static void
adjust_frame_related_expr (rtx last_sp_set, rtx insn,
HOST_WIDE_INT this_adjust)
{
rtx note = find_reg_note (last_sp_set, REG_FRAME_RELATED_EXPR, NULL_RTX);
rtx new_expr = NULL_RTX;
if (note == NULL_RTX && RTX_FRAME_RELATED_P (insn))
return;
if (note
&& GET_CODE (XEXP (note, 0)) == SEQUENCE
&& XVECLEN (XEXP (note, 0), 0) >= 2)
{
rtx expr = XEXP (note, 0);
rtx last = XVECEXP (expr, 0, XVECLEN (expr, 0) - 1);
int i;
if (GET_CODE (last) == SET
&& RTX_FRAME_RELATED_P (last) == RTX_FRAME_RELATED_P (insn)
&& SET_DEST (last) == stack_pointer_rtx
&& GET_CODE (SET_SRC (last)) == PLUS
&& XEXP (SET_SRC (last), 0) == stack_pointer_rtx
&& CONST_INT_P (XEXP (SET_SRC (last), 1)))
{
XEXP (SET_SRC (last), 1)
= GEN_INT (INTVAL (XEXP (SET_SRC (last), 1)) + this_adjust);
return;
}
new_expr = gen_rtx_SEQUENCE (VOIDmode,
rtvec_alloc (XVECLEN (expr, 0) + 1));
for (i = 0; i < XVECLEN (expr, 0); i++)
XVECEXP (new_expr, 0, i) = XVECEXP (expr, 0, i);
}
else
{
new_expr = gen_rtx_SEQUENCE (VOIDmode, rtvec_alloc (2));
if (note)
XVECEXP (new_expr, 0, 0) = XEXP (note, 0);
else
{
rtx expr = copy_rtx (single_set_for_csa (last_sp_set));
XEXP (SET_SRC (expr), 1)
= GEN_INT (INTVAL (XEXP (SET_SRC (expr), 1)) - this_adjust);
RTX_FRAME_RELATED_P (expr) = 1;
XVECEXP (new_expr, 0, 0) = expr;
}
}
XVECEXP (new_expr, 0, XVECLEN (new_expr, 0) - 1)
= copy_rtx (single_set_for_csa (insn));
RTX_FRAME_RELATED_P (XVECEXP (new_expr, 0, XVECLEN (new_expr, 0) - 1))
= RTX_FRAME_RELATED_P (insn);
if (note)
XEXP (note, 0) = new_expr;
else
add_reg_note (last_sp_set, REG_FRAME_RELATED_EXPR, new_expr);
}
static rtx
gen_speculative_prefetch (rtx address, gcov_type delta, int write)
{
rtx tmp;
rtx sequence;
/* TODO: we do the prefetching for just one iteration ahead, which
often is not enough. */
start_sequence ();
if (offsettable_address_p (0, VOIDmode, address))
tmp = plus_constant (copy_rtx (address), delta);
else
{
tmp = simplify_gen_binary (PLUS, Pmode,
copy_rtx (address), GEN_INT (delta));
tmp = force_operand (tmp, NULL);
}
if (! (*insn_data[(int)CODE_FOR_prefetch].operand[0].predicate)
(tmp, insn_data[(int)CODE_FOR_prefetch].operand[0].mode))
tmp = force_reg (Pmode, tmp);
emit_insn (gen_prefetch (tmp, GEN_INT (write), GEN_INT (3)));
sequence = get_insns ();
end_sequence ();
return sequence;
}
/* Generate code for transformations 3 and 4 (with MODE and OPERATION,
operands OP1 and OP2, result TARGET, at most SUB subtractions, and
probability of taking the optimal path(s) PROB1 and PROB2). */
static rtx
gen_mod_subtract (enum machine_mode mode, enum rtx_code operation,
rtx target, rtx op1, rtx op2, int sub, int prob1, int prob2)
{
rtx tmp, tmp1, jump;
rtx end_label = gen_label_rtx ();
rtx sequence;
int i;
start_sequence ();
if (!REG_P (op2))
{
tmp = gen_reg_rtx (mode);
emit_move_insn (tmp, copy_rtx (op2));
}
else
tmp = op2;
emit_move_insn (target, copy_rtx (op1));
do_compare_rtx_and_jump (target, tmp, LTU, 0, mode, NULL_RTX,
NULL_RTX, end_label);
/* Add branch probability to jump we just created. */
jump = get_last_insn ();
REG_NOTES (jump) = gen_rtx_EXPR_LIST (REG_BR_PROB,
GEN_INT (prob1), REG_NOTES (jump));
for (i = 0; i < sub; i++)
{
tmp1 = expand_simple_binop (mode, MINUS, target, tmp, target,
