/* Instruction scheduling pass. Copyright (C) 1992, 1993, 1994 Free Software Foundation, Inc. Contributed by Michael Tiemann (tiemann@cygnus.com) Enhanced by, and currently maintained by, Jim Wilson (wilson@cygnus.com) This file is part of GNU CC. GNU CC 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 2, or (at your option) any later version. GNU CC 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 GNU CC; see the file COPYING. If not, write to the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */ /* Instruction scheduling pass. This pass implements list scheduling within basic blocks. It is run after flow analysis, but before register allocation. The scheduler works as follows: We compute insn priorities based on data dependencies. Flow analysis only creates a fraction of the data-dependencies we must observe: namely, only those dependencies which the combiner can be expected to use. For this pass, we must therefore create the remaining dependencies we need to observe: register dependencies, memory dependencies, dependencies to keep function calls in order, and the dependence between a conditional branch and the setting of condition codes are all dealt with here. The scheduler first traverses the data flow graph, starting with the last instruction, and proceeding to the first, assigning values to insn_priority as it goes. This sorts the instructions topologically by data dependence. Once priorities have been established, we order the insns using list scheduling. This works as follows: starting with a list of all the ready insns, and sorted according to priority number, we schedule the insn from the end of the list by placing its predecessors in the list according to their priority order. We consider this insn scheduled by setting the pointer to the "end" of the list to point to the previous insn. When an insn has no predecessors, we either queue it until sufficient time has elapsed or add it to the ready list. As the instructions are scheduled or when stalls are introduced, the queue advances and dumps insns into the ready list. When all insns down to the lowest priority have been scheduled, the critical path of the basic block has been made as short as possible. The remaining insns are then scheduled in remaining slots. Function unit conflicts are resolved during reverse list scheduling by tracking the time when each insn is committed to the schedule and from that, the time the function units it uses must be free. As insns on the ready list are considered for scheduling, those that would result in a blockage of the already committed insns are queued until no blockage will result. Among the remaining insns on the ready list to be considered, the first one with the largest potential for causing a subsequent blockage is chosen. The following list shows the order in which we want to break ties among insns in the ready list: 1. choose insn with lowest conflict cost, ties broken by 2. choose insn with the longest path to end of bb, ties broken by 3. choose insn that kills the most registers, ties broken by 4. choose insn that conflicts with the most ready insns, or finally 5. choose insn with lowest UID. Memory references complicate matters. Only if we can be certain that memory references are not part of the data dependency graph (via true, anti, or output dependence), can we move operations past memory references. To first approximation, reads can be done independently, while writes introduce dependencies. Better approximations will yield fewer dependencies. Dependencies set up by memory references are treated in exactly the same way as other dependencies, by using LOG_LINKS. Having optimized the critical path, we may have also unduly extended the lifetimes of some registers. If an operation requires that constants be loaded into registers, it is certainly desirable to load those constants as early as necessary, but no earlier. I.e., it will not do to load up a bunch of registers at the beginning of a basic block only to use them at the end, if they could be loaded later, since this may result in excessive register utilization. Note that since branches are never in basic blocks, but only end basic blocks, this pass will not do any branch scheduling. But that is ok, since we can use GNU's delayed branch scheduling pass to take care of this case. Also note that no further optimizations based on algebraic identities are performed, so this pass would be a good one to perform instruction splitting, such as breaking up a multiply instruction into shifts and adds where that is profitable. Given the memory aliasing analysis that this pass should perform, it should be possible to remove redundant stores to memory, and to load values from registers instead of hitting memory. This pass must update information that subsequent passes expect to be correct. Namely: reg_n_refs, reg_n_sets, reg_n_deaths, reg_n_calls_crossed, and reg_live_length. Also, basic_block_head, basic_block_end. The information in the line number notes is carefully retained by this pass. All other NOTE insns are grouped in their same relative order at the beginning of basic blocks that have been scheduled. */ #include #include "config.h" #include "rtl.h" #include "basic-block.h" #include "regs.h" #include "hard-reg-set.h" #include "flags.h" #include "insn-config.h" #include "insn-attr.h" #ifdef INSN_SCHEDULING /* Arrays set up by scheduling for the same respective purposes as similar-named arrays set up by flow analysis. We work with these arrays during the scheduling pass so we can compare values against unscheduled code. Values of these arrays are copied at the end of this pass into the arrays set up by flow analysis. */ static short *sched_reg_n_deaths; static int *sched_reg_n_calls_crossed; static int *sched_reg_live_length; /* Element N is the next insn that sets (hard or pseudo) register N within the current basic block; or zero, if there is no such insn. Needed for new registers which may be introduced by splitting insns. */ static rtx *reg_last_uses; static rtx *reg_last_sets; static regset reg_pending_sets; static int reg_pending_sets_all; /* Vector indexed by INSN_UID giving the original ordering of the insns. */ static int *insn_luid; #define INSN_LUID(INSN) (insn_luid[INSN_UID (INSN)]) /* Vector indexed by INSN_UID giving each instruction a priority. */ static int *insn_priority; #define INSN_PRIORITY(INSN) (insn_priority[INSN_UID (INSN)]) static short *insn_costs; #define INSN_COST(INSN) insn_costs[INSN_UID (INSN)] /* Vector indexed by INSN_UID giving an encoding of the function units used. */ static short *insn_units; #define INSN_UNIT(INSN) insn_units[INSN_UID (INSN)] /* Vector indexed by INSN_UID giving an encoding of the blockage range function. The unit and the range are encoded. */ static unsigned int *insn_blockage; #define INSN_BLOCKAGE(INSN) insn_blockage[INSN_UID (INSN)] #define UNIT_BITS 5 #define BLOCKAGE_MASK ((1 << BLOCKAGE_BITS) - 1) #define ENCODE_BLOCKAGE(U,R) \ ((((U) << UNIT_BITS) << BLOCKAGE_BITS \ | MIN_BLOCKAGE_COST (R)) << BLOCKAGE_BITS \ | MAX_BLOCKAGE_COST (R)) #define UNIT_BLOCKED(B) ((B) >> (2 * BLOCKAGE_BITS)) #define BLOCKAGE_RANGE(B) \ (((((B) >> BLOCKAGE_BITS) & BLOCKAGE_MASK) << (HOST_BITS_PER_INT / 2)) \ | (B) & BLOCKAGE_MASK) /* Encodings of the `_unit_blockage_range' function. */ #define MIN_BLOCKAGE_COST(R) ((R) >> (HOST_BITS_PER_INT / 2)) #define MAX_BLOCKAGE_COST(R) ((R) & ((1 << (HOST_BITS_PER_INT / 2)) - 1)) #define DONE_PRIORITY -1 #define MAX_PRIORITY 0x7fffffff #define TAIL_PRIORITY 0x7ffffffe #define LAUNCH_PRIORITY 0x7f000001 #define DONE_PRIORITY_P(INSN) (INSN_PRIORITY (INSN) < 0) #define LOW_PRIORITY_P(INSN) ((INSN_PRIORITY (INSN) & 0x7f000000) == 0) /* Vector indexed by INSN_UID giving number of insns referring to this insn. */ static int *insn_ref_count; #define INSN_REF_COUNT(INSN) (insn_ref_count[INSN_UID (INSN)]) /* Vector indexed by INSN_UID giving line-number note in effect for each insn. For line-number notes, this indicates whether the note may be reused. */ static rtx *line_note; #define LINE_NOTE(INSN) (line_note[INSN_UID (INSN)]) /* Vector indexed by basic block number giving the starting line-number for each basic block. */ static rtx *line_note_head; /* List of important notes we must keep around. This is a pointer to the last element in the list. */ static rtx note_list; /* Regsets telling whether a given register is live or dead before the last scheduled insn. Must scan the instructions once before scheduling to determine what registers are live or dead at the end of the block. */ static regset bb_dead_regs; static regset bb_live_regs; /* Regset telling whether a given register is live after the insn currently being scheduled. Before processing an insn, this is equal to bb_live_regs above. This is used so that we can find registers that are newly born/dead after processing an insn. */ static regset old_live_regs; /* The chain of REG_DEAD notes. REG_DEAD notes are removed from all insns during the initial scan and reused later. If there are not exactly as many REG_DEAD notes in the post scheduled code as there were in the prescheduled code then we trigger an abort because this indicates a bug. */ static rtx dead_notes; /* Queues, etc. */ /* An instruction is ready to be scheduled when all insns following it have already been scheduled. It is important to ensure that all insns which use its result will not be executed until its result has been computed. An insn is maintained in one of four structures: (P) the "Pending" set of insns which cannot be scheduled until their dependencies have been satisfied. (Q) the "Queued" set of insns that can be scheduled when sufficient time has passed. (R) the "Ready" list of unscheduled, uncommitted insns. (S) the "Scheduled" list of insns. Initially, all insns are either "Pending" or "Ready" depending on whether their dependencies are satisfied. Insns move from the "Ready" list to the "Scheduled" list as they are committed to the schedule. As this occurs, the insns in the "Pending" list have their dependencies satisfied and move to either the "Ready" list or the "Queued" set depending on whether sufficient time has passed to make them ready. As time passes, insns move from the "Queued" set to the "Ready" list. Insns may move from the "Ready" list to the "Queued" set if they are blocked due to a function unit conflict. The "Pending" list (P) are the insns in the LOG_LINKS of the unscheduled insns, i.e., those that are ready, queued, and pending. The "Queued" set (Q) is implemented by the variable `insn_queue'. The "Ready" list (R) is implemented by the variables `ready' and `n_ready'. The "Scheduled" list (S) is the new insn chain built by this pass. The transition (R->S) is implemented in the scheduling loop in `schedule_block' when the best insn to schedule is chosen. The transition (R->Q) is implemented in `schedule_select' when an insn is found to to have a function unit conflict with the already committed insns. The transitions (P->R and P->Q) are implemented in `schedule_insn' as insns move from the ready list to the scheduled list. The transition (Q->R) is implemented at the top of the scheduling loop in `schedule_block' as time passes or stalls are introduced. */ /* Implement a circular buffer to delay instructions until sufficient time has passed. INSN_QUEUE_SIZE is a power of two larger than MAX_BLOCKAGE and MAX_READY_COST computed by genattr.c. This is the longest time an isnsn may be queued. */ static rtx insn_queue[INSN_QUEUE_SIZE]; static int q_ptr = 0; static int q_size = 0; #define NEXT_Q(X) (((X)+1) & (INSN_QUEUE_SIZE-1)) #define NEXT_Q_AFTER(X,C) (((X)+C) & (INSN_QUEUE_SIZE-1)) /* Vector indexed by INSN_UID giving the minimum clock tick at which the insn becomes ready. This is used to note timing constraints for insns in the pending list. */ static int *insn_tick; #define INSN_TICK(INSN) (insn_tick[INSN_UID (INSN)]) /* Data structure for keeping track of register information during that register's life. */ struct sometimes { short offset; short bit; short live_length; short calls_crossed; }; /* Forward declarations. */ static rtx canon_rtx PROTO((rtx)); static int rtx_equal_for_memref_p PROTO((rtx, rtx)); static rtx find_symbolic_term PROTO((rtx)); static int memrefs_conflict_p PROTO((int, rtx, int, rtx, HOST_WIDE_INT)); static void add_dependence PROTO((rtx, rtx, enum reg_note)); static void remove_dependence PROTO((rtx, rtx)); static rtx find_insn_list PROTO((rtx, rtx)); static int insn_unit PROTO((rtx)); static unsigned int blockage_range PROTO((int, rtx)); static void clear_units PROTO((void)); static void prepare_unit PROTO((int)); static int actual_hazard_this_instance PROTO((int, int, rtx, int, int)); static void schedule_unit PROTO((int, rtx, int)); static int actual_hazard PROTO((int, rtx, int, int)); static int potential_hazard PROTO((int, rtx, int)); static int insn_cost PROTO((rtx, rtx, rtx)); static int priority PROTO((rtx)); static void free_pending_lists PROTO((void)); static void add_insn_mem_dependence PROTO((rtx *, rtx *, rtx, rtx)); static void flush_pending_lists PROTO((rtx)); static void sched_analyze_1 PROTO((rtx, rtx)); static void sched_analyze_2 PROTO((rtx, rtx)); static void sched_analyze_insn PROTO((rtx, rtx, rtx)); static int sched_analyze PROTO((rtx, rtx)); static void sched_note_set PROTO((int, rtx, int)); static int rank_for_schedule PROTO((rtx *, rtx *)); static void swap_sort PROTO((rtx *, int)); static void queue_insn PROTO((rtx, int)); static int birthing_insn PROTO((rtx)); static void adjust_priority PROTO((rtx)); static int schedule_insn PROTO((rtx, rtx *, int, int)); static int schedule_select PROTO((rtx *, int, int, FILE *)); static void create_reg_dead_note PROTO((rtx, rtx)); static void attach_deaths PROTO((rtx, rtx, int)); static void attach_deaths_insn PROTO((rtx)); static rtx unlink_notes PROTO((rtx, rtx)); static int new_sometimes_live PROTO((struct sometimes *, int, int, int)); static void finish_sometimes_live PROTO((struct sometimes *, int)); static void schedule_block PROTO((int, FILE *)); static rtx regno_use_in PROTO((int, rtx)); static void split_hard_reg_notes PROTO((rtx, rtx, rtx, rtx)); static void new_insn_dead_notes PROTO((rtx, rtx, rtx, rtx)); static void update_n_sets PROTO((rtx, int)); static void update_flow_info PROTO((rtx, rtx, rtx, rtx)); /* Main entry point of this file. */ void schedule_insns PROTO((FILE *)); #endif /* INSN_SCHEDULING */ #define SIZE_FOR_MODE(X) (GET_MODE_SIZE (GET_MODE (X))) /* Vector indexed by N giving the initial (unchanging) value known for pseudo-register N. */ static rtx *reg_known_value; /* Vector recording for each reg_known_value whether it is due to a REG_EQUIV note. Future passes (viz., reload) may replace the pseudo with the equivalent expression and so we account for the dependences that would be introduced if that happens. */ /* ??? This is a problem only on the Convex. The REG_EQUIV notes created in assign_parms mention the arg pointer, and there are explicit insns in the RTL that modify the arg pointer. Thus we must ensure that such insns don't get scheduled across each other because that would invalidate the REG_EQUIV notes. One could argue that the REG_EQUIV notes are wrong, but solving the problem in the scheduler will likely give better code, so we do it here. */ static char *reg_known_equiv_p; /* Indicates number of valid entries in reg_known_value. */ static int reg_known_value_size; static rtx canon_rtx (x) rtx x; { if (GET_CODE (x) == REG && REGNO (x) >= FIRST_PSEUDO_REGISTER && REGNO (x) <= reg_known_value_size) return reg_known_value[REGNO (x)]; else if (GET_CODE (x) == PLUS) { rtx x0 = canon_rtx (XEXP (x, 0)); rtx x1 = canon_rtx (XEXP (x, 1)); if (x0 != XEXP (x, 0) || x1 != XEXP (x, 1)) { /* We can tolerate LO_SUMs being offset here; these rtl are used for nothing other than comparisons. */ if (GET_CODE (x0) == CONST_INT) return plus_constant_for_output (x1, INTVAL (x0)); else if (GET_CODE (x1) == CONST_INT) return plus_constant_for_output (x0, INTVAL (x1)); return gen_rtx (PLUS, GET_MODE (x), x0, x1); } } return x; } /* Set up all info needed to perform alias analysis on memory references. */ void init_alias_analysis () { int maxreg = max_reg_num (); rtx insn; rtx note; rtx set; reg_known_value_size = maxreg; reg_known_value = (rtx *) oballoc ((maxreg-FIRST_PSEUDO_REGISTER) * sizeof (rtx)) - FIRST_PSEUDO_REGISTER; bzero ((char *) (reg_known_value + FIRST_PSEUDO_REGISTER), (maxreg-FIRST_PSEUDO_REGISTER) * sizeof (rtx)); reg_known_equiv_p = (char *) oballoc ((maxreg -FIRST_PSEUDO_REGISTER) * sizeof (char)) - FIRST_PSEUDO_REGISTER; bzero (reg_known_equiv_p + FIRST_PSEUDO_REGISTER, (maxreg - FIRST_PSEUDO_REGISTER) * sizeof (char)); /* Fill in the