/* Perform instruction reorganizations for delay slot filling. Copyright (C) 1992 Free Software Foundation, Inc. Contributed by Richard Kenner (kenner@nyu.edu). Hacked by Michael Tiemann (tiemann@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 reorganization pass. This pass runs after register allocation and final jump optimization. It should be the last pass to run before peephole. It serves primarily to fill delay slots of insns, typically branch and call insns. Other insns typically involve more complicated interactions of data dependencies and resource constraints, and are better handled by scheduling before register allocation (by the function `schedule_insns'). The Branch Penalty is the number of extra cycles that are needed to execute a branch insn. On an ideal machine, branches take a single cycle, and the Branch Penalty is 0. Several RISC machines approach branch delays differently: The MIPS and AMD 29000 have a single branch delay slot. Most insns (except other branches) can be used to fill this slot. When the slot is filled, two insns execute in two cycles, reducing the branch penalty to zero. The Motorola 88000 conditionally exposes its branch delay slot, so code is shorter when it is turned off, but will run faster when useful insns are scheduled there. The IBM ROMP has two forms of branch and call insns, both with and without a delay slot. Much like the 88k, insns not using the delay slot can be shorted (2 bytes vs. 4 bytes), but will run slowed. The SPARC always has a branch delay slot, but its effects can be annulled when the branch is not taken. This means that failing to find other sources of insns, we can hoist an insn from the branch target that would only be safe to execute knowing that the branch is taken. Three techniques for filling delay slots have been implemented so far: (1) `fill_simple_delay_slots' is the simplest, most efficient way to fill delay slots. This pass first looks for insns which come from before the branch and which are safe to execute after the branch. Then it searches after the insn requiring delay slots or, in the case of a branch, for insns that are after the point at which the branch merges into the fallthrough code, if such a point exists. When such insns are found, the branch penalty decreases and no code expansion takes place. (2) `fill_eager_delay_slots' is more complicated: it is used for scheduling conditional jumps, or for scheduling jumps which cannot be filled using (1). A machine need not have annulled jumps to use this strategy, but it helps (by keeping more options open). `fill_eager_delay_slots' tries to guess the direction the branch will go; if it guesses right 100% of the time, it can reduce the branch penalty as much as `fill_eager_delay_slots' does. If it guesses wrong 100% of the time, it might as well schedule nops (or on the m88k, unexpose the branch slot). When `fill_eager_delay_slots' takes insns from the fall-through path of the jump, usually there is no code expansion; when it takes insns from the branch target, there is code expansion if it is not the only way to reach that target. (3) `relax_delay_slots' uses a set of rules to simplify code that has been reorganized by (1) and (2). It finds cases where conditional test can be eliminated, jumps can be threaded, extra insns can be eliminated, etc. It is the job of (1) and (2) to do a good job of scheduling locally; `relax_delay_slots' takes care of making the various individual schedules work well together. It is especially tuned to handle the control flow interactions of branch insns. It does nothing for insns with delay slots that do not branch. On machines that use CC0, we are very conservative. We will not make a copy of an insn involving CC0 since we want to maintain a 1-1 correspondence between the insn that sets and uses CC0. The insns are allowed to be separated by placing an insn that sets CC0 (but not an insn that uses CC0; we could do this, but it doesn't seem worthwhile) in a delay slot. In that case, we point each insn at the other with REG_CC_USER and REG_CC_SETTER notes. Note that these restrictions affect very few machines because most RISC machines with delay slots will not use CC0 (the RT is the only known exception at this point). Not yet implemented: The Acorn Risc Machine can conditionally execute most insns, so it is profitable to move single insns into a position to execute based on the condition code of the previous insn. The HP-PA can conditionally nullify insns, providing a similar effect to the ARM, differing mostly in which insn is "in charge". */ #include #include "config.h" #include "rtl.h" #include "insn-config.h" #include "conditions.h" #include "hard-reg-set.h" #include "basic-block.h" #include "regs.h" #include "insn-flags.h" #include "recog.h" #include "flags.h" #include "output.h" #include "obstack.h" #include "insn-attr.h" #ifdef DELAY_SLOTS #define obstack_chunk_alloc xmalloc #define obstack_chunk_free free #ifndef ANNUL_IFTRUE_SLOTS #define eligible_for_annul_true(INSN, SLOTS, TRIAL) 0 #endif #ifndef ANNUL_IFFALSE_SLOTS #define eligible_for_annul_false(INSN, SLOTS, TRIAL) 0 #endif /* Insns which have delay slots that have not yet been filled. */ static struct obstack unfilled_slots_obstack; static rtx *unfilled_firstobj; /* Define macros to refer to the first and last slot containing unfilled insns. These are used because the list may move and its address should be recomputed at each use. */ #define unfilled_slots_base \ ((rtx *) obstack_base (&unfilled_slots_obstack)) #define unfilled_slots_next \ ((rtx *) obstack_next_free (&unfilled_slots_obstack)) /* This structure is used to indicate which hardware resources are set or needed by insns so far. */ struct resources { char memory; /* Insn sets or needs a memory location. */ char volatil; /* Insn sets or needs a volatile memory loc. */ char cc; /* Insn sets or needs the condition codes. */ HARD_REG_SET regs; /* Which registers are set or needed. */ }; /* Macro to clear all resources. */ #define CLEAR_RESOURCE(RES) \ do { (RES)->memory = (RES)->volatil = (RES)->cc = 0; \ CLEAR_HARD_REG_SET ((RES)->regs); } while (0) /* Indicates what resources are required at function end. */ static struct resources end_of_function_needs; /* Points to the label before the end of the function. */ static rtx end_of_function_label; /* This structure is used to record liveness information at the targets or fallthrough insns of branches. We will most likely need the information at targets again, so save them in a hash table rather than recomputing them each time. */ struct target_info { int uid; /* INSN_UID of target. */ struct target_info *next; /* Next info for same hash bucket. */ HARD_REG_SET live_regs; /* Registers live at target. */ int block; /* Basic block number containing target. */ int bb_tick; /* Generation count of basic block info. */ }; #define TARGET_HASH_PRIME 257 /* Define the hash table itself. */ static struct target_info **target_hash_table; /* For each basic block, we maintain a generation number of its basic block info, which is updated each time we move an insn from the target of a jump. This is the generation number indexed by block number. */ static int *bb_ticks; /* Mapping between INSN_UID's and position in the code since INSN_UID's do not always monotonically increase. */ static int *uid_to_ruid; /* Highest valid index in `uid_to_ruid'. */ static int max_uid; /* Forward references: */ static int redundant_insn_p (); static void update_block (); /* Given X, some rtl, and RES, a pointer to a `struct resource', mark which resources are references by the insn. If INCLUDE_CALLED_ROUTINE is TRUE, resources used by the called routine will be included for CALL_INSNs. */ static void mark_referenced_resources (x, res, include_called_routine) register rtx x; register struct resources *res; register int include_called_routine; { register enum rtx_code code = GET_CODE (x); register int i, j; register char *format_ptr; /* Handle leaf items for which we set resource flags. Also, special-case CALL, SET and CLOBBER operators. */ switch (code) { case CONST: case CONST_INT: case CONST_DOUBLE: case PC: case SYMBOL_REF: case LABEL_REF: return; case SUBREG: if (GET_CODE (SUBREG_REG (x)) != REG) mark_referenced_resources (SUBREG_REG (x), res, 0); else { int regno = REGNO (SUBREG_REG (x)) + SUBREG_WORD (x); int last_regno = regno + HARD_REGNO_NREGS (regno, GET_MODE (x)); for (i = regno; i < last_regno; i++) SET_HARD_REG_BIT (res->regs, i); } return; case REG: for (i = 0; i < HARD_REGNO_NREGS (REGNO (x), GET_MODE (x)); i++) SET_HARD_REG_BIT (res->regs, REGNO (x) + i); return; case MEM: /* If this memory shouldn't change, it really isn't referencing memory. */ if (! RTX_UNCHANGING_P (x)) res->memory = 1; res->volatil = MEM_VOLATILE_P (x); /* Mark registers used to access memory. */ mark_referenced_resources (XEXP (x, 0), res, 0); return; case CC0: res->cc = 1; return; case UNSPEC_VOLATILE: case ASM_INPUT: /* Traditional asm's are always volatile. */ res->volatil = 1; return; case ASM_OPERANDS: res->volatil = MEM_VOLATILE_P (x); /* 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. */ for (i = 0; i < ASM_OPERANDS_INPUT_LENGTH (x); i++) mark_referenced_resources (ASM_OPERANDS_INPUT (x, i), res, 0); return; case CALL: /* The first operand will be a (MEM (xxx)) but doesn't really reference memory. The second operand may be referenced, though. */ mark_referenced_resources (XEXP (XEXP (x, 0), 0), res, 0); mark_referenced_resources (XEXP (x, 1), res, 0); return; case SET: /* Usually, the first operand of SET is set, not referenced. But registers used to access memory are referenced. SET_DEST is also referenced if it is a ZERO_EXTRACT or SIGN_EXTRACT. */ mark_referenced_resources (SET_SRC (x), res, 0); x = SET_DEST (x); if (GET_CODE (x) == SIGN_EXTRACT || GET_CODE (x) == ZERO_EXTRACT) mark_referenced_resources (x, res, 0); else if (GET_CODE (x) == SUBREG) x = SUBREG_REG (x); if (GET_CODE (x) == MEM) mark_referenced_resources (XEXP (x, 0), res, 0); return; case CLOBBER: return; case CALL_INSN: if (include_called_routine) { /* A CALL references memory, the frame pointer if it exists, the stack pointer, any global registers and any registers given in USE insns immediately in front of the CALL. However, we may have moved some of the parameter loading insns into the delay slot of this CALL. If so, the USE's for them don't count and should be skipped. */ rtx insn = PREV_INSN (x); rtx sequence = 0; int seq_size = 0; int i; /* If we are part of a delay slot sequence, point at the SEQUENCE. */ if (NEXT_INSN (insn) != x) { sequence = PATTERN (NEXT_INSN (insn)); seq_size = XVECLEN (sequence, 0); if (GET_CODE (sequence) != SEQUENCE) abort (); } res->memory = 1; SET_HARD_REG_BIT (res->regs, STACK_POINTER_REGNUM); if (frame_pointer_needed) SET_HARD_REG_BIT (res->regs, FRAME_POINTER_REGNUM); for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (global_regs[i]) SET_HARD_REG_BIT (res->regs, i); /* Skip any labels between the CALL_INSN and possible USE insns. */ while (GET_CODE (insn) == CODE_LABEL) insn = PREV_INSN (insn); for ( ; (insn && GET_CODE (insn) == INSN && GET_CODE (PATTERN (insn)) == USE); insn = PREV_INSN (insn)) { for (i = 1; i < seq_size; i++) { rtx slot_pat = PATTERN (XVECEXP (sequence, 0, i)); if (GET_CODE (slot_pat) == SET && rtx_equal_p (SET_DEST (slot_pat), XEXP (PATTERN (insn), 0))) break; } if (i >= seq_size) mark_referenced_resources (XEXP (PATTERN (insn), 0), res, 0); } } /* ... fall through to other INSN processing ... */ case INSN: case JUMP_INSN: /* No special processing, just speed up. */ mark_referenced_resources (PATTERN (x), res, include_called_routine); return; } /* Process each sub-expression and flag what it needs. */ format_ptr = GET_RTX_FORMAT (code); for (i = 0; i < GET_RTX_LENGTH (code); i++) switch (*format_ptr++) { case 'e': mark_referenced_resources (XEXP (x, i), res, include_called_routine); break; case 'E': for (j = 0; j < XVECLEN (x, i); j++) mark_referenced_resources (XVECEXP (x, i, j), res, include_called_routine); break; } } /* Given an insn, INSN, and a pointer to a `struct resource', RES, indicate which resources are modified by the insn. If INCLUDE_CALLED_ROUTINE is TRUE, also mark resources potentially set by the called routine. We never mark the insn as modifying the condition code unless it explicitly SETs CC0 even though this is not totally correct. The reason for this is that we require a SET of CC0 to immediately precede the reference to CC0. So if some other insn sets CC0 as a side-effect, we know it cannot affect our computation and thus may be placed in a delay slot. */ static void mark_set_resources (insn, res, include_called_routine) register rtx insn; register struct resources *res; int include_called_routine; { register int i; switch (GET_CODE (insn)) { case NOTE: case BARRIER: case CODE_LABEL: /* These don't set any resources. */ return; case CALL_INSN: /* Called