0, OPTAB_WIDEN);
if (tmp1 != target)
emit_move_insn (target, tmp1);
do_compare_rtx_and_jump (target, tmp, LTU, 0, mode, NULL_RTX,
NULL_RTX, end_label);
/* Add branch probability to jump we just created. */
jump = get_last_insn ();
REG_NOTES (jump) = gen_rtx_EXPR_LIST (REG_BR_PROB,
GEN_INT (prob2), REG_NOTES (jump));
}
tmp1 = simplify_gen_binary (operation, mode, copy_rtx (target), copy_rtx (tmp));
tmp1 = force_operand (tmp1, target);
if (tmp1 != target)
emit_move_insn (target, tmp1);
emit_label (end_label);
sequence = get_insns ();
end_sequence ();
rebuild_jump_labels (sequence);
return sequence;
}
/* Generate code for transformation 2 (with MODE and OPERATION, operands OP1
and OP2, result TARGET and probability of taking the optimal path PROB). */
static rtx
gen_mod_pow2 (enum machine_mode mode, enum rtx_code operation, rtx target,
rtx op1, rtx op2, int prob)
{
rtx tmp, tmp1, tmp2, tmp3, jump;
rtx neq_label = gen_label_rtx ();
rtx end_label = gen_label_rtx ();
rtx sequence;
start_sequence ();
if (!REG_P (op2))
{
tmp = gen_reg_rtx (mode);
emit_move_insn (tmp, copy_rtx (op2));
}
else
tmp = op2;
tmp1 = expand_simple_binop (mode, PLUS, tmp, constm1_rtx, NULL_RTX,
0, OPTAB_WIDEN);
tmp2 = expand_simple_binop (mode, AND, tmp, tmp1, NULL_RTX,
0, OPTAB_WIDEN);
do_compare_rtx_and_jump (tmp2, const0_rtx, NE, 0, mode, NULL_RTX,
NULL_RTX, neq_label);
/* Add branch probability to jump we just created. */
jump = get_last_insn ();
REG_NOTES (jump) = gen_rtx_EXPR_LIST (REG_BR_PROB,
GEN_INT (REG_BR_PROB_BASE - prob),
REG_NOTES (jump));
tmp3 = expand_simple_binop (mode, AND, op1, tmp1, target,
0, OPTAB_WIDEN);
if (tmp3 != target)
emit_move_insn (copy_rtx (target), tmp3);
emit_jump_insn (gen_jump (end_label));
emit_barrier ();
emit_label (neq_label);
tmp1 = simplify_gen_binary (operation, mode, copy_rtx (op1), copy_rtx (tmp));
tmp1 = force_operand (tmp1, target);
if (tmp1 != target)
emit_move_insn (target, tmp1);
emit_label (end_label);
sequence = get_insns ();
end_sequence ();
rebuild_jump_labels (sequence);
return sequence;
}
/* Generate code for transformation 1 (with MODE and OPERATION, operands OP1
and OP2, whose value is expected to be VALUE, result TARGET and
probability of taking the optimal path PROB). */
static rtx
gen_divmod_fixed_value (enum machine_mode mode, enum rtx_code operation,
rtx target, rtx op1, rtx op2, gcov_type value,
int prob)
{
rtx tmp, tmp1, jump;
rtx neq_label = gen_label_rtx ();
rtx end_label = gen_label_rtx ();
rtx sequence;
start_sequence ();
if (!REG_P (op2))
{
tmp = gen_reg_rtx (mode);
emit_move_insn (tmp, copy_rtx (op2));
}
else
tmp = op2;
do_compare_rtx_and_jump (tmp, GEN_INT (value), NE, 0, mode, NULL_RTX,
NULL_RTX, neq_label);
/* Add branch probability to jump we just created. */
jump = get_last_insn ();
REG_NOTES (jump) = gen_rtx_EXPR_LIST (REG_BR_PROB,
GEN_INT (REG_BR_PROB_BASE - prob),
REG_NOTES (jump));
tmp1 = simplify_gen_binary (operation, mode,
copy_rtx (op1), GEN_INT (value));
tmp1 = force_operand (tmp1, target);
if (tmp1 != target)
emit_move_insn (copy_rtx (target), copy_rtx (tmp1));
emit_jump_insn (gen_jump (end_label));
emit_barrier ();
emit_label (neq_label);
tmp1 = simplify_gen_binary (operation, mode,