entries with known constant values. */ for (insn = get_insns (); insn; insn = NEXT_INSN (insn)) if ((set = single_set (insn)) != 0 && GET_CODE (SET_DEST (set)) == REG && REGNO (SET_DEST (set)) >= FIRST_PSEUDO_REGISTER && (((note = find_reg_note (insn, REG_EQUAL, 0)) != 0 && reg_n_sets[REGNO (SET_DEST (set))] == 1) || (note = find_reg_note (insn, REG_EQUIV, NULL_RTX)) != 0) && GET_CODE (XEXP (note, 0)) != EXPR_LIST) { int regno = REGNO (SET_DEST (set)); reg_known_value[regno] = XEXP (note, 0); reg_known_equiv_p[regno] = REG_NOTE_KIND (note) == REG_EQUIV; } /* Fill in the remaining entries. */ while (--maxreg >= FIRST_PSEUDO_REGISTER) if (reg_known_value[maxreg] == 0) reg_known_value[maxreg] = regno_reg_rtx[maxreg]; } /* Return 1 if X and Y are identical-looking rtx's. We use the data in reg_known_value above to see if two registers with different numbers are, in fact, equivalent. */ static int rtx_equal_for_memref_p (x, y) rtx x, y; { register int i; register int j; register enum rtx_code code; register char *fmt; if (x == 0 && y == 0) return 1; if (x == 0 || y == 0) return 0; x = canon_rtx (x); y = canon_rtx (y); if (x == y) return 1; code = GET_CODE (x); /* Rtx's of different codes cannot be equal. */ if (code != GET_CODE (y)) return 0; /* (MULT:SI x y) and (MULT:HI x y) are NOT equivalent. (REG:SI x) and (REG:HI x) are NOT equivalent. */ if (GET_MODE (x) != GET_MODE (y)) return 0; /* REG, LABEL_REF, and SYMBOL_REF can be compared nonrecursively. */ if (code == REG) return REGNO (x) == REGNO (y); if (code == LABEL_REF) return XEXP (x, 0) == XEXP (y, 0); if (code == SYMBOL_REF) return XSTR (x, 0) == XSTR (y, 0); /* For commutative operations, the RTX match if the operand match in any order. Also handle the simple binary and unary cases without a loop. */ if (code == EQ || code == NE || GET_RTX_CLASS (code) == 'c') return ((rtx_equal_for_memref_p (XEXP (x, 0), XEXP (y, 0)) && rtx_equal_for_memref_p (XEXP (x, 1), XEXP (y, 1))) || (rtx_equal_for_memref_p (XEXP (x, 0), XEXP (y, 1)) && rtx_equal_for_memref_p (XEXP (x, 1), XEXP (y, 0)))); else if (GET_RTX_CLASS (code) == '<' || GET_RTX_CLASS (code) == '2') return (rtx_equal_for_memref_p (XEXP (x, 0), XEXP (y, 0)) && rtx_equal_for_memref_p (XEXP (x, 1), XEXP (y, 1))); else if (GET_RTX_CLASS (code) == '1') return rtx_equal_for_memref_p (XEXP (x, 0), XEXP (y, 0)); /* Compare the elements. If any pair of corresponding elements fail to match, return 0 for the whole things. */ fmt = GET_RTX_FORMAT (code); for (i = GET_RTX_LENGTH (code) - 1; i >= 0; i--) { switch (fmt[i]) { case 'w': if (XWINT (x, i) != XWINT (y, i)) return 0; break; case 'n': case 'i': if (XINT (x, i) != XINT (y, i)) return 0; break; case 'V': case 'E': /* Two vectors must have the same length. */ if (XVECLEN (x, i) != XVECLEN (y, i)) return 0; /* And the corresponding elements must match. */ for (j = 0; j < XVECLEN (x, i); j++) if (rtx_equal_for_memref_p (XVECEXP (x, i, j), XVECEXP (y, i, j)) == 0) return 0; break; case 'e': if (rtx_equal_for_memref_p (XEXP (x, i), XEXP (y, i)) == 0) return 0; break; case 'S': case 's': if (strcmp (XSTR (x, i), XSTR (y, i))) return 0; break; case 'u': /* These are just backpointers, so they don't matter. */ break; case '0': break; /* It is believed that rtx's at this level will never contain anything but integers and other rtx's, except for within LABEL_REFs and SYMBOL_REFs. */ default: abort (); } } return 1; } /* Given an rtx X, find a SYMBOL_REF or LABEL_REF within X and return it, or return 0 if none found. */ static rtx find_symbolic_term (x) rtx x; { register int i; register enum rtx_code code; register char *fmt; code = GET_CODE (x); if (code == SYMBOL_REF || code == LABEL_REF) return x; if (GET_RTX_CLASS (code) == 'o') return 0; fmt = GET_RTX_FORMAT (code); for (i = GET_RTX_LENGTH (code) - 1; i >= 0; i--) { rtx t; if (fmt[i] == 'e') { t = find_symbolic_term (XEXP (x, i)); if (t != 0) return t; } else if (fmt[i] == 'E') break; } return 0; } /* Return nonzero if X and Y (memory addresses) could reference the same location in memory. C is an offset accumulator. When C is nonzero, we are testing aliases between X and Y + C. XSIZE is the size in bytes of the X reference, similarly YSIZE is the size in bytes for Y. If XSIZE or YSIZE is zero, we do not know the amount of memory being referenced (the reference was BLKmode), so make the most pessimistic assumptions. We recognize the following cases of non-conflicting memory: (1) addresses involving the frame pointer cannot conflict with addresses involving static variables. (2) static variables with different addresses cannot conflict. Nice to notice that varying addresses cannot conflict with fp if no local variables had their addresses taken, but that's too hard now. */ /* ??? In Fortran, references to a array parameter can never conflict with another array parameter. */ static int memrefs_conflict_p (xsize, x, ysize, y, c) rtx x, y; int xsize, ysize; HOST_WIDE_INT c; { if (GET_CODE (x) == HIGH) x = XEXP (x, 0); else if (GET_CODE (x) == LO_SUM) x = XEXP (x, 1); else x = canon_rtx (x); if (GET_CODE (y) == HIGH) y = XEXP (y, 0); else if (GET_CODE (y) == LO_SUM) y = XEXP (y, 1); else y = canon_rtx (y); if (rtx_equal_for_memref_p (x, y)) return (xsize == 0 || ysize == 0 || (c >= 0 && xsize > c) || (c < 0 && ysize+c > 0)); if (y == frame_pointer_rtx || y == hard_frame_pointer_rtx || y == stack_pointer_rtx) { rtx t = y; int tsize = ysize; y = x; ysize = xsize; x = t; xsize = tsize; } if (x == frame_pointer_rtx || x == hard_frame_pointer_rtx || x == stack_pointer_rtx) { rtx y1; if (CONSTANT_P (y)) return 0; if (GET_CODE (y) == PLUS && canon_rtx (XEXP (y, 0)) == x && (y1 = canon_rtx (XEXP (y, 1))) && GET_CODE (y1) == CONST_INT) { c += INTVAL (y1); return (xsize == 0 || ysize == 0 || (c >= 0 && xsize > c) || (c < 0 && ysize+c > 0)); } if (GET_CODE (y) == PLUS && (y1 = canon_rtx (XEXP (y, 0))) && CONSTANT_P (y1)) return 0; return 1; } if (GET_CODE (x) == PLUS) { /* The fact that X is canonicalized means that this PLUS rtx is canonicalized. */ rtx x0 = XEXP (x, 0); rtx x1 = XEXP (x, 1); if (GET_CODE (y) == PLUS) { /* The fact that Y is canonicalized means that this PLUS rtx is canonicalized. */ rtx y0 = XEXP (y, 0); rtx y1 = XEXP (y, 1); if (rtx_equal_for_memref_p (x1, y1)) return memrefs_conflict_p (xsize, x0, ysize, y0, c); if (rtx_equal_for_memref_p (x0, y0)) return memrefs_conflict_p (xsize, x1, ysize, y1, c); if (GET_CODE (x1) == CONST_INT) if (GET_CODE (y1) == CONST_INT) return memrefs_conflict_p (xsize, x0, ysize, y0, c - INTVAL (x1) + INTVAL (y1)); else return memrefs_conflict_p (xsize, x0, ysize, y, c - INTVAL (x1)); else if (GET_CODE (y1) == CONST_INT) return memrefs_conflict_p (xsize, x, ysize, y0, c + INTVAL (y1)); /* Handle case where we cannot understand iteration operators, but we notice that the base addresses are distinct objects. */ x = find_symbolic_term (x); if (x == 0) return 1; y = find_symbolic_term (y); if (y == 0) return 1; return rtx_equal_for_memref_p (x, y); } else if (GET_CODE (x1) == CONST_INT) return memrefs_conflict_p (xsize, x0, ysize, y, c - INTVAL (x1)); } else if (GET_CODE (y) == PLUS) { /* The fact that Y is canonicalized means that this PLUS rtx is canonicalized. */ rtx y0 = XEXP (y, 0); rtx y1 = XEXP (y, 1); if (GET_CODE (y1) == CONST_INT) return memrefs_conflict_p (xsize, x, ysize, y0, c + INTVAL (y1)); else return 1; } if (GET_CODE (x) == GET_CODE (y)) switch (GET_CODE (x)) { case MULT: { /* Handle cases where we expect the second operands to be the same, and check only whether the first operand would conflict or not. */ rtx x0, y0; rtx x1 = canon_rtx (XEXP (x, 1)); rtx y1 = canon_rtx (XEXP (y, 1)); if (! rtx_equal_for_memref_p (x1, y1)) return 1; x0 = canon_rtx (XEXP (x, 0)); y0 = canon_rtx (XEXP (y, 0)); if (rtx_equal_for_memref_p (x0, y0)) return (xsize == 0 || ysize == 0 || (c >= 0 && xsize > c) || (c < 0 && ysize+c > 0)); /* Can't properly adjust our sizes. */ if (GET_CODE (x1) != CONST_INT) return 1; xsize /= INTVAL (x1); ysize /= INTVAL (x1); c /= INTVAL (x1); return memrefs_conflict_p (xsize, x0, ysize, y0, c); } } if (CONSTANT_P (x)) { if (GET_CODE (x) == CONST_INT && GET_CODE (y) == CONST_INT) { c += (INTVAL (y) - INTVAL (x)); return (xsize == 0 || ysize == 0 || (c >= 0 && xsize > c) || (c < 0 && ysize+c > 0)); } if (GET_CODE (x) == CONST) { if (GET_CODE (y) == CONST) return memrefs_conflict_p (xsize, canon_rtx (XEXP (x, 0)), ysize, canon_rtx (XEXP (y, 0)), c); else return memrefs_conflict_p (xsize, canon_rtx (XEXP (x, 0)), ysize, y, c); } if (GET_CODE (y) == CONST) return memrefs_conflict_p (xsize, x, ysize, canon_rtx (XEXP (y, 0)), c); if (CONSTANT_P (y)) return (rtx_equal_for_memref_p (x, y) && (xsize == 0 || ysize == 0 || (c >= 0 && xsize > c) || (c < 0 && ysize+c > 0))); return 1; } return 1; } /* Functions to compute memory dependencies. Since we process the insns in execution order, we can build tables to keep track of what registers are fixed (and not aliased), what registers are varying in known ways, and what registers are varying in unknown ways. If both memory references are volatile, then there must always be a dependence between the two references, since their order can not be changed. A volatile and non-volatile reference can be interchanged though. A MEM_IN_STRUCT reference at a non-QImode varying address can never conflict with a non-MEM_IN_STRUCT reference at a fixed address. We must allow QImode aliasing because the ANSI C standard allows character pointers to alias anything. We are assuming that characters are always QImode here. */ /* Read dependence: X is read after read in MEM takes place. There can only be a dependence here if both reads are volatile. */ int read_dependence (mem, x) rtx mem; rtx x; { return MEM_VOLATILE_P (x) && MEM_VOLATILE_P (mem); } /* True dependence: X is read after store in MEM takes place. */ int true_dependence (mem, x) rtx mem; rtx x; { /* If X is an unchanging read, then it can't possibly conflict with any non-unchanging store. It may conflict with an unchanging write though, because there may be a single store to this address to initialize it. Just fall through to the code below to resolve the case where we have both an unchanging read and an unchanging write. This won't handle all cases optimally, but the possible performance loss should be negligible. */ if (RTX_UNCHANGING_P (x) && ! RTX_UNCHANGING_P (mem)) return 0; return ((MEM_VOLATILE_P (x) && MEM_VOLATILE_P (mem)) || (memrefs_conflict_p (SIZE_FOR_MODE (mem), XEXP (mem, 0), SIZE_FOR_MODE (x), XEXP (x, 0), 0) && ! (MEM_IN_STRUCT_P (mem) && rtx_addr_varies_p (mem) && GET_MODE (mem) != QImode && ! MEM_IN_STRUCT_P (x) && ! rtx_addr_varies_p (x)) && ! (MEM_IN_STRUCT_P (x) && rtx_addr_varies_p (x) && GET_MODE (x) != QImode && ! MEM_IN_STRUCT_P (mem) && ! rtx_addr_varies_p (mem)))); } /* Anti dependence: X is written after read in MEM takes place. */ int anti_dependence (mem, x) rtx mem; rtx x; { /* If MEM is an unchanging read, then it can't possibly conflict with the store to X, because there is at most one store to MEM, and it must have occured somewhere before MEM. */ if (RTX_UNCHANGING_P (mem)) return 0; return ((MEM_VOLATILE_P (x) && MEM_VOLATILE_P (mem)) || (memrefs_conflict_p (SIZE_FOR_MODE (mem), XEXP (mem, 0), SIZE_FOR_MODE (x), XEXP (x, 0), 0) && ! (MEM_IN_STRUCT_P (mem) && rtx_addr_varies_p (mem) && GET_MODE (mem) != QImode && ! MEM_IN_STRUCT_P (x) && ! rtx_addr_varies_p (x)) && ! (MEM_IN_STRUCT_P (x) && rtx_addr_varies_p (x) && GET_MODE (x) != QImode && ! MEM_IN_STRUCT_P (mem) && ! rtx_addr_varies_p (mem)))); } /* Output dependence: X is written after store in MEM takes place. */ int output_dependence (mem, x) rtx mem; rtx x; { return ((MEM_VOLATILE_P (x) && MEM_VOLATILE_P (mem)) || (memrefs_conflict_p (SIZE_FOR_MODE (mem), XEXP (mem, 0), SIZE_FOR_MODE (x), XEXP (x, 0), 0) && ! (MEM_IN_STRUCT_P (mem) && rtx_addr_varies_p (mem) && GET_MODE (mem) != QImode && ! MEM_IN_STRUCT_P (x) && ! rtx_addr_varies_p (x)) && ! (MEM_IN_STRUCT_P (x) && rtx_addr_varies_p (x) && GET_MODE (x) != QImode && ! MEM_IN_STRUCT_P (mem) && ! rtx_addr_varies_p (mem)))); } /* Helper functions for instruction scheduling. */ /* Add ELEM wrapped in an INSN_LIST with reg note kind DEP_TYPE to the LOG_LINKS of INSN, if not already there. DEP_TYPE indicates the type of dependence that this link represents. */ static void add_dependence (insn, elem, dep_type) rtx insn; rtx elem; enum reg_note dep_type; { rtx link, next; /* Don't depend an insn on itself. */ if (insn == elem) return; /* If elem is part of a sequence that must be scheduled together, then make the dependence point to the last insn of the sequence. When HAVE_cc0, it is possible for NOTEs to exist between users and setters of the condition codes, so we must skip past notes here. Otherwise, NOTEs are impossible here. */ next = NEXT_INSN (elem); #ifdef HAVE_cc0 while (next && GET_CODE (next) == NOTE) next = NEXT_INSN (next); #endif if (next && SCHED_GROUP_P (next)) { /* Notes will never intervene here though, so don't bother checking for them. */ /* We must reject CODE_LABELs, so that we don't get confused by one that has LABEL_PRESERVE_P set, which is represented by the same bit in the rtl as SCHED_GROUP_P. A CODE_LABEL can never be SCHED_GROUP_P. */ while (NEXT_INSN (next) && SCHED_GROUP_P (NEXT_INSN (next)) && GET_CODE (NEXT_INSN (next)) != CODE_LABEL) next = NEXT_INSN (next); /* Again, don't depend an insn on itself. */ if (insn == next) return; /* Make the dependence to NEXT, the last insn of the group, instead of the original ELEM. */ elem = next; } /* Check that we don't already have this dependence. */ for (link = LOG_LINKS (insn); link; link = XEXP (link, 1)) if (XEXP (link, 0) == elem) { /* If this is a more restrictive type of dependence than the existing one, then change the existing dependence to this type. */ if ((int) dep_type < (int) REG_NOTE_KIND (link)) PUT_REG_NOTE_KIND (link, dep_type); return; } /* Might want to check one level of transitivity to save conses. */ link = rtx_alloc (INSN_LIST); /* Insn dependency, not data dependency. */ PUT_REG_NOTE_KIND (link, dep_type); XEXP (link, 0) = elem; XEXP (link, 1) = LOG_LINKS (insn); LOG_LINKS (insn) = link; } /* Remove ELEM wrapped in an INSN_LIST from the LOG_LINKS of INSN. Abort if not found. */ static void remove_dependence (insn, elem) rtx insn; rtx elem; { rtx prev, link; int found = 0; for (prev = 0, link = LOG_LINKS (insn); link; prev = link, link = XEXP (link, 1)) { if (XEXP (link, 0) == elem) { if (prev) XEXP (prev, 1) = XEXP (link, 1); else LOG_LINKS (insn) = XEXP (link, 1); found = 1; } } if (! found) abort (); return; } #ifndef INSN_SCHEDULING void schedule_insns (dump_file) FILE *dump_file; { } #else #ifndef __GNUC__ #define __inline #endif /* Computation of memory dependencies. */ /* The *_insns and *_mems are paired lists. Each pending memory operation will have a pointer to the MEM rtx on one list and a pointer to the containing insn on the other list in the same place in the list. */ /* We can't use add_dependence like the old code did, because a single insn may have multiple memory accesses, and hence needs to be on the list once for each memory access. Add_dependence won't let you add an insn to a list more than once. */ /* An INSN_LIST containing all insns with pending read operations. */ static rtx pending_read_insns; /* An EXPR_LIST containing all MEM rtx's which are pending reads. */ static rtx pending_read_mems; /* An INSN_LIST containing all insns with pending write operations. */ static rtx pending_write_insns; /* An EXPR_LIST containing all MEM rtx's which are pending writes. */ static rtx pending_write_mems; /* Indicates the combined length of the two pending lists. We must prevent these lists from ever growing too large since the number of dependencies produced is at least O(N*N), and execution time is at least O(4*N*N), as a function of the length of these pending lists. */ static int pending_lists_length; /* An INSN_LIST containing all INSN_LISTs allocated but currently unused. */ static rtx unused_insn_list; /* An EXPR_LIST containing all EXPR_LISTs allocated but currently unused. */ static rtx unused_expr_list; /* The last insn upon which all memory references must depend. This is an insn which flushed the pending lists, creating a dependency between it and all previously pending memory references. This creates a barrier (or a checkpoint) which no memory reference is allowed to cross. This includes all non constant CALL_INSNs. When we do interprocedural alias analysis, this restriction can be relaxed. This may also be an INSN that writes memory if the pending lists grow too large. */ static rtx last_pending_memory_flush; /* The last function call we have seen. All hard regs, and, of