| /* Expands front end tree to back end RTL for GCC |
| Copyright (C) 1987-2015 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 |
| <http://www.gnu.org/licenses/>. */ |
| |
| /* This file handles the generation of rtl code from tree structure |
| above the level of expressions, using subroutines in exp*.c and emit-rtl.c. |
| The functions whose names start with `expand_' are called by the |
| expander to generate RTL instructions for various kinds of constructs. */ |
| |
| #include "config.h" |
| #include "system.h" |
| #include "coretypes.h" |
| #include "tm.h" |
| |
| #include "rtl.h" |
| #include "hard-reg-set.h" |
| #include "hash-set.h" |
| #include "machmode.h" |
| #include "vec.h" |
| #include "double-int.h" |
| #include "input.h" |
| #include "alias.h" |
| #include "symtab.h" |
| #include "wide-int.h" |
| #include "inchash.h" |
| #include "tree.h" |
| #include "fold-const.h" |
| #include "varasm.h" |
| #include "stor-layout.h" |
| #include "tm_p.h" |
| #include "flags.h" |
| #include "except.h" |
| #include "function.h" |
| #include "insn-config.h" |
| #include "hashtab.h" |
| #include "statistics.h" |
| #include "real.h" |
| #include "fixed-value.h" |
| #include "expmed.h" |
| #include "dojump.h" |
| #include "explow.h" |
| #include "calls.h" |
| #include "emit-rtl.h" |
| #include "stmt.h" |
| #include "expr.h" |
| #include "libfuncs.h" |
| #include "recog.h" |
| #include "diagnostic-core.h" |
| #include "output.h" |
| #include "langhooks.h" |
| #include "predict.h" |
| #include "insn-codes.h" |
| #include "optabs.h" |
| #include "target.h" |
| #include "cfganal.h" |
| #include "basic-block.h" |
| #include "tree-ssa-alias.h" |
| #include "internal-fn.h" |
| #include "gimple-expr.h" |
| #include "is-a.h" |
| #include "gimple.h" |
| #include "regs.h" |
| #include "alloc-pool.h" |
| #include "pretty-print.h" |
| #include "params.h" |
| #include "dumpfile.h" |
| #include "builtins.h" |
| |
| |
| /* Functions and data structures for expanding case statements. */ |
| |
| /* Case label structure, used to hold info on labels within case |
| statements. We handle "range" labels; for a single-value label |
| as in C, the high and low limits are the same. |
| |
| We start with a vector of case nodes sorted in ascending order, and |
| the default label as the last element in the vector. Before expanding |
| to RTL, we transform this vector into a list linked via the RIGHT |
| fields in the case_node struct. Nodes with higher case values are |
| later in the list. |
| |
| Switch statements can be output in three forms. A branch table is |
| used if there are more than a few labels and the labels are dense |
| within the range between the smallest and largest case value. If a |
| branch table is used, no further manipulations are done with the case |
| node chain. |
| |
| The alternative to the use of a branch table is to generate a series |
| of compare and jump insns. When that is done, we use the LEFT, RIGHT, |
| and PARENT fields to hold a binary tree. Initially the tree is |
| totally unbalanced, with everything on the right. We balance the tree |
| with nodes on the left having lower case values than the parent |
| and nodes on the right having higher values. We then output the tree |
| in order. |
| |
| For very small, suitable switch statements, we can generate a series |
| of simple bit test and branches instead. */ |
| |
| struct case_node |
| { |
| struct case_node *left; /* Left son in binary tree */ |
| struct case_node *right; /* Right son in binary tree; also node chain */ |
| struct case_node *parent; /* Parent of node in binary tree */ |
| tree low; /* Lowest index value for this label */ |
| tree high; /* Highest index value for this label */ |
| tree code_label; /* Label to jump to when node matches */ |
| int prob; /* Probability of taking this case. */ |
| /* Probability of reaching subtree rooted at this node */ |
| int subtree_prob; |
| }; |
| |
| typedef struct case_node case_node; |
| typedef struct case_node *case_node_ptr; |
| |
| extern basic_block label_to_block_fn (struct function *, tree); |
| |
| static bool check_unique_operand_names (tree, tree, tree); |
| static char *resolve_operand_name_1 (char *, tree, tree, tree); |
| static void balance_case_nodes (case_node_ptr *, case_node_ptr); |
| static int node_has_low_bound (case_node_ptr, tree); |
| static int node_has_high_bound (case_node_ptr, tree); |
| static int node_is_bounded (case_node_ptr, tree); |
| static void emit_case_nodes (rtx, case_node_ptr, rtx, int, tree); |
| |
| /* Return the rtx-label that corresponds to a LABEL_DECL, |
| creating it if necessary. */ |
| |
| rtx |
| label_rtx (tree label) |
| { |
| gcc_assert (TREE_CODE (label) == LABEL_DECL); |
| |
| if (!DECL_RTL_SET_P (label)) |
| { |
| rtx_code_label *r = gen_label_rtx (); |
| SET_DECL_RTL (label, r); |
| if (FORCED_LABEL (label) || DECL_NONLOCAL (label)) |
| LABEL_PRESERVE_P (r) = 1; |
| } |
| |
| return DECL_RTL (label); |
| } |
| |
| /* As above, but also put it on the forced-reference list of the |
| function that contains it. */ |
| rtx |
| force_label_rtx (tree label) |
| { |
| rtx_insn *ref = as_a <rtx_insn *> (label_rtx (label)); |
| tree function = decl_function_context (label); |
| |
| gcc_assert (function); |
| |
| forced_labels = gen_rtx_INSN_LIST (VOIDmode, ref, forced_labels); |
| return ref; |
| } |
| |
| /* Add an unconditional jump to LABEL as the next sequential instruction. */ |
| |
| void |
| emit_jump (rtx label) |
| { |
| do_pending_stack_adjust (); |
| emit_jump_insn (gen_jump (label)); |
| emit_barrier (); |
| } |
| |
| /* Handle goto statements and the labels that they can go to. */ |
| |
| /* Specify the location in the RTL code of a label LABEL, |
| which is a LABEL_DECL tree node. |
| |
| This is used for the kind of label that the user can jump to with a |
| goto statement, and for alternatives of a switch or case statement. |
| RTL labels generated for loops and conditionals don't go through here; |
| they are generated directly at the RTL level, by other functions below. |
| |
| Note that this has nothing to do with defining label *names*. |
| Languages vary in how they do that and what that even means. */ |
| |
| void |
| expand_label (tree label) |
| { |
| rtx_insn *label_r = as_a <rtx_insn *> (label_rtx (label)); |
| |
| do_pending_stack_adjust (); |
| emit_label (label_r); |
| if (DECL_NAME (label)) |
| LABEL_NAME (DECL_RTL (label)) = IDENTIFIER_POINTER (DECL_NAME (label)); |
| |
| if (DECL_NONLOCAL (label)) |
| { |
| expand_builtin_setjmp_receiver (NULL); |
| nonlocal_goto_handler_labels |
| = gen_rtx_INSN_LIST (VOIDmode, label_r, |
| nonlocal_goto_handler_labels); |
| } |
| |
| if (FORCED_LABEL (label)) |
| forced_labels = gen_rtx_INSN_LIST (VOIDmode, label_r, forced_labels); |
| |
| if (DECL_NONLOCAL (label) || FORCED_LABEL (label)) |
| maybe_set_first_label_num (label_r); |
| } |
| |
| /* Parse the output constraint pointed to by *CONSTRAINT_P. It is the |
| OPERAND_NUMth output operand, indexed from zero. There are NINPUTS |
| inputs and NOUTPUTS outputs to this extended-asm. Upon return, |
| *ALLOWS_MEM will be TRUE iff the constraint allows the use of a |
| memory operand. Similarly, *ALLOWS_REG will be TRUE iff the |
| constraint allows the use of a register operand. And, *IS_INOUT |
| will be true if the operand is read-write, i.e., if it is used as |
| an input as well as an output. If *CONSTRAINT_P is not in |
| canonical form, it will be made canonical. (Note that `+' will be |
| replaced with `=' as part of this process.) |
| |
| Returns TRUE if all went well; FALSE if an error occurred. */ |
| |
| bool |
| parse_output_constraint (const char **constraint_p, int operand_num, |
| int ninputs, int noutputs, bool *allows_mem, |
| bool *allows_reg, bool *is_inout) |
| { |
| const char *constraint = *constraint_p; |
| const char *p; |
| |