copy_rtx (op1), copy_rtx (tmp));
tmp1 = force_operand (tmp1, target);
if (tmp1 != target)
emit_move_insn (copy_rtx (target), copy_rtx (tmp1));
emit_label (end_label);
sequence = get_insns ();
end_sequence ();
rebuild_jump_labels (sequence);
return sequence;
}
static rtx
propagate_rtx (rtx x, machine_mode mode, rtx old_rtx, rtx new_rtx,
bool speed)
{
rtx tem;
bool collapsed;
int flags;
if (REG_P (new_rtx) && REGNO (new_rtx) < FIRST_PSEUDO_REGISTER)
return NULL_RTX;
flags = 0;
if (REG_P (new_rtx)
|| CONSTANT_P (new_rtx)
|| (GET_CODE (new_rtx) == SUBREG
&& REG_P (SUBREG_REG (new_rtx))
&& (GET_MODE_SIZE (mode)
<= GET_MODE_SIZE (GET_MODE (SUBREG_REG (new_rtx))))))
flags |= PR_CAN_APPEAR;
if (!varying_mem_p (new_rtx))
flags |= PR_HANDLE_MEM;
if (speed)
flags |= PR_OPTIMIZE_FOR_SPEED;
tem = x;
collapsed = propagate_rtx_1 (&tem, old_rtx, copy_rtx (new_rtx), flags);
if (tem == x || !collapsed)
return NULL_RTX;
/* gen_lowpart_common will not be able to process VOIDmode entities other
than CONST_INTs. */
if (GET_MODE (tem) == VOIDmode && !CONST_INT_P (tem))
return NULL_RTX;
if (GET_MODE (tem) == VOIDmode)
tem = rtl_hooks.gen_lowpart_no_emit (mode, tem);
else
gcc_assert (GET_MODE (tem) == mode);
return tem;
}
int
c54x_expand_movqi(rtx ops[])
{
int done = 0;
int i;
fprintf(stderr, "--->>>");
for(i=0; i < 2; i++) {
print_rtl(stderr, ops[i]);
}
fprintf(stderr, "<<<---\n");
if(ACC_REG_P(ops[0])) {
ops[0] = copy_rtx(ops[0]);
PUT_MODE(ops[0], PSImode);
fprintf(stderr, "+++");
print_rtl(stderr, ops[0]);
fprintf(stderr, "+++\n");
done = 1;
if(MEM_P(ops[1])) {
emit_insn(gen_ldm(ops[0], ops[1]));
} else if(REG_P(ops[1])) {
emit_insn(gen_ldu(ops[0], ops[1]));
} else if(CONSTANT_P(ops[1])) {
emit_insn(gen_ld_const(ops[0], ops[1], gen_reg_rtx(QImode)));
} else {
done = 2;
}
} else if( (REG_P(ops[0]) && (GET_CODE(ops[1]) == MEM && REG_P(XEXP(ops[1],0))))
|| (T_REG_P(ops[0]) && ARSP_REG_P(ops[1])) )
{
done = 2;
}
return done;
}
rtx
copy_rtx (rtx orig)
{
rtx copy;
int i, j;
RTX_CODE code;
const char *format_ptr;
code = GET_CODE (orig);
switch (code)
{
case REG:
case DEBUG_EXPR:
case VALUE:
case CONST_INT:
case CONST_DOUBLE:
case CONST_FIXED:
case CONST_VECTOR:
case SYMBOL_REF:
case CODE_LABEL:
case PC:
case CC0:
case RETURN:
case SIMPLE_RETURN:
case SCRATCH:
/* SCRATCH must be shared because they represent distinct values. */
return orig;
case CLOBBER:
if (REG_P (XEXP (orig, 0)) && REGNO (XEXP (orig, 0)) < FIRST_PSEUDO_REGISTER)
return orig;
break;
case CONST:
if (shared_const_p (orig))
return orig;
break;
/* A MEM with a constant address is not sharable. The problem is that
the constant address may need to be reloaded. If the mem is shared,
then reloading one copy of this mem will cause all copies to appear
to have been reloaded. */
default:
break;
}
/* Copy the various flags, fields, and other information. We assume
that all fields need copying, and then clear the fields that should
not be copied. That is the sensible default behavior, and forces
us to explicitly document why we are *not* copying a flag. */
copy = shallow_copy_rtx (orig);
/* We do not copy the USED flag, which is used as a mark bit during
walks over the RTL. */
RTX_FLAG (copy, used) = 0;