course, the last function call, must depend on this. */ static rtx last_function_call; /* The LOG_LINKS field of this is a list of insns which use a pseudo register that does not already cross a call. We create dependencies between each of those insn and the next call insn, to ensure that they won't cross a call after scheduling is done. */ static rtx sched_before_next_call; /* Pointer to the last instruction scheduled. Used by rank_for_schedule, so that insns independent of the last scheduled insn will be preferred over dependent instructions. */ static rtx last_scheduled_insn; /* Process an insn's memory dependencies. There are four kinds of dependencies: (0) read dependence: read follows read (1) true dependence: read follows write (2) anti dependence: write follows read (3) output dependence: write follows write We are careful to build only dependencies which actually exist, and use transitivity to avoid building too many links. */ /* Return the INSN_LIST containing INSN in LIST, or NULL if LIST does not contain INSN. */ __inline static rtx find_insn_list (insn, list) rtx insn; rtx list; { while (list) { if (XEXP (list, 0) == insn) return list; list = XEXP (list, 1); } return 0; } /* Compute the function units used by INSN. This caches the value returned by function_units_used. A function unit is encoded as the unit number if the value is non-negative and the compliment of a mask if the value is negative. A function unit index is the non-negative encoding. */ __inline static int insn_unit (insn) rtx insn; { register int unit = INSN_UNIT (insn); if (unit == 0) { recog_memoized (insn); /* A USE insn, or something else we don't need to understand. We can't pass these directly to function_units_used because it will trigger a fatal error for unrecognizable insns. */ if (INSN_CODE (insn) < 0) unit = -1; else { unit = function_units_used (insn); /* Increment non-negative values so we can cache zero. */ if (unit >= 0) unit++; } /* We only cache 16 bits of the result, so if the value is out of range, don't cache it. */ if (FUNCTION_UNITS_SIZE < HOST_BITS_PER_SHORT || unit >= 0 || (~unit & ((1 << (HOST_BITS_PER_SHORT - 1)) - 1)) == 0) INSN_UNIT (insn) = unit; } return (unit > 0 ? unit - 1 : unit); } /* Compute the blockage range for executing INSN on UNIT. This caches the value returned by the blockage_range_function for the unit. These values are encoded in an int where the upper half gives the minimum value and the lower half gives the maximum value. */ __inline static unsigned int blockage_range (unit, insn) int unit; rtx insn; { unsigned int blockage = INSN_BLOCKAGE (insn); unsigned int range; if (UNIT_BLOCKED (blockage) != unit + 1) { range = function_units[unit].blockage_range_function (insn); /* We only cache the blockage range for one unit and then only if the values fit. */ if (HOST_BITS_PER_INT >= UNIT_BITS + 2 * BLOCKAGE_BITS) INSN_BLOCKAGE (insn) = ENCODE_BLOCKAGE (unit + 1, range); } else range = BLOCKAGE_RANGE (blockage); return range; } /* A vector indexed by function unit instance giving the last insn to use the unit. The value of the function unit instance index for unit U instance I is (U + I * FUNCTION_UNITS_SIZE). */ static rtx unit_last_insn[FUNCTION_UNITS_SIZE * MAX_MULTIPLICITY]; /* A vector indexed by function unit instance giving the minimum time when the unit will unblock based on the maximum blockage cost. */ static int unit_tick[FUNCTION_UNITS_SIZE * MAX_MULTIPLICITY]; /* A vector indexed by function unit number giving the number of insns that remain to use the unit. */ static int unit_n_insns[FUNCTION_UNITS_SIZE]; /* Reset the function unit state to the null state. */ static void clear_units () { bzero ((char *) unit_last_insn, sizeof (unit_last_insn)); bzero ((char *) unit_tick, sizeof (unit_tick)); bzero ((char *) unit_n_insns, sizeof (unit_n_insns)); } /* Record an insn as one that will use the units encoded by UNIT. */ __inline static void prepare_unit (unit) int unit; { int i; if (unit >= 0) unit_n_insns[unit]++; else for (i = 0, unit = ~unit; unit; i++, unit >>= 1) if ((unit & 1) != 0) prepare_unit (i); } /* Return the actual hazard cost of executing INSN on the unit UNIT, instance INSTANCE at time CLOCK if the previous actual hazard cost was COST. */ __inline static int actual_hazard_this_instance (unit, instance, insn, clock, cost) int unit, instance, clock, cost; rtx insn; { int tick = unit_tick[instance]; if (tick - clock > cost) { /* The scheduler is operating in reverse, so INSN is the executing insn and the unit's last insn is the candidate insn. We want a more exact measure of the blockage if we execute INSN at CLOCK given when we committed the execution of the unit's last insn. The blockage value is given by either the unit's max blockage constant, blockage range function, or blockage function. Use the most exact form for the given unit. */ if (function_units[unit].blockage_range_function) { if (function_units[unit].blockage_function) tick += (function_units[unit].blockage_function (insn, unit_last_insn[instance]) - function_units[unit].max_blockage); else tick += ((int) MAX_BLOCKAGE_COST (blockage_range (unit, insn)) - function_units[unit].max_blockage); } if (tick - clock > cost) cost = tick - clock; } return cost; } /* Record INSN as having begun execution on the units encoded by UNIT at time CLOCK. */ __inline static void schedule_unit (unit, insn, clock) int unit, clock; rtx insn; { int i; if (unit >= 0) { int instance = unit; #if MAX_MULTIPLICITY > 1 /* Find the first free instance of the function unit and use that one. We assume that one is free. */ for (i = function_units[unit].multiplicity - 1; i > 0; i--) { if (! actual_hazard_this_instance (unit, instance, insn, clock, 0)) break; instance += FUNCTION_UNITS_SIZE; } #endif unit_last_insn[instance] = insn; unit_tick[instance] = (clock + function_units[unit].max_blockage); } else for (i = 0, unit = ~unit; unit; i++, unit >>= 1) if ((unit & 1) != 0) schedule_unit (i, insn, clock); } /* Return the actual hazard cost of executing INSN on the units encoded by UNIT at time CLOCK if the previous actual hazard cost was COST. */ __inline static int actual_hazard (unit, insn, clock, cost) int unit, clock, cost; rtx insn; { int i; if (unit >= 0) { /* Find the instance of the function unit with the minimum hazard. */ int instance = unit; int best_cost = actual_hazard_this_instance (unit, instance, insn, clock, cost); int this_cost; #if MAX_MULTIPLICITY > 1 if (best_cost > cost) { for (i = function_units[unit].multiplicity - 1; i > 0; i--) { instance += FUNCTION_UNITS_SIZE; this_cost = actual_hazard_this_instance (unit, instance, insn, clock, cost); if (this_cost < best_cost) { best_cost = this_cost; if (this_cost <= cost) break; } } } #endif cost = MAX (cost, best_cost); } else for (i = 0, unit = ~unit; unit; i++, unit >>= 1) if ((unit & 1) != 0) cost = actual_hazard (i, insn, clock, cost); return cost; } /* Return the potential hazard cost of executing an instruction on the units encoded by UNIT if the previous potential hazard cost was COST. An insn with a large blockage time is chosen in preference to one with a smaller time; an insn that uses a unit that is more likely to be used is chosen in preference to one with a unit that is less used. We are trying to minimize a subsequent actual hazard. */ __inline static int potential_hazard (unit, insn, cost) int unit, cost; rtx insn; { int i, ncost; unsigned int minb, maxb; if (unit >= 0) { minb = maxb = function_units[unit].max_blockage; if (maxb > 1) { if (function_units[unit].blockage_range_function) { maxb = minb = blockage_range (unit, insn); maxb = MAX_BLOCKAGE_COST (maxb); minb = MIN_BLOCKAGE_COST (minb); } if (maxb > 1) { /* Make the number of instructions left dominate. Make the minimum delay dominate the maximum delay. If all these are the same, use the unit number to add an arbitrary ordering. Other terms can be added. */ ncost = minb * 0x40 + maxb; ncost *= (unit_n_insns[unit] - 1) * 0x1000 + unit; if (ncost > cost) cost = ncost; } } } else for (i = 0, unit = ~unit; unit; i++, unit >>= 1) if ((unit & 1) != 0) cost = potential_hazard (i, insn, cost); return cost; } /* Compute cost of executing INSN given the dependence LINK on the insn USED. This is the number of virtual cycles taken between instruction issue and instruction results. */ __inline static int insn_cost (insn, link, used) rtx insn, link, used; { register int cost = INSN_COST (insn); if (cost == 0) { recog_memoized (insn); /* A USE insn, or something else we don't need to understand. We can't pass these directly to result_ready_cost because it will trigger a fatal error for unrecognizable insns. */ if (INSN_CODE (insn) < 0) { INSN_COST (insn) = 1; return 1; } else { cost = result_ready_cost (insn); if (cost < 1) cost = 1; INSN_COST (insn) = cost; } } /* A USE insn should never require the value used to be computed. This allows the computation of a function's result and parameter values to overlap the return and call. */ recog_memoized (used); if (INSN_CODE (used) < 0) LINK_COST_FREE (link) = 1; /* If some dependencies vary the cost, compute the adjustment. Most commonly, the adjustment is complete: either the cost is ignored (in the case of an output- or anti-dependence), or the cost is unchanged. These values are cached in the link as LINK_COST_FREE and LINK_COST_ZERO. */ if (LINK_COST_FREE (link)) cost = 1; #ifdef ADJUST_COST else if (! LINK_COST_ZERO (link)) { int ncost = cost; ADJUST_COST (used, link, insn, ncost); if (ncost <= 1) LINK_COST_FREE (link) = ncost = 1; if (cost == ncost) LINK_COST_ZERO (link) = 1; cost = ncost; } #endif return cost; } /* Compute the priority number for INSN. */ static int priority (insn) rtx insn; { if (insn && GET_RTX_CLASS (GET_CODE (insn)) == 'i') { int prev_priority; int max_priority; int this_priority = INSN_PRIORITY (insn); rtx prev; if (this_priority > 0) return this_priority; max_priority = 1; /* Nonzero if these insns must be scheduled together. */ if (SCHED_GROUP_P (insn)) { prev = insn; while (SCHED_GROUP_P (prev)) { prev = PREV_INSN (prev); INSN_REF_COUNT (prev) += 1; } } for (prev = LOG_LINKS (insn); prev; prev = XEXP (prev, 1)) { rtx x = XEXP (prev, 0); /* A dependence pointing to a note or deleted insn is always obsolete, because sched_analyze_insn will have created any necessary new dependences which replace it. Notes and deleted insns can be created when instructions are deleted by insn splitting, or by register allocation. */ if (GET_CODE (x) == NOTE || INSN_DELETED_P (x)) { remove_dependence (insn, x); continue; } /* Clear the link cost adjustment bits. */ LINK_COST_FREE (prev) = 0; #ifdef ADJUST_COST LINK_COST_ZERO (prev) = 0; #endif /* This priority calculation was chosen because it results in the least instruction movement, and does not hurt the performance of the resulting code compared to the old algorithm. This makes the sched algorithm more stable, which results in better code, because there is less register pressure, cross jumping is more likely to work, and debugging is easier. When all instructions have a latency of 1, there is no need to move any instructions. Subtracting one here ensures that in such cases all instructions will end up with a priority of one, and hence no scheduling will be done. The original code did not subtract the one, and added the insn_cost of the current instruction to its priority (e.g. move the insn_cost call down to the end). */ prev_priority = priority (x) + insn_cost (x, prev, insn) - 1; if (prev_priority > max_priority) max_priority = prev_priority; INSN_REF_COUNT (x) += 1; } prepare_unit (insn_unit (insn)); INSN_PRIORITY (insn) = max_priority; return INSN_PRIORITY (insn); } return 0; } /* Remove all INSN_LISTs and EXPR_LISTs from the pending lists and add them to the unused_*_list variables, so that they can be reused. */ static void free_pending_lists () { register rtx link, prev_link; if (pending_read_insns) { prev_link = pending_read_insns; link = XEXP (prev_link, 1); while (link) { prev_link = link; link = XEXP (link, 1); } XEXP (prev_link, 1) = unused_insn_list; unused_insn_list = pending_read_insns; pending_read_insns = 0; } if (pending_write_insns) { prev_link = pending_write_insns; link = XEXP (prev_link, 1); while (link) { prev_link = link; link = XEXP (link, 1); } XEXP (prev_link, 1) = unused_insn_list; unused_insn_list = pending_write_insns; pending_write_insns = 0; } if (pending_read_mems) { prev_link = pending_read_mems; link = XEXP (prev_link, 1); while (link) { prev_link = link; link = XEXP (link, 1); } XEXP (prev_link, 1) = unused_expr_list; unused_expr_list = pending_read_mems; pending_read_mems = 0; } if (pending_write_mems) { prev_link = pending_write_mems; link = XEXP (prev_link, 1); while (link) { prev_link = link; link = XEXP (link, 1); } XEXP (prev_link, 1) = unused_expr_list; unused_expr_list = pending_write_mems; pending_write_mems = 0; } } /* Add an INSN and MEM reference pair to a pending INSN_LIST and MEM_LIST. The MEM is a memory reference contained within INSN, which we are saving so that we can do memory aliasing on it. */ static void add_insn_mem_dependence (insn_list, mem_list, insn, mem) rtx *insn_list, *mem_list, insn, mem; { register rtx link; if (unused_insn_list) { link = unused_insn_list; unused_insn_list = XEXP (link, 1); } else link = rtx_alloc (INSN_LIST); XEXP (link, 0) = insn; XEXP (link, 1) = *insn_list; *insn_list = link; if (unused_expr_list) { link = unused_expr_list; unused_expr_list = XEXP (link, 1); } else link = rtx_alloc (EXPR_LIST); XEXP (link, 0) = mem; XEXP (link, 1) = *mem_list; *mem_list = link; pending_lists_length++; } /* Make a dependency between every memory reference on the pending lists and INSN, thus flushing the pending lists. */ static void flush_pending_lists (insn) rtx insn; { rtx link; while (pending_read_insns) { add_dependence (insn, XEXP (pending_read_insns, 0), REG_DEP_ANTI); link = pending_read_insns; pending_read_insns = XEXP (pending_read_insns, 1); XEXP (link, 1) = unused_insn_list; unused_insn_list = link; link = pending_read_mems; pending_read_mems = XEXP (pending_read_mems, 1); XEXP (link, 1) = unused_expr_list; unused_expr_list = link; } while (pending_write_insns) { add_dependence (insn, XEXP (pending_write_insns, 0), REG_DEP_ANTI); link = pending_write_insns; pending_write_insns = XEXP (pending_write_insns, 1); XEXP (link, 1) = unused_insn_list; unused_insn_list = link; link = pending_write_mems; pending_write_mems = XEXP (pending_write_mems, 1); XEXP (link, 1) = unused_expr_list; unused_expr_list = link; } pending_lists_length = 0; if (last_pending_memory_flush) add_dependence (insn, last_pending_memory_flush, REG_DEP_ANTI); last_pending_memory_flush = insn; } /* Analyze a single SET or CLOBBER rtx, X, creating all dependencies generated by the write to the destination of X, and reads of everything mentioned. */ static void sched_analyze_1 (x, insn) rtx x; rtx insn; { register int regno; register rtx dest = SET_DEST (x); if (dest == 0) return; while (GET_CODE (dest) == STRICT_LOW_PART || GET_CODE (dest) == SUBREG || GET_CODE (dest) == ZERO_EXTRACT || GET_CODE (dest) == SIGN_EXTRACT) { if (GET_CODE (dest) == ZERO_EXTRACT || GET_CODE (dest) == SIGN_EXTRACT) { /* The second and third arguments are values read by this insn. */ sched_analyze_2 (XEXP (dest, 1), insn); sched_analyze_2 (XEXP (dest, 2), insn); } dest = SUBREG_REG (dest); } if (GET_CODE (dest) == REG) { register int i; regno = REGNO (dest); /* A hard reg in a wide mode may really be multiple registers. If so, mark all of them just like the first. */ if (regno < FIRST_PSEUDO_REGISTER) { i = HARD_REGNO_NREGS (regno, GET_MODE (dest)); while (--i >= 0) { rtx u; for (u = reg_last_uses[regno+i]; u; u = XEXP (u, 1)) add_dependence (insn, XEXP (u, 0), REG_DEP_ANTI); reg_last_uses[regno + i] = 0; if (reg_last_sets[regno + i]) add_dependence (insn, reg_last_sets[regno + i], REG_DEP_OUTPUT); reg_pending_sets[(regno + i) / REGSET_ELT_BITS] |= (REGSET_ELT_TYPE) 1 << ((regno + i) % REGSET_ELT_BITS); if ((call_used_regs[i] || global_regs[i]) && last_function_call) /* Function calls clobber all call_used regs. */ add_dependence (insn, last_function_call, REG_DEP_ANTI); } } else { rtx u; for (u = reg_last_uses[regno]; u; u = XEXP (u, 1)) add_dependence (insn, XEXP (u, 0), REG_DEP_ANTI); reg_last_uses[regno] = 0; if (reg_last_sets[regno]) add_dependence (insn, reg_last_sets[regno], REG_DEP_OUTPUT); reg_pending_sets[regno / REGSET_ELT_BITS] |= (REGSET_ELT_TYPE) 1 << (regno % REGSET_ELT_BITS); /* Pseudos that are REG_EQUIV to something may be replaced by that during reloading. We need only add dependencies for the address in the REG_EQUIV note. */ if (! reload_completed && reg_known_equiv_p[regno] && GET_CODE (reg_known_value[regno]) == MEM) sched_analyze_2 (XEXP (reg_known_value[regno], 0), insn); /* Don't let it cross a call after scheduling if it doesn't already cross one. */ if (reg_n_calls_crossed[regno] == 0 && last_function_call) add_dependence (insn, last_function_call, REG_DEP_ANTI); } } else if (GET_CODE (dest) == MEM) { /* Writing memory. */ if (pending_lists_length > 32) { /* Flush all pending reads and writes to prevent the pending lists from getting any larger. Insn scheduling runs too slowly when these lists get long. The number 32 was chosen because it seems like a reasonable number. When compiling GCC with itself, this flush occurs 8 times for sparc, and 10 times for m88k using the number 32. */ flush_pending_lists (insn); } else { rtx pending, pending_mem; pending = pending_read_insns; pending_mem = pending_read_mems; while (pending) { /* If a dependency already exists, don't create a new one. */ if (! find_insn_list (XEXP (pending, 0), LOG_LINKS (insn))) if (anti_dependence (XEXP (pending_mem, 0), dest)) add_dependence (insn, XEXP (pending, 0), REG_DEP_ANTI); pending = XEXP (pending, 1); pending_mem = XEXP (pending_mem, 1); } pending = pending_write_insns; pending_mem = pending_write_mems; while (pending) { /* If a dependency already exists, don't create a new one. */ if (! find_insn_list (XEXP (pending, 0), LOG_LINKS (insn))) if (output_dependence (XEXP (pending_mem, 0), dest)) add_dependence (insn, XEXP (pending, 0), REG_DEP_OUTPUT); pending = XEXP (pending, 1); pending_mem = XEXP (pending_mem, 1); } if (last_pending_memory_flush) add_dependence (insn, last_pending_memory_flush, REG_DEP_ANTI); add_insn_mem_dependence (&pending_write_insns, &pending_write_mems, insn, dest); } sched_analyze_2 (XEXP (dest, 0), insn); } /* Analyze reads. */ if (GET_CODE (x) == SET) sched_analyze_2 (SET_SRC (x), insn); } /* Analyze the uses of memory and registers in rtx X in INSN. */ static void sched_analyze_2 (x, insn) rtx x; rtx insn; { register int i; register int j; register enum rtx_code code; register char *fmt; if (x == 0) return; code = GET_CODE (x); switch (code) { case CONST_INT: case CONST_DOUBLE: case SYMBOL_REF: case CONST: case LABEL_REF: /* Ignore constants. Note that we must handle CONST_DOUBLE here because it may have a cc0_rtx in its CONST_DOUBLE_CHAIN field, but this does not mean that this insn is using cc0. */ return; #ifdef HAVE_cc0 case CC0: { rtx link, prev; /* There may be a note before this insn now, but all notes will be removed before we actually try to schedule the insns, so it won't cause a problem later. We must avoid it here though. */ /* User of CC0 depends on immediately preceding insn. */ SCHED_GROUP_P (insn) = 1; /* Make a copy of all dependencies on the immediately previous insn, and add to this insn. This is so that all the dependencies will apply to the group. Remove an explicit dependence on this insn as SCHED_GROUP_P now represents it. */ prev = PREV_INSN (insn); while (GET_CODE (prev) == NOTE) prev = PREV_INSN (prev); if (find_insn_list (prev, LOG_LINKS (insn))) remove_dependence (insn, prev); for (link = LOG_LINKS (prev); link; link = XEXP (link, 1)) add_dependence (insn, XEXP (link, 0), REG_NOTE_KIND (link)); return; } #endif case REG: { int regno = REGNO (x); if (regno < FIRST_PSEUDO_REGISTER) { int i; i = HARD_REGNO_NREGS (regno, GET_MODE (x)); while (--i >= 0) { reg_last_uses[regno + i] = gen_rtx (INSN_LIST, VOIDmode, insn, reg_last_uses[regno + i]); if (reg_last_sets[regno + i]) add_dependence (insn, reg_last_sets[regno + i], 0); if ((call_used_regs[regno + i] || global_regs[regno + i]) && last_function_call) /* Function calls clobber all call_used regs. */ add_dependence (insn, last_function_call, REG_DEP_ANTI); } } else { reg_last_uses[regno] = gen_rtx (INSN_LIST, VOIDmode, insn, reg_last_uses[regno]); if (reg_last_sets[regno]) add_dependence (insn, reg_last_sets[regno], 0); /* Pseudos that are REG_EQUIV to something may be replaced by that during reloading. We need only add dependencies for the address in the REG_EQUIV note. */ if (! reload_completed && reg_known_equiv_p[regno] && GET_CODE (reg_known_value[regno]) == MEM) sched_analyze_2 (XEXP (reg_known_value[regno], 0), insn); /* If the register does not already cross any calls, then add this insn to the sched_before_next_call list so that it will still not cross calls after scheduling. */ if (reg_n_calls_crossed[regno] == 0) add_dependence (sched_before_next_call, insn, REG_DEP_ANTI); } return; } case MEM: { /* Reading memory. */ rtx pending, pending_mem; pending = pending_read_insns; pending_mem = pending_read_mems; while (pending) { /* If a dependency already exists, don't create a new one. */ if (! find_insn_list (XEXP (pending, 0), LOG_LINKS (insn))) if (read_dependence (XEXP (pending_mem, 0), x)) add_dependence (insn, XEXP (pending, 0), REG_DEP_ANTI); pending = XEXP (pending, 1); pending_mem = XEXP (pending_mem, 1); } pending = pending_write_insns; pending_mem = pending_write_mems; while (pending) { /* If a dependency already exists, don't create a new one. */ if (! find_insn_list (XEXP (pending, 0), LOG_LINKS (insn))) if (true_dependence (XEXP (pending_mem, 0), x)) add_dependence (insn, XEXP (pending, 0), 0); pending = XEXP (pending, 1); pending_mem = XEXP (pending_mem, 1); } if (last_pending_memory_flush) add_dependence (insn, last_pending_memory_flush, REG_DEP_ANTI); /* Always add these dependencies to pending_reads, since this insn may be followed by a write. */ add_insn_mem_dependence (&pending_read_insns, &pending_read_mems, insn, x); /* Take advantage of tail recursion here. */ sched_analyze_2 (XEXP (x, 0), insn); return; } case ASM_OPERANDS: case ASM_INPUT: case UNSPEC_VOLATILE: case TRAP_IF: { rtx u; /* Traditional and volatile asm instructions must be considered to use and clobber all hard registers, all pseudo-registers and all of memory. So must TRAP_IF and UNSPEC_VOLATILE operations. Consider for instance a volatile asm that changes the fpu rounding mode. An insn should not be moved across this even if it only uses pseudo-regs because it might give an incorrectly rounded result. */ if (code != ASM_OPERANDS || MEM_VOLATILE_P (x)) { int max_reg = max_reg_num (); for (i = 0; i < max_reg; i++) { for (u = reg_last_uses[i]; u; u = XEXP (u, 1)) add_dependence (insn, XEXP (u, 0), REG_DEP_ANTI); reg_last_uses[i] = 0; if (reg_last_sets[i]) add_dependence (insn, reg_last_sets[i], 0); } reg_pending_sets_all = 1; flush_pending_lists (insn); } /* For all ASM_OPERANDS, we must traverse the vector of input operands. We can not just fall through here since then we would be confused by the ASM_INPUT rtx inside ASM_OPERANDS, which do not indicate traditional asms unlike their normal usage. */ if (code == ASM_OPERANDS) { for (j = 0; j < ASM_OPERANDS_INPUT_LENGTH (x); j++) sched_analyze_2 (ASM_OPERANDS_INPUT (x, j), insn); return; } break; } case PRE_DEC: case POST_DEC: case PRE_INC: case POST_INC: /* These both read and modify the result. We must handle them as writes to get proper dependencies for following instructions. We must handle them as reads to get proper dependencies from this to previous instructions. Thus we need to pass them to both sched_analyze_1 and sched_analyze_2. We must call sched_analyze_2 first in order to get the proper antecedent for the read. */ sched_analyze_2 (XEXP (x, 0), insn); sched_analyze_1 (x, insn); return; } /* Other cases: walk the insn. */ fmt = GET_RTX_FORMAT (code); for (i = GET_RTX_LENGTH (code) - 1; i >= 0; i--) { if (fmt[i] == 'e') sched_analyze_2 (XEXP (x, i), insn); else if (fmt[i] == 'E') for (j = 0; j < XVECLEN (x, i); j++) sched_analyze_2 (XVECEXP (x, i, j), insn); } } /* Analyze an INSN with pattern X to find all dependencies. */ static void sched_analyze_insn (x, insn, loop_notes) rtx x, insn; rtx loop_notes; { register RTX_CODE code = GET_CODE (x); rtx link; int maxreg = max_reg_num (); int i; if (code == SET || code == CLOBBER) sched_analyze_1 (x, insn); else if (code == PARALLEL) { register int i; for (i = XVECLEN (x, 0) - 1; i >= 0; i--) { code = GET_CODE (XVECEXP (x, 0, i)); if (code == SET || code == CLOBBER) sched_analyze_1 (XVECEXP (x, 0, i), insn); else sched_analyze_2 (XVECEXP (x, 0, i), insn); } } else sched_analyze_2 (x, insn); /* Mark registers CLOBBERED or used by called function. */ if (GET_CODE (insn) == CALL_INSN) for (link = CALL_INSN_FUNCTION_USAGE (insn); link; link = XEXP (link, 1)) { if (GET_CODE (XEXP (link, 0)) == CLOBBER) sched_analyze_1 (XEXP (link, 0), insn); else sched_analyze_2 (XEXP (link, 0), insn); } /* If there is a LOOP_{BEG,END} note in the middle of a basic block, then we must be sure that no instructions are scheduled across it. Otherwise, the reg_n_refs info (which depends on loop_depth) would become incorrect. */ if (loop_notes) { int max_reg = max_reg_num (); rtx link; for (i = 0; i < max_reg; i++) { rtx u; for (u = reg_last_uses[i]; u; u = XEXP (u, 1)) add_dependence (insn, XEXP (u, 0), REG_DEP_ANTI); reg_last_uses[i] = 0; if (reg_last_sets[i]) add_dependence (insn, reg_last_sets[i], 0); } reg_pending_sets_all = 1; flush_pending_lists (insn); link = loop_notes; while (XEXP (link, 1)) link = XEXP (link, 1); XEXP (link, 1) = REG_NOTES (insn); REG_NOTES (insn) = loop_notes; } /* After reload, it is possible for an instruction to have a REG_DEAD note for a register that actually dies a few instructions earlier. For example, this can happen with SECONDARY_MEMORY_NEEDED reloads. In this case, we must consider the insn to use the register mentioned in the REG_DEAD note. Otherwise, we may accidentally move this insn after another insn that sets the register, thus getting obviously invalid rtl. This confuses reorg which believes that REG_DEAD notes are still meaningful. ??? We would get better code if we fixed reload to put the REG_DEAD notes in the right places, but that may not be worth the effort. */ if (reload_completed) { rtx note; for (note = REG_NOTES (insn); note; note = XEXP (note, 1)) if (REG_NOTE_KIND (note) == REG_DEAD) sched_analyze_2 (XEXP (note, 0), insn); } for (i = 0; i < regset_size; i++) { REGSET_ELT_TYPE sets = reg_pending_sets[i]; if (sets) { register int bit; for (bit = 0; bit < REGSET_ELT_BITS; bit++) if (sets & ((REGSET_ELT_TYPE) 1 << bit)) reg_last_sets[i * REGSET_ELT_BITS + bit] = insn; reg_pending_sets[i] = 0; } } if (reg_pending_sets_all) { for (i = 0; i < maxreg; i++) reg_last_sets[i] = insn; reg_pending_sets_all = 0; } /* Handle function calls and function returns created by the epilogue threading code. */ if (GET_CODE (insn) == CALL_INSN || GET_CODE (insn) == JUMP_INSN) { rtx dep_insn; rtx prev_dep_insn; /* When scheduling instructions, we make sure calls don't lose their accompanying USE insns by depending them one on another in order. Also, we must do the same thing for returns created by the epilogue threading code. Note this code works only in this special case, because other passes make no guarantee that they will never emit an instruction between a USE and a RETURN. There is such a guarantee for USE instructions immediately before a call. */ prev_dep_insn = insn; dep_insn = PREV_INSN (insn); while (GET_CODE (dep_insn) == INSN && GET_CODE (PATTERN (dep_insn)) == USE && GET_CODE (XEXP (PATTERN (dep_insn), 0)) == REG) { SCHED_GROUP_P (prev_dep_insn) = 1; /* Make a copy of all dependencies on dep_insn, and add to insn. This is so that all of the dependencies will apply to the group. */ for (link = LOG_LINKS (dep_insn); link; link = XEXP (link, 1)) add_dependence (insn, XEXP (link, 0), REG_NOTE_KIND (link)); prev_dep_insn = dep_insn; dep_insn = PREV_INSN (dep_insn); } } } /* Analyze every insn between HEAD and TAIL inclusive, creating LOG_LINKS for every dependency. */ static int sched_analyze (head, tail) rtx head, tail; { register rtx insn; register int n_insns = 0; register rtx u; register int luid = 0; rtx loop_notes = 0; for (insn = head; ; insn = NEXT_INSN (insn)) { INSN_LUID (insn) = luid++; if (GET_CODE (insn) == INSN || GET_CODE (insn) == JUMP_INSN) { sched_analyze_insn (PATTERN (insn), insn, loop_notes); loop_notes = 0; n_insns += 1; } else if (GET_CODE (insn) == CALL_INSN) { rtx x; register int i; /* Any instruction using a hard register which may get clobbered by a call needs to be marked as dependent on this call. This prevents a use of a hard return reg from being moved past a void call (i.e. it does not explicitly set the hard return reg). */ /* If this call is followed by a NOTE_INSN_SETJMP, then assume that all registers, not just hard registers, may be clobbered by this call. */ /* Insn, being a CALL_INSN, magically depends on `last_function_call' already. */ if (NEXT_INSN (insn) && GET_CODE (NEXT_INSN (insn)) == NOTE && NOTE_LINE_NUMBER (NEXT_INSN (insn)) == NOTE_INSN_SETJMP) { int max_reg = max_reg_num (); for (i = 0; i < max_reg; i++) { for (u = reg_last_uses[i]; u; u = XEXP (u, 1)) add_dependence (insn, XEXP (u, 0), REG_DEP_ANTI); reg_last_uses[i] = 0; if (reg_last_sets[i]) add_dependence (insn, reg_last_sets[i], 0); } reg_pending_sets_all = 1; /* Add a fake REG_NOTE which we will later convert back into a NOTE_INSN_SETJMP note. */ REG_NOTES (insn) = gen_rtx (EXPR_LIST, REG_DEAD, GEN_INT (NOTE_INSN_SETJMP), REG_NOTES (insn)); } else { for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (call_used_regs[i] || global_regs[i]) { for (u = reg_last_uses[i]; u; u = XEXP (u, 1)) add_dependence (insn, XEXP (u, 0), REG_DEP_ANTI); reg_last_uses[i] = 0; if (reg_last_sets[i]) add_dependence (insn, reg_last_sets[i], REG_DEP_ANTI); reg_pending_sets[i / REGSET_ELT_BITS] |= (REGSET_ELT_TYPE) 1 << (i % REGSET_ELT_BITS); } } /* For each insn which shouldn't cross a call, add a dependence between that insn and this call insn. */ x = LOG_LINKS (sched_before_next_call); while (x) { add_dependence (insn, XEXP (x, 0), REG_DEP_ANTI); x = XEXP (x, 1); } LOG_LINKS (sched_before_next_call) = 0; sched_analyze_insn (PATTERN (insn), insn, loop_notes); loop_notes = 0; /* We don't need to flush memory for a function call which does not involve memory. */ if (! CONST_CALL_P (insn)) { /* In the absence of interprocedural alias analysis, we must flush all pending reads and writes, and start new dependencies starting from here. */ flush_pending_lists (insn); } /* Depend this function call (actually, the user of this function call) on all hard register clobberage. */ last_function_call = insn; n_insns += 1; } else if (GET_CODE (insn) == NOTE && (NOTE_LINE_NUMBER (insn) == NOTE_INSN_LOOP_BEG || NOTE_LINE_NUMBER (insn) == NOTE_INSN_LOOP_END)) loop_notes = gen_rtx (EXPR_LIST, REG_DEAD, GEN_INT (NOTE_LINE_NUMBER (insn)), loop_notes); if (insn == tail) return n_insns; } } /* Called when we see a set of a register. If death is true, then we are scanning backwards. Mark that register as unborn. If nobody says otherwise, that is how things will remain. If death is false, then we are scanning forwards. Mark that register as being born. */ static void sched_note_set (b, x, death) int b; rtx x; int death; { register int regno; register rtx reg = SET_DEST (x); int subreg_p = 0; if (reg == 0) return; while (GET_CODE (reg) == SUBREG || GET_CODE (reg) == STRICT_LOW_PART || GET_CODE (reg) == SIGN_EXTRACT || GET_CODE (reg) == ZERO_EXTRACT) { /* Must treat modification of just one hardware register of a multi-reg value or just a byte field of a register exactly the same way that mark_set_1 in flow.c does, i.e. anything except a paradoxical subreg does not kill the entire register. */ if (GET_CODE (reg) != SUBREG || REG_SIZE (SUBREG_REG (reg)) > REG_SIZE (reg)) subreg_p = 1; reg = SUBREG_REG (reg); } if (GET_CODE (reg) != REG) return; /* Global registers are always live, so the code below does not apply to them. */ regno = REGNO (reg); if (regno >= FIRST_PSEUDO_REGISTER || ! global_regs[regno]) { register int offset = regno / REGSET_ELT_BITS; register REGSET_ELT_TYPE bit = (REGSET_ELT_TYPE) 1 << (regno % REGSET_ELT_BITS); if (death) { /* If we only set part of the register, then this set does not kill it. */ if (subreg_p) return; /* Try killing this register. */ if (regno < FIRST_PSEUDO_REGISTER) { int j = HARD_REGNO_NREGS (regno, GET_MODE (reg)); while (--j >= 0) { offset = (regno + j) / REGSET_ELT_BITS; bit = (REGSET_ELT_TYPE) 1 << ((regno + j) % REGSET_ELT_BITS); bb_live_regs[offset] &= ~bit; bb_dead_regs[offset] |= bit; } } else { bb_live_regs[offset] &= ~bit; bb_dead_regs[offset] |= bit; } } else { /* Make the register live again. */ if (regno < FIRST_PSEUDO_REGISTER) { int j = HARD_REGNO_NREGS (regno, GET_MODE (reg)); while (--j >= 0) { offset = (regno + j) / REGSET_ELT_BITS; bit = (REGSET_ELT_TYPE) 1 << ((regno + j) % REGSET_ELT_BITS); bb_live_regs[offset] |= bit; bb_dead_regs[offset] &= ~bit; } } else { bb_live_regs[offset] |= bit; bb_dead_regs[offset] &= ~bit; } } } } /* Macros and functions