routine modifies the condition code, memory, any registers that aren't saved across calls, global registers and anything explicitly CLOBBERed immediately after the CALL_INSN. */ if (include_called_routine) { rtx next = NEXT_INSN (insn); res->cc = res->memory = 1; for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (call_used_regs[i] || global_regs[i]) SET_HARD_REG_BIT (res->regs, i); /* Skip any possible labels between the CALL_INSN and CLOBBERs. */ while (GET_CODE (next) == CODE_LABEL) next = NEXT_INSN (next); for (; (next && GET_CODE (next) == INSN && GET_CODE (PATTERN (next)) == CLOBBER); next = NEXT_INSN (next)) mark_referenced_resources (XEXP (PATTERN (next), 0), res, 0); } /* ... and also what it's RTL says it modifies, if anything. */ case JUMP_INSN: case INSN: { register rtx body = PATTERN (insn); register rtx note; /* An insn consisting of just a CLOBBER (or USE) is just for flow and doesn't actually do anything, so we don't check for it. If the source of a SET is a CALL, this is actually done by the called routine. So only include it if we are to include the effects of the calling routine. */ if (GET_CODE (body) == SET && (include_called_routine || GET_CODE (SET_SRC (body)) != CALL)) mark_referenced_resources (SET_DEST (body), res, 0); else if (GET_CODE (body) == PARALLEL) { for (i = 0; i < XVECLEN (body, 0); i++) if ((GET_CODE (XVECEXP (body, 0, i)) == SET && (include_called_routine || GET_CODE (SET_SRC (XVECEXP (body, 0, i))) != CALL)) || GET_CODE (XVECEXP (body, 0, i)) == CLOBBER) mark_referenced_resources (SET_DEST (XVECEXP (body, 0, i)), res, 0); } else if (GET_CODE (body) == SEQUENCE) for (i = 0; i < XVECLEN (body, 0); i++) if (! (INSN_ANNULLED_BRANCH_P (XVECEXP (body, 0, 0)) && INSN_FROM_TARGET_P (XVECEXP (body, 0, i)))) mark_set_resources (XVECEXP (body, 0, i), res, include_called_routine); #ifdef AUTO_INC_DEC /* If any register are incremented or decremented in an address, they are set here. */ for (note = REG_NOTES (insn); note; note = XEXP (note, 1)) if (REG_NOTE_KIND (note) == REG_INC) mark_referenced_resources (XEXP (note, 0), res, 0); #endif #ifdef PUSH_ROUNDING /* An insn that has a PRE_DEC on SP will not have a REG_INC note. Until we fix this correctly, consider all insns as modifying SP on such machines. So far, we don't have delay slot scheduling on any machines with PUSH_ROUNDING. */ SET_HARD_REG_BIT (res->regs, STACK_POINTER_REGNUM); #endif return; } default: abort (); } } /* Return TRUE if this insn should stop the search for insn to fill delay slots. LABELS_P indicates that labels should terminate the search. In all cases, jumps terminate the search. */ static int stop_search_p (insn, labels_p) rtx insn; int labels_p; { if (insn == 0) return 1; switch (GET_CODE (insn)) { case NOTE: case CALL_INSN: return 0; case CODE_LABEL: return labels_p; case JUMP_INSN: case BARRIER: return 1; case INSN: /* OK unless it contains a delay slot or is an `asm' insn of some type. We don't know anything about these. */ return (GET_CODE (PATTERN (insn)) == SEQUENCE || GET_CODE (PATTERN (insn)) == ASM_INPUT || asm_noperands (PATTERN (insn)) >= 0); default: abort (); } } /* Return TRUE if any resources are marked in both RES1 and RES2 or if either resource set contains a volatile memory reference. Otherwise, return FALSE. */ static int resource_conflicts_p (res1, res2) struct resources *res1, *res2; { if ((res1->cc && res2->cc) || (res1->memory && res2->memory) || res1->volatil || res2->volatil) return 1; #ifdef HARD_REG_SET return (res1->regs & res2->regs) != HARD_CONST (0); #else { int i; for (i = 0; i < HARD_REG_SET_LONGS; i++) if ((res1->regs[i] & res2->regs[i]) != 0) return 1; return 0; } #endif } /* Return TRUE if any resource marked in RES, a `struct resources', is referenced by INSN. If INCLUDE_CALLED_ROUTINE is set, return if the called routine is using those resources. We compute this by computing all the resources referenced by INSN and seeing if this conflicts with RES. It might be faster to directly check ourselves, and this is the way it used to work, but it means duplicating a large block of complex code. */ static int insn_references_resource_p (insn, res, include_called_routine) register rtx insn; register struct resources *res; int include_called_routine; { struct resources insn_res; CLEAR_RESOURCE (&insn_res); mark_referenced_resources (insn, &insn_res, include_called_routine); return resource_conflicts_p (&insn_res, res); } /* Return TRUE if INSN modifies resources that are marked in RES. INCLUDE_CALLED_ROUTINE is set if the actions of that routine should be included. CC0 is only modified if it is explicitly set; see comments in front of mark_set_resources for details. */ static int insn_sets_resource_p (insn, res, include_called_routine) register rtx insn; register struct resources *res; int include_called_routine; { struct resources insn_sets; CLEAR_RESOURCE (&insn_sets); mark_set_resources (insn, &insn_sets, include_called_routine); return resource_conflicts_p (&insn_sets, res); } /* Find a label at the end of the function or before a RETURN. If there is none, make one. */ static rtx find_end_label () { rtx insn; /* If we found one previously, return it. */ if (end_of_function_label) return end_of_function_label; /* Otherwise, see if there is a label at the end of the function. If there is, it must be that RETURN insns aren't needed, so that is our return label and we don't have to do anything else. */ insn = get_last_insn (); while (GET_CODE (insn) == NOTE || (GET_CODE (insn) == INSN && (GET_CODE (PATTERN (insn)) == USE || GET_CODE (PATTERN (insn)) == CLOBBER))) insn = PREV_INSN (insn); if (GET_CODE (insn) == CODE_LABEL) end_of_function_label = insn; else { /* Otherwise, make a new label and emit a RETURN and BARRIER, if needed. */ end_of_function_label = gen_label_rtx (); LABEL_NUSES (end_of_function_label) = 0; emit_label (end_of_function_label); #ifdef HAVE_return if (HAVE_return) { emit_jump_insn (gen_return ()); emit_barrier (); } #endif } /* Show one additional use for this label so it won't go away until we are done. */ ++LABEL_NUSES (end_of_function_label); return end_of_function_label; } /* Put INSN and LIST together in a SEQUENCE rtx of LENGTH, and replace the pattern of INSN with the SEQUENCE. Chain the insns so that NEXT_INSN of each insn in the sequence points to the next and NEXT_INSN of the last insn in the sequence points to the first insn after the sequence. Similarly for PREV_INSN. This makes it easier to scan all insns. Returns the SEQUENCE that replaces INSN. */ static rtx emit_delay_sequence (insn, list, length, avail) rtx insn; rtx list; int length; int avail; { register int i = 1; register rtx li; int had_barrier = 0; /* Allocate the the rtvec to hold the insns and the SEQUENCE. */ rtvec seqv = rtvec_alloc (length + 1); rtx seq = gen_rtx (SEQUENCE, VOIDmode, seqv); rtx seq_insn = make_insn_raw (seq); rtx first = get_insns (); rtx last = get_last_insn (); /* Make a copy of the insn having delay slots. */ rtx delay_insn = copy_rtx (insn); /* If INSN is followed by a BARRIER, delete the BARRIER since it will only confuse further processing. Update LAST in case it was the last insn. We will put the BARRIER back in later. */ if (NEXT_INSN (insn) && GET_CODE (NEXT_INSN (insn)) == BARRIER) { delete_insn (NEXT_INSN (insn)); last = get_last_insn (); had_barrier = 1; } /* Splice our SEQUENCE into the insn stream where INSN used to be. */ NEXT_INSN (seq_insn) = NEXT_INSN (insn); PREV_INSN (seq_insn) = PREV_INSN (insn); if (insn == last) set_new_first_and_last_insn (first, seq_insn); else PREV_INSN (NEXT_INSN (seq_insn)) = seq_insn; if (insn == first) set_new_first_and_last_insn (seq_insn, last); else NEXT_INSN (PREV_INSN (seq_insn)) = seq_insn; /* Build our SEQUENCE and rebuild the insn chain. */ XVECEXP (seq, 0, 0) = delay_insn; INSN_DELETED_P (delay_insn) = 0; PREV_INSN (delay_insn) = PREV_INSN (seq_insn); for (li = list; li; li = XEXP (li, 1), i++) { rtx tem = XEXP (li, 0); rtx note; /* Show that this copy of the insn isn't deleted. */ INSN_DELETED_P (tem) = 0; XVECEXP (seq, 0, i) = tem; PREV_INSN (tem) = XVECEXP (seq, 0, i - 1); NEXT_INSN (XVECEXP (seq, 0, i - 1)) = tem; /* Remove any REG_DEAD notes because we can't rely on them now that the insn has been moved. */ for (note = REG_NOTES (tem); note; note = XEXP (note, 1)) if (REG_NOTE_KIND (note) == REG_DEAD) XEXP (note, 0) = const0_rtx; } NEXT_INSN (XVECEXP (seq, 0, length)) = NEXT_INSN (seq_insn); /* If the previous insn is a SEQUENCE, update the NEXT_INSN pointer on the last insn in that SEQUENCE to point to us. Similarly for the first insn in the following insn if it is a SEQUENCE. */ if (PREV_INSN (seq_insn) && GET_CODE (PREV_INSN (seq_insn)) == INSN && GET_CODE (PATTERN (PREV_INSN (seq_insn))) == SEQUENCE) NEXT_INSN (XVECEXP (PATTERN (PREV_INSN (seq_insn)), 0, XVECLEN (PATTERN (PREV_INSN (seq_insn)), 0) - 1)) = seq_insn; if (NEXT_INSN (seq_insn) && GET_CODE (NEXT_INSN (seq_insn)) == INSN && GET_CODE (PATTERN (NEXT_INSN (seq_insn))) == SEQUENCE) PREV_INSN (XVECEXP (PATTERN (NEXT_INSN (seq_insn)), 0, 0)) = seq_insn; /* If there used to be a BARRIER, put it back. */ if (had_barrier) emit_barrier_after (seq_insn); if (i != length + 1) abort (); return seq_insn; } /* Add INSN to DELAY_LIST and return the head of the new list. The list must be in the order in which the insns are to be executed. */ static rtx add_to_delay_list (insn, delay_list) rtx insn; rtx delay_list; { /* If we have an empty list, just make a new list element. */ if (delay_list == 0) return gen_rtx (INSN_LIST, VOIDmode, insn, NULL_RTX); /* Otherwise this must be an INSN_LIST. Add INSN to the end of the list. */ XEXP (delay_list, 1) = add_to_delay_list (insn, XEXP (delay_list, 1)); return delay_list; } /* Delete INSN from the the delay slot of the insn that it is in. This may produce an insn without anything in its delay slots. */ static void delete_from_delay_slot (insn) rtx insn; { rtx trial, seq_insn, seq, prev; rtx delay_list = 0; int i; /* We first must find the insn containing the SEQUENCE with INSN in its delay slot. Do this by finding an insn, TRIAL, where PREV_INSN (NEXT_INSN (TRIAL)) != TRIAL. */ for (trial = insn; PREV_INSN (NEXT_INSN (trial)) == trial; trial = NEXT_INSN (trial)) ; seq_insn = PREV_INSN (NEXT_INSN (trial)); seq = PATTERN (seq_insn); /* Create a delay list consisting of all the insns other than the one we are deleting (unless we were the only one). */ if (XVECLEN (seq, 0) > 2) for (i = 1; i < XVECLEN (seq, 0); i++) if (XVECEXP (seq, 0, i) != insn) delay_list = add_to_delay_list (XVECEXP (seq, 0, i), delay_list); /* Delete the old SEQUENCE, re-emit the insn that used to have the delay list, and rebuild the delay list if non-empty. */ prev = PREV_INSN (seq_insn); trial = XVECEXP (seq, 0, 0); delete_insn (seq_insn); add_insn_after (trial, prev); if (GET_CODE (trial) == JUMP_INSN && (simplejump_p (trial) || GET_CODE (PATTERN (trial)) == RETURN)) emit_barrier_after (trial); /* If there are any delay insns, remit them. Otherwise clear the annul flag. */ if (delay_list) trial = emit_delay_sequence (trial, delay_list, XVECLEN (seq, 0) - 2, 0); else INSN_ANNULLED_BRANCH_P (trial) = 0; INSN_FROM_TARGET_P (insn) = 0; /* Show we need to fill this insn again. */ obstack_ptr_grow (&unfilled_slots_obstack, trial); } /* Delete INSN, a JUMP_INSN. If it is a conditional jump, we must track down the insn that sets CC0 for it and delete it too. */ static void delete_scheduled_jump (insn) rtx insn; { /* Delete the insn that sets cc0 for us. On machines without cc0, we could delete the insn that sets the condition code, but it is hard to find it. Since this case is rare anyway, don't bother trying; there would likely be other insns that became dead anyway, which we wouldn't know to delete. */ #ifdef HAVE_cc0 if (reg_mentioned_p (cc0_rtx, insn)) { rtx note = find_reg_note (insn, REG_CC_SETTER, NULL_RTX); /* If a reg-note was found, it points to an insn to set CC0. This insn is in the delay list of some other insn. So delete it from the delay list it was in. */ if (note) { if (! FIND_REG_INC_NOTE (XEXP (note, 0), NULL_RTX) && sets_cc0_p (PATTERN (XEXP (note, 0))) == 1) delete_from_delay_slot (XEXP (note, 0)); } else { /* The insn setting CC0 is our previous insn, but it may be in a delay slot. It will be the last insn in the delay slot, if it is. */ rtx trial = previous_insn (insn); if (GET_CODE (trial) == NOTE) trial = prev_nonnote_insn (trial); if (sets_cc0_p (PATTERN (trial)) != 1 || FIND_REG_INC_NOTE (trial, 0)) return; if (PREV_INSN (NEXT_INSN (trial)) == trial) delete_insn (trial); else delete_from_delay_slot (trial); } } #endif delete_insn (insn); } /* Counters for delay-slot filling. */ #define NUM_REORG_FUNCTIONS 2 #define MAX_DELAY_HISTOGRAM 3 #define MAX_REORG_PASSES 2 static int num_insns_needing_delays[NUM_REORG_FUNCTIONS][MAX_REORG_PASSES]; static int num_filled_delays[NUM_REORG_FUNCTIONS][MAX_DELAY_HISTOGRAM+1][MAX_REORG_PASSES]; static