format_ptr = GET_RTX_FORMAT (GET_CODE (copy));
for (i = 0; i < GET_RTX_LENGTH (GET_CODE (copy)); i++)
switch (*format_ptr++)
{
case 'e':
if (XEXP (orig, i) != NULL)
XEXP (copy, i) = copy_rtx (XEXP (orig, i));
break;
case 'E':
case 'V':
if (XVEC (orig, i) != NULL)
{
XVEC (copy, i) = rtvec_alloc (XVECLEN (orig, i));
for (j = 0; j < XVECLEN (copy, i); j++)
XVECEXP (copy, i, j) = copy_rtx (XVECEXP (orig, i, j));
}
break;
case 't':
case 'w':
case 'i':
case 's':
case 'S':
case 'T':
case 'u':
case 'B':
case '0':
/* These are left unchanged. */
break;
default:
gcc_unreachable ();
}
return copy;
}
rtx
copy_rtx (rtx orig)
{
rtx copy;
int i, j;
RTX_CODE code;
const char *format_ptr;
code = GET_CODE (orig);
switch (code)
{
case REG:
case DEBUG_EXPR:
case VALUE:
CASE_CONST_ANY:
case SYMBOL_REF:
case CODE_LABEL:
case PC:
case CC0:
case RETURN:
case SIMPLE_RETURN:
case SCRATCH:
/* SCRATCH must be shared because they represent distinct values. */
return orig;
case CLOBBER:
/* Share clobbers of hard registers (like cc0), but do not share pseudo reg
clobbers or clobbers of hard registers that originated as pseudos.
This is needed to allow safe register renaming. */
if (REG_P (XEXP (orig, 0)) && REGNO (XEXP (orig, 0)) < FIRST_PSEUDO_REGISTER
&& ORIGINAL_REGNO (XEXP (orig, 0)) == REGNO (XEXP (orig, 0)))
return orig;
break;
case CLOBBER_HIGH:
gcc_assert (REG_P (XEXP (orig, 0)));
return orig;
case CONST:
if (shared_const_p (orig))
return orig;
break;
/* A MEM with a constant address is not sharable. The problem is that
the constant address may need to be reloaded. If the mem is shared,
then reloading one copy of this mem will cause all copies to appear
to have been reloaded. */
default:
break;
}
/* Copy the various flags, fields, and other information. We assume
that all fields need copying, and then clear the fields that should
not be copied. That is the sensible default behavior, and forces
us to explicitly document why we are *not* copying a flag. */
copy = shallow_copy_rtx (orig);
format_ptr = GET_RTX_FORMAT (GET_CODE (copy));
for (i = 0; i < GET_RTX_LENGTH (GET_CODE (copy)); i++)
switch (*format_ptr++)
{
case 'e':
if (XEXP (orig, i) != NULL)
XEXP (copy, i) = copy_rtx (XEXP (orig, i));
break;
case 'E':
case 'V':
if (XVEC (orig, i) != NULL)
{
XVEC (copy, i) = rtvec_alloc (XVECLEN (orig, i));
for (j = 0; j < XVECLEN (copy, i); j++)
XVECEXP (copy, i, j) = copy_rtx (XVECEXP (orig, i, j));
}
break;
case 't':
case 'w':
case 'i':
case 'p':
case 's':
case 'S':
case 'T':
case 'u':
case 'B':
case '0':
/* These are left unchanged. */
break;
default:
gcc_unreachable ();
}
return copy;
}
static void
process_rtx (rtx desc, int lineno)
{
switch (GET_CODE (desc))
{
case DEFINE_INSN:
queue_pattern (desc, &define_insn_tail, read_rtx_filename, lineno);
break;
case DEFINE_COND_EXEC:
queue_pattern (desc, &define_cond_exec_tail, read_rtx_filename, lineno);
break;
case DEFINE_ATTR:
queue_pattern (desc, &define_attr_tail, read_rtx_filename, lineno);
break;
case DEFINE_PREDICATE:
case DEFINE_SPECIAL_PREDICATE:
case DEFINE_CONSTRAINT:
case DEFINE_REGISTER_CONSTRAINT:
case DEFINE_MEMORY_CONSTRAINT:
case DEFINE_ADDRESS_CONSTRAINT:
queue_pattern (desc, &define_pred_tail, read_rtx_filename, lineno);