| /* Assume the constraint doesn't allow the use of either a register |
| or memory. */ |
| *allows_mem = false; |
| *allows_reg = false; |
| |
| /* Allow the `=' or `+' to not be at the beginning of the string, |
| since it wasn't explicitly documented that way, and there is a |
| large body of code that puts it last. Swap the character to |
| the front, so as not to uglify any place else. */ |
| p = strchr (constraint, '='); |
| if (!p) |
| p = strchr (constraint, '+'); |
| |
| /* If the string doesn't contain an `=', issue an error |
| message. */ |
| if (!p) |
| { |
| error ("output operand constraint lacks %<=%>"); |
| return false; |
| } |
| |
| /* If the constraint begins with `+', then the operand is both read |
| from and written to. */ |
| *is_inout = (*p == '+'); |
| |
| /* Canonicalize the output constraint so that it begins with `='. */ |
| if (p != constraint || *is_inout) |
| { |
| char *buf; |
| size_t c_len = strlen (constraint); |
| |
| if (p != constraint) |
| warning (0, "output constraint %qc for operand %d " |
| "is not at the beginning", |
| *p, operand_num); |
| |
| /* Make a copy of the constraint. */ |
| buf = XALLOCAVEC (char, c_len + 1); |
| strcpy (buf, constraint); |
| /* Swap the first character and the `=' or `+'. */ |
| buf[p - constraint] = buf[0]; |
| /* Make sure the first character is an `='. (Until we do this, |
| it might be a `+'.) */ |
| buf[0] = '='; |
| /* Replace the constraint with the canonicalized string. */ |
| *constraint_p = ggc_alloc_string (buf, c_len); |
| constraint = *constraint_p; |
| } |
| |
| /* Loop through the constraint string. */ |
| for (p = constraint + 1; *p; p += CONSTRAINT_LEN (*p, p)) |
| switch (*p) |
| { |
| case '+': |
| case '=': |
| error ("operand constraint contains incorrectly positioned " |
| "%<+%> or %<=%>"); |
| return false; |
| |
| case '%': |
| if (operand_num + 1 == ninputs + noutputs) |
| { |
| error ("%<%%%> constraint used with last operand"); |
| return false; |
| } |
| break; |
| |
| case '?': case '!': case '*': case '&': case '#': |
| case '$': case '^': |
| case 'E': case 'F': case 'G': case 'H': |
| case 's': case 'i': case 'n': |
| case 'I': case 'J': case 'K': case 'L': case 'M': |
| case 'N': case 'O': case 'P': case ',': |
| break; |
| |
| case '0': case '1': case '2': case '3': case '4': |
| case '5': case '6': case '7': case '8': case '9': |
| case '[': |
| error ("matching constraint not valid in output operand"); |
| return false; |
| |
| case '<': case '>': |
| /* ??? Before flow, auto inc/dec insns are not supposed to exist, |
| excepting those that expand_call created. So match memory |
| and hope. */ |
| *allows_mem = true; |
| break; |
| |
| case 'g': case 'X': |
| *allows_reg = true; |
| *allows_mem = true; |
| break; |
| |
| default: |
| if (!ISALPHA (*p)) |
| break; |
| enum constraint_num cn = lookup_constraint (p); |
| if (reg_class_for_constraint (cn) != NO_REGS |
| || insn_extra_address_constraint (cn)) |
| *allows_reg = true; |
| else if (insn_extra_memory_constraint (cn)) |
| *allows_mem = true; |
| else |
| insn_extra_constraint_allows_reg_mem (cn, allows_reg, allows_mem); |
| break; |
| } |
| |
| return true; |
| } |
| |
| /* Similar, but for input constraints. */ |
| |
| bool |
| parse_input_constraint (const char **constraint_p, int input_num, |
| int ninputs, int noutputs, int ninout, |
| const char * const * constraints, |
| bool *allows_mem, bool *allows_reg) |
| { |
| const char *constraint = *constraint_p; |
| const char *orig_constraint = constraint; |
| size_t c_len = strlen (constraint); |
| size_t j; |
| bool saw_match = false; |
| |
| /* Assume the constraint doesn't allow the use of either |
| a register or memory. */ |
| *allows_mem = false; |
| *allows_reg = false; |
| |
| /* Make sure constraint has neither `=', `+', nor '&'. */ |
| |
| for (j = 0; j < c_len; j += CONSTRAINT_LEN (constraint[j], constraint+j)) |
| switch (constraint[j]) |
| { |
| case '+': case '=': case '&': |
| if (constraint == orig_constraint) |
| { |
| error ("input operand constraint contains %qc", constraint[j]); |
| return false; |
| } |
| break; |
| |
| case '%': |
| if (constraint == orig_constraint |
| && input_num + 1 == ninputs - ninout) |
| { |
| error ("%<%%%> constraint used with last operand"); |
| return false; |
| } |
| break; |
| |
| case '<': case '>': |
| case '?': case '!': case '*': case '#': |
| case '$': case '^': |
| case 'E': case 'F': case 'G': case 'H': |
| case 's': case 'i': case 'n': |
| case 'I': case 'J': case 'K': case 'L': case 'M': |
| case 'N': case 'O': case 'P': case ',': |
| break; |
| |
| /* Whether or not a numeric constraint allows a register is |
| decided by the matching constraint, and so there is no need |
| to do anything special with them. We must handle them in |
| the default case, so that we don't unnecessarily force |
| operands to memory. */ |
| case '0': case '1': case '2': case '3': case '4': |
| case '5': case '6': case '7': case '8': case '9': |
| { |
| char *end; |
| unsigned long match; |
| |
| saw_match = true; |
| |
| match = strtoul (constraint + j, &end, 10); |
| if (match >= (unsigned long) noutputs) |
| { |
| error ("matching constraint references invalid operand number"); |
| return false; |
| } |
| |
| /* Try and find the real constraint for this dup. Only do this |
| if the matching constraint is the only alternative. */ |
| if (*end == '\0' |
| && (j == 0 || (j == 1 && constraint[0] == '%'))) |
| { |
| constraint = constraints[match]; |
| *constraint_p = constraint; |
| c_len = strlen (constraint); |
| j = 0; |
| /* ??? At the end of the loop, we will skip the first part of |
| the matched constraint. This assumes not only that the |
| other constraint is an output constraint, but also that |
| the '=' or '+' come first. */ |
| break; |
| } |
| else |
| j = end - constraint; |
| /* Anticipate increment at end of loop. */ |
| j--; |
| } |
| /* Fall through. */ |
| |
| case 'g': case 'X': |
| *allows_reg = true; |
| *allows_mem = true; |
| break; |
| |
| default: |
| if (! ISALPHA (constraint[j])) |
| { |
| error ("invalid punctuation %qc in constraint", constraint[j]); |
| return false; |
| } |
| enum constraint_num cn = lookup_constraint (constraint + j); |
| if (reg_class_for_constraint (cn) != NO_REGS |
| || insn_extra_address_constraint (cn)) |
| *allows_reg = true; |
| else if (insn_extra_memory_constraint (cn)) |
| *allows_mem = true; |
| else |
| insn_extra_constraint_allows_reg_mem (cn, allows_reg, allows_mem); |
| break; |
| } |
| |
| if (saw_match && !*allows_reg) |
| warning (0, "matching constraint does not allow a register"); |
| |
| return true; |
| } |
| |
| /* Return DECL iff there's an overlap between *REGS and DECL, where DECL |
| can be an asm-declared register. Called via walk_tree. */ |
| |
| static tree |
| decl_overlaps_hard_reg_set_p (tree *declp, int *walk_subtrees ATTRIBUTE_UNUSED, |
| void *data) |
| { |
| tree decl = *declp; |
| const HARD_REG_SET *const regs = (const HARD_REG_SET *) data; |
| |
| if (TREE_CODE (decl) == VAR_DECL) |
| { |
| if (DECL_HARD_REGISTER (decl) |
| && REG_P (DECL_RTL (decl)) |
| && REGNO (DECL_RTL (decl)) < FIRST_PSEUDO_REGISTER) |
| { |
| rtx reg = DECL_RTL (decl); |
| |
| if (overlaps_hard_reg_set_p (*regs, GET_MODE (reg), REGNO (reg))) |
| return decl; |
| } |
| walk_subtrees = 0; |
| } |
| else if (TYPE_P (decl) || TREE_CODE (decl) == PARM_DECL) |
| walk_subtrees = 0; |
| return NULL_TREE; |
| } |
| |
| /* If there is an overlap between *REGS and DECL, return the first overlap |
| found. */ |
| tree |
| tree_overlaps_hard_reg_set (tree decl, HARD_REG_SET *regs) |
| { |
| return walk_tree (&decl, decl_overlaps_hard_reg_set_p, regs, NULL); |
| } |
| |
| |
| /* A subroutine of expand_asm_operands. Check that all operand names |
| are unique. Return true if so. We rely on the fact that these names |