int reorg_pass_number; static void note_delay_statistics (slots_filled, index) int slots_filled, index; { num_insns_needing_delays[index][reorg_pass_number]++; if (slots_filled > MAX_DELAY_HISTOGRAM) slots_filled = MAX_DELAY_HISTOGRAM; num_filled_delays[index][slots_filled][reorg_pass_number]++; } #if defined(ANNUL_IFFALSE_SLOTS) || defined(ANNUL_IFTRUE_SLOTS) /* Optimize the following cases: 1. When a conditional branch skips over only one instruction, use an annulling branch and put that insn in the delay slot. Use either a branch that annuls when the condition if true or invert the test with a branch that annuls when the condition is false. This saves insns, since otherwise we must copy an insn from the L1 target. (orig) (skip) (otherwise) Bcc.n L1 Bcc',a L1 Bcc,a L1' insn insn insn2 L1: L1: L1: insn2 insn2 insn2 insn3 insn3 L1': insn3 2. When a conditional branch skips over only one instruction, and after that, it unconditionally branches somewhere else, perform the similar optimization. This saves executing the second branch in the case where the inverted condition is true. Bcc.n L1 Bcc',a L2 insn insn L1: L1: Bra L2 Bra L2 INSN is a JUMP_INSN. This should be expanded to skip over N insns, where N is the number of delay slots required. */ static rtx optimize_skip (insn) register rtx insn; { register rtx trial = next_nonnote_insn (insn); rtx next_trial = next_active_insn (trial); rtx delay_list = 0; rtx target_label; if (trial == 0 || GET_CODE (trial) != INSN || GET_CODE (PATTERN (trial)) == SEQUENCE || recog_memoized (trial) < 0 || (! eligible_for_annul_false (insn, 0, trial) && ! eligible_for_annul_true (insn, 0, trial))) return 0; /* There are two cases where we are just executing one insn (we assume here that a branch requires only one insn; this should be generalized at some point): Where the branch goes around a single insn or where we have one insn followed by a branch to the same label we branch to. In both of these cases, inverting the jump and annulling the delay slot give the same effect in fewer insns. */ if ((next_trial == next_active_insn (JUMP_LABEL (insn))) || (next_trial != 0 && GET_CODE (next_trial) == JUMP_INSN && JUMP_LABEL (insn) == JUMP_LABEL (next_trial) && (simplejump_p (next_trial) || GET_CODE (PATTERN (next_trial)) == RETURN))) { if (eligible_for_annul_false (insn, 0, trial)) { if (invert_jump (insn, JUMP_LABEL (insn))) INSN_FROM_TARGET_P (trial) = 1; else if (! eligible_for_annul_true (insn, 0, trial)) return 0; } delay_list = add_to_delay_list (trial, NULL_RTX); next_trial = next_active_insn (trial); update_block (trial, trial); delete_insn (trial); /* Also, if we are targeting an unconditional branch, thread our jump to the target of that branch. Don't change this into a RETURN here, because it may not accept what we have in the delay slot. We'll fix this up later. */ if (next_trial && GET_CODE (next_trial) == JUMP_INSN && (simplejump_p (next_trial) || GET_CODE (PATTERN (next_trial)) == RETURN)) { target_label = JUMP_LABEL (next_trial); if (target_label == 0) target_label = find_end_label (); redirect_jump (insn, target_label); } INSN_ANNULLED_BRANCH_P (insn) = 1; } return delay_list; } #endif /* Return truth value of the statement that this branch is mostly taken. If we think that the branch is extremely likely to be taken, we return 2. If the branch is slightly more likely to be taken, return 1. Otherwise, return 0. CONDITION, if non-zero, is the condition that JUMP_INSN is testing. */ static int mostly_true_jump (jump_insn, condition) rtx jump_insn, condition; { rtx target_label = JUMP_LABEL (jump_insn); rtx insn; /* If this is a conditional return insn, assume it won't return. */ if (target_label == 0) return 0; /* If TARGET_LABEL has no jumps between it and the end of the function, this is essentially a conditional return, so predict it as false. */ for (insn = NEXT_INSN (target_label); insn && GET_CODE (insn) != JUMP_INSN; insn = NEXT_INSN (insn)) ; if (insn == 0) return 0; /* If this is the test of a loop, it is very likely true. We scan backwards from the target label. If we find a NOTE_INSN_LOOP_BEG before the next real insn, we assume the branch is to the top of the loop. */ for (insn = PREV_INSN (target_label); insn && GET_CODE (insn) == NOTE; insn = PREV_INSN (insn)) if (NOTE_LINE_NUMBER (insn) == NOTE_INSN_LOOP_BEG) return 2; /* If we couldn't figure out what this jump was, assume it won't be taken. This should be rare. */ if (condition == 0) return 0; /* EQ tests are usually false and NE tests are usually true. Also, most quantities are positive, so we can make the appropriate guesses about signed comparisons against zero. */ switch (GET_CODE (condition)) { case CONST_INT: /* Unconditional branch. */ return 1; case EQ: return 0; case NE: return 1; case LE: case LT: if (XEXP (condition, 1) == const0_rtx) return 0; break; case GE: case GT: if (XEXP (condition, 1) == const0_rtx) return 1; break; } /* Predict backward branches usually take, forward branches usually not. If we don't know whether this is forward or backward, assume the branch will be taken, since most are. */ return (INSN_UID (jump_insn) > max_uid || INSN_UID (target_label) > max_uid || (uid_to_ruid[INSN_UID (jump_insn)] > uid_to_ruid[INSN_UID (target_label)]));; } /* Return the condition under which INSN will branch to TARGET. If TARGET is zero, return the condition under which INSN will return. If INSN is an unconditional branch, return const_true_rtx. If INSN isn't a simple type of jump, or it doesn't go to TARGET, return 0. */ static rtx get_branch_condition (insn, target) rtx insn; rtx target; { rtx pat = PATTERN (insn); rtx src; if (GET_CODE (pat) == RETURN) return target == 0 ? const_true_rtx : 0; else if (GET_CODE (pat) != SET || SET_DEST (pat) != pc_rtx) return 0; src = SET_SRC (pat); if (GET_CODE (src) == LABEL_REF && XEXP (src, 0) == target) return const_true_rtx; else if (GET_CODE (src) == IF_THEN_ELSE && ((target == 0 && GET_CODE (XEXP (src, 1)) == RETURN) || (GET_CODE (XEXP (src, 1)) == LABEL_REF && XEXP (XEXP (src, 1), 0) == target)) && XEXP (src, 2) == pc_rtx) return XEXP (src, 0); else if (GET_CODE (src) == IF_THEN_ELSE && ((target == 0 && GET_CODE (XEXP (src, 2)) == RETURN) || (GET_CODE (XEXP (src, 2)) == LABEL_REF && XEXP (XEXP (src, 2), 0) == target)) && XEXP (src, 1) == pc_rtx) return gen_rtx (reverse_condition (GET_CODE (XEXP (src, 0))), GET_MODE (XEXP (src, 0)), XEXP (XEXP (src, 0), 0), XEXP (XEXP (src, 0), 1)); return 0; } /* Return non-zero if CONDITION is more strict than the condition of INSN, i.e., if INSN will always branch if CONDITION is true. */ static int condition_dominates_p (condition, insn) rtx condition; rtx insn; { rtx other_condition = get_branch_condition (insn, JUMP_LABEL (insn)); enum rtx_code code = GET_CODE (condition); enum rtx_code other_code; if (rtx_equal_p (condition, other_condition) || other_condition == const_true_rtx) return 1; else if (condition == const_true_rtx || other_condition == 0) return 0; other_code = GET_CODE (other_condition); if (GET_RTX_LENGTH (code) != 2 || GET_RTX_LENGTH (other_code) != 2 || ! rtx_equal_p (XEXP (condition, 0), XEXP (other_condition, 0)) || ! rtx_equal_p (XEXP (condition, 1), XEXP (other_condition, 1))) return 0; return comparison_dominates_p (code, other_code); } /* INSN branches to an insn whose pattern SEQ is a SEQUENCE. Given that the condition tested by INSN is CONDITION and the resources shown in OTHER_NEEDED are needed after INSN, see whether INSN can take all the insns from SEQ's delay list, in addition to whatever insns it may execute (in DELAY_LIST). SETS and NEEDED are denote resources already set and needed while searching for delay slot insns. Return the concatenated delay list if possible, otherwise, return 0. SLOTS_TO_FILL is the total number of slots required by INSN, and PSLOTS_FILLED points to the number filled so far (also the number of insns in DELAY_LIST). It is updated with the number that have been filled from the SEQUENCE, if any. PANNUL_P points to a non-zero value if we already know that we need to annul INSN. If this routine determines that annulling is needed, it may set that value non-zero. PNEW_THREAD points to a location that is to receive the place at which execution should continue. */ static rtx steal_delay_list_from_target (insn, condition, seq, delay_list, sets, needed, other_needed, slots_to_fill, pslots_filled, pannul_p, pnew_thread) rtx insn, condition; rtx seq; rtx delay_list; struct resources *sets, *needed, *other_needed; int slots_to_fill; int *pslots_filled; int *pannul_p; rtx *pnew_thread; { rtx temp; int slots_remaining = slots_to_fill - *pslots_filled; int total_slots_filled = *pslots_filled; rtx new_delay_list = 0; int must_annul = *pannul_p; int i; /* We can't do anything if there are more delay slots in SEQ than we can handle, or if we don't know that it will be a taken branch. We know that it will be a taken branch if it is either an unconditional branch or a conditional branch with a stricter branch condition. */ if (XVECLEN (seq, 0) - 1 > slots_remaining || ! condition_dominates_p (condition, XVECEXP (seq, 0, 0))) return delay_list; for (i = 1; i < XVECLEN (seq, 0); i++) { rtx trial = XVECEXP (seq, 0, i); if (insn_references_resource_p (trial, sets, 0) || insn_sets_resource_p (trial, needed, 0) || insn_sets_resource_p (trial, sets, 0) #ifdef HAVE_cc0 /* If TRIAL sets CC0, we can't copy it, so we can't steal this delay list. */ || find_reg_note (trial, REG_CC_USER, NULL_RTX) #endif /* If TRIAL is from the fallthrough code of an annulled branch insn in SEQ, we cannot use it. */ || (INSN_ANNULLED_BRANCH_P (XVECEXP (seq, 0, 0)) && ! INSN_FROM_TARGET_P (trial))) return delay_list; /* If this insn was already done (usually in a previous delay slot), pretend we put it in our delay slot. */ if (redundant_insn_p (trial, insn, new_delay_list)) continue; if (! must_annul && ((condition == const_true_rtx || (! insn_sets_resource_p (trial, other_needed, 0) && ! may_trap_p (PATTERN (trial))))) ? eligible_for_delay (insn, total_slots_filled, trial) : (must_annul = 1, eligible_for_annul_false (insn, total_slots_filled, trial))) { temp = copy_rtx (trial); INSN_FROM_TARGET_P (temp) = 1; new_delay_list = add_to_delay_list (temp, new_delay_list); total_slots_filled++; if (--slots_remaining == 0) break; } else return delay_list; } /* Show the place to which we will be branching. */ *pnew_thread = next_active_insn (JUMP_LABEL (XVECEXP (seq, 0, 0))); /* Add any new insns to the delay list and update the count of the number of slots filled. */ *pslots_filled = total_slots_filled; *pannul_p = must_annul; if (delay_list == 0) return new_delay_list; for (temp = new_delay_list; temp; temp = XEXP (temp, 1)) delay_list = add_to_delay_list (XEXP (temp, 0), delay_list); return delay_list; } /* Similar to steal_delay_list_from_target except that SEQ is on the fallthrough path of INSN. Here we only do something if the delay insn of SEQ is an unconditional branch. In that case we steal its delay slot for INSN since unconditional branches are much easier to fill. */ static rtx steal_delay_list_from_fallthrough (insn, condition, seq, delay_list, sets, needed, other_needed, slots_to_fill, pslots_filled, pannul_p) rtx insn, condition; rtx seq; rtx delay_list; struct resources *sets, *needed, *other_needed; int slots_to_fill; int *pslots_filled; int *pannul_p; { int i; /* We can't do anything if SEQ's delay insn isn't an unconditional branch. */ if (! simplejump_p (XVECEXP (seq, 0, 0)) && GET_CODE (PATTERN (XVECEXP (seq, 0, 0))) != RETURN) return delay_list; for (i = 1; i < XVECLEN (seq, 0); i++) { rtx trial = XVECEXP (seq, 0, i); /* If TRIAL sets CC0, stealing it will move it too far from the use of CC0. */ if (insn_references_resource_p (trial, sets, 0) || insn_sets_resource_p (trial, needed, 0) || insn_sets_resource_p (trial, sets, 0) #ifdef HAVE_cc0 || sets_cc0_p (PATTERN (trial)) #endif ) break; /* If this insn was already done, we don't need it. */ if (redundant_insn_p (trial, insn, delay_list)) { delete_from_delay_slot (trial); continue; } if (! *pannul_p && ((condition == const_true_rtx || (! insn_sets_resource_p (trial, other_needed, 0) && ! may_trap_p (PATTERN (trial))))) ? eligible_for_delay (insn, *pslots_filled, trial) : (*pannul_p = 1, eligible_for_annul_true (insn, *pslots_filled, trial))) { delete_from_delay_slot (trial); delay_list = add_to_delay_list (trial, delay_list); if (++(*pslots_filled) == slots_to_fill) break; } else break; } return delay_list; } /* Try merging insns starting at THREAD which match exactly the insns in INSN's delay list. If all insns were matched and the insn was previously annulling, the annul bit will be cleared. For each insn that is merged, if the branch is or will be non-annulling, we delete the merged insn. */ static void try_merge_delay_insns (insn, thread) rtx insn, thread; { rtx trial, next_trial; rtx delay_insn = XVECEXP (PATTERN (insn), 0, 0); int annul_p = INSN_ANNULLED_BRANCH_P (delay_insn); int slot_number = 1; int num_slots = XVECLEN (PATTERN (insn), 0); rtx next_to_match = XVECEXP (PATTERN (insn), 0, slot_number); struct resources set, needed; rtx merged_insns = 0; int i; CLEAR_RESOURCE (&needed); CLEAR_RESOURCE (&set); /* If this is not an annulling branch, take into account anything needed in NEXT_TO_MATCH. This prevents two increments from being incorrectly folded into one. If we are annulling, this would be the correct thing to do. (The alternative, looking at things set in NEXT_TO_MATCH will essentially disable this optimization. This method is somewhat of a kludge, but I don't see a better way.) */ if (! annul_p) mark_referenced_resources (next_to_match, &needed, 1); for (trial = thread; !stop_search_p (trial, 1); trial = next_trial) { rtx pat = PATTERN (trial); next_trial = next_nonnote_insn (trial); /* TRIAL must be a CALL_INSN or INSN. Skip USE and CLOBBER. */ if (GET_CODE (trial) == INSN && (GET_CODE (pat) == USE || GET_CODE (pat) == CLOBBER)) continue; if (GET_CODE (next_to_match) == GET_CODE (trial) #ifdef HAVE_cc0 /* We can't share an insn that sets cc0. */ && ! sets_cc0_p (pat) #endif && ! insn_references_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &needed, 1) && (trial = try_split (pat, trial, 0)) != 0 && rtx_equal_p (PATTERN (next_to_match), PATTERN (trial)) /* Have to test this condition if annul condition is different from (and less restrictive than) non-annulling one. */ && eligible_for_delay (delay_insn, slot_number - 1, trial)) { next_trial = next_nonnote_insn (trial); if (! annul_p) { update_block (trial, thread); delete_insn (trial); INSN_FROM_TARGET_P (next_to_match) = 0; } else merged_insns = gen_rtx (INSN_LIST, VOIDmode, trial, merged_insns); if (++slot_number == num_slots) break; next_to_match = XVECEXP (PATTERN (insn), 0, slot_number); if (! annul_p) mark_referenced_resources (next_to_match, &needed, 1); } mark_set_resources (trial, &set, 1); mark_referenced_resources (trial, &needed, 1); } /* See if we stopped on a filled insn. If we did, try to see if its delay slots match. */ if (slot_number != num_slots && trial && GET_CODE (trial) == INSN && GET_CODE (PATTERN (trial)) == SEQUENCE && ! INSN_ANNULLED_BRANCH_P (XVECEXP (PATTERN (trial), 0, 0))) { rtx pat = PATTERN (trial); for (i = 1; i < XVECLEN (pat, 0); i++) { rtx dtrial = XVECEXP (pat, 0, i); if (! insn_references_resource_p (dtrial, &set, 1) && ! insn_sets_resource_p (dtrial, &set, 1) && ! insn_sets_resource_p (dtrial, &needed, 1) #ifdef HAVE_cc0 && ! sets_cc0_p (PATTERN (dtrial)) #endif && rtx_equal_p (PATTERN (next_to_match), PATTERN (dtrial)) && eligible_for_delay (delay_insn, slot_number - 1, dtrial)) { if (! annul_p) { update_block (dtrial, thread); delete_from_delay_slot (dtrial); INSN_FROM_TARGET_P (next_to_match) = 0; } else merged_insns = gen_rtx (INSN_LIST, SImode, dtrial, merged_insns); if (++slot_number == num_slots) break; next_to_match = XVECEXP (PATTERN (insn), 0, slot_number); } } } /* If all insns in the delay slot have been matched and we were previously annulling the branch, we need not any more. In that case delete all the merged insns. Also clear the INSN_FROM_TARGET_P bit of each insn the the delay list so that we know that it isn't only being used at the target. */ if (next_to_match == 0 && annul_p) { for (; merged_insns; merged_insns = XEXP (merged_insns, 1)) { if (GET_MODE (merged_insns) == SImode) { update_block (XEXP (merged_insns, 0), thread); delete_from_delay_slot (XEXP (merged_insns, 0)); } else { update_block (XEXP (merged_insns, 0), thread); delete_insn (XEXP (merged_insns, 0)); } } INSN_ANNULLED_BRANCH_P (delay_insn) = 0; for (i = 0; i < XVECLEN (PATTERN (insn), 0); i++) INSN_FROM_TARGET_P (XVECEXP (PATTERN (insn), 0, i)) = 0; } } /* See if INSN is redundant with an insn in front of TARGET. Often this is called when INSN is a candidate for a delay slot of TARGET. DELAY_LIST are insns that will be placed in delay slots of TARGET in front of INSN. Often INSN will be redundant with an insn in a delay slot of some previous insn. This happens when we have a series of branches to the same label; in that case the first insn at the target might want to go into each of the delay slots. If we are not careful, this routine can take up a significant fraction of the total compilation time (4%), but only wins rarely. Hence we speed this routine up by making two passes. The first pass goes back until it hits a label and sees if it find an insn with an identical pattern. Only in this (relatively rare) event does it check for data conflicts. We do not split insns we encounter. This could cause us not to find a redundant insn, but the cost of splitting seems greater than the possible gain in rare cases. */ static int redundant_insn_p (insn, target, delay_list) rtx insn; rtx target; rtx delay_list; { rtx target_main = target; rtx ipat = PATTERN (insn); rtx trial, pat; struct resources needed, set; int i; /* Scan backwards looking for a match. */ for (trial = PREV_INSN (target); trial; trial = PREV_INSN (trial)) { if (GET_CODE (trial) == CODE_LABEL) return 0; if (GET_CODE (trial) != INSN && GET_CODE (trial) != JUMP_INSN && GET_CODE (trial) != JUMP_INSN) continue; pat = PATTERN (trial); if (GET_CODE (pat) == USE || GET_CODE (pat) == CLOBBER) continue; if (GET_CODE (pat) == SEQUENCE) { /* Stop for a CALL and its delay slots because it difficult to track its resource needs correctly. */ if (GET_CODE (XVECEXP (pat, 0, 0)) == CALL_INSN) return 0; for (i = XVECLEN (pat, 0) - 1; i > 0; i--) if (GET_CODE (XVECEXP (pat, 0, i)) == GET_CODE (insn) && rtx_equal_p (PATTERN (XVECEXP (pat, 0, i)), ipat)) break; /* If found a match, exit this loop early. */ if (i > 0) break; } else if (GET_CODE (trial) == GET_CODE (insn) && rtx_equal_p (pat, ipat)) break; } /* If we didn't find an insn that matches, return 0. */ if (trial == 0) return 0; /* See what resources this insn sets and needs. If they overlap, or if this insn references CC0, it can't be redundant. */ CLEAR_RESOURCE (&needed); CLEAR_RESOURCE (&set); mark_set_resources (insn, &set, 1); mark_referenced_resources (insn, &needed, 1); /* If TARGET is a SEQUENCE, get the main insn. */ if (GET_CODE (target) == INSN && GET_CODE (PATTERN (target)) == SEQUENCE) target_main = XVECEXP (PATTERN (target), 0, 0); if (resource_conflicts_p (&needed, &set) #ifdef HAVE_cc0 || reg_mentioned_p (cc0_rtx, ipat) #endif /* The insn requiring the delay may not set anything needed or set by INSN. */ || insn_sets_resource_p (target_main, &needed, 1) || insn_sets_resource_p (target_main, &set, 1)) return 0; /* Insns we pass may not set either NEEDED or SET, so merge them for simpler tests. */ needed.memory |= set.memory; IOR_HARD_REG_SET (needed.regs, set.regs); /* This insn isn't redundant if it conflicts with an insn that either is or will be in a delay slot of TARGET. */ while (delay_list) { if (insn_sets_resource_p (XEXP (delay_list, 0), &needed, 1)) return 0; delay_list = XEXP (delay_list, 1); } if (GET_CODE (target) == INSN && GET_CODE (PATTERN (target)) == SEQUENCE) for (i = 1; i < XVECLEN (PATTERN (target), 0); i++) if (insn_sets_resource_p (XVECEXP (PATTERN (target), 0, i), &needed, 1)) return 0; /* Scan backwards until we reach a label or an insn that uses something INSN sets or sets something insn uses or sets. */ for (trial = PREV_INSN (target); trial && GET_CODE (trial) != CODE_LABEL; trial = PREV_INSN (trial)) { if (GET_CODE (trial) != INSN && GET_CODE (trial) != CALL_INSN && GET_CODE (trial) != JUMP_INSN) continue; pat = PATTERN (trial); if (GET_CODE (pat) == USE || GET_CODE (pat) == CLOBBER) continue; if (GET_CODE (pat) == SEQUENCE) { /* If this is a CALL_INSN and its delay slots, it is hard to track the resource needs properly, so give up. */ if (GET_CODE (XVECEXP (pat, 0, 0)) == CALL_INSN) return 0; /* See if any of the insns in the delay slot match, updating resource requirements as we go. */ for (i = XVECLEN (pat, 0) - 1; i > 0; i--) { rtx candidate = XVECEXP (pat, 0, i); /* If an insn will be annulled if the branch is false, it isn't considered as a possible duplicate insn. */ if (rtx_equal_p (PATTERN (candidate), ipat) && ! (INSN_ANNULLED_BRANCH_P (XVECEXP (pat, 0, 0)) && INSN_FROM_TARGET_P (candidate))) { /* Show that this insn will be used in the sequel. */ INSN_FROM_TARGET_P (candidate) = 0; return 1; } /* Unless this is an annulled insn from the target of a branch, we must stop if it sets anything needed or set by INSN. */ if ((! INSN_ANNULLED_BRANCH_P (XVECEXP (pat, 0, 0)) || ! INSN_FROM_TARGET_P (candidate)) && insn_sets_resource_p (candidate, &needed, 1)) return 0; } /* If the insn requiring the delay slot conflicts with INSN, we must stop. */ if (insn_sets_resource_p (XVECEXP (pat, 0, 0), &needed, 1)) return 0; } else { /* See if TRIAL is the same as INSN. */ pat = PATTERN (trial); if (rtx_equal_p (pat, ipat)) return 1; /* Can't go any further if TRIAL conflicts with INSN. */ if (insn_sets_resource_p (trial, &needed, 1)) return 0; } } return 0; } /* Return 1 if THREAD can only be executed in one way. If LABEL is non-zero, it is the target of the branch insn being scanned. If ALLOW_FALLTHROUGH is non-zero, we are allowed to fall into this thread; otherwise, we are not. If LABEL is used more than one or we pass a label other than LABEL before finding an active insn, we do not own this thread. */ static int own_thread_p (thread, label, allow_fallthrough) rtx thread; rtx label; int allow_fallthrough; { rtx active_insn; rtx insn; /* We don't own the function end. */ if (thread == 0) return 0; /* Get the first active insn, or THREAD, if it is an active insn. */ active_insn = next_active_insn (PREV_INSN (thread)); for (insn = thread; insn != active_insn; insn = NEXT_INSN (insn)) if (GET_CODE (insn) == CODE_LABEL && (insn != label || LABEL_NUSES (insn) != 1)) return 0; if (allow_fallthrough) return 1; /* Ensure that we reach a BARRIER before any insn or label. */ for (insn = prev_nonnote_insn (thread); insn == 0 || GET_CODE (insn) != BARRIER; insn = prev_nonnote_insn (insn)) if (insn == 0 || GET_CODE (insn) == CODE_LABEL || (GET_CODE (insn) == INSN && GET_CODE (PATTERN (insn)) != USE && GET_CODE (PATTERN (insn)) != CLOBBER)) return 0; return 1; } /* Find the number of the basic block that starts closest to INSN. Return -1 if we couldn't find such a basic block. */ static int find_basic_block (insn) rtx insn; { int i; /* Scan backwards to the previous BARRIER. Then see if we can find a label that starts a basic block. Return the basic block number. */ for (insn = prev_nonnote_insn (insn); insn && GET_CODE (insn) != BARRIER; insn = prev_nonnote_insn (insn)) ; /* The start of the function is basic block zero. */ if (insn == 0) return 0; /* See if any of the upcoming CODE_LABELs start a basic block. If we reach anything other than a CODE_LABEL or note, we can't find this code. */ for (insn = next_nonnote_insn (insn); insn && GET_CODE (insn) == CODE_LABEL; insn = next_nonnote_insn (insn)) { for (i = 0; i < n_basic_blocks; i++) if (insn == basic_block_head[i]) return i; } return -1; } /* Called when INSN is being moved from a location near the target of a jump. We leave a marker of the form (use (INSN)) immediately in front of WHERE for mark_target_live_regs. These markers will be deleted when reorg finishes. We used to try to update the live status of registers if WHERE is at the start of a basic block, but that can't work since we may remove a BARRIER in relax_delay_slots. */ static void update_block (insn, where) rtx insn; rtx where; { int b; /* Ignore if this was in a delay slot and it came from the target of a branch. */ if (INSN_FROM_TARGET_P (insn)) return; emit_insn_before (gen_rtx (USE, VOIDmode, insn), where); /* INSN might be making a value live in a block where it didn't use to be. So recompute liveness information for this block. */ b = find_basic_block (insn); if (b != -1) bb_ticks[b]++; } /* Marks registers possibly live at the current place being scanned by mark_target_live_regs. Used only by next two function. */ static HARD_REG_SET current_live_regs; /* Marks registers for which we have seen a REG_DEAD note but no assignment. Also only used by the next two functions. */ static HARD_REG_SET pending_dead_regs; /* Utility function called from mark_target_live_regs via note_stores. It deadens any CLOBBERed registers and livens any SET registers. */ static void update_live_status (dest, x) rtx dest; rtx x; { int first_regno, last_regno; int i; if (GET_CODE (dest) != REG && (GET_CODE (dest) != SUBREG || GET_CODE (SUBREG_REG (dest)) != REG)) return; if (GET_CODE (dest) == SUBREG) first_regno = REGNO (SUBREG_REG (dest)) + SUBREG_WORD (dest); else first_regno = REGNO (dest); last_regno = first_regno + HARD_REGNO_NREGS (first_regno, GET_MODE (dest)); if (GET_CODE (x) == CLOBBER) for (i = first_regno; i < last_regno; i++) CLEAR_HARD_REG_BIT (current_live_regs, i); else for (i = first_regno; i < last_regno; i++) { SET_HARD_REG_BIT (current_live_regs, i); CLEAR_HARD_REG_BIT (pending_dead_regs, i); } } /* Similar to next_insn, but ignores insns in the delay slots of an annulled branch. */ static rtx next_insn_no_annul (insn) rtx insn; { if (insn) { /* If INSN is an annulled branch, skip any insns from the target of the branch. */ if (INSN_ANNULLED_BRANCH_P (insn) && NEXT_INSN (PREV_INSN (insn)) != insn) while (INSN_FROM_TARGET_P (NEXT_INSN (insn))) insn = NEXT_INSN (insn); insn = NEXT_INSN (insn); if (insn && GET_CODE (insn) == INSN && GET_CODE (PATTERN (insn)) == SEQUENCE) insn = XVECEXP (PATTERN (insn), 0, 0); } return insn; } /* Set the resources that are live at TARGET. If TARGET is zero, we refer to the end of the current function and can return our precomputed value. Otherwise, we try to find out what is live by consulting the basic block information. This is tricky, because we must consider the actions of reload and jump optimization, which occur after the basic block information has been computed. Accordingly, we proceed as follows:: We find the previous BARRIER and look at all immediately following labels (with no intervening active insns) to see if any of them start a basic block. If we hit the start of the function first, we use block 0. Once we have found a basic block and a corresponding first insns, we can accurately compute the live status from basic_block_live_regs and reg_renumber. (By starting at a label following a BARRIER, we are immune to actions taken by reload and jump.) Then we scan all insns between that point and our target. For each CLOBBER (or for call-clobbered regs when we pass a CALL_INSN), mark the appropriate registers are dead. For a SET, mark them as live. We have to be careful when using REG_DEAD notes because they are not updated by such things as find_equiv_reg. So keep track of registers marked as dead that haven't been assigned to, and mark them dead at the next CODE_LABEL since reload and jump won't propagate values across labels. If we cannot find the start of a basic block (should be a very rare case, if it can happen at all), mark everything as potentially live. Next, scan forward from TARGET looking for things set or clobbered before they are used. These are not live. Because we can be called many times on the same target, save our results in a hash table indexed by INSN_UID. */ static void mark_target_live_regs (target, res) rtx target; struct resources *res; { int b = -1; int i; struct target_info *tinfo; rtx insn, next; rtx jump_insn = 0; HARD_REG_SET scratch; struct resources set, needed; int jump_count = 0; /* Handle end of function. */ if (target == 0) { *res = end_of_function_needs; return; } /* We have to assume memory is needed, but the CC isn't. */ res->memory = 1; res->volatil = 0; res->cc = 0; /* See if we have computed this value already. */ for (tinfo = target_hash_table[INSN_UID (target) % TARGET_HASH_PRIME]; tinfo; tinfo = tinfo->next) if (tinfo->uid == INSN_UID (target)) break; /* Start by getting the basic block number. If we have saved information, we can get it from there unless the insn at the start of the basic block has been deleted. */ if (tinfo && tinfo->block != -1 && ! INSN_DELETED_P (basic_block_head[tinfo->block])) b = tinfo->block; if (b == -1) b = find_basic_block (target); if (tinfo) { /* If the information is up-to-date, use it. Otherwise, we will update it below. */ if (b == tinfo->block && b != -1 && tinfo->bb_tick == bb_ticks[b]) { COPY_HARD_REG_SET (res->regs, tinfo->live_regs); return; } } else { /* Allocate a place to put our results and chain it into the hash table. */ tinfo = (struct target_info *) oballoc (sizeof (struct target_info)); tinfo->uid = INSN_UID (target); tinfo->block = b; tinfo->next = target_hash_table[INSN_UID (target) % TARGET_HASH_PRIME]; target_hash_table[INSN_UID (target) % TARGET_HASH_PRIME] = tinfo; } CLEAR_HARD_REG_SET (pending_dead_regs); /* If we found a basic block, get the live registers from it and update them with anything set or killed between its start and the insn before TARGET. Otherwise, we must assume everything is live. */ if (b != -1) { regset regs_live = basic_block_live_at_start[b]; int offset, j; REGSET_ELT_TYPE bit; int regno; rtx start_insn, stop_insn; /* Compute hard regs live at start of block -- this is the real hard regs marked live, plus live pseudo regs that have been renumbered to hard regs. */ #ifdef HARD_REG_SET current_live_regs = *regs_live; #else COPY_HARD_REG_SET (current_live_regs, regs_live); #endif for (offset = 0, i = 0; offset < regset_size; offset++) { if (regs_live[offset] == 0) i += REGSET_ELT_BITS; else for (bit = 1; bit && i < max_regno; bit <<= 1, i++) if ((regs_live[offset] & bit) && (regno = reg_renumber[i]) >= 0) for (j = regno; j < regno + HARD_REGNO_NREGS (regno, PSEUDO_REGNO_MODE (i)); j++) SET_HARD_REG_BIT (current_live_regs, j); } /* Get starting and ending insn, handling the case where each might be a SEQUENCE. */ start_insn = (b == 0 ? get_insns () : basic_block_head[b]); stop_insn = target; if (GET_CODE (start_insn) == INSN && GET_CODE (PATTERN (start_insn)) == SEQUENCE) start_insn = XVECEXP (PATTERN (start_insn), 0, 0); if (GET_CODE (stop_insn) == INSN && GET_CODE (PATTERN (stop_insn)) == SEQUENCE) stop_insn = next_insn (PREV_INSN (stop_insn)); for (insn = start_insn; insn != stop_insn; insn = next_insn_no_annul (insn)) { rtx link; rtx real_insn = insn; /* If this insn is from the target of a branch, it isn't going to be used in the sequel. If it is used in both cases, this test will not be true. */ if (INSN_FROM_TARGET_P (insn)) continue; /* If this insn is a USE made by update_block, we care about the underlying insn. */ if (GET_CODE (insn) == INSN && GET_CODE (PATTERN (insn)) == USE && (GET_CODE (XEXP (PATTERN (insn), 0)) == INSN || GET_CODE (XEXP (PATTERN (insn), 0)) == CALL_INSN || GET_CODE (XEXP (PATTERN (insn), 0)) == JUMP_INSN)) real_insn = XEXP (PATTERN (insn), 0); if (GET_CODE (real_insn) == CALL_INSN) { /* CALL clobbers all call-used regs that aren't fixed except sp, ap, and fp. Do this before setting the result of the call live. */ for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (call_used_regs[i] && i != STACK_POINTER_REGNUM && i != FRAME_POINTER_REGNUM && i != ARG_POINTER_REGNUM #if ARG_POINTER_REGNUM != FRAME_POINTER_REGNUM && ! (i == ARG_POINTER_REGNUM && fixed_regs[i]) #endif #ifdef PIC_OFFSET_TABLE_REGNUM && ! (i == PIC_OFFSET_TABLE_REGNUM && flag_pic) #endif ) CLEAR_HARD_REG_BIT (current_live_regs, i); /* A CALL_INSN sets any global register live, since it may have been modified by the call. */ for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (global_regs[i]) SET_HARD_REG_BIT (current_live_regs, i); } /* Mark anything killed in an insn to be deadened at the next label. Ignore USE insns; the only REG_DEAD notes will be for parameters. But they might be early. A CALL_INSN will usually clobber registers used for parameters. It isn't worth bothering with the unlikely case when it won't. */ if ((GET_CODE (real_insn) == INSN && GET_CODE (PATTERN (real_insn)) != USE) || GET_CODE (real_insn) == JUMP_INSN || GET_CODE (real_insn) == CALL_INSN) { for (link = REG_NOTES (real_insn); link; link = XEXP (link, 1)) if (REG_NOTE_KIND (link) == REG_DEAD && GET_CODE (XEXP (link, 0)) == REG && REGNO (XEXP (link, 0)) < FIRST_PSEUDO_REGISTER) { int first_regno = REGNO (XEXP (link, 0)); int last_regno = (first_regno + HARD_REGNO_NREGS (first_regno, GET_MODE (XEXP (link, 0)))); for (i = first_regno; i < last_regno; i++) SET_HARD_REG_BIT (pending_dead_regs, i); } note_stores (PATTERN (real_insn), update_live_status); /* If any registers were unused after this insn, kill them. These notes will always be accurate. */ for (link = REG_NOTES (real_insn); link; link = XEXP (link, 1)) if (REG_NOTE_KIND (link) == REG_UNUSED && GET_CODE (XEXP (link, 0)) == REG && REGNO (XEXP (link, 0)) < FIRST_PSEUDO_REGISTER) { int first_regno = REGNO (XEXP (link, 0)); int last_regno = (first_regno + HARD_REGNO_NREGS (first_regno, GET_MODE (XEXP (link, 0)))); for (i = first_regno; i < last_regno; i++) CLEAR_HARD_REG_BIT (current_live_regs, i); } } if (GET_CODE (real_insn) == CODE_LABEL) { /* A label clobbers the pending dead registers since neither reload nor jump will propagate a value across a label. */ AND_COMPL_HARD_REG_SET (current_live_regs, pending_dead_regs); CLEAR_HARD_REG_SET (pending_dead_regs); } } COPY_HARD_REG_SET (res->regs, current_live_regs); tinfo->block = b; tinfo->bb_tick = bb_ticks[b]; } else /* We didn't find the start of a basic block. Assume everything in use. This should happen only extremely rarely. */ SET_HARD_REG_SET (res->regs); /* Now step forward from TARGET looking for registers that are set before they are used. These are dead. If we pass a label, any pending dead registers that weren't yet used can be made dead. Stop when we pass a conditional JUMP_INSN; follow the first few unconditional branches. */ CLEAR_RESOURCE (&set); CLEAR_RESOURCE (&needed); for (insn = target; insn; insn = next) { rtx main_insn = insn; next = NEXT_INSN (insn); switch (GET_CODE (insn)) { case CODE_LABEL: AND_COMPL_HARD_REG_SET (pending_dead_regs, needed.regs); AND_COMPL_HARD_REG_SET (res->regs, pending_dead_regs); CLEAR_HARD_REG_SET (pending_dead_regs); continue; case BARRIER: case NOTE: continue; case INSN: if (GET_CODE (PATTERN (insn)) == USE || GET_CODE (PATTERN (insn)) == CLOBBER) continue; if (GET_CODE (PATTERN (insn)) == SEQUENCE) main_insn = XVECEXP (PATTERN (insn), 0, 0); } if (GET_CODE (main_insn) == JUMP_INSN) { if (jump_count++ < 10 && (simplejump_p (main_insn) || GET_CODE (PATTERN (main_insn)) == RETURN)) { next = next_active_insn (JUMP_LABEL (main_insn)); if (jump_insn == 0) jump_insn = insn; } else break; } mark_referenced_resources (insn, &needed, 1); mark_set_resources (insn, &set, 1); COPY_HARD_REG_SET (scratch, set.regs); AND_COMPL_HARD_REG_SET (scratch, needed.regs); AND_COMPL_HARD_REG_SET (res->regs, scratch); } /* If we hit an unconditional branch, we have another way of finding out what is live: we can see what is live at the branch target and include anything used but not set before the branch. The only things that are live are those that are live using the above test and the test below. Don't try this if we expired our jump count above, since that would mean there may be an infinite loop in the function being compiled. */ if (jump_insn && jump_count < 10) { rtx jump_target = (GET_CODE (jump_insn) == INSN ? JUMP_LABEL (XVECEXP (PATTERN (jump_insn), 0, 0)) : JUMP_LABEL (jump_insn)); struct resources new_resources; rtx stop_insn = next_active_insn (jump_insn); mark_target_live_regs (next_active_insn (jump_target), &new_resources); CLEAR_RESOURCE (&set); CLEAR_RESOURCE (&needed); /* Include JUMP_INSN in the needed registers. */ for (insn = target; insn != stop_insn; insn = next_active_insn (insn)) { mark_referenced_resources (insn, &needed, 1); COPY_HARD_REG_SET (scratch, needed.regs); AND_COMPL_HARD_REG_SET (scratch, set.regs); IOR_HARD_REG_SET (new_resources.regs, scratch); mark_set_resources (insn, &set, 1); } AND_HARD_REG_SET (res->regs, new_resources.regs); } COPY_HARD_REG_SET (tinfo->live_regs, res->regs); } /* Scan a function looking for insns that need a delay slot and find insns to put into the delay slot. NON_JUMPS_P is non-zero if we are to only try to fill non-jump insns (such as calls). We do these first since we don't want jump insns (that are easier to fill) to get the only insns that could be used for non-jump insns. When it is zero, only try to fill JUMP_INSNs. When slots are filled in this manner, the insns (including the delay_insn) are put together in a SEQUENCE rtx. In this fashion, it is possible to tell whether a delay slot has really been filled or not. `final' knows how to deal with this, by communicating through FINAL_SEQUENCE. */ static void fill_simple_delay_slots (first, non_jumps_p) rtx first; { register rtx insn, pat, trial, next_trial; register int i; int num_unfilled_slots = unfilled_slots_next - unfilled_slots_base; struct resources needed, set; register int slots_to_fill, slots_filled; rtx delay_list; for (i = 0; i < num_unfilled_slots; i++) { /* Get the next insn to fill. If it has already had any slots assigned, we can't do anything with it. Maybe we'll improve this later. */ insn = unfilled_slots_base[i]; if (insn == 0 || INSN_DELETED_P (insn) || (GET_CODE (insn) == INSN && GET_CODE (PATTERN (insn)) == SEQUENCE) || (GET_CODE (insn) == JUMP_INSN && non_jumps_p) || (GET_CODE (insn) != JUMP_INSN && ! non_jumps_p)) continue; slots_to_fill = num_delay_slots (insn); if (slots_to_fill == 0) abort (); /* This insn needs, or can use, some delay slots. SLOTS_TO_FILL says how many. After initialization, scan backwards from the insn to search for a potential delay-slot candidate. Stop searching when a label or jump is hit. For each candidate, if it is to go into the delay slot (moved