break;
case INCLUDE:
process_include (desc, lineno);
break;
case DEFINE_INSN_AND_SPLIT:
{
const char *split_cond;
rtx split;
rtvec attr;
int i;
struct queue_elem *insn_elem;
struct queue_elem *split_elem;
/* Create a split with values from the insn_and_split. */
split = rtx_alloc (DEFINE_SPLIT);
i = XVECLEN (desc, 1);
XVEC (split, 0) = rtvec_alloc (i);
while (--i >= 0)
{
XVECEXP (split, 0, i) = copy_rtx (XVECEXP (desc, 1, i));
remove_constraints (XVECEXP (split, 0, i));
}
/* If the split condition starts with "&&", append it to the
insn condition to create the new split condition. */
split_cond = XSTR (desc, 4);
if (split_cond[0] == '&' && split_cond[1] == '&')
{
copy_rtx_ptr_loc (split_cond + 2, split_cond);
split_cond = join_c_conditions (XSTR (desc, 2), split_cond + 2);
}
XSTR (split, 1) = split_cond;
XVEC (split, 2) = XVEC (desc, 5);
XSTR (split, 3) = XSTR (desc, 6);
/* Fix up the DEFINE_INSN. */
attr = XVEC (desc, 7);
PUT_CODE (desc, DEFINE_INSN);
XVEC (desc, 4) = attr;
/* Queue them. */
insn_elem
= queue_pattern (desc, &define_insn_tail, read_rtx_filename,
lineno);
split_elem
= queue_pattern (split, &other_tail, read_rtx_filename, lineno);
insn_elem->split = split_elem;
break;
}
default:
queue_pattern (desc, &other_tail, read_rtx_filename, lineno);
break;
}
}
static bool
doloop_optimize (struct loop *loop)
{
enum machine_mode mode;
rtx doloop_seq, doloop_pat, doloop_reg;
rtx iterations, count;
rtx iterations_max;
rtx start_label;
rtx condition;
unsigned level, est_niter;
struct niter_desc *desc;
unsigned word_mode_size;
unsigned HOST_WIDE_INT word_mode_max;
if (dump_file)
fprintf (dump_file, "Doloop: Processing loop %d.\n", loop->num);
/* APPLE LOCAL begin lno */
/* Ignore large loops. */
if (loop->ninsns > (unsigned) PARAM_VALUE (PARAM_MAX_DOLOOP_INSNS))
{
if (dump_file)
fprintf (dump_file,
"Doloop: The loop is too large.\n");
return false;
}
/* APPLE LOCAL end lno */
iv_analysis_loop_init (loop);
/* Find the simple exit of a LOOP. */
desc = get_simple_loop_desc (loop);
/* Check that loop is a candidate for a low-overhead looping insn. */
if (!doloop_valid_p (loop, desc))
{
if (dump_file)
fprintf (dump_file,
"Doloop: The loop is not suitable.\n");
return false;
}
mode = desc->mode;
est_niter = 3;
if (desc->const_iter)
est_niter = desc->niter;
/* If the estimate on number of iterations is reliable (comes from profile
feedback), use it. Do not use it normally, since the expected number
of iterations of an unrolled loop is 2. */
if (loop->header->count)
est_niter = expected_loop_iterations (loop);
if (est_niter < 3)
{
if (dump_file)
fprintf (dump_file,
"Doloop: Too few iterations (%u) to be profitable.\n",
est_niter);
return false;
}
count = copy_rtx (desc->niter_expr);
iterations = desc->const_iter ? desc->niter_expr : const0_rtx;
iterations_max = GEN_INT (desc->niter_max);
level = get_loop_level (loop) + 1;
/* Generate looping insn. If the pattern FAILs then give up trying
to modify the loop since there is some aspect the back-end does
not like. */
start_label = block_label (desc->in_edge->dest);
doloop_reg = gen_reg_rtx (mode);
doloop_seq = gen_doloop_end (doloop_reg, iterations, iterations_max,
GEN_INT (level), start_label);