| are identifiers, and so have been canonicalized by get_identifier, |
| so all we need are pointer comparisons. */ |
| |
| static bool |
| check_unique_operand_names (tree outputs, tree inputs, tree labels) |
| { |
| tree i, j, i_name = NULL_TREE; |
| |
| for (i = outputs; i ; i = TREE_CHAIN (i)) |
| { |
| i_name = TREE_PURPOSE (TREE_PURPOSE (i)); |
| if (! i_name) |
| continue; |
| |
| for (j = TREE_CHAIN (i); j ; j = TREE_CHAIN (j)) |
| if (simple_cst_equal (i_name, TREE_PURPOSE (TREE_PURPOSE (j)))) |
| goto failure; |
| } |
| |
| for (i = inputs; i ; i = TREE_CHAIN (i)) |
| { |
| i_name = TREE_PURPOSE (TREE_PURPOSE (i)); |
| if (! i_name) |
| continue; |
| |
| for (j = TREE_CHAIN (i); j ; j = TREE_CHAIN (j)) |
| if (simple_cst_equal (i_name, TREE_PURPOSE (TREE_PURPOSE (j)))) |
| goto failure; |
| for (j = outputs; j ; j = TREE_CHAIN (j)) |
| if (simple_cst_equal (i_name, TREE_PURPOSE (TREE_PURPOSE (j)))) |
| goto failure; |
| } |
| |
| for (i = labels; i ; i = TREE_CHAIN (i)) |
| { |
| i_name = TREE_PURPOSE (i); |
| if (! i_name) |
| continue; |
| |
| for (j = TREE_CHAIN (i); j ; j = TREE_CHAIN (j)) |
| if (simple_cst_equal (i_name, TREE_PURPOSE (j))) |
| goto failure; |
| for (j = inputs; j ; j = TREE_CHAIN (j)) |
| if (simple_cst_equal (i_name, TREE_PURPOSE (TREE_PURPOSE (j)))) |
| goto failure; |
| } |
| |
| return true; |
| |
| failure: |
| error ("duplicate asm operand name %qs", TREE_STRING_POINTER (i_name)); |
| return false; |
| } |
| |
| /* A subroutine of expand_asm_operands. Resolve the names of the operands |
| in *POUTPUTS and *PINPUTS to numbers, and replace the name expansions in |
| STRING and in the constraints to those numbers. */ |
| |
| tree |
| resolve_asm_operand_names (tree string, tree outputs, tree inputs, tree labels) |
| { |
| char *buffer; |
| char *p; |
| const char *c; |
| tree t; |
| |
| check_unique_operand_names (outputs, inputs, labels); |
| |
| /* Substitute [<name>] in input constraint strings. There should be no |
| named operands in output constraints. */ |
| for (t = inputs; t ; t = TREE_CHAIN (t)) |
| { |
| c = TREE_STRING_POINTER (TREE_VALUE (TREE_PURPOSE (t))); |
| if (strchr (c, '[') != NULL) |
| { |
| p = buffer = xstrdup (c); |
| while ((p = strchr (p, '[')) != NULL) |
| p = resolve_operand_name_1 (p, outputs, inputs, NULL); |
| TREE_VALUE (TREE_PURPOSE (t)) |
| = build_string (strlen (buffer), buffer); |
| free (buffer); |
| } |
| } |
| |
| /* Now check for any needed substitutions in the template. */ |
| c = TREE_STRING_POINTER (string); |
| while ((c = strchr (c, '%')) != NULL) |
| { |
| if (c[1] == '[') |
| break; |
| else if (ISALPHA (c[1]) && c[2] == '[') |
| break; |
| else |
| { |
| c += 1 + (c[1] == '%'); |
| continue; |
| } |
| } |
| |
| if (c) |
| { |
| /* OK, we need to make a copy so we can perform the substitutions. |
| Assume that we will not need extra space--we get to remove '[' |
| and ']', which means we cannot have a problem until we have more |
| than 999 operands. */ |
| buffer = xstrdup (TREE_STRING_POINTER (string)); |
| p = buffer + (c - TREE_STRING_POINTER (string)); |
| |
| while ((p = strchr (p, '%')) != NULL) |
| { |
| if (p[1] == '[') |
| p += 1; |
| else if (ISALPHA (p[1]) && p[2] == '[') |
| p += 2; |
| else |
| { |
| p += 1 + (p[1] == '%'); |
| continue; |
| } |
| |
| p = resolve_operand_name_1 (p, outputs, inputs, labels); |
| } |
| |
| string = build_string (strlen (buffer), buffer); |
| free (buffer); |
| } |
| |
| return string; |
| } |
| |
| /* A subroutine of resolve_operand_names. P points to the '[' for a |
| potential named operand of the form [<name>]. In place, replace |
| the name and brackets with a number. Return a pointer to the |
| balance of the string after substitution. */ |
| |
| static char * |
| resolve_operand_name_1 (char *p, tree outputs, tree inputs, tree labels) |
| { |
| char *q; |
| int op; |
| tree t; |
| |
| /* Collect the operand name. */ |
| q = strchr (++p, ']'); |
| if (!q) |
| { |
| error ("missing close brace for named operand"); |
| return strchr (p, '\0'); |
| } |
| *q = '\0'; |
| |
| /* Resolve the name to a number. */ |
| for (op = 0, t = outputs; t ; t = TREE_CHAIN (t), op++) |
| { |
| tree name = TREE_PURPOSE (TREE_PURPOSE (t)); |
| if (name && strcmp (TREE_STRING_POINTER (name), p) == 0) |
| goto found; |
| } |
| for (t = inputs; t ; t = TREE_CHAIN (t), op++) |
| { |
| tree name = TREE_PURPOSE (TREE_PURPOSE (t)); |
| if (name && strcmp (TREE_STRING_POINTER (name), p) == 0) |
| goto found; |
| } |
| for (t = labels; t ; t = TREE_CHAIN (t), op++) |
| { |
| tree name = TREE_PURPOSE (t); |
| if (name && strcmp (TREE_STRING_POINTER (name), p) == 0) |
| goto found; |
| } |
| |
| error ("undefined named operand %qs", identifier_to_locale (p)); |
| op = 0; |
| |
| found: |
| /* Replace the name with the number. Unfortunately, not all libraries |
| get the return value of sprintf correct, so search for the end of the |
| generated string by hand. */ |
| sprintf (--p, "%d", op); |
| p = strchr (p, '\0'); |
| |
| /* Verify the no extra buffer space assumption. */ |
| gcc_assert (p <= q); |
| |
| /* Shift the rest of the buffer down to fill the gap. */ |
| memmove (p, q + 1, strlen (q + 1) + 1); |
| |
| return p; |
| } |
| |
| |
| /* Generate RTL to return directly from the current function. |
| (That is, we bypass any return value.) */ |
| |
| void |
| expand_naked_return (void) |
| { |
| rtx end_label; |
| |
| clear_pending_stack_adjust (); |
| do_pending_stack_adjust (); |
| |
| end_label = naked_return_label; |
| if (end_label == 0) |
| end_label = naked_return_label = gen_label_rtx (); |
| |
| emit_jump (end_label); |
| } |
| |
| /* Generate code to jump to LABEL if OP0 and OP1 are equal in mode MODE. PROB |
| is the probability of jumping to LABEL. */ |
| static void |
| do_jump_if_equal (machine_mode mode, rtx op0, rtx op1, rtx label, |
| int unsignedp, int prob) |
| { |
| gcc_assert (prob <= REG_BR_PROB_BASE); |
| do_compare_rtx_and_jump (op0, op1, EQ, unsignedp, mode, |
| NULL_RTX, NULL_RTX, label, prob); |
| } |
| |
| /* Do the insertion of a case label into case_list. The labels are |
| fed to us in descending order from the sorted vector of case labels used |
| in the tree part of the middle end. So the list we construct is |
| sorted in ascending order. |
| |
| LABEL is the case label to be inserted. LOW and HIGH are the bounds |
| against which the index is compared to jump to LABEL and PROB is the |
| estimated probability LABEL is reached from the switch statement. */ |
| |
| static struct case_node * |
| add_case_node (struct case_node *head, tree low, tree high, |
| tree label, int prob, alloc_pool case_node_pool) |
| { |
| struct case_node *r; |
| |
| gcc_checking_assert (low); |
| gcc_checking_assert (high && (TREE_TYPE (low) == TREE_TYPE (high))); |
| |
| /* Add this label to the chain. */ |
| r = (struct case_node *) pool_alloc (case_node_pool); |
| r->low = low; |
| r->high = high; |
| r->code_label = label; |
| r->parent = r->left = NULL; |
| r->prob = prob; |
| r->subtree_prob = prob; |
| r->right = head; |
| return r; |
| } |
| |
| /* Dump ROOT, a list or tree of case nodes, to file. */ |
| |
| static void |
| dump_case_nodes (FILE *f, struct case_node *root, |
| int indent_step, int indent_level) |
| { |
| if (root == 0) |
| return; |
| indent_level++; |
| |
| dump_case_nodes (f, root->left, indent_step, indent_level); |
| |
| fputs (";; ", f); |
| fprintf (f, "%*s", indent_step * indent_level, ""); |
| print_dec (root->low, f, TYPE_SIGN (TREE_TYPE (root->low))); |
| if (!tree_int_cst_equal (root->low, root->high)) |
| { |
| fprintf (f, " ... "); |
| print_dec (root->high, f, TYPE_SIGN (TREE_TYPE (root->high))); |
| } |
| fputs ("\n", f); |
| |
| dump_case_nodes (f, root->right, indent_step, indent_level); |
| } |
| |
| #ifndef HAVE_casesi |
| #define HAVE_casesi 0 |
| #endif |
| |
| #ifndef HAVE_tablejump |