forward in execution sequence), it must not need or set any resources that were set by later insns and must not set any resources that are needed for those insns. The delay slot insn itself sets resources unless it is a call (in which case the called routine, not the insn itself, is doing the setting). */ slots_filled = 0; delay_list = 0; CLEAR_RESOURCE (&needed); CLEAR_RESOURCE (&set); mark_set_resources (insn, &set, 0); mark_referenced_resources (insn, &needed, 0); for (trial = prev_nonnote_insn (insn); ! stop_search_p (trial, 1); trial = next_trial) { next_trial = prev_nonnote_insn (trial); /* This must be an INSN or CALL_INSN. */ pat = PATTERN (trial); /* USE and CLOBBER at this level was just for flow; ignore it. */ if (GET_CODE (pat) == USE || GET_CODE (pat) == CLOBBER) continue; /* Check for resource conflict first, to avoid unnecessary splitting. */ if (! insn_references_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &needed, 1) #ifdef HAVE_cc0 /* Can't separate set of cc0 from its use. */ && ! (reg_mentioned_p (cc0_rtx, pat) && ! sets_cc0_p (cc0_rtx, pat)) #endif ) { trial = try_split (pat, trial, 1); next_trial = prev_nonnote_insn (trial); if (eligible_for_delay (insn, slots_filled, trial)) { /* In this case, we are searching backward, so if we find insns to put on the delay list, we want to put them at the head, rather than the tail, of the list. */ delay_list = gen_rtx (INSN_LIST, VOIDmode, trial, delay_list); update_block (trial, trial); delete_insn (trial); if (slots_to_fill == ++slots_filled) break; continue; } } mark_set_resources (trial, &set, 1); mark_referenced_resources (trial, &needed, 1); } if (slots_filled == slots_to_fill) /* happy. */ ; /* If all needed slots haven't been filled, we come here. */ /* Try to optimize case of jumping around a single insn. */ #if defined(ANNUL_IFFALSE_SLOTS) || defined(ANNUL_IFTRUE_SLOTS) else if (delay_list == 0 && GET_CODE (insn) == JUMP_INSN && condjump_p (insn)) { delay_list = optimize_skip (insn); if (delay_list) slots_filled += 1; } #endif /* @@ This would be a good place to optimize: call _foo call _foo nop add %o7,.-L1,%o7 b,a L1 nop Someday... */ /* Try to get insns from beyond the insn needing the delay slot. These insns can neither set or reference resources set in insns being skipped, cannot set resources in the insn being skipped, and, if this is a CALL_INSN (or a CALL_INSN is passed), cannot trap (because the call might not return). If this is a conditional jump, see if it merges back to us early enough for us to pick up insns from the merge point. Don't do this if there is another branch to our label unless we pass all of them. Another similar merge is if we jump to the same place that a later unconditional jump branches to. In that case, we don't care about the number of uses of our label. */ else if (GET_CODE (insn) != JUMP_INSN || (condjump_p (insn) && ! simplejump_p (insn) && JUMP_LABEL (insn) != 0)) { rtx target = 0; int maybe_never = 0; int passed_label = 0; int target_uses; struct resources needed_at_jump; CLEAR_RESOURCE (&needed); CLEAR_RESOURCE (&set); if (GET_CODE (insn) == CALL_INSN) { mark_set_resources (insn, &set, 1); mark_referenced_resources (insn, &needed, 1); maybe_never = 1; } else if (GET_CODE (insn) == JUMP_INSN) { /* Get our target and show how many more uses we want to see before we hit the label. */ target = JUMP_LABEL (insn); target_uses = LABEL_NUSES (target) - 1; } for (trial = next_nonnote_insn (insn); trial; trial = next_trial) { rtx pat, trial_delay; next_trial = next_nonnote_insn (trial); if (GET_CODE (trial) == CODE_LABEL) { passed_label = 1; /* If this is our target, see if we have seen all its uses. If so, indicate we have passed our target and ignore it. All other labels cause us to stop our search. */ if (trial == target && target_uses == 0) { target = 0; continue; } else break; } else if (GET_CODE (trial) == BARRIER) break; /* We must have an INSN, JUMP_INSN, or CALL_INSN. */ pat = PATTERN (trial); /* Stand-alone USE and CLOBBER are just for flow. */ if (GET_CODE (pat) == USE || GET_CODE (pat) == CLOBBER) continue; /* If this already has filled delay slots, get the insn needing the delay slots. */ if (GET_CODE (pat) == SEQUENCE) trial_delay = XVECEXP (pat, 0, 0); else trial_delay = trial; /* If this is a jump insn to our target, indicate that we have seen another jump to it. If we aren't handling a conditional jump, stop our search. Otherwise, compute the needs at its target and add them to NEEDED. */ if (GET_CODE (trial_delay) == JUMP_INSN) { if (target == 0) break; else if (JUMP_LABEL (trial_delay) == target) target_uses--; else { mark_target_live_regs (next_active_insn (JUMP_LABEL (trial_delay)), &needed_at_jump); needed.memory |= needed_at_jump.memory; IOR_HARD_REG_SET (needed.regs, needed_at_jump.regs); } } /* See if we have a resource problem before we try to split. */ if (target == 0 && GET_CODE (pat) != SEQUENCE && ! insn_references_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &needed, 1) #ifdef HAVE_cc0 && ! (reg_mentioned_p (cc0_rtx, pat) && ! sets_cc0_p (pat)) #endif && ! (maybe_never && may_trap_p (pat)) && (trial = try_split (pat, trial, 0)) && eligible_for_delay (insn, slots_filled, trial)) { next_trial = next_nonnote_insn (trial); delay_list = add_to_delay_list (trial, delay_list); #ifdef HAVE_cc0 if (reg_mentioned_p (cc0_rtx, pat)) link_cc0_insns (trial); #endif if (passed_label) update_block (trial, trial); delete_insn (trial); if (slots_to_fill == ++slots_filled) break; continue; } mark_set_resources (trial, &set, 1); mark_referenced_resources (trial, &needed, 1); /* Ensure we don't put insns between the setting of cc and the comparison by moving a setting of cc into an earlier delay slot since these insns could clobber the condition code. */ set.cc = 1; /* If this is a call or jump, we might not get here. */ if (GET_CODE (trial) == CALL_INSN || GET_CODE (trial) == JUMP_INSN) maybe_never = 1; } /* If there are slots left to fill and our search was stopped by an unconditional branch, try the insn at the branch target. We can redirect the branch if it works. */ if (slots_to_fill != slots_filled && trial && GET_CODE (trial) == JUMP_INSN && simplejump_p (trial) && (target == 0 || JUMP_LABEL (trial) == target) && (next_trial = next_active_insn (JUMP_LABEL (trial))) != 0 && ! (GET_CODE (next_trial) == INSN && GET_CODE (PATTERN (next_trial)) == SEQUENCE) && ! insn_references_resource_p (next_trial, &set, 1) && ! insn_sets_resource_p (next_trial, &set, 1) && ! insn_sets_resource_p (next_trial, &needed, 1) #ifdef HAVE_cc0 && ! (reg_mentioned_p (cc0_rtx, PATTERN (next_trial)) && ! sets_cc0_p (PATTERN (next_trial))) #endif && ! (maybe_never && may_trap_p (PATTERN (next_trial))) && (next_trial = try_split (PATTERN (next_trial), next_trial, 0)) && eligible_for_delay (insn, slots_filled, next_trial)) { rtx new_label = next_active_insn (next_trial); if (new_label != 0) new_label = get_label_before (new_label); delay_list = add_to_delay_list (copy_rtx (next_trial), delay_list); slots_filled++; redirect_jump (trial, new_label); /* If we merged because we both jumped to the same place, redirect the original insn also. */ if (target) redirect_jump (insn, new_label); } } if (delay_list) unfilled_slots_base[i] = emit_delay_sequence (insn, delay_list, slots_filled, slots_to_fill); if (slots_to_fill == slots_filled) unfilled_slots_base[i] = 0; note_delay_statistics (slots_filled, 0); } #ifdef DELAY_SLOTS_FOR_EPILOGUE /* See if the epilogue needs any delay slots. Try to fill them if so. The only thing we can do is scan backwards from the end of the function. If we did this in a previous pass, it is incorrect to do it again. */ if (current_function_epilogue_delay_list) return; slots_to_fill = DELAY_SLOTS_FOR_EPILOGUE; if (slots_to_fill == 0) return; slots_filled = 0; CLEAR_RESOURCE (&needed); CLEAR_RESOURCE (&set); for (trial = get_last_insn (); ! stop_search_p (trial, 1); trial = PREV_INSN (trial)) { if (GET_CODE (trial) == NOTE) continue; pat = PATTERN (trial); if (GET_CODE (pat) == USE || GET_CODE (pat) == CLOBBER) continue; if (! insn_references_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &needed, 1) #ifdef HAVE_cc0 /* Don't want to mess with cc0 here. */ && ! reg_mentioned_p (cc0_rtx, pat) #endif ) { trial = try_split (pat, trial, 1); if (ELIGIBLE_FOR_EPILOGUE_DELAY (trial, slots_filled)) { /* Here as well we are searching backward, so put the insns we find on the head of the list. */ current_function_epilogue_delay_list = gen_rtx (INSN_LIST, VOIDmode, trial, current_function_epilogue_delay_list); mark_referenced_resources (trial, &end_of_function_needs, 1); update_block (trial, trial); delete_insn (trial); /* Clear deleted bit so final.c will output the insn. */ INSN_DELETED_P (trial) = 0; if (slots_to_fill == ++slots_filled) break; continue; } } mark_set_resources (trial, &set, 1); mark_referenced_resources (trial, &needed, 1); } note_delay_statistics (slots_filled, 0); #endif } /* Try to find insns to place in delay slots. INSN is the jump needing SLOTS_TO_FILL delay slots. It tests CONDITION or is an unconditional branch if CONDITION is const_true_rtx. *PSLOTS_FILLED is updated with the number of slots that we have filled. THREAD is a flow-of-control, either the insns to be executed if the branch is true or if the branch is false, THREAD_IF_TRUE says which. OPPOSITE_THREAD is the thread in the opposite direction. It is used to see if any potential delay slot insns set things needed there. LIKELY is non-zero if it is extremely likely that the branch will be taken and THREAD_IF_TRUE is set. This is used for the branch at the end of a loop back up to the top. OWN_THREAD and OWN_OPPOSITE_THREAD are true if we are the only user of the thread. I.e., it is the fallthrough code of our jump or the target of the jump when we are the only jump going there. If OWN_THREAD is false, it must be the "true" thread of a jump. In that case, we can only take insns from the head of the thread for our delay slot. We then adjust the jump to point after the insns we have taken. */ static rtx fill_slots_from_thread (insn, condition, thread, opposite_thread, likely, thread_if_true, own_thread, own_opposite_thread, slots_to_fill, pslots_filled) rtx insn; rtx condition; rtx thread, opposite_thread; int likely; int thread_if_true; int own_thread, own_opposite_thread; int slots_to_fill, *pslots_filled; { rtx new_thread; rtx delay_list = 0; struct resources opposite_needed, set, needed; rtx trial; int lose = 0; int must_annul = 0; /* Validate our arguments. */ if ((condition == const_true_rtx && ! thread_if_true) || (! own_thread && ! thread_if_true)) abort (); /* If our thread is the end of subroutine, we can't get any delay insns from that. */ if (thread == 0) return 0; /* If this is an unconditional branch, nothing is needed at the opposite thread. Otherwise, compute what is needed there. */ if (condition == const_true_rtx) CLEAR_RESOURCE (&opposite_needed); else mark_target_live_regs (opposite_thread, &opposite_needed); /* If the insn at THREAD can be split, do it here to avoid having to update THREAD and NEW_THREAD if it is done in the loop below. Also initialize NEW_THREAD. */ new_thread = thread = try_split (PATTERN (thread), thread, 0); /* Scan insns at THREAD. We are looking for an insn that can be removed from THREAD (it neither sets nor references resources that were set ahead of it and it doesn't set anything needs by the insns ahead of it) and that either can be placed in an annulling insn or aren't needed at OPPOSITE_THREAD. */ CLEAR_RESOURCE (&needed); CLEAR_RESOURCE (&set); /* If we do not own this thread, we must stop as soon as we find something that we can't put in a delay slot, since all we can do is branch into THREAD at a later point. Therefore, labels stop the search if this is not the `true' thread. */ for (trial = thread; ! stop_search_p (trial, ! thread_if_true) && (! lose || own_thread); trial = next_nonnote_insn (trial)) { rtx pat; /* If we have passed a label, we no longer own this thread. */ if (GET_CODE (trial) == CODE_LABEL) { own_thread = 0; continue; } pat = PATTERN (trial); if (GET_CODE (pat) == USE || GET_CODE (pat) == CLOBBER) continue; /* If TRIAL conflicts with the insns ahead of it, we lose. Also, don't separate or copy insns that set and use CC0. */ if (! insn_references_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &set, 1) && ! insn_sets_resource_p (trial, &needed, 1) #ifdef HAVE_cc0 && ! (reg_mentioned_p (cc0_rtx, pat) && (! own_thread || ! sets_cc0_p (pat))) #endif ) { /* If TRIAL is redundant with some insn before INSN, we don't actually need to add it to the delay list; we can merely pretend we did. */ if (redundant_insn_p (trial, insn, delay_list)) { if (own_thread) { update_block (trial, thread); delete_insn (trial); } else new_thread = next_active_insn (trial); continue; } /* There are two ways we can win: If TRIAL doesn't set anything needed at