word_mode_size = GET_MODE_BITSIZE (word_mode);
word_mode_max
= ((unsigned HOST_WIDE_INT) 1 << (word_mode_size - 1) << 1) - 1;
if (! doloop_seq
&& mode != word_mode
/* Before trying mode different from the one in that # of iterations is
computed, we must be sure that the number of iterations fits into
the new mode. */
&& (word_mode_size >= GET_MODE_BITSIZE (mode)
|| desc->niter_max <= word_mode_max))
{
if (word_mode_size > GET_MODE_BITSIZE (mode))
{
count = simplify_gen_unary (ZERO_EXTEND, word_mode,
count, mode);
iterations = simplify_gen_unary (ZERO_EXTEND, word_mode,
iterations, mode);
iterations_max = simplify_gen_unary (ZERO_EXTEND, word_mode,
iterations_max, mode);
}
else
{
count = lowpart_subreg (word_mode, count, mode);
iterations = lowpart_subreg (word_mode, iterations, mode);
iterations_max = lowpart_subreg (word_mode, iterations_max, mode);
}
PUT_MODE (doloop_reg, word_mode);
doloop_seq = gen_doloop_end (doloop_reg, iterations, iterations_max,
GEN_INT (level), start_label);
}
if (! doloop_seq)
{
if (dump_file)
fprintf (dump_file,
"Doloop: Target unwilling to use doloop pattern!\n");
return false;
}
/* If multiple instructions were created, the last must be the
jump instruction. Also, a raw define_insn may yield a plain
pattern. */
doloop_pat = doloop_seq;
if (INSN_P (doloop_pat))
{
while (NEXT_INSN (doloop_pat) != NULL_RTX)
doloop_pat = NEXT_INSN (doloop_pat);
if (JUMP_P (doloop_pat))
doloop_pat = PATTERN (doloop_pat);
else
doloop_pat = NULL_RTX;
}
if (! doloop_pat
|| ! (condition = doloop_condition_get (doloop_pat)))
{
if (dump_file)
fprintf (dump_file, "Doloop: Unrecognizable doloop pattern!\n");
return false;
}
doloop_modify (loop, desc, doloop_seq, condition, count);
return true;
}
static void
doloop_modify (struct loop *loop, struct niter_desc *desc,
rtx doloop_seq, rtx condition, rtx count)
{
rtx counter_reg;
rtx tmp, noloop = NULL_RTX;
rtx sequence;
rtx jump_insn;
rtx jump_label;
int nonneg = 0, irr;
bool increment_count;
basic_block loop_end = desc->out_edge->src;
enum machine_mode mode;
jump_insn = BB_END (loop_end);
if (dump_file)
{
fprintf (dump_file, "Doloop: Inserting doloop pattern (");
if (desc->const_iter)
fprintf (dump_file, HOST_WIDEST_INT_PRINT_DEC, desc->niter);
else
fputs ("runtime", dump_file);
fputs (" iterations).\n", dump_file);
}
/* Discard original jump to continue loop. The original compare
result may still be live, so it cannot be discarded explicitly. */
delete_insn (jump_insn);
counter_reg = XEXP (condition, 0);
if (GET_CODE (counter_reg) == PLUS)
counter_reg = XEXP (counter_reg, 0);
mode = GET_MODE (counter_reg);
increment_count = false;
switch (GET_CODE (condition))
{
case NE:
/* Currently only NE tests against zero and one are supported. */
if (XEXP (condition, 1) == const1_rtx)
{
increment_count = true;
noloop = const1_rtx;
}
else if (XEXP (condition, 1) == const0_rtx)
noloop = const0_rtx;
else
abort ();
break;
case GE:
/* Currently only GE tests against zero are supported. */
if (XEXP (condition, 1) != const0_rtx)
abort ();
noloop = constm1_rtx;
/* The iteration count does not need incrementing for a GE test. */
increment_count = false;
/* Determine if the iteration counter will be non-negative.