for keeping the priority queue sorted, and dealing with queueing and unqueueing of instructions. */ #define SCHED_SORT(READY, NEW_READY, OLD_READY) \ do { if ((NEW_READY) - (OLD_READY) == 1) \ swap_sort (READY, NEW_READY); \ else if ((NEW_READY) - (OLD_READY) > 1) \ qsort (READY, NEW_READY, sizeof (rtx), rank_for_schedule); } \ while (0) /* Returns a positive value if y is preferred; returns a negative value if x is preferred. Should never return 0, since that will make the sort unstable. */ static int rank_for_schedule (x, y) rtx *x, *y; { rtx tmp = *y; rtx tmp2 = *x; rtx link; int tmp_class, tmp2_class; int value; /* Choose the instruction with the highest priority, if different. */ if (value = INSN_PRIORITY (tmp) - INSN_PRIORITY (tmp2)) return value; if (last_scheduled_insn) { /* Classify the instructions into three classes: 1) Data dependent on last schedule insn. 2) Anti/Output dependent on last scheduled insn. 3) Independent of last scheduled insn, or has latency of one. Choose the insn from the highest numbered class if different. */ link = find_insn_list (tmp, LOG_LINKS (last_scheduled_insn)); if (link == 0 || insn_cost (tmp, link, last_scheduled_insn) == 1) tmp_class = 3; else if (REG_NOTE_KIND (link) == 0) /* Data dependence. */ tmp_class = 1; else tmp_class = 2; link = find_insn_list (tmp2, LOG_LINKS (last_scheduled_insn)); if (link == 0 || insn_cost (tmp2, link, last_scheduled_insn) == 1) tmp2_class = 3; else if (REG_NOTE_KIND (link) == 0) /* Data dependence. */ tmp2_class = 1; else tmp2_class = 2; if (value = tmp_class - tmp2_class) return value; } /* If insns are equally good, sort by INSN_LUID (original insn order), so that we make the sort stable. This minimizes instruction movement, thus minimizing sched's effect on debugging and cross-jumping. */ return INSN_LUID (tmp) - INSN_LUID (tmp2); } /* Resort the array A in which only element at index N may be out of order. */ __inline static void swap_sort (a, n) rtx *a; int n; { rtx insn = a[n-1]; int i = n-2; while (i >= 0 && rank_for_schedule (a+i, &insn) >= 0) { a[i+1] = a[i]; i -= 1; } a[i+1] = insn; } static int max_priority; /* Add INSN to the insn queue so that it fires at least N_CYCLES before the currently executing insn. */ __inline static void queue_insn (insn, n_cycles) rtx insn; int n_cycles; { int next_q = NEXT_Q_AFTER (q_ptr, n_cycles); NEXT_INSN (insn) = insn_queue[next_q]; insn_queue[next_q] = insn; q_size += 1; } /* Return nonzero if PAT is the pattern of an insn which makes a register live. */ __inline static int birthing_insn_p (pat) rtx pat; { int j; if (reload_completed == 1) return 0; if (GET_CODE (pat) == SET && GET_CODE (SET_DEST (pat)) == REG) { rtx dest = SET_DEST (pat); int i = REGNO (dest); int offset = i / REGSET_ELT_BITS; REGSET_ELT_TYPE bit = (REGSET_ELT_TYPE) 1 << (i % REGSET_ELT_BITS); /* It would be more accurate to use refers_to_regno_p or reg_mentioned_p to determine when the dest is not live before this insn. */ if (bb_live_regs[offset] & bit) return (reg_n_sets[i] == 1); return 0; } if (GET_CODE (pat) == PARALLEL) { for (j = 0; j < XVECLEN (pat, 0); j++) if (birthing_insn_p (XVECEXP (pat, 0, j))) return 1; } return 0; } /* PREV is an insn that is ready to execute. Adjust its priority if that will help shorten register lifetimes. */ __inline static void adjust_priority (prev) rtx prev; { /* Trying to shorten register lives after reload has completed is useless and wrong. It gives inaccurate schedules. */ if (reload_completed == 0) { rtx note; int n_deaths = 0; /* ??? This code has no effect, because REG_DEAD notes are removed before we ever get here. */ for (note = REG_NOTES (prev); note; note = XEXP (note, 1)) if (REG_NOTE_KIND (note) == REG_DEAD) n_deaths += 1; /* Defer scheduling insns which kill registers, since that shortens register lives. Prefer scheduling insns which make registers live for the same reason. */ switch (n_deaths) { default: INSN_PRIORITY (prev) >>= 3; break; case 3: INSN_PRIORITY (prev) >>= 2; break; case 2: case 1: INSN_PRIORITY (prev) >>= 1; break; case 0: if (birthing_insn_p (PATTERN (prev))) { int max = max_priority; if (max > INSN_PRIORITY (prev)) INSN_PRIORITY (prev) = max; } break; } } } /* INSN is the "currently executing insn". Launch each insn which was waiting on INSN (in the backwards dataflow sense). READY is a vector of insns which are ready to fire. N_READY is the number of elements in READY. CLOCK is the current virtual cycle. */ static int schedule_insn (insn, ready, n_ready, clock) rtx insn; rtx *ready; int n_ready; int clock; { rtx link; int new_ready = n_ready; if (MAX_BLOCKAGE > 1) schedule_unit (insn_unit (insn), insn, clock); if (LOG_LINKS (insn) == 0) return n_ready; /* This is used by the function adjust_priority above. */ if (n_ready > 0) max_priority = MAX (INSN_PRIORITY (ready[0]), INSN_PRIORITY (insn)); else max_priority = INSN_PRIORITY (insn); for (link = LOG_LINKS (insn); link != 0; link = XEXP (link, 1)) { rtx prev = XEXP (link, 0); int cost = insn_cost (prev, link, insn); if ((INSN_REF_COUNT (prev) -= 1) != 0) { /* We satisfied one requirement to fire PREV. Record the earliest time when PREV can fire. No need to do this if the cost is 1, because PREV can fire no sooner than the next cycle. */ if (cost > 1) INSN_TICK (prev) = MAX (INSN_TICK (prev), clock + cost); } else { /* We satisfied the last requirement to fire PREV. Ensure that all timing requirements are satisfied. */ if (INSN_TICK (prev) - clock > cost) cost = INSN_TICK (prev) - clock; /* Adjust the priority of PREV and either put it on the ready list or queue it. */ adjust_priority (prev); if (cost <= 1) ready[new_ready++] = prev; else queue_insn (prev, cost); } } return new_ready; } /* Given N_READY insns in the ready list READY at time CLOCK, queue those that are blocked due to function unit hazards and rearrange the remaining ones to minimize subsequent function unit hazards. */ static int schedule_select (ready, n_ready, clock, file) rtx *ready; int n_ready, clock; FILE *file; { int pri = INSN_PRIORITY (ready[0]); int i, j, k, q, cost, best_cost, best_insn = 0, new_ready = n_ready; rtx insn; /* Work down the ready list in groups of instructions with the same priority value. Queue insns in the group that are blocked and select among those that remain for the one with the largest potential hazard. */ for (i = 0; i < n_ready; i = j) { int opri = pri; for (j = i + 1; j < n_ready; j++) if ((pri = INSN_PRIORITY (ready[j])) != opri) break; /* Queue insns in the group that are blocked. */ for (k = i, q = 0; k < j; k++) { insn = ready[k]; if ((cost = actual_hazard (insn_unit (insn), insn, clock, 0)) != 0) { q++; ready[k] = 0; queue_insn (insn, cost); if (file) fprintf (file, "\n;; blocking insn %d for %d cycles", INSN_UID (insn), cost); } } new_ready -= q; /* Check the next group if all insns were queued. */ if (j - i - q == 0) continue; /* If more than one remains, select the first one with the largest potential hazard. */ else if (j - i - q > 1) { best_cost = -1; for (k = i; k < j; k++) { if ((insn = ready[k]) == 0) continue; if ((cost = potential_hazard (insn_unit (insn), insn, 0)) > best_cost) { best_cost = cost; best_insn = k; } } } /* We have found a suitable insn to schedule. */ break; } /* Move the best insn to be front of the ready list. */ if (best_insn != 0) { if (file) { fprintf (file, ", now"); for (i = 0; i < n_ready; i++) if (ready[i]) fprintf (file, " %d", INSN_UID (ready[i])); fprintf (file, "\n;; insn %d has a greater potential hazard", INSN_UID (ready[best_insn])); } for (i = best_insn; i > 0; i--) { insn = ready[i-1]; ready[i-1] = ready[i]; ready[i] = insn; } } /* Compact the ready list. */ if (new_ready < n_ready) for (i = j = 0; i < n_ready; i++) if (ready[i]) ready[j++] = ready[i]; return new_ready; } /* Add a REG_DEAD note for REG to INSN, reusing a REG_DEAD note from the dead_notes list. */ static void create_reg_dead_note (reg, insn) rtx reg, insn; { rtx link; /* The number of registers killed after scheduling must be the same as the number of registers killed before scheduling. The number of REG_DEAD notes may not be conserved, i.e. two SImode hard register REG_DEAD notes might become one DImode hard register REG_DEAD note, but the number of registers killed will be conserved. We carefully remove REG_DEAD notes from the dead_notes list, so that there will be none left at the end. If we run out early, then there is a bug somewhere in flow, combine and/or sched. */ if (dead_notes == 0) { #if 1 abort (); #else link = rtx_alloc (EXPR_LIST); PUT_REG_NOTE_KIND (link, REG_DEAD); #endif } else { /* Number of regs killed by REG. */ int regs_killed = (REGNO (reg) >= FIRST_PSEUDO_REGISTER ? 1 : HARD_REGNO_NREGS (REGNO (reg), GET_MODE (reg))); /* Number of regs killed by REG_DEAD notes taken off the list. */ int reg_note_regs; link = dead_notes; reg_note_regs = (REGNO (XEXP (link, 0)) >= FIRST_PSEUDO_REGISTER ? 1 : HARD_REGNO_NREGS (REGNO (XEXP (link, 0)), GET_MODE (XEXP (link, 0)))); while (reg_note_regs < regs_killed) { link = XEXP (link, 1); reg_note_regs += (REGNO (XEXP (link, 0)) >= FIRST_PSEUDO_REGISTER ? 1 : HARD_REGNO_NREGS (REGNO (XEXP (link, 0)), GET_MODE (XEXP (link, 0)))); } dead_notes = XEXP (link, 1); /* If we took too many regs kills off, put the extra ones back. */ while (reg_note_regs > regs_killed) { rtx temp_reg, temp_link; temp_reg = gen_rtx (REG, word_mode, 0); temp_link = rtx_alloc (EXPR_LIST); PUT_REG_NOTE_KIND (temp_link, REG_DEAD); XEXP (temp_link, 0) = temp_reg; XEXP (temp_link, 1) = dead_notes; dead_notes = temp_link; reg_note_regs--; } } XEXP (link, 0) = reg; XEXP (link, 1) = REG_NOTES (insn); REG_NOTES (insn) = link; } /* Subroutine on attach_deaths_insn--handles the recursive search through INSN. If SET_P is true, then x is being modified by the insn. */ static void attach_deaths (x, insn, set_p) rtx x; rtx insn; int set_p; { register int i; register int j; register enum rtx_code code; register char *fmt; if (x == 0) return; code = GET_CODE (x); switch (code) { case CONST_INT: case CONST_DOUBLE: case LABEL_REF: case SYMBOL_REF: case CONST: case CODE_LABEL: case PC: case CC0: /* Get rid of the easy cases first. */ return; case REG: { /* If the register dies in this insn, queue that note, and mark this register as needing to die. */ /* This code is very similar to mark_used_1 (if set_p is false) and mark_set_1 (if set_p is true) in flow.c. */ register int regno = REGNO (x); register int offset = regno / REGSET_ELT_BITS; register REGSET_ELT_TYPE bit = (REGSET_ELT_TYPE) 1 << (regno % REGSET_ELT_BITS); REGSET_ELT_TYPE all_needed = (old_live_regs[offset] & bit); REGSET_ELT_TYPE some_needed = (old_live_regs[offset] & bit); if (set_p) return; if (regno < FIRST_PSEUDO_REGISTER) { int n; n = HARD_REGNO_NREGS (regno, GET_MODE (x)); while (--n > 0) { some_needed |= (old_live_regs[(regno + n) / REGSET_ELT_BITS] & ((REGSET_ELT_TYPE) 1 << ((regno + n) % REGSET_ELT_BITS))); all_needed &= (old_live_regs[(regno + n) / REGSET_ELT_BITS] & ((REGSET_ELT_TYPE) 1 << ((regno + n) % REGSET_ELT_BITS))); } } /* If it wasn't live before we started, then add a REG_DEAD note. We must check the previous lifetime info not the current info, because we may have to execute this code several times, e.g. once for a clobber (which doesn't add a note) and later for a use (which does add a note). Always make the register live. We must do this even if it was live before, because this may be an insn which sets and uses the same register, in which case the register has already been killed, so we must make it live again. Global registers are always live, and should never have a REG_DEAD note added for them, so none of the code below applies to them. */ if (regno >= FIRST_PSEUDO_REGISTER || ! global_regs[regno]) { /* Never add REG_DEAD notes for the FRAME_POINTER_REGNUM or the STACK_POINTER_REGNUM, since these are always considered to be live. Similarly for ARG_POINTER_REGNUM if it is fixed. */ if (regno != FRAME_POINTER_REGNUM #if HARD_FRAME_POINTER_REGNUM != FRAME_POINTER_REGNUM && ! (regno == HARD_FRAME_POINTER_REGNUM) #endif #if ARG_POINTER_REGNUM != FRAME_POINTER_REGNUM && ! (regno == ARG_POINTER_REGNUM && fixed_regs[regno]) #endif && regno != STACK_POINTER_REGNUM) { if (! all_needed && ! dead_or_set_p (insn, x)) { /* Check for the case where the register dying partially overlaps the register set by this insn. */ if (regno < FIRST_PSEUDO_REGISTER && HARD_REGNO_NREGS (regno, GET_MODE (x)) > 1) { int n = HARD_REGNO_NREGS (regno, GET_MODE (x)); while (--n >= 0) some_needed |= dead_or_set_regno_p (insn, regno + n); } /* If none of the words in X is needed, make a REG_DEAD note. Otherwise, we must make partial REG_DEAD notes. */ if (! some_needed) create_reg_dead_note (x, insn); else { int i; /* Don't make a REG_DEAD note for a part of a register that is set in the insn. */ for (i = HARD_REGNO_NREGS (regno, GET_MODE (x)) - 1; i >= 0; i--) if ((old_live_regs[(regno + i) / REGSET_ELT_BITS] & ((REGSET_ELT_TYPE) 1 << ((regno +i) % REGSET_ELT_BITS))) == 0 && ! dead_or_set_regno_p (insn, regno + i)) create_reg_dead_note (gen_rtx (REG, reg_raw_mode[regno + i], regno + i), insn); } } } if (regno < FIRST_PSEUDO_REGISTER) { int j = HARD_REGNO_NREGS (regno, GET_MODE (x)); while (--j >= 0) { offset = (regno + j) / REGSET_ELT_BITS; bit = (REGSET_ELT_TYPE) 1 << ((regno + j) % REGSET_ELT_BITS); bb_dead_regs[offset] &= ~bit; bb_live_regs[offset] |= bit; } } else { bb_dead_regs[offset] &= ~bit; bb_live_regs[offset] |= bit; } } return; } case MEM: /* Handle tail-recursive case. */ attach_deaths (XEXP (x, 0), insn, 0); return; case SUBREG: case STRICT_LOW_PART: /* These two cases preserve the value of SET_P, so handle them separately. */ attach_deaths (XEXP (x, 0), insn, set_p); return; case ZERO_EXTRACT: case SIGN_EXTRACT: /* This case preserves the value of SET_P for the first operand, but clears it for the other two. */ attach_deaths (XEXP (x, 0), insn, set_p); attach_deaths (XEXP (x, 1), insn, 0); attach_deaths (XEXP (x, 2), insn, 0); return; default: /* Other cases: walk the insn. */ fmt = GET_RTX_FORMAT (code); for (i = GET_RTX_LENGTH (code) - 1; i >= 0; i--) { if (fmt[i] == 'e') attach_deaths (XEXP (x, i), insn, 0); else if (fmt[i] == 'E') for (j = 0; j < XVECLEN (x, i); j++) attach_deaths (XVECEXP (x, i, j), insn, 0); } } } /* After INSN has executed, add register death notes for each register that is dead after INSN. */ static void attach_deaths_insn (insn) rtx insn; { rtx x = PATTERN (insn); register RTX_CODE code = GET_CODE (x); rtx link; if (code == SET) { attach_deaths (SET_SRC (x), insn, 0); /* A register might die here even if it is the destination, e.g. it is the target of a volatile read and is otherwise unused. Hence we must always call attach_deaths for the SET_DEST. */ attach_deaths (SET_DEST (x), insn, 1); } else if (code == PARALLEL) { register int i; for (i = XVECLEN (x, 0) - 1; i >= 0; i--) { code = GET_CODE (XVECEXP (x, 0, i)); if (code == SET) { attach_deaths (SET_SRC (XVECEXP (x, 0, i)), insn, 0); attach_deaths (SET_DEST (XVECEXP (x, 0, i)), insn, 1); } /* Flow does not add REG_DEAD notes to registers that die in clobbers, so we can't either. */ else if (code != CLOBBER) attach_deaths (XVECEXP (x, 0, i), insn, 0); } } /* If this is a CLOBBER, only add REG_DEAD notes to registers inside a MEM being clobbered, just like flow. */ else if (code == CLOBBER && GET_CODE (XEXP (x, 0)) == MEM) attach_deaths (XEXP (XEXP (x, 0), 0), insn, 0); /* Otherwise don't add a death note to things being clobbered. */ else if (code != CLOBBER) attach_deaths (x, insn, 0); /* Make death notes for things used in the called function. */ if (GET_CODE (insn) == CALL_INSN) for (link = CALL_INSN_FUNCTION_USAGE (insn); link; link = XEXP (link, 1)) attach_deaths (XEXP (XEXP (link, 0), 0), insn, GET_CODE (XEXP (link, 0)) == CLOBBER); } /* Delete notes beginning with INSN and maybe put them in the chain of notes ended by NOTE_LIST. Returns the insn following the notes. */ static rtx unlink_notes (insn, tail) rtx insn, tail; { rtx prev = PREV_INSN (insn); while (insn != tail && GET_CODE (insn) == NOTE) { rtx next = NEXT_INSN (insn); /* Delete the note from its current position. */ if (prev) NEXT_INSN (prev) = next; if (next) PREV_INSN (next) = prev; if (write_symbols != NO_DEBUG && NOTE_LINE_NUMBER (insn) > 0) /* Record line-number notes so they can be reused. */ LINE_NOTE (insn) = insn; /* Don't save away NOTE_INSN_SETJMPs, because they must remain immediately after the call they follow. We use a fake (REG_DEAD (const_int -1)) note to remember them. Likewise with NOTE_INSN_LOOP_BEG and NOTE_INSN_LOOP_END. */ else if (NOTE_LINE_NUMBER (insn) != NOTE_INSN_SETJMP && NOTE_LINE_NUMBER (insn) != NOTE_INSN_LOOP_BEG && NOTE_LINE_NUMBER (insn) != NOTE_INSN_LOOP_END) { /* Insert