| #define HAVE_tablejump 0 |
| #endif |
| |
| /* Return the smallest number of different values for which it is best to use a |
| jump-table instead of a tree of conditional branches. */ |
| |
| static unsigned int |
| case_values_threshold (void) |
| { |
| unsigned int threshold = PARAM_VALUE (PARAM_CASE_VALUES_THRESHOLD); |
| |
| if (threshold == 0) |
| threshold = targetm.case_values_threshold (); |
| |
| return threshold; |
| } |
| |
| /* Return true if a switch should be expanded as a decision tree. |
| RANGE is the difference between highest and lowest case. |
| UNIQ is number of unique case node targets, not counting the default case. |
| COUNT is the number of comparisons needed, not counting the default case. */ |
| |
| static bool |
| expand_switch_as_decision_tree_p (tree range, |
| unsigned int uniq ATTRIBUTE_UNUSED, |
| unsigned int count) |
| { |
| int max_ratio; |
| |
| /* If neither casesi or tablejump is available, or flag_jump_tables |
| over-ruled us, we really have no choice. */ |
| if (!HAVE_casesi && !HAVE_tablejump) |
| return true; |
| if (!flag_jump_tables) |
| return true; |
| #ifndef ASM_OUTPUT_ADDR_DIFF_ELT |
| if (flag_pic) |
| return true; |
| #endif |
| |
| /* If the switch is relatively small such that the cost of one |
| indirect jump on the target are higher than the cost of a |
| decision tree, go with the decision tree. |
| |
| If range of values is much bigger than number of values, |
| or if it is too large to represent in a HOST_WIDE_INT, |
| make a sequence of conditional branches instead of a dispatch. |
| |
| The definition of "much bigger" depends on whether we are |
| optimizing for size or for speed. If the former, the maximum |
| ratio range/count = 3, because this was found to be the optimal |
| ratio for size on i686-pc-linux-gnu, see PR11823. The ratio |
| 10 is much older, and was probably selected after an extensive |
| benchmarking investigation on numerous platforms. Or maybe it |
| just made sense to someone at some point in the history of GCC, |
| who knows... */ |
| max_ratio = optimize_insn_for_size_p () ? 3 : 10; |
| if (count < case_values_threshold () |
| || ! tree_fits_uhwi_p (range) |
| || compare_tree_int (range, max_ratio * count) > 0) |
| return true; |
| |
| return false; |
| } |
| |
| /* Generate a decision tree, switching on INDEX_EXPR and jumping to |
| one of the labels in CASE_LIST or to the DEFAULT_LABEL. |
| DEFAULT_PROB is the estimated probability that it jumps to |
| DEFAULT_LABEL. |
| |
| We generate a binary decision tree to select the appropriate target |
| code. This is done as follows: |
| |
| If the index is a short or char that we do not have |
| an insn to handle comparisons directly, convert it to |
| a full integer now, rather than letting each comparison |
| generate the conversion. |
| |
| Load the index into a register. |
| |
| The list of cases is rearranged into a binary tree, |
| nearly optimal assuming equal probability for each case. |
| |
| The tree is transformed into RTL, eliminating redundant |
| test conditions at the same time. |
| |
| If program flow could reach the end of the decision tree |
| an unconditional jump to the default code is emitted. |
| |
| The above process is unaware of the CFG. The caller has to fix up |
| the CFG itself. This is done in cfgexpand.c. */ |
| |
| static void |
| emit_case_decision_tree (tree index_expr, tree index_type, |
| struct case_node *case_list, rtx default_label, |
| int default_prob) |
| { |
| rtx index = expand_normal (index_expr); |
| |
| if (GET_MODE_CLASS (GET_MODE (index)) == MODE_INT |
| && ! have_insn_for (COMPARE, GET_MODE (index))) |
| { |
| int unsignedp = TYPE_UNSIGNED (index_type); |
| machine_mode wider_mode; |
| for (wider_mode = GET_MODE (index); wider_mode != VOIDmode; |
| wider_mode = GET_MODE_WIDER_MODE (wider_mode)) |
| if (have_insn_for (COMPARE, wider_mode)) |
| { |
| index = convert_to_mode (wider_mode, index, unsignedp); |
| break; |
| } |
| } |
| |
| do_pending_stack_adjust (); |
| |
| if (MEM_P (index)) |
| { |
| index = copy_to_reg (index); |
| if (TREE_CODE (index_expr) == SSA_NAME) |
| set_reg_attrs_for_decl_rtl (SSA_NAME_VAR (index_expr), index); |
| } |
| |
| balance_case_nodes (&case_list, NULL); |
| |
| if (dump_file && (dump_flags & TDF_DETAILS)) |
| { |
| int indent_step = ceil_log2 (TYPE_PRECISION (index_type)) + 2; |
| fprintf (dump_file, ";; Expanding GIMPLE switch as decision tree:\n"); |
| dump_case_nodes (dump_file, case_list, indent_step, 0); |
| } |
| |
| emit_case_nodes (index, case_list, default_label, default_prob, index_type); |
| if (default_label) |
| emit_jump (default_label); |
| } |
| |
| /* Return the sum of probabilities of outgoing edges of basic block BB. */ |
| |
| static int |
| get_outgoing_edge_probs (basic_block bb) |
| { |
| edge e; |
| edge_iterator ei; |
| int prob_sum = 0; |
| if (!bb) |
| return 0; |
| FOR_EACH_EDGE (e, ei, bb->succs) |
| prob_sum += e->probability; |
| return prob_sum; |
| } |
| |
| /* Computes the conditional probability of jumping to a target if the branch |
| instruction is executed. |
| TARGET_PROB is the estimated probability of jumping to a target relative |
| to some basic block BB. |
| BASE_PROB is the probability of reaching the branch instruction relative |
| to the same basic block BB. */ |
| |
| static inline int |
| conditional_probability (int target_prob, int base_prob) |
| { |
| if (base_prob > 0) |
| { |
| gcc_assert (target_prob >= 0); |
| gcc_assert (target_prob <= base_prob); |
| return GCOV_COMPUTE_SCALE (target_prob, base_prob); |
| } |
| return -1; |
| } |
| |
| /* Generate a dispatch tabler, switching on INDEX_EXPR and jumping to |
| one of the labels in CASE_LIST or to the DEFAULT_LABEL. |
| MINVAL, MAXVAL, and RANGE are the extrema and range of the case |
| labels in CASE_LIST. STMT_BB is the basic block containing the statement. |
| |
| First, a jump insn is emitted. First we try "casesi". If that |
| fails, try "tablejump". A target *must* have one of them (or both). |
| |
| Then, a table with the target labels is emitted. |
| |
| The process is unaware of the CFG. The caller has to fix up |
| the CFG itself. This is done in cfgexpand.c. */ |
| |
| static void |
| emit_case_dispatch_table (tree index_expr, tree index_type, |
| struct case_node *case_list, rtx default_label, |
| tree minval, tree maxval, tree range, |
| basic_block stmt_bb) |
| { |
| int i, ncases; |
| struct case_node *n; |
| rtx *labelvec; |
| rtx fallback_label = label_rtx (case_list->code_label); |
| rtx_code_label *table_label = gen_label_rtx (); |
| bool has_gaps = false; |
| edge default_edge = stmt_bb ? EDGE_SUCC (stmt_bb, 0) : NULL; |
| int default_prob = default_edge ? default_edge->probability : 0; |
| int base = get_outgoing_edge_probs (stmt_bb); |
| bool try_with_tablejump = false; |
| |
| int new_default_prob = conditional_probability (default_prob, |
| base); |
| |
| if (! try_casesi (index_type, index_expr, minval, range, |
| table_label, default_label, fallback_label, |
| new_default_prob)) |
| { |
| /* Index jumptables from zero for suitable values of minval to avoid |
| a subtraction. For the rationale see: |
| "http://gcc.gnu.org/ml/gcc-patches/2001-10/msg01234.html". */ |
| if (optimize_insn_for_speed_p () |
| && compare_tree_int (minval, 0) > 0 |
| && compare_tree_int (minval, 3) < 0) |
| { |
| minval = build_int_cst (index_type, 0); |
| range = maxval; |
| has_gaps = true; |
| } |
| try_with_tablejump = true; |
| } |
| |
| /* Get table of labels to jump to, in order of case index. */ |
| |
| ncases = tree_to_shwi (range) + 1; |
| labelvec = XALLOCAVEC (rtx, ncases); |
| memset (labelvec, 0, ncases * sizeof (rtx)); |
| |
| for (n = case_list; n; n = n->right) |
| { |
| /* Compute the low and high bounds relative to the minimum |
| value since that should fit in a HOST_WIDE_INT while the |
| actual values may not. */ |
