_NAME in unreachable blocks.
 398      that any released names not have dangling references in the IL.  */
 399   find_implicit_erroneous_behaviour ();
 400   find_explicit_erroneous_behaviour ();
 401
 402   free_original_copy_tables ();
 403
 404   /* We scramble the CFG and loop structures a bit, clean up
 405      appropriately.  We really should incrementally update the
 406      loop structures, in theory it shouldn't be that hard.  */
 407   if (cfg_altered)
 408     {
 409       free_dominance_info (CDI_DOMINATORS);
 410       free_dominance_info (CDI_POST_DOMINATORS);
 411       loops_state_set (LOOPS_NEED_FIXUP);
 412       return TODO_cleanup_cfg | TODO_update_ssa;
 413     }
 414   return 0;
 415 }
 416
 417 namespace {
 418 const pass_data pass_data_isolate_erroneous_paths =
 419 {
 420   GIMPLE_PASS, /* type */
 421   "isolate-paths", /* name */
 422   OPTGROUP_NONE, /* optinfo_flags */
 423   true, /* has_execute */
 424   TV_ISOLATE_ERRONEOUS_PATHS, /* tv_id */
 425   ( PROP_cfg | PROP_ssa ), /* properties_required */
 426   0, /* properties_provided */
 427   0, /* properties_destroyed */
 428   0, /* todo_flags_start */
 429   0, /* todo_flags_finish */
 430 };
 431
 432 class pass_isolate_erroneous_paths : public gimple_opt_pass
 433 {
 434 public:
 435   pass_isolate_erroneous_paths (gcc::context *ctxt)
 436     : gimple_opt_pass (pass_data_isolate_erroneous_paths, ctxt)
 437   {}
 438
 439   /* opt_pass methods: */
 440   opt_pass * clone () { return new pass_isolate_erroneous_paths (m_ctxt); }
 441   virtual bool gate (function *)
 442     {
 443       /* If we do not have a suitable builtin function for the trap statement,
 444          then do not perform the optimization.  */
 445       return (flag_isolate_erroneous_paths_dereference != 0
 446               || flag_isolate_erroneous_paths_attribute != 0);
 447     }
 448
 449   virtual unsigned int execute (function *)
 450     {
 451       return gimple_ssa_isolate_erroneous_paths ();
 452     }
 453
 454 }; // class pass_isolate_erroneous_paths
 455 }
 456
 457 gimple_opt_pass *
 458 make_pass_isolate_erroneous_paths (gcc::context *ctxt)
 459 {
 460   return new pass_isolate_erroneous_paths (ctxt);
 461 }