the note at the end of the notes list. */ PREV_INSN (insn) = note_list; if (note_list) NEXT_INSN (note_list) = insn; note_list = insn; } insn = next; } return insn; } /* Constructor for `sometimes' data structure. */ static int new_sometimes_live (regs_sometimes_live, offset, bit, sometimes_max) struct sometimes *regs_sometimes_live; int offset, bit; int sometimes_max; { register struct sometimes *p; register int regno = offset * REGSET_ELT_BITS + bit; /* There should never be a register greater than max_regno here. If there is, it means that a define_split has created a new pseudo reg. This is not allowed, since there will not be flow info available for any new register, so catch the error here. */ if (regno >= max_regno) abort (); p = ®s_sometimes_live[sometimes_max]; p->offset = offset; p->bit = bit; p->live_length = 0; p->calls_crossed = 0; sometimes_max++; return sometimes_max; } /* Count lengths of all regs we are currently tracking, and find new registers no longer live. */ static void finish_sometimes_live (regs_sometimes_live, sometimes_max) struct sometimes *regs_sometimes_live; int sometimes_max; { int i; for (i = 0; i < sometimes_max; i++) { register struct sometimes *p = ®s_sometimes_live[i]; int regno; regno = p->offset * REGSET_ELT_BITS + p->bit; sched_reg_live_length[regno] += p->live_length; sched_reg_n_calls_crossed[regno] += p->calls_crossed; } } /* Use modified list scheduling to rearrange insns in basic block B. FILE, if nonzero, is where we dump interesting output about this pass. */ static void schedule_block (b, file) int b; FILE *file; { rtx insn, last; rtx *ready, link; int i, j, n_ready = 0, new_ready, n_insns = 0; int sched_n_insns = 0; int clock; #define NEED_NOTHING 0 #define NEED_HEAD 1 #define NEED_TAIL 2 int new_needs; /* HEAD and TAIL delimit the region being scheduled. */ rtx head = basic_block_head[b]; rtx tail = basic_block_end[b]; /* PREV_HEAD and NEXT_TAIL are the boundaries of the insns being scheduled. When the insns have been ordered, these insns delimit where the new insns are to be spliced back into the insn chain. */ rtx next_tail; rtx prev_head; /* Keep life information accurate. */ register struct sometimes *regs_sometimes_live; int sometimes_max; if (file) fprintf (file, ";;\t -- basic block number %d from %d to %d --\n", b, INSN_UID (basic_block_head[b]), INSN_UID (basic_block_end[b])); i = max_reg_num (); reg_last_uses = (rtx *) alloca (i * sizeof (rtx)); bzero ((char *) reg_last_uses, i * sizeof (rtx)); reg_last_sets = (rtx *) alloca (i * sizeof (rtx)); bzero ((char *) reg_last_sets, i * sizeof (rtx)); reg_pending_sets = (regset) alloca (regset_bytes); bzero ((char *) reg_pending_sets, regset_bytes); reg_pending_sets_all = 0; clear_units (); /* Remove certain insns at the beginning from scheduling, by advancing HEAD. */ /* At the start of a function, before reload has run, don't delay getting parameters from hard registers into pseudo registers. */ if (reload_completed == 0 && b == 0) { while (head != tail && GET_CODE (head) == NOTE && NOTE_LINE_NUMBER (head) != NOTE_INSN_FUNCTION_BEG) head = NEXT_INSN (head); while (head != tail && GET_CODE (head) == INSN && GET_CODE (PATTERN (head)) == SET) { rtx src = SET_SRC (PATTERN (head)); while (GET_CODE (src) == SUBREG || GET_CODE (src) == SIGN_EXTEND || GET_CODE (src) == ZERO_EXTEND || GET_CODE (src) == SIGN_EXTRACT || GET_CODE (src) == ZERO_EXTRACT) src = XEXP (src, 0); if (GET_CODE (src) != REG || REGNO (src) >= FIRST_PSEUDO_REGISTER) break; /* Keep this insn from ever being scheduled. */ INSN_REF_COUNT (head) = 1; head = NEXT_INSN (head); } } /* Don't include any notes or labels at the beginning of the basic block, or notes at the ends of basic blocks. */ while (head != tail) { if (GET_CODE (head) == NOTE) head = NEXT_INSN (head); else if (GET_CODE (tail) == NOTE) tail = PREV_INSN (tail); else if (GET_CODE (head) == CODE_LABEL) head = NEXT_INSN (head); else break; } /* If the only insn left is a NOTE or a CODE_LABEL, then there is no need to schedule this block. */ if (head == tail && (GET_CODE (head) == NOTE || GET_CODE (head) == CODE_LABEL)) return; #if 0 /* This short-cut doesn't work. It does not count call insns crossed by registers in reg_sometimes_live. It does not mark these registers as dead if they die in this block. It does not mark these registers live (or create new reg_sometimes_live entries if necessary) if they are born in this block. The easy solution is to just always schedule a block. This block only has one insn, so this won't slow down this pass by much. */ if (head == tail) return; #endif /* Now HEAD through TAIL are the insns actually to be rearranged; Let PREV_HEAD and NEXT_TAIL enclose them. */ prev_head = PREV_INSN (head); next_tail = NEXT_INSN (tail); /* Initialize basic block data structures. */ dead_notes = 0; pending_read_insns = 0; pending_read_mems = 0; pending_write_insns = 0; pending_write_mems = 0; pending_lists_length = 0; last_pending_memory_flush = 0; last_function_call = 0; last_scheduled_insn = 0; LOG_LINKS (sched_before_next_call) = 0; n_insns += sched_analyze (head, tail); if (n_insns == 0) { free_pending_lists (); return; } /* Allocate vector to hold insns to be rearranged (except those insns which are controlled by an insn with SCHED_GROUP_P set). All these insns are included between ORIG_HEAD and ORIG_TAIL, as those variables ultimately are set up. */ ready = (rtx *) alloca ((n_insns+1) * sizeof (rtx)); /* TAIL is now the last of the insns to be rearranged. Put those insns into the READY vector. */ insn = tail; /* For all branches, calls, uses, and cc0 setters, force them to remain in order at the end of the block by adding dependencies and giving the last a high priority. There may be notes present, and prev_head may also be a note. Branches must obviously remain at the end. Calls should remain at the end since moving them results in worse register allocation. Uses remain at the end to ensure proper register allocation. cc0 setters remaim at the end because they can't be moved away from their cc0 user. */ last = 0; while (GET_CODE (insn) == CALL_INSN || GET_CODE (insn) == JUMP_INSN || (GET_CODE (insn) == INSN && (GET_CODE (PATTERN (insn)) == USE #ifdef HAVE_cc0 || sets_cc0_p (PATTERN (insn)) #endif )) || GET_CODE (insn) == NOTE) { if (GET_CODE (insn) != NOTE) { priority (insn); if (last == 0) { ready[n_ready++] = insn; INSN_PRIORITY (insn) = TAIL_PRIORITY - i; INSN_REF_COUNT (insn) = 0; } else if (! find_insn_list (insn, LOG_LINKS (last))) { add_dependence (last, insn, REG_DEP_ANTI); INSN_REF_COUNT (insn)++; } last = insn; /* Skip over insns that are part of a group. */ while (SCHED_GROUP_P (insn)) { insn = prev_nonnote_insn (insn); priority (insn); } } insn = PREV_INSN (insn); /* Don't overrun the bounds of the basic block. */ if (insn == prev_head) break; } /* Assign priorities to instructions. Also check whether they are in priority order already. If so then I will be nonnegative. We use this shortcut only before reloading. */ #if 0 i = reload_completed ? DONE_PRIORITY : MAX_PRIORITY; #endif for (; insn != prev_head; insn = PREV_INSN (insn)) { if (GET_RTX_CLASS (GET_CODE (insn)) == 'i') { priority (insn); if (INSN_REF_COUNT (insn) == 0) { if (last == 0) ready[n_ready++] = insn; else { /* Make this dependent on the last of the instructions that must remain in order at the end of the block. */ add_dependence (last, insn, REG_DEP_ANTI); INSN_REF_COUNT (insn) = 1; } } if (SCHED_GROUP_P (insn)) { while (SCHED_GROUP_P (insn)) { insn = PREV_INSN (insn); while (GET_CODE (insn) == NOTE) insn = PREV_INSN (insn); priority (insn); } continue; } #if 0 if (i < 0) continue; if (INSN_PRIORITY (insn) < i) i = INSN_PRIORITY (insn); else if (INSN_PRIORITY (insn) > i) i = DONE_PRIORITY; #endif } } #if 0 /* This short-cut doesn't work. It does not count call insns crossed by registers in reg_sometimes_live. It does not mark these registers as dead if they die in this block. It does not mark these registers live (or create new reg_sometimes_live entries if necessary) if they are born in this block. The easy solution is to just always schedule a block. These blocks tend to be very short, so this doesn't slow down this pass by much. */ /* If existing order is good, don't bother to reorder. */ if (i != DONE_PRIORITY) { if (file) fprintf (file, ";; already scheduled\n"); if (reload_completed == 0) { for (i = 0; i < sometimes_max; i++) regs_sometimes_live[i].live_length += n_insns; finish_sometimes_live (regs_sometimes_live, sometimes_max); } free_pending_lists (); return; } #endif /* Scan all the insns to be scheduled, removing NOTE insns and register death notes. Line number NOTE insns end up in NOTE_LIST. Register death notes end up in DEAD_NOTES. Recreate the register life information for the end of this basic block. */ if (reload_completed == 0) { bcopy ((char *) basic_block_live_at_start[b], (char *) bb_live_regs, regset_bytes); bzero ((char *) bb_dead_regs, regset_bytes); if (b == 0) { /* This is the first block in the function. There may be insns before head that we can't schedule. We still need to examine them though for accurate register lifetime analysis. */ /* We don't want to remove any REG_DEAD notes as the code below does. */ for (insn = basic_block_head[b]; insn != head; insn = NEXT_INSN (insn)) if (GET_RTX_CLASS (GET_CODE (insn)) == 'i') { /* See if the register gets born here. */ /* We must check for registers being born before we check for registers dying. It is possible for a register to be born and die in the same insn, e.g. reading from a volatile memory location into an otherwise unused register. Such a register must be marked as dead after this insn. */ if (GET_CODE (PATTERN (insn)) == SET || GET_CODE (PATTERN (insn)) == CLOBBER) sched_note_set (b, PATTERN (insn), 0); else if (GET_CODE (PATTERN (insn)) == PARALLEL) { int j; for (j = XVECLEN (PATTERN (insn), 0) - 1; j >= 0; j--) if (GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == SET || GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == CLOBBER) sched_note_set (b, XVECEXP (PATTERN (insn), 0, j), 0); /* ??? This code is obsolete and should be deleted. It is harmless though, so we will leave it in for now. */ for (j = XVECLEN (PATTERN (insn), 0) - 1; j >= 0; j--) if (GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == USE) sched_note_set (b, XVECEXP (PATTERN (insn), 0, j), 0); } for (link = REG_NOTES (insn); link; link = XEXP (link, 1)) { if ((REG_NOTE_KIND (link) == REG_DEAD || REG_NOTE_KIND (link) == REG_UNUSED) /* Verify that the REG_NOTE has a legal value. */ && GET_CODE (XEXP (link, 0)) == REG) { register int regno = REGNO (XEXP (link, 0)); register int offset = regno / REGSET_ELT_BITS; register REGSET_ELT_TYPE bit = (REGSET_ELT_TYPE) 1 << (regno % REGSET_ELT_BITS); if (regno < FIRST_PSEUDO_REGISTER) { int j = HARD_REGNO_NREGS (regno, GET_MODE (XEXP (link, 0))); while (--j >= 0) { offset = (regno + j) / REGSET_ELT_BITS; bit = ((REGSET_ELT_TYPE) 1 << ((regno + j) % REGSET_ELT_BITS)); bb_live_regs[offset] &= ~bit; bb_dead_regs[offset] |= bit; } } else { bb_live_regs[offset] &= ~bit; bb_dead_regs[offset] |= bit; } } } } } } /* If debugging information is being produced, keep track of the line number notes for each insn. */ if (write_symbols != NO_DEBUG) { /* We must use the true line number for the first insn in the block that was computed and saved at the start of this pass. We can't use the current line number, because scheduling of the previous block may have changed the current line number. */ rtx line = line_note_head[b]; for (insn = basic_block_head[b]; insn != next_tail; insn = NEXT_INSN (insn)) if (GET_CODE (insn) == NOTE && NOTE_LINE_NUMBER (insn) > 0) line = insn; else LINE_NOTE (insn) = line; } for (insn = head; insn != next_tail; insn = NEXT_INSN (insn)) { rtx prev, next, link; /* Farm out notes. This is needed to keep the debugger from getting completely deranged. */ if (GET_CODE (insn) == NOTE) { prev = insn; insn = unlink_notes (insn, next_tail); if (prev == tail) abort (); if (prev == head) abort (); if (insn == next_tail) abort (); } if (reload_completed == 0 && GET_RTX_CLASS (GET_CODE (insn)) == 'i') { /* See if the register gets born here. */ /* We must check for registers being born before we check for registers dying. It is possible for a register to be born and die in the same insn, e.g. reading from a volatile memory location into an otherwise unused register. Such a register must be marked as dead after this insn. */ if (GET_CODE (PATTERN (insn)) == SET || GET_CODE (PATTERN (insn)) == CLOBBER) sched_note_set (b, PATTERN (insn), 0); else if (GET_CODE (PATTERN (insn)) == PARALLEL) { int j; for (j = XVECLEN (PATTERN (insn), 0) - 1; j >= 0; j--) if (GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == SET || GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == CLOBBER) sched_note_set (b, XVECEXP (PATTERN (insn), 0, j), 0); /* ??? This code is obsolete and should be deleted. It is harmless though, so we will leave it in for now. */ for (j = XVECLEN (PATTERN (insn), 0) - 1; j >= 0; j--) if (GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == USE) sched_note_set (b, XVECEXP (PATTERN (insn), 0, j), 0); } /* Need to know what registers this insn kills. */ for (prev = 0, link = REG_NOTES (insn); link; link = next) { next = XEXP (link, 1); if ((REG_NOTE_KIND (link) == REG_DEAD || REG_NOTE_KIND (link) == REG_UNUSED) /* Verify that the REG_NOTE has a legal value. */ && GET_CODE (XEXP (link, 0)) == REG) { register int regno = REGNO (XEXP (link, 0)); register int offset = regno / REGSET_ELT_BITS; register REGSET_ELT_TYPE bit = (REGSET_ELT_TYPE) 1 << (regno % REGSET_ELT_BITS); /* Only unlink REG_DEAD notes; leave REG_UNUSED notes alone. */ if (REG_NOTE_KIND (link) == REG_DEAD) { if (prev) XEXP (prev, 1) = next; else REG_NOTES (insn) = next; XEXP (link, 1) = dead_notes; dead_notes = link; } else prev = link; if (regno < FIRST_PSEUDO_REGISTER) { int j = HARD_REGNO_NREGS (regno, GET_MODE (XEXP (link, 0))); while (--j >= 0) { offset = (regno + j) / REGSET_ELT_BITS; bit = ((REGSET_ELT_TYPE) 1 << ((regno + j) % REGSET_ELT_BITS)); bb_live_regs[offset] &= ~bit; bb_dead_regs[offset] |= bit; } } else { bb_live_regs[offset] &= ~bit; bb_dead_regs[offset] |= bit; } } else prev = link; } } } if (reload_completed == 0) { /* Keep track of register lives. */ old_live_regs = (regset) alloca (regset_bytes); regs_sometimes_live = (struct sometimes *) alloca (max_regno * sizeof (struct sometimes)); sometimes_max = 0; /* Start with registers live at end. */ for (j = 0; j < regset_size; j++) { REGSET_ELT_TYPE live = bb_live_regs[j]; old_live_regs[j] = live; if (live) { register int bit; for (bit = 0; bit < REGSET_ELT_BITS; bit++) if (live & ((REGSET_ELT_TYPE) 1 << bit)) sometimes_max = new_sometimes_live (regs_sometimes_live, j, bit, sometimes_max); } } } SCHED_SORT (ready, n_ready, 1); if (file) { fprintf (file, ";; ready list initially:\n;; "); for (i = 0; i < n_ready; i++) fprintf (file, "%d ", INSN_UID (ready[i])); fprintf (file, "\n\n"); for (insn = head; insn != next_tail; insn = NEXT_INSN (insn)) if (INSN_PRIORITY (insn) > 0) fprintf (file, ";; insn[%4d]: priority = %4d, ref_count = %4d\n", INSN_UID (insn), INSN_PRIORITY (insn), INSN_REF_COUNT (insn)); } /* Now HEAD and TAIL are going to become disconnected entirely from the insn chain. */ tail = 0; /* Q_SIZE will always be zero here. */ q_ptr = 0; clock = 0; bzero ((char *) insn_queue, sizeof (insn_queue)); /* Now, perform list scheduling. */ /* Where we start inserting insns is after TAIL. */ last = next_tail; new_needs = (NEXT_INSN (prev_head) == basic_block_head[b] ? NEED_HEAD : NEED_NOTHING); if (PREV_INSN (next_tail) == basic_block_end[b]) new_needs |= NEED_TAIL; new_ready = n_ready; while (sched_n_insns < n_insns) { q_ptr = NEXT_Q (q_ptr); clock++; /* Add all pending insns that can be scheduled without stalls to the ready list. */ for (insn = insn_queue[q_ptr]; insn; insn = NEXT_INSN (insn)) { if (file) fprintf (file, ";; launching %d before %d with no stalls at T-%d\n", INSN_UID (insn), INSN_UID (last), clock); ready[new_ready++] = insn; q_size -= 1; } insn_queue[q_ptr] = 0; /* If there are no ready insns, stall until one is ready and add all of the pending insns at that point to the ready list. */ if (new_ready == 0) { register int stalls; for (stalls = 1; stalls < INSN_QUEUE_SIZE; stalls++) if (insn = insn_queue[NEXT_Q_AFTER (q_ptr, stalls)]) { for (; insn; insn = NEXT_INSN (insn)) { if (file) fprintf (file, ";; launching %d before %d with %d stalls at T-%d\n", INSN_UID (insn), INSN_UID (last), stalls, clock); ready[new_ready++] = insn; q_size -= 1; } insn_queue[NEXT_Q_AFTER (q_ptr, stalls)] = 0; break; } q_ptr = NEXT_Q_AFTER (q_ptr, stalls); clock += stalls; } /* There