| HOST_WIDE_INT i_low |
| = tree_to_uhwi (fold_build2 (MINUS_EXPR, index_type, |
| n->low, minval)); |
| HOST_WIDE_INT i_high |
| = tree_to_uhwi (fold_build2 (MINUS_EXPR, index_type, |
| n->high, minval)); |
| HOST_WIDE_INT i; |
| |
| for (i = i_low; i <= i_high; i ++) |
| labelvec[i] |
| = gen_rtx_LABEL_REF (Pmode, label_rtx (n->code_label)); |
| } |
| |
| /* Fill in the gaps with the default. We may have gaps at |
| the beginning if we tried to avoid the minval subtraction, |
| so substitute some label even if the default label was |
| deemed unreachable. */ |
| if (!default_label) |
| default_label = fallback_label; |
| for (i = 0; i < ncases; i++) |
| if (labelvec[i] == 0) |
| { |
| has_gaps = true; |
| labelvec[i] = gen_rtx_LABEL_REF (Pmode, default_label); |
| } |
| |
| if (has_gaps) |
| { |
| /* There is at least one entry in the jump table that jumps |
| to default label. The default label can either be reached |
| through the indirect jump or the direct conditional jump |
| before that. Split the probability of reaching the |
| default label among these two jumps. */ |
| new_default_prob = conditional_probability (default_prob/2, |
| base); |
| default_prob /= 2; |
| base -= default_prob; |
| } |
| else |
| { |
| base -= default_prob; |
| default_prob = 0; |
| } |
| |
| if (default_edge) |
| default_edge->probability = default_prob; |
| |
| /* We have altered the probability of the default edge. So the probabilities |
| of all other edges need to be adjusted so that it sums up to |
| REG_BR_PROB_BASE. */ |
| if (base) |
| { |
| edge e; |
| edge_iterator ei; |
| FOR_EACH_EDGE (e, ei, stmt_bb->succs) |
| e->probability = GCOV_COMPUTE_SCALE (e->probability, base); |
| } |
| |
| if (try_with_tablejump) |
| { |
| bool ok = try_tablejump (index_type, index_expr, minval, range, |
| table_label, default_label, new_default_prob); |
| gcc_assert (ok); |
| } |
| /* Output the table. */ |
| emit_label (table_label); |
| |
| if (CASE_VECTOR_PC_RELATIVE || flag_pic) |
| emit_jump_table_data (gen_rtx_ADDR_DIFF_VEC (CASE_VECTOR_MODE, |
| gen_rtx_LABEL_REF (Pmode, |
| table_label), |
| gen_rtvec_v (ncases, labelvec), |
| const0_rtx, const0_rtx)); |
| else |
| emit_jump_table_data (gen_rtx_ADDR_VEC (CASE_VECTOR_MODE, |
| gen_rtvec_v (ncases, labelvec))); |
| |
| /* Record no drop-through after the table. */ |
| emit_barrier (); |
| } |
| |
| /* Reset the aux field of all outgoing edges of basic block BB. */ |
| |
| static inline void |
| reset_out_edges_aux (basic_block bb) |
| { |
| edge e; |
| edge_iterator ei; |
| FOR_EACH_EDGE (e, ei, bb->succs) |
| e->aux = (void *)0; |
| } |
| |
| /* Compute the number of case labels that correspond to each outgoing edge of |
| STMT. Record this information in the aux field of the edge. */ |
| |
| static inline void |
| compute_cases_per_edge (gswitch *stmt) |
| { |
| basic_block bb = gimple_bb (stmt); |
| reset_out_edges_aux (bb); |
| int ncases = gimple_switch_num_labels (stmt); |
| for (int i = ncases - 1; i >= 1; --i) |
| { |
| tree elt = gimple_switch_label (stmt, i); |
| tree lab = CASE_LABEL (elt); |
| basic_block case_bb = label_to_block_fn (cfun, lab); |
| edge case_edge = find_edge (bb, case_bb); |
| case_edge->aux = (void *)((intptr_t)(case_edge->aux) + 1); |
| } |
| } |
| |
| /* Terminate a case (Pascal/Ada) or switch (C) statement |
| in which ORIG_INDEX is the expression to be tested. |
| If ORIG_TYPE is not NULL, it is the original ORIG_INDEX |
| type as given in the source before any compiler conversions. |
| Generate the code to test it and jump to the right place. */ |
| |
| void |
| expand_case (gswitch *stmt) |
| { |
| tree minval = NULL_TREE, maxval = NULL_TREE, range = NULL_TREE; |
| rtx default_label = NULL_RTX; |
| unsigned int count, uniq; |
| int i; |
| int ncases = gimple_switch_num_labels (stmt); |
| tree index_expr = gimple_switch_index (stmt); |
| tree index_type = TREE_TYPE (index_expr); |
| tree elt; |
| basic_block bb = gimple_bb (stmt); |
| |
| /* A list of case labels; it is first built as a list and it may then |
| be rearranged into a nearly balanced binary tree. */ |
| struct case_node *case_list = 0; |
| |
| /* A pool for case nodes. */ |
| alloc_pool case_node_pool; |
| |
| /* An ERROR_MARK occurs for various reasons including invalid data type. |
| ??? Can this still happen, with GIMPLE and all? */ |
| if (index_type == error_mark_node) |
| return; |
| |
| /* cleanup_tree_cfg removes all SWITCH_EXPR with their index |
| expressions being INTEGER_CST. */ |
| gcc_assert (TREE_CODE (index_expr) != INTEGER_CST); |
| |
| case_node_pool = create_alloc_pool ("struct case_node pool", |
| sizeof (struct case_node), |
| 100); |
| |
| do_pending_stack_adjust (); |
| |
| /* Find the default case target label. */ |
| default_label = label_rtx (CASE_LABEL (gimple_switch_default_label (stmt))); |
| edge default_edge = EDGE_SUCC (bb, 0); |
| int default_prob = default_edge->probability; |
| |
| /* Get upper and lower bounds of case values. */ |
| elt = gimple_switch_label (stmt, 1); |
| minval = fold_convert (index_type, CASE_LOW (elt)); |
| elt = gimple_switch_label (stmt, ncases - 1); |
| if (CASE_HIGH (elt)) |
| maxval = fold_convert (index_type, CASE_HIGH (elt)); |
| else |
| maxval = fold_convert (index_type, CASE_LOW (elt)); |
| |
| /* Compute span of values. */ |
| range = fold_build2 (MINUS_EXPR, index_type, maxval, minval); |
| |
| /* Listify the labels queue and gather some numbers to decide |
| how to expand this switch(). */ |
| uniq = 0; |
| count = 0; |
| hash_set<tree> seen_labels; |
| compute_cases_per_edge (stmt); |
| |
| for (i = ncases - 1; i >= 1; --i) |
| { |
| elt = gimple_switch_label (stmt, i); |
| tree low = CASE_LOW (elt); |
| gcc_assert (low); |
| tree high = CASE_HIGH (elt); |
| gcc_assert (! high || tree_int_cst_lt (low, high)); |
| tree lab = CASE_LABEL (elt); |
| |
| /* Count the elements. |
| A range counts double, since it requires two compares. */ |
| count++; |
| if (high) |
| count++; |
| |
| /* If we have not seen this label yet, then increase the |
| number of unique case node targets seen. */ |
| if (!seen_labels.add (lab)) |
| uniq++; |
| |
| /* The bounds on the case range, LOW and HIGH, have to be converted |
| to case's index type TYPE. Note that the original type of the |
| case index in the source code is usually "lost" during |
| gimplification due to type promotion, but the case labels retain the |
| original type. Make sure to drop overflow flags. */ |
| low = fold_convert (index_type, low); |
| if (TREE_OVERFLOW (low)) |
| low = wide_int_to_tree (index_type, low); |
| |
| /* The canonical from of a case label in GIMPLE is that a simple case |
| has an empty CASE_HIGH. For the casesi and tablejump expanders, |
| the back ends want simple cases to have high == low. */ |
| if (! high) |
| high = low; |
| high = fold_convert (index_type, high); |
| if (TREE_OVERFLOW (high)) |
| high = wide_int_to_tree (index_type, high); |
| |
| basic_block case_bb = label_to_block_fn (cfun, lab); |
| edge case_edge = find_edge (bb, case_bb); |
| case_list = add_case_node ( |
| case_list, low, high, lab, |
| case_edge->probability / (intptr_t)(case_edge->aux), |
| case_node_pool); |
| } |
| reset_out_edges_aux (bb); |
| |
| /* cleanup_tree_cfg removes all SWITCH_EXPR with a single |
| destination, such as one with a default case only. |
| It also removes cases that are out of range for the switch |
| type, so we should never get a zero here. */ |
| gcc_assert (count > 0); |
| |
| rtx_insn *before_case = get_last_insn (); |
| |
| /* Decide how to expand this switch. |
| The two options at this point are a dispatch table (casesi or |
| tablejump) or a decision tree. */ |
| |
| if (expand_switch_as_decision_tree_p (range, uniq, count)) |
| emit_case_decision_tree (index_expr, index_type, |
| case_list, default_label, |
| default_prob); |
| else |