Note that the maximum value loaded is iterations_max - 1. */
if (desc->niter_max
<= ((unsigned HOST_WIDEST_INT) 1
<< (GET_MODE_BITSIZE (mode) - 1)))
nonneg = 1;
break;
/* Abort if an invalid doloop pattern has been generated. */
default:
abort ();
}
if (increment_count)
count = simplify_gen_binary (PLUS, mode, count, const1_rtx);
/* Insert initialization of the count register into the loop header. */
start_sequence ();
tmp = force_operand (count, counter_reg);
convert_move (counter_reg, tmp, 1);
sequence = get_insns ();
end_sequence ();
emit_insn_after (sequence, BB_END (loop_preheader_edge (loop)->src));
if (desc->noloop_assumptions)
{
rtx ass = copy_rtx (desc->noloop_assumptions);
basic_block preheader = loop_preheader_edge (loop)->src;
basic_block set_zero
= loop_split_edge_with (loop_preheader_edge (loop), NULL_RTX);
basic_block new_preheader
= loop_split_edge_with (loop_preheader_edge (loop), NULL_RTX);
basic_block bb;
edge te;
gcov_type cnt;
/* Expand the condition testing the assumptions and if it does not pass,
reset the count register to 0. */
add_test (XEXP (ass, 0), preheader, set_zero);
EDGE_SUCC (preheader, 0)->flags &= ~EDGE_FALLTHRU;
cnt = EDGE_SUCC (preheader, 0)->count;
EDGE_SUCC (preheader, 0)->probability = 0;
EDGE_SUCC (preheader, 0)->count = 0;
irr = EDGE_SUCC (preheader, 0)->flags & EDGE_IRREDUCIBLE_LOOP;
te = make_edge (preheader, new_preheader, EDGE_FALLTHRU | irr);
te->probability = REG_BR_PROB_BASE;
te->count = cnt;
set_immediate_dominator (CDI_DOMINATORS, new_preheader, preheader);
set_zero->count = 0;
set_zero->frequency = 0;
for (ass = XEXP (ass, 1); ass; ass = XEXP (ass, 1))
{
bb = loop_split_edge_with (te, NULL_RTX);
te = EDGE_SUCC (bb, 0);
add_test (XEXP (ass, 0), bb, set_zero);
make_edge (bb, set_zero, irr);
}
start_sequence ();
convert_move (counter_reg, noloop, 0);
sequence = get_insns ();
end_sequence ();
emit_insn_after (sequence, BB_END (set_zero));
}
/* Some targets (eg, C4x) need to initialize special looping
registers. */
#ifdef HAVE_doloop_begin
{
rtx init;
unsigned level = get_loop_level (loop) + 1;
init = gen_doloop_begin (counter_reg,
desc->const_iter ? desc->niter_expr : const0_rtx,
desc->niter_max,
GEN_INT (level));
if (init)
{
start_sequence ();
emit_insn (init);
sequence = get_insns ();
end_sequence ();
emit_insn_after (sequence, BB_END (loop_preheader_edge (loop)->src));
}
}
#endif
/* Insert the new low-overhead looping insn. */
emit_jump_insn_after (doloop_seq, BB_END (loop_end));
jump_insn = BB_END (loop_end);
jump_label = block_label (desc->in_edge->dest);
JUMP_LABEL (jump_insn) = jump_label;
LABEL_NUSES (jump_label)++;
/* Ensure the right fallthru edge is marked, for case we have reversed
the condition. */
desc->in_edge->flags &= ~EDGE_FALLTHRU;
desc->out_edge->flags |= EDGE_FALLTHRU;
/* Add a REG_NONNEG note if the actual or estimated maximum number
of iterations is non-negative. */
if (nonneg)
{
REG_NOTES (jump_insn)
= gen_rtx_EXPR_LIST (REG_NONNEG, NULL_RTX, REG_NOTES (jump_insn));
}
}
static void
process_rtx (rtx desc, int lineno)
{
switch (GET_CODE (desc))
{
case DEFINE_INSN:
queue_pattern (desc, &define_insn_tail, read_rtx_filename, lineno);
break;
case DEFINE_COND_EXEC:
queue_pattern (desc, &define_cond_exec_tail, read_rtx_filename, lineno);
break;
case DEFINE_ATTR:
queue_pattern (desc, &define_attr_tail, read_rtx_filename, lineno);
break;
case INCLUDE:
process_include (desc, lineno);
break;
case DEFINE_INSN_AND_SPLIT:
{
const char *split_cond;
rtx split;
rtvec attr;
int i;
/* Create a split with values from the insn_and_split. */
split = rtx_alloc (DEFINE_SPLIT);
i = XVECLEN (desc, 1);
XVEC (split, 0) = rtvec_alloc (i);
while (--i >= 0)
{
XVECEXP (split, 0, i) = copy_rtx (XVECEXP (desc, 1, i));
remove_constraints (XVECEXP (split, 0, i));
}
/* If the split condition starts with "&&", append it to the
insn condition to create the new split condition. */
split_cond = XSTR (desc, 4);
if (split_cond[0] == '&' && split_cond[1] == '&')
split_cond = concat (XSTR (desc, 2), split_cond, NULL);
XSTR (split, 1) = split_cond;
XVEC (split, 2) = XVEC (desc, 5);
XSTR (split, 3) = XSTR (desc, 6);
/* Fix up the DEFINE_INSN. */
attr = XVEC (desc, 7);
PUT_CODE (desc, DEFINE_INSN);
XVEC (desc, 4) = attr;
/* Queue them. */
queue_pattern (desc, &define_insn_tail, read_rtx_filename, lineno);
queue_pattern (split, &other_tail, read_rtx_filename, lineno);
break;
}
default:
queue_pattern (desc, &other_tail, read_rtx_filename, lineno);
break;
}
}
/* Find values inside INSN for that we want to measure histograms for
division/modulo optimization and stores them to VALUES. */
static void
insn_divmod_values_to_profile (rtx insn, histogram_values *values)
{
rtx set, set_src, op1, op2;
enum machine_mode mode;
histogram_value hist;
if (!INSN_P (insn))
return;
set = single_set (insn);
if (!set)
return;
mode = GET_MODE (SET_DEST (set));
if (!INTEGRAL_MODE_P (mode))
return;
set_src = SET_SRC (set);
switch (GET_CODE (set_src))
{
case DIV:
case MOD:
case UDIV:
case UMOD:
op1 = XEXP (set_src, 0);
op2 = XEXP (set_src, 1);
if (side_effects_p (op2))
return;
/* Check for a special case where the divisor is power of 2. */
if ((GET_CODE (set_src) == UMOD) && !CONSTANT_P (op2))
{
hist = ggc_alloc (sizeof (*hist));
hist->value = op2;
hist->seq = NULL_RTX;
hist->mode = mode;
hist->insn = insn;
hist->type = HIST_TYPE_POW2;
hist->hdata.pow2.may_be_other = 1;
VEC_safe_push (histogram_value, *values, hist);
}
/* Check whether the divisor is not in fact a constant. */
if (!CONSTANT_P (op2))
{
hist = ggc_alloc (sizeof (*hist));
hist->value = op2;
hist->mode = mode;
hist->seq = NULL_RTX;
hist->insn = insn;
hist->type = HIST_TYPE_SINGLE_VALUE;
VEC_safe_push (histogram_value, *values, hist);
}
/* For mod, check whether it is not often a noop (or replaceable by
a few subtractions). */
if (GET_CODE (set_src) == UMOD && !side_effects_p (op1))
{
rtx tmp;
hist = ggc_alloc (sizeof (*hist));
start_sequence ();
tmp = simplify_gen_binary (DIV, mode, copy_rtx (op1), copy_rtx (op2));
hist->value = force_operand (tmp, NULL_RTX);
hist->seq = get_insns ();
end_sequence ();
hist->mode = mode;
hist->insn = insn;
hist->type = HIST_TYPE_INTERVAL;
hist->hdata.intvl.int_start = 0;
hist->hdata.intvl.steps = 2;
hist->hdata.intvl.may_be_less = 1;
hist->hdata.intvl.may_be_more = 1;
VEC_safe_push (histogram_value, *values, hist);
}
return;
default:
return;
}
}