should be some instructions waiting to fire. */ if (new_ready == 0) abort (); if (file) { fprintf (file, ";; ready list at T-%d:", clock); for (i = 0; i < new_ready; i++) fprintf (file, " %d (%x)", INSN_UID (ready[i]), INSN_PRIORITY (ready[i])); } /* Sort the ready list and choose the best insn to schedule. Select which insn should issue in this cycle and queue those that are blocked by function unit hazards. N_READY holds the number of items that were scheduled the last time, minus the one instruction scheduled on the last loop iteration; it is not modified for any other reason in this loop. */ SCHED_SORT (ready, new_ready, n_ready); if (MAX_BLOCKAGE > 1) { new_ready = schedule_select (ready, new_ready, clock, file); if (new_ready == 0) { if (file) fprintf (file, "\n"); /* We must set n_ready here, to ensure that sorting always occurs when we come back to the SCHED_SORT line above. */ n_ready = 0; continue; } } n_ready = new_ready; last_scheduled_insn = insn = ready[0]; /* The first insn scheduled becomes the new tail. */ if (tail == 0) tail = insn; if (file) { fprintf (file, ", now"); for (i = 0; i < n_ready; i++) fprintf (file, " %d", INSN_UID (ready[i])); fprintf (file, "\n"); } if (DONE_PRIORITY_P (insn)) abort (); if (reload_completed == 0) { /* Process this insn, and each insn linked to this one which must be immediately output after this insn. */ do { /* First we kill registers set by this insn, and then we make registers used by this insn live. This is the opposite order used above because we are traversing the instructions backwards. */ /* Strictly speaking, we should scan REG_UNUSED notes and make every register mentioned there live, however, we will just kill them again immediately below, so there doesn't seem to be any reason why we bother to do this. */ /* See if this is the last notice we must take of a register. */ if (GET_CODE (PATTERN (insn)) == SET || GET_CODE (PATTERN (insn)) == CLOBBER) sched_note_set (b, PATTERN (insn), 1); else if (GET_CODE (PATTERN (insn)) == PARALLEL) { int j; for (j = XVECLEN (PATTERN (insn), 0) - 1; j >= 0; j--) if (GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == SET || GET_CODE (XVECEXP (PATTERN (insn), 0, j)) == CLOBBER) sched_note_set (b, XVECEXP (PATTERN (insn), 0, j), 1); } /* This code keeps life analysis information up to date. */ if (GET_CODE (insn) == CALL_INSN) { register struct sometimes *p; /* A call kills all call used and global registers, except for those mentioned in the call pattern which will be made live again later. */ for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (call_used_regs[i] || global_regs[i]) { register int offset = i / REGSET_ELT_BITS; register REGSET_ELT_TYPE bit = (REGSET_ELT_TYPE) 1 << (i % REGSET_ELT_BITS); bb_live_regs[offset] &= ~bit; bb_dead_regs[offset] |= bit; } /* Regs live at the time of a call instruction must not go in a register clobbered by calls. Record this for all regs now live. Note that insns which are born or die in a call do not cross a call, so this must be done after the killings (above) and before the births (below). */ p = regs_sometimes_live; for (i = 0; i < sometimes_max; i++, p++) if (bb_live_regs[p->offset] & ((REGSET_ELT_TYPE) 1 << p->bit)) p->calls_crossed += 1; } /* Make every register used live, and add REG_DEAD notes for registers which were not live before we started. */ attach_deaths_insn (insn); /* Find registers now made live by that instruction. */ for (i = 0; i < regset_size; i++) { REGSET_ELT_TYPE diff = bb_live_regs[i] & ~old_live_regs[i]; if (diff) { register int bit; old_live_regs[i] |= diff; for (bit = 0; bit < REGSET_ELT_BITS; bit++) if (diff & ((REGSET_ELT_TYPE) 1 << bit)) sometimes_max = new_sometimes_live (regs_sometimes_live, i, bit, sometimes_max); } } /* Count lengths of all regs we are worrying about now, and handle registers no longer live. */ for (i = 0; i < sometimes_max; i++) { register struct sometimes *p = ®s_sometimes_live[i]; int regno = p->offset*REGSET_ELT_BITS + p->bit; p->live_length += 1; if ((bb_live_regs[p->offset] & ((REGSET_ELT_TYPE) 1 << p->bit)) == 0) { /* This is the end of one of this register's lifetime segments. Save the lifetime info collected so far, and clear its bit in the old_live_regs entry. */ sched_reg_live_length[regno] += p->live_length; sched_reg_n_calls_crossed[regno] += p->calls_crossed; old_live_regs[p->offset] &= ~((REGSET_ELT_TYPE) 1 << p->bit); /* Delete the reg_sometimes_live entry for this reg by copying the last entry over top of it. */ *p = regs_sometimes_live[--sometimes_max]; /* ...and decrement i so that this newly copied entry will be processed. */ i--; } } link = insn; insn = PREV_INSN (insn); } while (SCHED_GROUP_P (link)); /* Set INSN back to the insn we are scheduling now. */ insn = ready[0]; } /* Schedule INSN. Remove it from the ready list. */ ready += 1; n_ready -= 1; sched_n_insns += 1; NEXT_INSN (insn) = last; PREV_INSN (last) = insn; last = insn; /* Check to see if we need to re-emit any notes here. */ { rtx note; for (note = REG_NOTES (insn); note; note = XEXP (note, 1)) { if (REG_NOTE_KIND (note) == REG_DEAD && GET_CODE (XEXP (note, 0)) == CONST_INT) { if (INTVAL (XEXP (note, 0)) == NOTE_INSN_SETJMP) emit_note_after (INTVAL (XEXP (note, 0)), insn); else last = emit_note_before (INTVAL (XEXP (note, 0)), last); remove_note (insn, note); } } } /* Everything that precedes INSN now either becomes "ready", if it can execute immediately before INSN, or "pending", if there must be a delay. Give INSN high enough priority that at least one (maybe more) reg-killing insns can be launched ahead of all others. Mark INSN as scheduled by changing its priority to -1. */ INSN_PRIORITY (insn) = LAUNCH_PRIORITY; new_ready = schedule_insn (insn, ready, n_ready, clock); INSN_PRIORITY (insn) = DONE_PRIORITY; /* Schedule all prior insns that must not be moved. */ if (SCHED_GROUP_P (insn)) { /* Disable these insns from being launched. */ link = insn; while (SCHED_GROUP_P (link)) { /* Disable these insns from being launched by anybody. */ link = PREV_INSN (link); INSN_REF_COUNT (link) = 0; } /* None of these insns can move forward into delay slots. */ while (SCHED_GROUP_P (insn)) { insn = PREV_INSN (insn); new_ready = schedule_insn (insn, ready, new_ready, clock); INSN_PRIORITY (insn) = DONE_PRIORITY; sched_n_insns += 1; NEXT_INSN (insn) = last; PREV_INSN (last) = insn; last = insn; } } } if (q_size != 0) abort (); if (reload_completed == 0) finish_sometimes_live (regs_sometimes_live, sometimes_max); /* HEAD is now the first insn in the chain of insns that been scheduled by the loop above. TAIL is the last of those insns. */ head = insn; /* NOTE_LIST is the end of a chain of notes previously found among the insns. Insert them at the beginning of the insns. */ if (note_list != 0) { rtx note_head = note_list; while (PREV_INSN (note_head)) note_head = PREV_INSN (note_head); PREV_INSN (head) = note_list; NEXT_INSN (note_list) = head; head = note_head; } /* There should be no REG_DEAD notes leftover at the end. In practice, this can occur as the result of bugs in flow, combine.c, and/or sched.c. The values of the REG_DEAD notes remaining are meaningless, because dead_notes is just used as a free list. */ #if 1 if (dead_notes != 0) abort (); #endif if (new_needs & NEED_HEAD) basic_block_head[b] = head; PREV_INSN (head) = prev_head; NEXT_INSN (prev_head) = head; if (new_needs & NEED_TAIL) basic_block_end[b] = tail; NEXT_INSN (tail) = next_tail; PREV_INSN (next_tail) = tail; /* Restore the line-number notes of each insn. */ if (write_symbols != NO_DEBUG) { rtx line, note, prev, new; int notes = 0; head = basic_block_head[b]; next_tail = NEXT_INSN (basic_block_end[b]); /* Determine the current line-number. We want to know the current line number of the first insn of the block here, in case it is different from the true line number that was saved earlier. If different, then we need a line number note before the first insn of this block. If it happens to be the same, then we don't want to emit another line number note here. */ for (line = head; line; line = PREV_INSN (line)) if (GET_CODE (line) == NOTE && NOTE_LINE_NUMBER (line) > 0) break; /* Walk the insns keeping track of the current line-number and inserting the line-number notes as needed. */ for (insn = head; insn != next_tail; insn = NEXT_INSN (insn)) if (GET_CODE (insn) == NOTE && NOTE_LINE_NUMBER (insn) > 0) line = insn; /* This used to emit line number notes before every non-deleted note. However, this confuses a debugger, because line notes not separated by real instructions all end up at the same address. I can find no use for line number notes before other notes, so none are emitted. */ else if (GET_CODE (insn) != NOTE && (note = LINE_NOTE (insn)) != 0 && note != line && (line == 0 || NOTE_LINE_NUMBER (note) != NOTE_LINE_NUMBER (line) || NOTE_SOURCE_FILE (note) != NOTE_SOURCE_FILE (line))) { line = note; prev = PREV_INSN (insn); if (LINE_NOTE (note)) { /* Re-use the original line-number note. */ LINE_NOTE (note) = 0; PREV_INSN (note) = prev; NEXT_INSN (prev) = note; PREV_INSN (insn) = note; NEXT_INSN (note) = insn; } else { notes++; new = emit_note_after (NOTE_LINE_NUMBER (note), prev); NOTE_SOURCE_FILE (new) = NOTE_SOURCE_FILE (note); } } if (file && notes) fprintf (file, ";; added %d line-number notes\n", notes); } if (file) { fprintf (file, ";; total time = %d\n;; new basic block head = %d\n;; new basic block end = %d\n\n", clock, INSN_UID (basic_block_head[b]), INSN_UID (basic_block_end[b])); } /* Yow! We're done! */ free_pending_lists (); return; } /* Subroutine of split_hard_reg_notes. Searches X for any reference to REGNO, returning the rtx of the reference found if any. Otherwise, returns 0. */ static rtx regno_use_in (regno, x) int regno; rtx x; { register char *fmt; int i, j; rtx tem; if (GET_CODE (x) == REG && REGNO (x) == regno) return x; fmt = GET_RTX_FORMAT (GET_CODE (x)); for (i = GET_RTX_LENGTH (GET_CODE (x)) - 1; i >= 0; i--) { if (fmt[i] == 'e') { if (tem = regno_use_in (regno, XEXP (x, i))) return tem; } else if (fmt[i] == 'E') for (j = XVECLEN (x, i) - 1; j >= 0; j--) if (tem = regno_use_in (regno , XVECEXP (x, i, j))) return tem; } return 0; } /* Subroutine of update_flow_info. Determines whether any new REG_NOTEs are needed for the hard register mentioned in the note. This can happen if the reference to the hard register in the original insn was split into several smaller hard register references in the split insns. */ static void split_hard_reg_notes (note, first, last, orig_insn) rtx note, first, last, orig_insn; { rtx reg, temp, link; int n_regs, i, new_reg; rtx insn; /* Assume that this is a REG_DEAD note. */ if (REG_NOTE_KIND (note) != REG_DEAD) abort (); reg = XEXP (note, 0); n_regs = HARD_REGNO_NREGS (REGNO (reg), GET_MODE (reg)); for (i = 0; i < n_regs; i++) { new_reg = REGNO (reg) + i; /* Check for references to new_reg in the split insns. */ for (insn = last; ; insn = PREV_INSN (insn)) { if (GET_RTX_CLASS (GET_CODE (insn)) == 'i' && (temp = regno_use_in (new_reg, PATTERN (insn)))) { /* Create a new reg dead note here. */ link = rtx_alloc (EXPR_LIST); PUT_REG_NOTE_KIND (link, REG_DEAD); XEXP (link, 0) = temp; XEXP (link, 1) = REG_NOTES (insn); REG_NOTES (insn) = link; /* If killed multiple registers here, then add in the excess. */ i += HARD_REGNO_NREGS (REGNO (temp), GET_MODE (temp)) - 1; break; } /* It isn't mentioned anywhere, so no new reg note is needed for this register. */ if (insn == first) break; } } } /* Subroutine of update_flow_info. Determines whether a SET or CLOBBER in an insn created by splitting needs a REG_DEAD or REG_UNUSED note added. */ static void new_insn_dead_notes (pat, insn, last, orig_insn) rtx pat, insn, last, orig_insn; { rtx dest, tem, set; /* PAT is either a CLOBBER or a SET here. */ dest = XEXP (pat, 0); while (GET_CODE (dest) == ZERO_EXTRACT || GET_CODE (dest) == SUBREG || GET_CODE (dest) == STRICT_LOW_PART || GET_CODE (dest) == SIGN_EXTRACT) dest = XEXP (dest, 0); if (GET_CODE (dest) == REG) { for (tem = last; tem != insn; tem = PREV_INSN (tem)) { if (GET_RTX_CLASS (GET_CODE (tem)) == 'i' && reg_overlap_mentioned_p (dest, PATTERN (tem)) && (set = single_set (tem))) { rtx tem_dest = SET_DEST (set); while (GET_CODE (tem_dest) == ZERO_EXTRACT || GET_CODE (tem_dest) == SUBREG || GET_CODE (tem_dest) == STRICT_LOW_PART || GET_CODE (tem_dest) == SIGN_EXTRACT) tem_dest = XEXP (tem_dest, 0); if (! rtx_equal_p (tem_dest, dest)) { /* Use the same scheme as combine.c, don't put both REG_DEAD and REG_UNUSED notes on the same insn. */ if (! find_regno_note (tem, REG_UNUSED, REGNO (dest)) && ! find_regno_note (tem, REG_DEAD, REGNO (dest))) { rtx note = rtx_alloc (EXPR_LIST); PUT_REG_NOTE_KIND (note, REG_DEAD); XEXP (note, 0) = dest; XEXP (note, 1) = REG_NOTES (tem); REG_NOTES (tem) = note; } /* The reg only dies in one insn, the last one that uses it. */ break; } else if (reg_overlap_mentioned_p (dest, SET_SRC (set))) /* We found an instruction that both uses the register, and sets it, so no new REG_NOTE is needed for this set. */ break; } } /* If this is a set, it must die somewhere, unless it is the dest of the original insn, and hence is live after the original insn. Abort if it isn't supposed to be live after the original insn. If this is a clobber, then just add a REG_UNUSED note. */ if (tem == insn) { int live_after_orig_insn = 0; rtx pattern = PATTERN (orig_insn); int i; if (GET_CODE (pat) == CLOBBER) { rtx note = rtx_alloc (EXPR_LIST); PUT_REG_NOTE_KIND (note, REG_UNUSED); XEXP (note, 0) = dest; XEXP (note, 1) = REG_NOTES (insn); REG_NOTES (insn) = note; return; } /* The original insn could have multiple sets, so search the insn for all sets. */ if (GET_CODE (pattern) == SET) { if (reg_overlap_mentioned_p (dest, SET_DEST (pattern))) live_after_orig_insn = 1; } else if (GET_CODE (pattern) == PARALLEL) { for (i = 0; i < XVECLEN (pattern, 0); i++) if (GET_CODE (XVECEXP (pattern, 0, i)) == SET && reg_overlap_mentioned_p (dest, SET_DEST (XVECEXP (pattern, 0, i)))) live_after_orig_insn = 1; } if (! live_after_orig_insn) abort (); } } } /* Subroutine of update_flow_info. Update the value of reg_n_sets for all registers modified by X. INC is -1 if the containing insn is being deleted, and is 1 if the containing insn is a newly generated insn. */ static void update_n_sets (x, inc) rtx x; int inc; { rtx dest = SET_DEST (x); while (GET_CODE (dest) == STRICT_LOW_PART || GET_CODE (dest) == SUBREG || GET_CODE (dest) == ZERO_EXTRACT || GET_CODE (dest) == SIGN_EXTRACT) dest = SUBREG_REG (dest); if (GET_CODE (dest) == REG) { int regno = REGNO (dest); if (regno < FIRST_PSEUDO_REGISTER) { register int i; int endregno = regno + HARD_REGNO_NREGS (regno, GET_MODE (dest)); for (i = regno; i < endregno; i++) reg_n_sets[i] += inc; } else reg_n_sets[regno] += inc; } } /* Updates all flow-analysis related quantities (including REG_NOTES) for the insns from FIRST to LAST inclusive that were created by splitting ORIG_INSN. NOTES are the original REG_NOTES. */ static void update_flow_info (notes, first, last, orig_insn) rtx notes; rtx first, last; rtx orig_insn; { rtx insn, note; rtx next; rtx orig_dest, temp; rtx set; /* Get and save the destination set by the original insn. */ orig_dest = single_set (orig_insn); if (orig_dest) orig_dest = SET_DEST (orig_dest); /* Move REG_NOTES from the original insn to where they now belong. */ for (note = notes; note; note = next) { next = XEXP (note, 1); switch (REG_NOTE_KIND (note)) { case REG_DEAD: case REG_UNUSED: /* Move these notes from the original insn to the last new insn where the register is now set. */ for (insn = last; ; insn = PREV_INSN (insn)) { if (GET_RTX_CLASS (GET_CODE (insn)) == 'i' && reg_mentioned_p (XEXP (note, 0), PATTERN (insn))) { /* If this note refers to a multiple word hard register, it may have been split into several smaller hard register references, so handle it specially. */ temp = XEXP (note, 0); if (REG_NOTE_KIND (note) == REG_DEAD && GET_CODE (temp) == REG && REGNO (temp) < FIRST_PSEUDO_REGISTER && HARD_REGNO_NREGS (REGNO (temp), GET_MODE (temp)) > 1) split_hard_reg_notes (note, first, last, orig_insn); else { XEXP (note, 1) = REG_NOTES (insn); REG_NOTES (insn) = note; } /* Sometimes need to convert REG_UNUSED notes to REG_DEAD notes. */ /* ??? This won't handle multiple word registers correctly, but should be good enough for now. */ if (REG_NOTE_KIND (note) == REG_UNUSED && ! dead_or_set_p (insn, XEXP (note, 0))) PUT_REG_NOTE_KIND (note, REG_DEAD); /* The reg only dies in one insn, the last one that uses it. */ break; } /* It must die somewhere, fail it we couldn't find where it died. If this is a REG_UNUSED note, then it must be a temporary register that was not needed by this instantiation of the pattern, so we can