| emit_case_dispatch_table (index_expr, index_type, |
| case_list, default_label, |
| minval, maxval, range, bb); |
| |
| reorder_insns (NEXT_INSN (before_case), get_last_insn (), before_case); |
| |
| free_temp_slots (); |
| free_alloc_pool (case_node_pool); |
| } |
| |
| /* Expand the dispatch to a short decrement chain if there are few cases |
| to dispatch to. Likewise if neither casesi nor tablejump is available, |
| or if flag_jump_tables is set. Otherwise, expand as a casesi or a |
| tablejump. The index mode is always the mode of integer_type_node. |
| Trap if no case matches the index. |
| |
| DISPATCH_INDEX is the index expression to switch on. It should be a |
| memory or register operand. |
| |
| DISPATCH_TABLE is a set of case labels. The set should be sorted in |
| ascending order, be contiguous, starting with value 0, and contain only |
| single-valued case labels. */ |
| |
| void |
| expand_sjlj_dispatch_table (rtx dispatch_index, |
| vec<tree> dispatch_table) |
| { |
| tree index_type = integer_type_node; |
| machine_mode index_mode = TYPE_MODE (index_type); |
| |
| int ncases = dispatch_table.length (); |
| |
| do_pending_stack_adjust (); |
| rtx_insn *before_case = get_last_insn (); |
| |
| /* Expand as a decrement-chain if there are 5 or fewer dispatch |
| labels. This covers more than 98% of the cases in libjava, |
| and seems to be a reasonable compromise between the "old way" |
| of expanding as a decision tree or dispatch table vs. the "new |
| way" with decrement chain or dispatch table. */ |
| if (dispatch_table.length () <= 5 |
| || (!HAVE_casesi && !HAVE_tablejump) |
| || !flag_jump_tables) |
| { |
| /* Expand the dispatch as a decrement chain: |
| |
| "switch(index) {case 0: do_0; case 1: do_1; ...; case N: do_N;}" |
| |
| ==> |
| |
| if (index == 0) do_0; else index--; |
| if (index == 0) do_1; else index--; |
| ... |
| if (index == 0) do_N; else index--; |
| |
| This is more efficient than a dispatch table on most machines. |
| The last "index--" is redundant but the code is trivially dead |
| and will be cleaned up by later passes. */ |
| rtx index = copy_to_mode_reg (index_mode, dispatch_index); |
| rtx zero = CONST0_RTX (index_mode); |
| for (int i = 0; i < ncases; i++) |
| { |
| tree elt = dispatch_table[i]; |
| rtx lab = label_rtx (CASE_LABEL (elt)); |
| do_jump_if_equal (index_mode, index, zero, lab, 0, -1); |
| force_expand_binop (index_mode, sub_optab, |
| index, CONST1_RTX (index_mode), |
| index, 0, OPTAB_DIRECT); |
| } |
| } |
| else |
| { |
| /* Similar to expand_case, but much simpler. */ |
| struct case_node *case_list = 0; |
| alloc_pool case_node_pool = create_alloc_pool ("struct sjlj_case pool", |
| sizeof (struct case_node), |
| ncases); |
| tree index_expr = make_tree (index_type, dispatch_index); |
| tree minval = build_int_cst (index_type, 0); |
| tree maxval = CASE_LOW (dispatch_table.last ()); |
| tree range = maxval; |
| rtx_code_label *default_label = gen_label_rtx (); |
| |
| for (int i = ncases - 1; i >= 0; --i) |
| { |
| tree elt = dispatch_table[i]; |
| tree low = CASE_LOW (elt); |
| tree lab = CASE_LABEL (elt); |
| case_list = add_case_node (case_list, low, low, lab, 0, case_node_pool); |
| } |
| |
| emit_case_dispatch_table (index_expr, index_type, |
| case_list, default_label, |
| minval, maxval, range, |
| BLOCK_FOR_INSN (before_case)); |
| emit_label (default_label); |
| free_alloc_pool (case_node_pool); |
| } |
| |
| /* Dispatching something not handled? Trap! */ |
| expand_builtin_trap (); |
| |
| reorder_insns (NEXT_INSN (before_case), get_last_insn (), before_case); |
| |
| free_temp_slots (); |
| } |
| |
| |
| /* Take an ordered list of case nodes |
| and transform them into a near optimal binary tree, |
| on the assumption that any target code selection value is as |
| likely as any other. |
| |
| The transformation is performed by splitting the ordered |
| list into two equal sections plus a pivot. The parts are |
| then attached to the pivot as left and right branches. Each |
| branch is then transformed recursively. */ |
| |
| static void |
| balance_case_nodes (case_node_ptr *head, case_node_ptr parent) |
| { |
| case_node_ptr np; |
| |
| np = *head; |
| if (np) |
| { |
| int i = 0; |
| int ranges = 0; |
| case_node_ptr *npp; |
| case_node_ptr left; |
| |
| /* Count the number of entries on branch. Also count the ranges. */ |
| |
| while (np) |
| { |
| if (!tree_int_cst_equal (np->low, np->high)) |
| ranges++; |
| |
| i++; |
| np = np->right; |
| } |
| |
| if (i > 2) |
| { |
| /* Split this list if it is long enough for that to help. */ |
| npp = head; |
| left = *npp; |
| |
| /* If there are just three nodes, split at the middle one. */ |
| if (i == 3) |
| npp = &(*npp)->right; |
| else |
| { |
| /* Find the place in the list that bisects the list's total cost, |
| where ranges count as 2. |
| Here I gets half the total cost. */ |
| i = (i + ranges + 1) / 2; |
| while (1) |
| { |
| /* Skip nodes while their cost does not reach that amount. */ |
| if (!tree_int_cst_equal ((*npp)->low, (*npp)->high)) |
| i--; |
| i--; |
| if (i <= 0) |
| break; |
| npp = &(*npp)->right; |
| } |
| } |
| *head = np = *npp; |
| *npp = 0; |
| np->parent = parent; |
| np->left = left; |
| |
| /* Optimize each of the two split parts. */ |
| balance_case_nodes (&np->left, np); |
| balance_case_nodes (&np->right, np); |
| np->subtree_prob = np->prob; |
| np->subtree_prob += np->left->subtree_prob; |
| np->subtree_prob += np->right->subtree_prob; |
| } |
| else |
| { |
| /* Else leave this branch as one level, |
| but fill in `parent' fields. */ |
| np = *head; |
| np->parent = parent; |
| np->subtree_prob = np->prob; |
| for (; np->right; np = np->right) |
| { |
| np->right->parent = np; |
| (*head)->subtree_prob += np->right->subtree_prob; |
| } |
| } |
| } |
| } |
| |
| /* Search the parent sections of the case node tree |
| to see if a test for the lower bound of NODE would be redundant. |
| INDEX_TYPE is the type of the index expression. |
| |
| The instructions to generate the case decision tree are |
| output in the same order as nodes are processed so it is |
| known that if a parent node checks the range of the current |
| node minus one that the current node is bounded at its lower |
| span. Thus the test would be redundant. */ |
| |
| static int |
| node_has_low_bound (case_node_ptr node, tree index_type) |
| { |
| tree low_minus_one; |
| case_node_ptr pnode; |
| |
| /* If the lower bound of this node is the lowest value in the index type, |
| we need not test it. */ |
| |
| if (tree_int_cst_equal (node->low, TYPE_MIN_VALUE (index_type))) |
| return 1; |
| |
| /* If this node has a left branch, the value at the left must be less |
| than that at this node, so it cannot be bounded at the bottom and |
| we need not bother testing any further. */ |
| |
| if (node->left) |
| return 0; |
| |
| low_minus_one = fold_build2 (MINUS_EXPR, TREE_TYPE (node->low), |
| node->low, |
| build_int_cst (TREE_TYPE (node->low), 1)); |
| |
| /* If the subtraction above overflowed, we can't verify anything. |
| Otherwise, look for a parent that tests our value - 1. */ |
| |
| if (! tree_int_cst_lt (low_minus_one, node->low)) |
| return 0; |
| |
| for (pnode = node->parent; pnode; pnode = pnode->parent) |
| if (tree_int_cst_equal (low_minus_one, pnode->high)) |
| return 1; |
| |
| return 0; |
| } |
| |
| /* Search the parent sections of the case node tree |
| to see if a test for the upper bound of NODE would be redundant. |
| INDEX_TYPE is the type of the index expression. |
| |
| The instructions to generate the case decision tree are |
| output in the same order as nodes are processed so it is |
| known that if a parent node checks the range of the current |
| node plus one that the current node is bounded at its upper |
| span. Thus the test would be redundant. */ |
| |
| static int |