the opposite thread and can't trap, or if it can go into an annulled delay slot. */ if (condition == const_true_rtx || (! insn_sets_resource_p (trial, &opposite_needed, 1) && ! may_trap_p (pat))) { trial = try_split (pat, trial, 0); pat = PATTERN (trial); if (eligible_for_delay (insn, *pslots_filled, trial)) goto winner; } else if (0 #ifdef ANNUL_IFTRUE_SLOTS || ! thread_if_true #endif #ifdef ANNUL_IFFALSE_SLOTS || thread_if_true #endif ) { trial = try_split (pat, trial, 0); pat = PATTERN (trial); if ((thread_if_true ? eligible_for_annul_false (insn, *pslots_filled, trial) : eligible_for_annul_true (insn, *pslots_filled, trial))) { rtx temp; must_annul = 1; winner: #ifdef HAVE_cc0 if (reg_mentioned_p (cc0_rtx, pat)) link_cc0_insns (trial); #endif /* If we own this thread, delete the insn. If this is the destination of a branch, show that a basic block status may have been updated. In any case, mark the new starting point of this thread. */ if (own_thread) { update_block (trial, thread); delete_insn (trial); } else new_thread = next_active_insn (trial); temp = own_thread ? trial : copy_rtx (trial); if (thread_if_true) INSN_FROM_TARGET_P (temp) = 1; delay_list = add_to_delay_list (temp, delay_list); if (slots_to_fill == ++(*pslots_filled)) { /* Even though we have filled all the slots, we may be branching to a location that has a redundant insn. Skip any if so. */ while (new_thread && ! own_thread && ! insn_sets_resource_p (new_thread, &set, 1) && ! insn_sets_resource_p (new_thread, &needed, 1) && ! insn_references_resource_p (new_thread, &set, 1) && redundant_insn_p (new_thread, insn, delay_list)) new_thread = next_active_insn (new_thread); break; } continue; } } } /* This insn can't go into a delay slot. */ lose = 1; mark_set_resources (trial, &set, 1); mark_referenced_resources (trial, &needed, 1); /* Ensure we don't put insns between the setting of cc and the comparison by moving a setting of cc into an earlier delay slot since these insns could clobber the condition code. */ set.cc = 1; /* If this insn is a register-register copy and the next insn has a use of our destination, change it to use our source. That way, it will become a candidate for our delay slot the next time through this loop. This case occurs commonly in loops that scan a list. We could check for more complex cases than those tested below, but it doesn't seem worth it. It might also be a good idea to try to swap the two insns. That might do better. We can't do this if the next insn modifies our source, because that would make the replacement into the insn invalid. This also prevents updating the contents of a PRE_INC. */ if (GET_CODE (trial) == INSN && GET_CODE (pat) == SET && GET_CODE (SET_SRC (pat)) == REG && GET_CODE (SET_DEST (pat)) == REG) { rtx next = next_nonnote_insn (trial); if (next && GET_CODE (next) == INSN && GET_CODE (PATTERN (next)) != USE && ! reg_set_p (SET_DEST (pat), next) && reg_referenced_p (SET_DEST (pat), PATTERN (next))) validate_replace_rtx (SET_DEST (pat), SET_SRC (pat), next); } } /* If we stopped on a branch insn that has delay slots, see if we can steal some of the insns in those slots. */ if (trial && GET_CODE (trial) == INSN && GET_CODE (PATTERN (trial)) == SEQUENCE && GET_CODE (XVECEXP (PATTERN (trial), 0, 0)) == JUMP_INSN) { /* If this is the `true' thread, we will want to follow the jump, so we can only do this if we have taken everything up to here. */ if (thread_if_true && trial == new_thread) delay_list = steal_delay_list_from_target (insn, condition, PATTERN (trial), delay_list, &set, &needed, &opposite_needed, slots_to_fill, pslots_filled, &must_annul, &new_thread); else if (! thread_if_true) delay_list = steal_delay_list_from_fallthrough (insn, condition, PATTERN (trial), delay_list, &set, &needed, &opposite_needed, slots_to_fill, pslots_filled, &must_annul); } /* If we haven't found anything for this delay slot and it is very likely that the branch will be taken, see if the insn at our target increments or decrements a register with an increment that does not depend on the destination register. If so, try to place the opposite arithmetic insn after the jump insn and put the arithmetic insn in the delay slot. If we can't do this, return. */ if (delay_list == 0 && likely && new_thread && GET_CODE (new_thread) == INSN) { rtx pat = PATTERN (new_thread); rtx dest; rtx src; trial = new_thread; pat = PATTERN (trial); if (GET_CODE (trial) != INSN || GET_CODE (pat) != SET || ! eligible_for_delay (insn, 0, trial)) return 0; dest = SET_DEST (pat), src = SET_SRC (pat); if ((GET_CODE (src) == PLUS || GET_CODE (src) == MINUS) && rtx_equal_p (XEXP (src, 0), dest) && ! reg_overlap_mentioned_p (dest, XEXP (src, 1))) { rtx other = XEXP (src, 1); rtx new_arith; rtx ninsn; /* If this is a constant adjustment, use the same code with the negated constant. Otherwise, reverse the sense of the arithmetic. */ if (GET_CODE (other) == CONST_INT) new_arith = gen_rtx (GET_CODE (src), GET_MODE (src), dest, negate_rtx (GET_MODE (src), other)); else new_arith = gen_rtx (GET_CODE (src) == PLUS ? MINUS : PLUS, GET_MODE (src), dest, other); ninsn = emit_insn_after (gen_rtx (SET, VOIDmode, dest, new_arith), insn); if (recog_memoized (ninsn) < 0 || (insn_extract (ninsn), ! constrain_operands (INSN_CODE (ninsn), 1))) { delete_insn (ninsn); return 0; } if (own_thread) { update_block (trial, thread); delete_insn (trial); } else new_thread = next_active_insn (trial); ninsn = own_thread ? trial : copy_rtx (trial); if (thread_if_true) INSN_FROM_TARGET_P (ninsn) = 1; delay_list = add_to_delay_list (ninsn, NULL_RTX); (*pslots_filled)++; } } if (delay_list && must_annul) INSN_ANNULLED_BRANCH_P (insn) = 1; /* If we are to branch into the middle of this thread, find an appropriate label or make a new one if none, and redirect INSN to it. If we hit the end of the function, use the end-of-function label. */ if (new_thread != thread) { rtx label; if (! thread_if_true) abort (); if (new_thread && GET_CODE (new_thread) == JUMP_INSN && (simplejump_p (new_thread) || GET_CODE (PATTERN (new_thread)) == RETURN)) new_thread = follow_jumps (JUMP_LABEL (new_thread)); if (new_thread == 0) label = find_end_label (); else if (GET_CODE (new_thread) == CODE_LABEL) label = new_thread; else label = get_label_before (new_thread); redirect_jump (insn, label); } return delay_list; } /* Make another attempt to find insns to place in delay slots. We previously looked for insns located in front of the delay insn and, for non-jump delay insns, located behind the delay insn. Here only try to schedule jump insns and try to move insns from either the target or the following insns into the delay slot. If annulling is supported, we will be likely to do this. Otherwise, we can do this only if safe. */ static void fill_eager_delay_slots (first) rtx first; { register rtx insn; register int i; int num_unfilled_slots = unfilled_slots_next - unfilled_slots_base; for (i = 0; i < num_unfilled_slots; i++) { rtx condition; rtx target_label, insn_at_target, fallthrough_insn; rtx delay_list = 0; int own_target; int own_fallthrough; int prediction, slots_to_fill, slots_filled; insn = unfilled_slots_base[i]; if (insn == 0 || INSN_DELETED_P (insn) || GET_CODE (insn) != JUMP_INSN || ! condjump_p (insn)) continue; slots_to_fill = num_delay_slots (insn); if (slots_to_fill == 0) abort (); slots_filled = 0; target_label = JUMP_LABEL (insn); condition = get_branch_condition (insn, target_label); if (condition == 0) continue; /* Get the next active fallthough and target insns and see if we own them. Then see whether the branch is likely true. We don't need to do a lot of this for unconditional branches. */ insn_at_target = next_active_insn (target_label); own_target = own_thread_p (target_label, target_label, 0); if (condition == const_true_rtx) { own_fallthrough = 0; fallthrough_insn = 0; prediction = 2; } else { fallthrough_insn = next_active_insn (insn); own_fallthrough = own_thread_p (NEXT_INSN (insn), NULL_RTX, 1); prediction = mostly_true_jump (insn, condition); } /* If this insn is expected to branch, first try to get insns from our target, then our fallthrough insns. If it is not, expected to branch, try the other order. */ if (prediction) { delay_list = fill_slots_from_thread (insn, condition, insn_at_target, fallthrough_insn, prediction == 2, 1, own_target, own_fallthrough, slots_to_fill, &slots_filled); if (delay_list == 0 && own_fallthrough) { /* Even though we didn't find anything for delay slots, we might have found a redundant insn which we deleted from the thread that was filled. So we have to recompute the next insn at the target. */ target_label = JUMP_LABEL (insn); insn_at_target = next_active_insn (target_label); delay_list = fill_slots_from_thread (insn, condition, fallthrough_insn, insn_at_target, 0, 0, own_fallthrough, own_target, slots_to_fill, &slots_filled); } } else { if (own_fallthrough) delay_list = fill_slots_from_thread (insn, condition, fallthrough_insn, insn_at_target, 0, 0, own_fallthrough, own_target, slots_to_fill, &slots_filled); if (delay_list == 0) delay_list = fill_slots_from_thread (insn, condition, insn_at_target, next_active_insn (insn), 0, 1, own_target, own_fallthrough, slots_to_fill, &slots_filled); } if (delay_list) unfilled_slots_base[i] = emit_delay_sequence (insn, delay_list, slots_filled, slots_to_fill); if (slots_to_fill == slots_filled) unfilled_slots_base[i] = 0; note_delay_statistics (slots_filled, 1); } } /* Once we have tried two ways to fill a delay slot, make a pass over the code to try to improve the results and to do such things as more jump threading. */ static void relax_delay_slots (first) rtx first; { register rtx insn, next, pat; register rtx trial, delay_insn, target_label; /* Look at every JUMP_INSN and see if we can improve it. */ for (insn = first; insn; insn = next) { rtx other; next = next_active_insn (insn); /* If this is a jump insn, see if it now jumps to a jump, jumps to the next insn, or jumps to a label that is not the last of a group of consecutive labels. */ if (GET_CODE (insn) == JUMP_INSN && (target_label = JUMP_LABEL (insn)) != 0) { target_label = follow_jumps (target_label); target_label = prev_label (next_active_insn (target_label)); if (target_label == 0) target_label = find_end_label (); if (next_active_insn (target_label) == next) { delete_jump (insn); continue; } if (target_label != JUMP_LABEL (insn)) redirect_jump (insn, target_label); /* See if this jump branches around a unconditional jump. If so, invert this jump and point it to the target of the second jump. */ if (next && GET_CODE (next) == JUMP_INSN && (simplejump_p (next) || GET_CODE (PATTERN (next)) == RETURN) && next_active_insn (target_label) == next_active_insn (next) && no_labels_between_p (insn, next)) { rtx label = JUMP_LABEL (next); /* Be careful how we do this to avoid deleting code or labels that are momentarily dead. See similar optimization in jump.c. We also need to ensure we properly handle the case when invert_jump fails. */ ++LABEL_NUSES (target_label); if (label) ++LABEL_NUSES (label); if (invert_jump (insn, label)) { delete_insn (next); next = insn; } if (label) --LABEL_NUSES (label); if (--LABEL_NUSES (target_label) == 0) delete_insn (target_label); continue; } } /* If this is an unconditional jump and the previous insn is a conditional jump, try reversing the condition of the previous insn and swapping our targets. The next pass might be able to fill the slots. Don't do this if we expect the conditional branch to be true, because we would then be making the more common case longer. */ if (GET_CODE (insn) == JUMP_INSN && (simplejump_p (insn) || GET_CODE (PATTERN (insn)) == RETURN) && (other = prev_active_insn (insn)) != 0 && condjump_p (other) && no_labels_between_p (other, insn) && ! mostly_true_jump (other, get_branch_condition (other, JUMP_LABEL (other)))) { rtx other_target = JUMP_LABEL (other); /* Increment the count of OTHER_TARGET, so it doesn't get deleted as we move the label. */ if (other_target) ++LABEL_NUSES (other_target); if (invert_jump (other, target_label)) redirect_jump (insn, other_target); if (other_target) --LABEL_NUSES (other_target); } /* Now look only at cases where we have filled a delay slot. */ if (GET_CODE (insn) != INSN || GET_CODE (PATTERN (insn)) != SEQUENCE) continue; pat = PATTERN (insn); delay_insn = XVECEXP (pat, 0, 0); /* See if the first insn in the delay slot is redundant with some previous insn. Remove it from the delay slot if so; then set up to reprocess this insn. */ if (redundant_insn_p (XVECEXP (pat, 0, 1), delay_insn, 0)) { delete_from_delay_slot (XVECEXP (pat, 0, 1)); next = prev_active_insn (next); continue; } /* Now look only at the cases where we have a filled JUMP_INSN. */ if (GET_CODE (XVECEXP (PATTERN (insn), 0, 0)) != JUMP_INSN || ! condjump_p (XVECEXP (PATTERN (insn), 0, 0))) continue; target_label = JUMP_LABEL (delay_insn); if (target_label) { /* If this jump goes to another