safely ignore it. */ if (insn == first) { if (REG_NOTE_KIND (note) != REG_UNUSED) abort (); break; } } break; case REG_WAS_0: /* This note applies to the dest of the original insn. Find the first new insn that now has the same dest, and move the note there. */ if (! orig_dest) abort (); for (insn = first; ; insn = NEXT_INSN (insn)) { if (GET_RTX_CLASS (GET_CODE (insn)) == 'i' && (temp = single_set (insn)) && rtx_equal_p (SET_DEST (temp), orig_dest)) { XEXP (note, 1) = REG_NOTES (insn); REG_NOTES (insn) = note; /* The reg is only zero before one insn, the first that uses it. */ break; } /* It must be set somewhere, fail if we couldn't find where it was set. */ if (insn == last) abort (); } break; case REG_EQUAL: case REG_EQUIV: /* A REG_EQUIV or REG_EQUAL note on an insn with more than one set is meaningless. Just drop the note. */ if (! orig_dest) break; case REG_NO_CONFLICT: /* These notes apply to the dest of the original insn. Find the last new insn that now has the same dest, and move the note there. */ if (! orig_dest) abort (); for (insn = last; ; insn = PREV_INSN (insn)) { if (GET_RTX_CLASS (GET_CODE (insn)) == 'i' && (temp = single_set (insn)) && rtx_equal_p (SET_DEST (temp), orig_dest)) { XEXP (note, 1) = REG_NOTES (insn); REG_NOTES (insn) = note; /* Only put this note on one of the new insns. */ break; } /* The original dest must still be set someplace. Abort if we couldn't find it. */ if (insn == first) abort (); } break; case REG_LIBCALL: /* Move a REG_LIBCALL note to the first insn created, and update the corresponding REG_RETVAL note. */ XEXP (note, 1) = REG_NOTES (first); REG_NOTES (first) = note; insn = XEXP (note, 0); note = find_reg_note (insn, REG_RETVAL, NULL_RTX); if (note) XEXP (note, 0) = first; break; case REG_RETVAL: /* Move a REG_RETVAL note to the last insn created, and update the corresponding REG_LIBCALL note. */ XEXP (note, 1) = REG_NOTES (last); REG_NOTES (last) = note; insn = XEXP (note, 0); note = find_reg_note (insn, REG_LIBCALL, NULL_RTX); if (note) XEXP (note, 0) = last; break; case REG_NONNEG: /* This should be moved to whichever instruction is a JUMP_INSN. */ for (insn = last; ; insn = PREV_INSN (insn)) { if (GET_CODE (insn) == JUMP_INSN) { XEXP (note, 1) = REG_NOTES (insn); REG_NOTES (insn) = note; /* Only put this note on one of the new insns. */ break; } /* Fail if we couldn't find a JUMP_INSN. */ if (insn == first) abort (); } break; case REG_INC: /* This should be moved to whichever instruction now has the increment operation. */ abort (); case REG_LABEL: /* Should be moved to the new insn(s) which use the label. */ for (insn = first; insn != NEXT_INSN (last); insn = NEXT_INSN (insn)) if (GET_RTX_CLASS (GET_CODE (insn)) == 'i' && reg_mentioned_p (XEXP (note, 0), PATTERN (insn))) REG_NOTES (insn) = gen_rtx (EXPR_LIST, REG_LABEL, XEXP (note, 0), REG_NOTES (insn)); break; case REG_CC_SETTER: case REG_CC_USER: /* These two notes will never appear until after reorg, so we don't have to handle them here. */ default: abort (); } } /* Each new insn created, except the last, has a new set. If the destination is a register, then this reg is now live across several insns, whereas previously the dest reg was born and died within the same insn. To reflect this, we now need a REG_DEAD note on the insn where this dest reg dies. Similarly, the new insns may have clobbers that need REG_UNUSED notes. */ for (insn = first; insn != last; insn = NEXT_INSN (insn)) { rtx pat; int i; pat = PATTERN (insn); if (GET_CODE (pat) == SET || GET_CODE (pat) == CLOBBER) new_insn_dead_notes (pat, insn, last, orig_insn); else if (GET_CODE (pat) == PARALLEL) { for (i = 0; i < XVECLEN (pat, 0); i++) if (GET_CODE (XVECEXP (pat, 0, i)) == SET || GET_CODE (XVECEXP (pat, 0, i)) == CLOBBER) new_insn_dead_notes (XVECEXP (pat, 0, i), insn, last, orig_insn); } } /* If any insn, except the last, uses the register set by the last insn, then we need a new REG_DEAD note on that insn. In this case, there would not have been a REG_DEAD note for this register in the original insn because it was used and set within one insn. There is no new REG_DEAD note needed if the last insn uses the register that it is setting. */ set = single_set (last); if (set) { rtx dest = SET_DEST (set); while (GET_CODE (dest) == ZERO_EXTRACT || GET_CODE (dest) == SUBREG || GET_CODE (dest) == STRICT_LOW_PART || GET_CODE (dest) == SIGN_EXTRACT) dest = XEXP (dest, 0); if (GET_CODE (dest) == REG && ! reg_overlap_mentioned_p (dest, SET_SRC (set))) { for (insn = PREV_INSN (last); ; insn = PREV_INSN (insn)) { if (GET_RTX_CLASS (GET_CODE (insn)) == 'i' && reg_mentioned_p (dest, PATTERN (insn)) && (set = single_set (insn))) { rtx insn_dest = SET_DEST (set); while (GET_CODE (insn_dest) == ZERO_EXTRACT || GET_CODE (insn_dest) == SUBREG || GET_CODE (insn_dest) == STRICT_LOW_PART || GET_CODE (insn_dest) == SIGN_EXTRACT) insn_dest = XEXP (insn_dest, 0); if (insn_dest != dest) { note = rtx_alloc (EXPR_LIST); PUT_REG_NOTE_KIND (note, REG_DEAD); XEXP (note, 0) = dest; XEXP (note, 1) = REG_NOTES (insn); REG_NOTES (insn) = note; /* The reg only dies in one insn, the last one that uses it. */ break; } } if (insn == first) break; } } } /* If the original dest is modifying a multiple register target, and the original instruction was split such that the original dest is now set by two or more SUBREG sets, then the split insns no longer kill the destination of the original insn. In this case, if there exists an instruction in the same basic block, before the split insn, which uses the original dest, and this use is killed by the original insn, then we must remove the REG_DEAD note on this insn, because it is now superfluous. This does not apply when a hard register gets split, because the code knows how to handle overlapping hard registers properly. */ if (orig_dest && GET_CODE (orig_dest) == REG) { int found_orig_dest = 0; int found_split_dest = 0; for (insn = first; ; insn = NEXT_INSN (insn)) { set = single_set (insn); if (set) { if (GET_CODE (SET_DEST (set)) == REG && REGNO (SET_DEST (set)) == REGNO (orig_dest)) { found_orig_dest = 1; break; } else if (GET_CODE (SET_DEST (set)) == SUBREG && SUBREG_REG (SET_DEST (set)) == orig_dest) { found_split_dest = 1; break; } } if (insn == last) break; } if (found_split_dest) { /* Search backwards from FIRST, looking for the first insn that uses the original dest. Stop if we pass a CODE_LABEL or a JUMP_INSN. If we find an insn, and it has a REG_DEAD note, then delete the note. */ for (insn = first; insn; insn = PREV_INSN (insn)) { if (GET_CODE (insn) == CODE_LABEL || GET_CODE (insn) == JUMP_INSN) break; else if (GET_RTX_CLASS (GET_CODE (insn)) == 'i' && reg_mentioned_p (orig_dest, insn)) { note = find_regno_note (insn, REG_DEAD, REGNO (orig_dest)); if (note) remove_note (insn, note); } } } else if (! found_orig_dest) { /* This should never happen. */ abort (); } } /* Update reg_n_sets. This is necessary to prevent local alloc from converting REG_EQUAL notes to REG_EQUIV when splitting has modified a reg from set once to set multiple times. */ { rtx x = PATTERN (orig_insn); RTX_CODE code = GET_CODE (x); if (code == SET || code == CLOBBER) update_n_sets (x, -1); else if (code == PARALLEL) { int i; for (i = XVECLEN (x, 0) - 1; i >= 0; i--) { code = GET_CODE (XVECEXP (x, 0, i)); if (code == SET || code == CLOBBER) update_n_sets (XVECEXP (x, 0, i), -1); } } for (insn = first; ; insn = NEXT_INSN (insn)) { x = PATTERN (insn); code = GET_CODE (x); if (code == SET || code == CLOBBER) update_n_sets (x, 1); else if (code == PARALLEL) { int i; for (i = XVECLEN (x, 0) - 1; i >= 0; i--) { code = GET_CODE (XVECEXP (x, 0, i)); if (code == SET || code == CLOBBER) update_n_sets (XVECEXP (x, 0, i), 1); } } if (insn == last) break; } } } /* The one entry point in this file. DUMP_FILE is the dump file for this pass. */ void schedule_insns (dump_file) FILE *dump_file; { int max_uid = MAX_INSNS_PER_SPLIT * (get_max_uid () + 1); int b; rtx insn; /* Taking care of this degenerate case makes the rest of this code simpler. */ if (n_basic_blocks == 0) return; /* Create an insn here so that we can hang dependencies off of it later. */ sched_before_next_call = gen_rtx (INSN, VOIDmode, 0, NULL_RTX, NULL_RTX, NULL_RTX, 0, NULL_RTX, 0); /* Initialize the unused_*_lists. We can't use the ones left over from the previous function, because gcc has freed that memory. We can use the ones left over from the first sched pass in the second pass however, so only clear them on the first sched pass. The first pass is before reload if flag_schedule_insns is set, otherwise it is afterwards. */ if (reload_completed == 0 || ! flag_schedule_insns) { unused_insn_list = 0; unused_expr_list = 0; } /* We create no insns here, only reorder them, so we remember how far we can cut back the stack on exit. */ /* Allocate data for this pass. See comments, above, for what these vectors do. */ insn_luid = (int *) alloca (max_uid * sizeof (int)); insn_priority = (int *) alloca (max_uid * sizeof (int)); insn_tick = (int *) alloca (max_uid * sizeof (int)); insn_costs = (short *) alloca (max_uid * sizeof (short)); insn_units = (short *) alloca (max_uid * sizeof (short)); insn_blockage = (unsigned int *) alloca (max_uid * sizeof (unsigned int)); insn_ref_count = (int *) alloca (max_uid * sizeof (int)); if (reload_completed == 0) { sched_reg_n_deaths = (short *) alloca (max_regno * sizeof (short)); sched_reg_n_calls_crossed = (int *) alloca (max_regno * sizeof (int)); sched_reg_live_length = (int *) alloca (max_regno * sizeof (int)); bb_dead_regs = (regset) alloca (regset_bytes); bb_live_regs = (regset) alloca (regset_bytes); bzero ((char *) sched_reg_n_calls_crossed, max_regno * sizeof (int)); bzero ((char *) sched_reg_live_length, max_regno * sizeof (int)); bcopy ((char *) reg_n_deaths, (char *) sched_reg_n_deaths, max_regno * sizeof (short)); init_alias_analysis (); } else { sched_reg_n_deaths = 0; sched_reg_n_calls_crossed = 0; sched_reg_live_length = 0; bb_dead_regs = 0; bb_live_regs = 0; if (! flag_schedule_insns) init_alias_analysis (); } if (write_symbols != NO_DEBUG) { rtx line; line_note = (rtx *) alloca (max_uid * sizeof (rtx)); bzero ((char *) line_note, max_uid * sizeof (rtx)); line_note_head = (rtx *) alloca (n_basic_blocks * sizeof (rtx)); bzero ((char *) line_note_head, n_basic_blocks * sizeof (rtx)); /* Determine the line-number at the start of each basic block. This must be computed and saved now, because after a basic block's predecessor has been scheduled, it is impossible to accurately determine the correct line number for the first insn of the block. */ for (b = 0; b < n_basic_blocks; b++) for (line = basic_block_head[b]; line; line = PREV_INSN (line)) if (GET_CODE (line) == NOTE && NOTE_LINE_NUMBER (line) > 0) { line_note_head[b] = line; break; } } bzero ((char *) insn_luid, max_uid * sizeof (int)); bzero ((char *) insn_priority, max_uid * sizeof (int)); bzero ((char *) insn_tick, max_uid * sizeof (int)); bzero ((char *) insn_costs, max_uid * sizeof (short)); bzero ((char *) insn_units, max_uid * sizeof (short)); bzero ((char *) insn_blockage, max_uid * sizeof (unsigned int)); bzero ((char *) insn_ref_count, max_uid * sizeof (int)); /* Schedule each basic block, block by block. */ /* ??? Add a NOTE after the last insn of the last basic block. It is not known why this is done. */ insn = basic_block_end[n_basic_blocks-1]; if (NEXT_INSN (insn) == 0 || (GET_CODE (insn) != NOTE && GET_CODE (insn) != CODE_LABEL /* Don't emit a NOTE if it would end up between an unconditional jump and a BARRIER. */ && ! (GET_CODE (insn) == JUMP_INSN && GET_CODE (NEXT_INSN (insn)) == BARRIER))) emit_note_after (NOTE_INSN_DELETED, basic_block_end[n_basic_blocks-1]); for (b = 0; b < n_basic_blocks; b++) { rtx insn, next; note_list = 0; for (insn = basic_block_head[b]; ; insn = next) { rtx prev; rtx set; /* Can't use `next_real_insn' because that might go across CODE_LABELS and short-out basic blocks. */ next = NEXT_INSN (insn); if (GET_CODE (insn) != INSN) { if (insn == basic_block_end[b]) break; continue; } /* Don't split no-op move insns. These should silently disappear later in final. Splitting such insns would break the code that handles REG_NO_CONFLICT blocks. */ set = single_set (insn); if (set && rtx_equal_p (SET_SRC (set), SET_DEST (set))) { if (insn == basic_block_end[b]) break; /* Nops get in the way while scheduling, so delete them now if register allocation has already been done. It is too risky to try to do this before register allocation, and there are unlikely to be very many nops then anyways. */ if (reload_completed) { PUT_CODE (insn, NOTE); NOTE_LINE_NUMBER (insn) = NOTE_INSN_DELETED; NOTE_SOURCE_FILE (insn) = 0; } continue; } /* Split insns here to get max fine-grain parallelism. */ prev = PREV_INSN (insn); if (reload_completed == 0) { rtx last, first = PREV_INSN (insn); rtx notes = REG_NOTES (insn); last = try_split (PATTERN (insn), insn, 1); if (last != insn) { /* try_split returns the NOTE that INSN became. */ first = NEXT_INSN (first); update_flow_info (notes, first, last, insn); PUT_CODE (insn, NOTE); NOTE_SOURCE_FILE (insn) = 0; NOTE_LINE_NUMBER (insn) = NOTE_INSN_DELETED; if (insn == basic_block_head[b]) basic_block_head[b] = first; if (insn == basic_block_end[b]) { basic_block_end[b] = last; break; } } } if (insn == basic_block_end[b]) break; } schedule_block (b, dump_file); #ifdef USE_C_ALLOCA alloca (0); #endif } /* Reposition the prologue and epilogue notes in case we moved the prologue/epilogue insns. */ if (reload_completed) reposition_prologue_and_epilogue_notes (get_insns ()); if (write_symbols != NO_DEBUG) { rtx line = 0; rtx insn = get_insns (); int active_insn = 0; int notes = 0; /* Walk the insns deleting redundant line-number notes. Many of these are already present. The remainder tend to occur at basic block boundaries. */ for (insn = get_last_insn (); insn; insn = PREV_INSN (insn)) if (GET_CODE (insn) == NOTE && NOTE_LINE_NUMBER (insn) > 0) { /* If there are no active insns following, INSN is redundant. */ if (active_insn == 0) { notes++; NOTE_SOURCE_FILE (insn) = 0; NOTE_LINE_NUMBER (insn) = NOTE_INSN_DELETED; } /* If the line number is unchanged, LINE is redundant. */ else if (line && NOTE_LINE_NUMBER (line) == NOTE_LINE_NUMBER (insn) && NOTE_SOURCE_FILE (line) == NOTE_SOURCE_FILE (insn)) { notes++; NOTE_SOURCE_FILE (line) = 0; NOTE_LINE_NUMBER (line) = NOTE_INSN_DELETED; line = insn; } else line = insn; active_insn = 0; } else if (! ((GET_CODE (insn) == NOTE && NOTE_LINE_NUMBER (insn) == NOTE_INSN_DELETED) || (GET_CODE (insn) == INSN && (GET_CODE (PATTERN (insn)) == USE || GET_CODE (PATTERN (insn)) == CLOBBER)))) active_insn++; if (dump_file && notes) fprintf (dump_file, ";; deleted %d line-number notes\n", notes); } if (reload_completed == 0) { int regno; for (regno = 0; regno < max_regno; regno++) if (sched_reg_live_length[regno]) { if (dump_file) { if (reg_live_length[regno] > sched_reg_live_length[regno]) fprintf (dump_file, ";; register %d life shortened from %d to %d\n", regno, reg_live_length[regno], sched_reg_live_length[regno]); /* Negative values are special; don't overwrite the current reg_live_length value if it is negative. */ else if (reg_live_length[regno] < sched_reg_live_length[regno] && reg_live_length[regno] >= 0) fprintf (dump_file, ";; register %d life extended from %d to %d\n", regno, reg_live_length[regno], sched_reg_live_length[regno]); if (! reg_n_calls_crossed[regno] && sched_reg_n_calls_crossed[regno]) fprintf (dump_file, ";; register %d now crosses calls\n", regno); else if (reg_n_calls_crossed[regno] && ! sched_reg_n_calls_crossed[regno] && reg_basic_block[regno] != REG_BLOCK_GLOBAL) fprintf (dump_file, ";; register %d no longer crosses calls\n", regno); } /* Negative values are special; don't overwrite the current reg_live_length value if it is negative. */ if (reg_live_length[regno] >= 0) reg_live_length[regno] = sched_reg_live_length[regno]; /* We can't change the value of reg_n_calls_crossed to zero for pseudos which are live in more than one block. This is because combine might have made an optimization which invalidated basic_block_live_at_start and reg_n_calls_crossed, but it does not update them. If we update reg_n_calls_crossed here, the two variables are now inconsistent, and this might confuse the caller-save code into saving a register that doesn't need to be saved. This is only a problem when we zero calls crossed for a pseudo live in multiple basic blocks. Alternatively, we could try to correctly update basic block live at start here in sched, but that seems complicated. */ if (sched_reg_n_calls_crossed[regno] || reg_basic_block[regno] != REG_BLOCK_GLOBAL) reg_n_calls_crossed[regno] = sched_reg_n_calls_crossed[regno]; } } } #endif /* INSN_SCHEDULING */