| node_has_high_bound (case_node_ptr node, tree index_type) |
| { |
| tree high_plus_one; |
| case_node_ptr pnode; |
| |
| /* If there is no upper bound, obviously no test is needed. */ |
| |
| if (TYPE_MAX_VALUE (index_type) == NULL) |
| return 1; |
| |
| /* If the upper bound of this node is the highest value in the type |
| of the index expression, we need not test against it. */ |
| |
| if (tree_int_cst_equal (node->high, TYPE_MAX_VALUE (index_type))) |
| return 1; |
| |
| /* If this node has a right branch, the value at the right must be greater |
| than that at this node, so it cannot be bounded at the top and |
| we need not bother testing any further. */ |
| |
| if (node->right) |
| return 0; |
| |
| high_plus_one = fold_build2 (PLUS_EXPR, TREE_TYPE (node->high), |
| node->high, |
| build_int_cst (TREE_TYPE (node->high), 1)); |
| |
| /* If the addition above overflowed, we can't verify anything. |
| Otherwise, look for a parent that tests our value + 1. */ |
| |
| if (! tree_int_cst_lt (node->high, high_plus_one)) |
| return 0; |
| |
| for (pnode = node->parent; pnode; pnode = pnode->parent) |
| if (tree_int_cst_equal (high_plus_one, pnode->low)) |
| return 1; |
| |
| return 0; |
| } |
| |
| /* Search the parent sections of the |
| case node tree to see if both tests for the upper and lower |
| bounds of NODE would be redundant. */ |
| |
| static int |
| node_is_bounded (case_node_ptr node, tree index_type) |
| { |
| return (node_has_low_bound (node, index_type) |
| && node_has_high_bound (node, index_type)); |
| } |
| |
| |
| /* Emit step-by-step code to select a case for the value of INDEX. |
| The thus generated decision tree follows the form of the |
| case-node binary tree NODE, whose nodes represent test conditions. |
| INDEX_TYPE is the type of the index of the switch. |
| |
| Care is taken to prune redundant tests from the decision tree |
| by detecting any boundary conditions already checked by |
| emitted rtx. (See node_has_high_bound, node_has_low_bound |
| and node_is_bounded, above.) |
| |
| Where the test conditions can be shown to be redundant we emit |
| an unconditional jump to the target code. As a further |
| optimization, the subordinates of a tree node are examined to |
| check for bounded nodes. In this case conditional and/or |
| unconditional jumps as a result of the boundary check for the |
| current node are arranged to target the subordinates associated |
| code for out of bound conditions on the current node. |
| |
| We can assume that when control reaches the code generated here, |
| the index value has already been compared with the parents |
| of this node, and determined to be on the same side of each parent |
| as this node is. Thus, if this node tests for the value 51, |
| and a parent tested for 52, we don't need to consider |
| the possibility of a value greater than 51. If another parent |
| tests for the value 50, then this node need not test anything. */ |
| |
| static void |
| emit_case_nodes (rtx index, case_node_ptr node, rtx default_label, |
| int default_prob, tree index_type) |
| { |
| /* If INDEX has an unsigned type, we must make unsigned branches. */ |
| int unsignedp = TYPE_UNSIGNED (index_type); |
| int probability; |
| int prob = node->prob, subtree_prob = node->subtree_prob; |
| machine_mode mode = GET_MODE (index); |
| machine_mode imode = TYPE_MODE (index_type); |
| |
| /* Handle indices detected as constant during RTL expansion. */ |
| if (mode == VOIDmode) |
| mode = imode; |
| |
| /* See if our parents have already tested everything for us. |
| If they have, emit an unconditional jump for this node. */ |
| if (node_is_bounded (node, index_type)) |
| emit_jump (label_rtx (node->code_label)); |
| |
| else if (tree_int_cst_equal (node->low, node->high)) |
| { |
| probability = conditional_probability (prob, subtree_prob + default_prob); |
| /* Node is single valued. First see if the index expression matches |
| this node and then check our children, if any. */ |
| do_jump_if_equal (mode, index, |
| convert_modes (mode, imode, |
| expand_normal (node->low), |
| unsignedp), |
| label_rtx (node->code_label), unsignedp, probability); |
| /* Since this case is taken at this point, reduce its weight from |
| subtree_weight. */ |
| subtree_prob -= prob; |
| if (node->right != 0 && node->left != 0) |
| { |
| /* This node has children on both sides. |
| Dispatch to one side or the other |
| by comparing the index value with this node's value. |
| If one subtree is bounded, check that one first, |
| so we can avoid real branches in the tree. */ |
| |
| if (node_is_bounded (node->right, index_type)) |
| { |
| probability = conditional_probability ( |
| node->right->prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| GT, NULL_RTX, mode, unsignedp, |
| label_rtx (node->right->code_label), |
| probability); |
| emit_case_nodes (index, node->left, default_label, default_prob, |
| index_type); |
| } |
| |
| else if (node_is_bounded (node->left, index_type)) |
| { |
| probability = conditional_probability ( |
| node->left->prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| LT, NULL_RTX, mode, unsignedp, |
| label_rtx (node->left->code_label), |
| probability); |
| emit_case_nodes (index, node->right, default_label, default_prob, index_type); |
| } |
| |
| /* If both children are single-valued cases with no |
| children, finish up all the work. This way, we can save |
| one ordered comparison. */ |
| else if (tree_int_cst_equal (node->right->low, node->right->high) |
| && node->right->left == 0 |
| && node->right->right == 0 |
| && tree_int_cst_equal (node->left->low, node->left->high) |
| && node->left->left == 0 |
| && node->left->right == 0) |
| { |
| /* Neither node is bounded. First distinguish the two sides; |
| then emit the code for one side at a time. */ |
| |
| /* See if the value matches what the right hand side |
| wants. */ |
| probability = conditional_probability ( |
| node->right->prob, |
| subtree_prob + default_prob); |
| do_jump_if_equal (mode, index, |
| convert_modes (mode, imode, |
| expand_normal (node->right->low), |
| unsignedp), |
| label_rtx (node->right->code_label), |
| unsignedp, probability); |
| |
| /* See if the value matches what the left hand side |
| wants. */ |
| probability = conditional_probability ( |
| node->left->prob, |
| subtree_prob + default_prob); |
| do_jump_if_equal (mode, index, |
| convert_modes (mode, imode, |
| expand_normal (node->left->low), |
| unsignedp), |
| label_rtx (node->left->code_label), |
| unsignedp, probability); |
| } |
| |
| else |
| { |
| /* Neither node is bounded. First distinguish the two sides; |
| then emit the code for one side at a time. */ |
| |
| tree test_label |
| = build_decl (curr_insn_location (), |
| LABEL_DECL, NULL_TREE, void_type_node); |
| |
| /* The default label could be reached either through the right |
| subtree or the left subtree. Divide the probability |
| equally. */ |
| probability = conditional_probability ( |
| node->right->subtree_prob + default_prob/2, |
| subtree_prob + default_prob); |
| /* See if the value is on the right. */ |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| GT, NULL_RTX, mode, unsignedp, |
| label_rtx (test_label), |
| probability); |
| default_prob /= 2; |
| |
| /* Value must be on the left. |
| Handle the left-hand subtree. */ |
| emit_case_nodes (index, node->left, default_label, default_prob, index_type); |
| /* If left-hand subtree does nothing, |
| go to default. */ |
| if (default_label) |
| emit_jump (default_label); |
| |
| /* Code branches here for the right-hand subtree. */ |
| expand_label (test_label); |
| emit_case_nodes (index, node->right, default_label, default_prob, index_type); |
| } |
| } |
| |
| else if (node->right != 0 && node->left == 0) |
| { |
| /* Here we have a right child but no left so we issue a conditional |
| branch to default and process the right child. |