unconditional jump, thread it, but don't convert a jump into a RETURN here. */ trial = follow_jumps (target_label); trial = prev_label (next_active_insn (trial)); if (trial == 0 && target_label != 0) trial = find_end_label (); if (trial != target_label) { redirect_jump (delay_insn, trial); target_label = trial; } /* If the first insn at TARGET_LABEL is redundant with a previous insn, redirect the jump to the following insn process again. */ trial = next_active_insn (target_label); if (trial && GET_CODE (PATTERN (trial)) != SEQUENCE && redundant_insn_p (trial, insn, 0)) { trial = next_active_insn (trial); if (trial == 0) target_label = find_end_label (); else target_label = get_label_before (trial); redirect_jump (delay_insn, target_label); next = insn; continue; } /* Similarly, if it is an unconditional jump with one insn in its delay list and that insn is redundant, thread the jump. */ if (trial && GET_CODE (PATTERN (trial)) == SEQUENCE && XVECLEN (PATTERN (trial), 0) == 2 && GET_CODE (XVECEXP (PATTERN (trial), 0, 0)) == JUMP_INSN && (simplejump_p (XVECEXP (PATTERN (trial), 0, 0)) || GET_CODE (PATTERN (XVECEXP (PATTERN (trial), 0, 0))) == RETURN) && redundant_insn_p (XVECEXP (PATTERN (trial), 0, 1), insn, 0)) { target_label = JUMP_LABEL (XVECEXP (PATTERN (trial), 0, 0)); if (target_label == 0) target_label = find_end_label (); redirect_jump (delay_insn, target_label); next = insn; continue; } } if (! INSN_ANNULLED_BRANCH_P (delay_insn) && prev_active_insn (target_label) == insn #ifdef HAVE_cc0 /* If the last insn in the delay slot sets CC0 for some insn, various code assumes that it is in a delay slot. We could put it back where it belonged and delete the register notes, but it doesn't seem worthwhile in this uncommon case. */ && ! find_reg_note (XVECEXP (pat, 0, XVECLEN (pat, 0) - 1), REG_CC_USER, NULL_RTX) #endif ) { int i; /* All this insn does is execute its delay list and jump to the following insn. So delete the jump and just execute the delay list insns. We do this by deleting the INSN containing the SEQUENCE, then re-emitting the insns separately, and then deleting the jump. This allows the count of the jump target to be properly decremented. */ /* Clear the from target bit, since these insns are no longer in delay slots. */ for (i = 0; i < XVECLEN (pat, 0); i++) INSN_FROM_TARGET_P (XVECEXP (pat, 0, i)) = 0; trial = PREV_INSN (insn); delete_insn (insn); emit_insn_after (pat, trial); delete_scheduled_jump (delay_insn); continue; } /* See if this jump (with its delay slots) branches around another jump (without delay slots). If so, invert this jump and point it to the target of the second jump. We cannot do this for annulled jumps, though. Again, don't convert a jump to a RETURN here. */ if (! INSN_ANNULLED_BRANCH_P (delay_insn) && next && GET_CODE (next) == JUMP_INSN && (simplejump_p (next) || GET_CODE (PATTERN (next)) == RETURN) && next_active_insn (target_label) == next_active_insn (next) && no_labels_between_p (insn, next)) { rtx label = JUMP_LABEL (next); rtx old_label = JUMP_LABEL (delay_insn); if (label == 0) label = find_end_label (); /* Be careful how we do this to avoid deleting code or labels that are momentarily dead. See similar optimization in jump.c */ if (old_label) ++LABEL_NUSES (old_label); if (invert_jump (delay_insn, label)) { delete_insn (next); next = insn; } if (old_label && --LABEL_NUSES (old_label) == 0) delete_insn (old_label); continue; } /* If we own the thread opposite the way this insn branches, see if we can merge its delay slots with following insns. */ if (INSN_FROM_TARGET_P (XVECEXP (pat, 0, 1)) && own_thread_p (NEXT_INSN (insn), 0, 1)) try_merge_delay_insns (insn, next); else if (! INSN_FROM_TARGET_P (XVECEXP (pat, 0, 1)) && own_thread_p (target_label, target_label, 0)) try_merge_delay_insns (insn, next_active_insn (target_label)); /* If we get here, we haven't deleted INSN. But we may have deleted NEXT, so recompute it. */ next = next_active_insn (insn); } } #ifdef HAVE_return /* Look for filled jumps to the end of function label. We can try to convert them into RETURN insns if the insns in the delay slot are valid for the RETURN as well. */ static void make_return_insns (first) rtx first; { rtx insn, jump_insn, pat; rtx real_return_label = end_of_function_label; int slots, i; /* See if there is a RETURN insn in the function other than the one we made for END_OF_FUNCTION_LABEL. If so, set up anything we can't change into a RETURN to jump to it. */ for (insn = first; insn; insn = NEXT_INSN (insn)) if (GET_CODE (insn) == JUMP_INSN && GET_CODE (PATTERN (insn)) == RETURN) { real_return_label = get_label_before (insn); break; } /* Show an extra usage of REAL_RETURN_LABEL so it won't go away if it was equal to END_OF_FUNCTION_LABEL. */ LABEL_NUSES (real_return_label)++; /* Clear the list of insns to fill so we can use it. */ obstack_free (&unfilled_slots_obstack, unfilled_firstobj); for (insn = first; insn; insn = NEXT_INSN (insn)) { /* Only look at filled JUMP_INSNs that go to the end of function label. */ if (GET_CODE (insn) != INSN || GET_CODE (PATTERN (insn)) != SEQUENCE || GET_CODE (XVECEXP (PATTERN (insn), 0, 0)) != JUMP_INSN || JUMP_LABEL (XVECEXP (PATTERN (insn), 0, 0)) != end_of_function_label) continue; pat = PATTERN (insn); jump_insn = XVECEXP (pat, 0, 0); /* If we can't make the jump into a RETURN, redirect it to the best RETURN and go on to the next insn. */ if (! redirect_jump (jump_insn, NULL_RTX)) { redirect_jump (jump_insn, real_return_label); continue; } /* See if this RETURN can accept the insns current in its delay slot. It can if it has more or an equal number of slots and the contents of each is valid. */ slots = num_delay_slots (jump_insn); if (slots >= XVECLEN (pat, 0) - 1) { for (i = 1; i < XVECLEN (pat, 0); i++) if (! ( #ifdef ANNUL_IFFALSE_SLOTS (INSN_ANNULLED_BRANCH_P (jump_insn) && INSN_FROM_TARGET_P (XVECEXP (pat, 0, i))) ? eligible_for_annul_false (jump_insn, i - 1, XVECEXP (pat, 0, i)) : #endif #ifdef ANNUL_IFTRUE_SLOTS (INSN_ANNULLED_BRANCH_P (jump_insn) && ! INSN_FROM_TARGET_P (XVECEXP (pat, 0, i))) ? eligible_for_annul_true (jump_insn, i - 1, XVECEXP (pat, 0, i)) : #endif eligible_for_delay (jump_insn, i -1, XVECEXP (pat, 0, i)))) break; } else i = 0; if (i == XVECLEN (pat, 0)) continue; /* We have to do something with this insn. If it is an unconditional RETURN, delete the SEQUENCE and output the individual insns, followed by the RETURN. Then set things up so we try to find insns for its delay slots, if it needs some. */ if (GET_CODE (PATTERN (jump_insn)) == RETURN) { rtx prev = PREV_INSN (insn); delete_insn (insn); for (i = 1; i < XVECLEN (pat, 0); i++) prev = emit_insn_after (PATTERN (XVECEXP (pat, 0, i)), prev); insn = emit_jump_insn_after (PATTERN (jump_insn), prev); emit_barrier_after (insn); if (slots) obstack_ptr_grow (&unfilled_slots_obstack, insn); } else /* It is probably more efficient to keep this with its current delay slot as a branch to a RETURN. */ redirect_jump (jump_insn, real_return_label); } /* Now delete REAL_RETURN_LABEL if we never used it. Then try to fill any new delay slots we have created. */ if (--LABEL_NUSES (real_return_label) == 0) delete_insn (real_return_label); fill_simple_delay_slots (first, 1); fill_simple_delay_slots (first, 0); } #endif /* Try to find insns to place in delay slots. */ void dbr_schedule (first, file) rtx first; FILE *file; { rtx insn, next; int i; #if 0 int old_flag_no_peephole = flag_no_peephole; /* Execute `final' once in prescan mode to delete any insns that won't be used. Don't let final try to do any peephole optimization--it will ruin dataflow information for this pass. */ flag_no_peephole = 1; final (first, 0, NO_DEBUG, 1, 1); flag_no_peephole = old_flag_no_peephole; #endif /* Find the highest INSN_UID and allocate and initialize our map from INSN_UID's to position in code. */ for (max_uid = 0, insn = first; insn; insn = NEXT_INSN (insn)) if (INSN_UID (insn) > max_uid) max_uid = INSN_UID (insn); uid_to_ruid = (int *) alloca ((max_uid + 1) * sizeof (int *)); for (i = 0, insn = first; insn; i++, insn = NEXT_INSN (insn)) uid_to_ruid[INSN_UID (insn)] = i; /* Initialize the list of insns that need filling. */ if (unfilled_firstobj == 0) { gcc_obstack_init (&unfilled_slots_obstack); unfilled_firstobj = (rtx *) obstack_alloc (&unfilled_slots_obstack, 0); } for (insn = next_active_insn (first); insn; insn = next_active_insn (insn)) { rtx target; INSN_ANNULLED_BRANCH_P (insn) = 0; INSN_FROM_TARGET_P (insn) = 0; /* Skip vector tables. We can't get attributes for them. */ if (GET_CODE (insn) == JUMP_INSN && (GET_CODE (PATTERN (insn)) == ADDR_VEC || GET_CODE (PATTERN (insn)) == ADDR_DIFF_VEC)) continue; if (num_delay_slots (insn) > 0) obstack_ptr_grow (&unfilled_slots_obstack, insn); /* Ensure all jumps go to the last of a set of consecutive labels. */ if (GET_CODE (insn) == JUMP_INSN && condjump_p (insn) && JUMP_LABEL (insn) != 0 && ((target = prev_label (next_active_insn (JUMP_LABEL (insn)))) != JUMP_LABEL (insn))) redirect_jump (insn, target); } /* Indicate what resources are required to be valid at the end of the current function. The condition code never is and memory always is. If the frame pointer is needed, it is and so is the stack pointer unless EXIT_IGNORE_STACK is non-zero. If the frame pointer is not needed, the stack pointer is. Registers used to return the function value are needed. Registers holding global variables are needed. */ end_of_function_needs.cc = 0; end_of_function_needs.memory = 1; CLEAR_HARD_REG_SET (end_of_function_needs.regs); if (frame_pointer_needed) { SET_HARD_REG_BIT (end_of_function_needs.regs, FRAME_POINTER_REGNUM); #ifdef EXIT_IGNORE_STACK if (! EXIT_IGNORE_STACK) #endif SET_HARD_REG_BIT (end_of_function_needs.regs, STACK_POINTER_REGNUM); } else SET_HARD_REG_BIT (end_of_function_needs.regs, STACK_POINTER_REGNUM); if (current_function_return_rtx != 0 && GET_CODE (current_function_return_rtx) == REG) mark_referenced_resources (current_function_return_rtx, &end_of_function_needs, 0); for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) if (global_regs[i]) SET_HARD_REG_BIT (end_of_function_needs.regs, i); /* Show we haven't computed an end-of-function label yet. */ end_of_function_label = 0; /* Allocate and initialize the tables used by mark_target_live_regs. */ target_hash_table = (struct target_info **) alloca ((TARGET_HASH_PRIME * sizeof (struct target_info *))); bzero (target_hash_table, TARGET_HASH_PRIME * sizeof (struct target_info *)); bb_ticks = (int *) alloca (n_basic_blocks * sizeof (int)); bzero (bb_ticks, n_basic_blocks * sizeof (int)); /* Initialize the statistics for this function. */ bzero (num_insns_needing_delays, sizeof num_insns_needing_delays); bzero (num_filled_delays, sizeof num_filled_delays); /* Now do the delay slot filling. Try everything twice in case earlier changes make more slots fillable. */ for (reorg_pass_number = 0; reorg_pass_number < MAX_REORG_PASSES; reorg_pass_number++) { fill_simple_delay_slots (first, 1); fill_simple_delay_slots (first, 0); fill_eager_delay_slots (first); relax_delay_slots (first); } /* Delete any USE insns made by update_block; subsequent passes don't need them or know how to deal with them. */ for (insn = first; insn; insn = next) { next = NEXT_INSN (insn); if (GET_CODE (insn) == INSN && GET_CODE (PATTERN (insn)) == USE && (GET_CODE (XEXP (PATTERN (insn), 0)) == INSN || GET_CODE (XEXP (PATTERN (insn), 0)) == JUMP_INSN || GET_CODE (XEXP (PATTERN (insn), 0)) == CALL_INSN)) next = delete_insn (insn); } /* If we made an end of function label, indicate that it is now safe to delete it by undoing our prior adjustment to LABEL_NUSES. If it is now unused, delete it. */ if (end_of_function_label && --LABEL_NUSES (end_of_function_label) == 0) delete_insn (end_of_function_label); #ifdef HAVE_return if (HAVE_return && end_of_function_label != 0) make_return_insns (first); #endif obstack_free (&unfilled_slots_obstack, unfilled_firstobj); /* It is not clear why the line below is needed, but it does seem to be. */ unfilled_firstobj = (rtx *) obstack_alloc (&unfilled_slots_obstack, 0); /* Reposition the prologue and epilogue notes in case we moved the prologue/epilogue insns. */ reposition_prologue_and_epilogue_notes (first); if (file) { register int i, j, need_comma; for (reorg_pass_number = 0; reorg_pass_number < MAX_REORG_PASSES; reorg_pass_number++) { fprintf (file, ";; Reorg pass #%d:\n", reorg_pass_number + 1); for (i = 0; i < NUM_REORG_FUNCTIONS; i++) { need_comma = 0; fprintf (file, ";; Reorg function #%d\n", i); fprintf (file, ";; %d insns needing delay slots\n;; ", num_insns_needing_delays[i][reorg_pass_number]); for (j = 0; j < MAX_DELAY_HISTOGRAM; j++) if (num_filled_delays[i][j][reorg_pass_number]) { if (need_comma) fprintf (file, ", "); need_comma = 1; fprintf (file, "%d got %d delays", num_filled_delays[i][j][reorg_pass_number], j); } fprintf (file, "\n"); } } } } #endif /* DELAY_SLOTS */