| |
| Omit the conditional branch to default if the right child |
| does not have any children and is single valued; it would |
| cost too much space to save so little time. */ |
| |
| if (node->right->right || node->right->left |
| || !tree_int_cst_equal (node->right->low, node->right->high)) |
| { |
| if (!node_has_low_bound (node, index_type)) |
| { |
| probability = conditional_probability ( |
| default_prob/2, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| LT, NULL_RTX, mode, unsignedp, |
| default_label, |
| probability); |
| default_prob /= 2; |
| } |
| |
| emit_case_nodes (index, node->right, default_label, default_prob, index_type); |
| } |
| else |
| { |
| probability = conditional_probability ( |
| node->right->subtree_prob, |
| subtree_prob + default_prob); |
| /* We cannot process node->right normally |
| since we haven't ruled out the numbers less than |
| this node's value. So handle node->right explicitly. */ |
| do_jump_if_equal (mode, index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->right->low), |
| unsignedp), |
| label_rtx (node->right->code_label), unsignedp, probability); |
| } |
| } |
| |
| else if (node->right == 0 && node->left != 0) |
| { |
| /* Just one subtree, on the left. */ |
| if (node->left->left || node->left->right |
| || !tree_int_cst_equal (node->left->low, node->left->high)) |
| { |
| if (!node_has_high_bound (node, index_type)) |
| { |
| probability = conditional_probability ( |
| default_prob/2, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| GT, NULL_RTX, mode, unsignedp, |
| default_label, |
| probability); |
| default_prob /= 2; |
| } |
| |
| emit_case_nodes (index, node->left, default_label, |
| default_prob, index_type); |
| } |
| else |
| { |
| probability = conditional_probability ( |
| node->left->subtree_prob, |
| subtree_prob + default_prob); |
| /* We cannot process node->left normally |
| since we haven't ruled out the numbers less than |
| this node's value. So handle node->left explicitly. */ |
| do_jump_if_equal (mode, index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->left->low), |
| unsignedp), |
| label_rtx (node->left->code_label), unsignedp, probability); |
| } |
| } |
| } |
| else |
| { |
| /* Node is a range. These cases are very similar to those for a single |
| value, except that we do not start by testing whether this node |
| is the one to branch to. */ |
| |
| if (node->right != 0 && node->left != 0) |
| { |
| /* Node has subtrees on both sides. |
| If the right-hand subtree is bounded, |
| test for it first, since we can go straight there. |
| Otherwise, we need to make a branch in the control structure, |
| then handle the two subtrees. */ |
| tree test_label = 0; |
| |
| if (node_is_bounded (node->right, index_type)) |
| { |
| /* Right hand node is fully bounded so we can eliminate any |
| testing and branch directly to the target code. */ |
| probability = conditional_probability ( |
| node->right->subtree_prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| GT, NULL_RTX, mode, unsignedp, |
| label_rtx (node->right->code_label), |
| probability); |
| } |
| else |
| { |
| /* Right hand node requires testing. |
| Branch to a label where we will handle it later. */ |
| |
| test_label = build_decl (curr_insn_location (), |
| LABEL_DECL, NULL_TREE, void_type_node); |
| probability = conditional_probability ( |
| node->right->subtree_prob + default_prob/2, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| GT, NULL_RTX, mode, unsignedp, |
| label_rtx (test_label), |
| probability); |
| default_prob /= 2; |
| } |
| |
| /* Value belongs to this node or to the left-hand subtree. */ |
| |
| probability = conditional_probability ( |
| prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->low), |
| unsignedp), |
| GE, NULL_RTX, mode, unsignedp, |
| label_rtx (node->code_label), |
| probability); |
| |
| /* Handle the left-hand subtree. */ |
| emit_case_nodes (index, node->left, default_label, default_prob, index_type); |
| |
| /* If right node had to be handled later, do that now. */ |
| |
| if (test_label) |
| { |
| /* If the left-hand subtree fell through, |
| don't let it fall into the right-hand subtree. */ |
| if (default_label) |
| emit_jump (default_label); |
| |
| expand_label (test_label); |
| emit_case_nodes (index, node->right, default_label, default_prob, index_type); |
| } |
| } |
| |
| else if (node->right != 0 && node->left == 0) |
| { |
| /* Deal with values to the left of this node, |
| if they are possible. */ |
| if (!node_has_low_bound (node, index_type)) |
| { |
| probability = conditional_probability ( |
| default_prob/2, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->low), |
| unsignedp), |
| LT, NULL_RTX, mode, unsignedp, |
| default_label, |
| probability); |
| default_prob /= 2; |
| } |
| |
| /* Value belongs to this node or to the right-hand subtree. */ |
| |
| probability = conditional_probability ( |
| prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| LE, NULL_RTX, mode, unsignedp, |
| label_rtx (node->code_label), |
| probability); |
| |
| emit_case_nodes (index, node->right, default_label, default_prob, index_type); |
| } |
| |
| else if (node->right == 0 && node->left != 0) |
| { |
| /* Deal with values to the right of this node, |
| if they are possible. */ |
| if (!node_has_high_bound (node, index_type)) |
| { |
| probability = conditional_probability ( |
| default_prob/2, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| GT, NULL_RTX, mode, unsignedp, |
| default_label, |
| probability); |
| default_prob /= 2; |
| } |
| |
| /* Value belongs to this node or to the left-hand subtree. */ |
| |
| probability = conditional_probability ( |
| prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->low), |
| unsignedp), |
| GE, NULL_RTX, mode, unsignedp, |
| label_rtx (node->code_label), |
| probability); |
| |
| emit_case_nodes (index, node->left, default_label, default_prob, index_type); |
| } |
| |
| else |
| { |
| /* Node has no children so we check low and high bounds to remove |
| redundant tests. Only one of the bounds can exist, |
| since otherwise this node is bounded--a case tested already. */ |
| int high_bound = node_has_high_bound (node, index_type); |
| int low_bound = node_has_low_bound (node, index_type); |
| |
| if (!high_bound && low_bound) |
| { |
| probability = conditional_probability ( |
| default_prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->high), |
| unsignedp), |
| GT, NULL_RTX, mode, unsignedp, |
| default_label, |
| probability); |
| } |
| |
| else if (!low_bound && high_bound) |
| { |
| probability = conditional_probability ( |
| default_prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (index, |
| convert_modes |
| (mode, imode, |
| expand_normal (node->low), |
| unsignedp), |
| LT, NULL_RTX, mode, unsignedp, |
| default_label, |
| probability); |
| } |
| else if (!low_bound && !high_bound) |
| { |
| /* Widen LOW and HIGH to the same width as INDEX. */ |
| tree type = lang_hooks.types.type_for_mode (mode, unsignedp); |
| tree low = build1 (CONVERT_EXPR, type, node->low); |
| tree high = build1 (CONVERT_EXPR, type, node->high); |
| rtx low_rtx, new_index, new_bound; |
| |
| /* Instead of doing two branches, emit one unsigned branch for |
| (index-low) > (high-low). */ |
| low_rtx = expand_expr (low, NULL_RTX, mode, EXPAND_NORMAL); |
| new_index = expand_simple_binop (mode, MINUS, index, low_rtx, |
| NULL_RTX, unsignedp, |
| OPTAB_WIDEN); |
| new_bound = expand_expr (fold_build2 (MINUS_EXPR, type, |
| high, low), |
| NULL_RTX, mode, EXPAND_NORMAL); |
| |
| probability = conditional_probability ( |
| default_prob, |
| subtree_prob + default_prob); |
| emit_cmp_and_jump_insns (new_index, new_bound, GT, NULL_RTX, |
| mode, 1, default_label, probability); |
| } |
| |
| emit_jump (label_rtx (node->code_label)); |
| } |
| } |
| } |