/* Analyze RTL for GNU compiler.
- Copyright (C) 1987-2017 Free Software Foundation, Inc.
+ Copyright (C) 1987-2018 Free Software Foundation, Inc.
This file is part of GCC.
static int computed_jump_p_1 (const_rtx);
static void parms_set (rtx, const_rtx, void *);
-static unsigned HOST_WIDE_INT cached_nonzero_bits (const_rtx, machine_mode,
+static unsigned HOST_WIDE_INT cached_nonzero_bits (const_rtx, scalar_int_mode,
const_rtx, machine_mode,
unsigned HOST_WIDE_INT);
-static unsigned HOST_WIDE_INT nonzero_bits1 (const_rtx, machine_mode,
+static unsigned HOST_WIDE_INT nonzero_bits1 (const_rtx, scalar_int_mode,
const_rtx, machine_mode,
unsigned HOST_WIDE_INT);
-static unsigned int cached_num_sign_bit_copies (const_rtx, machine_mode, const_rtx,
- machine_mode,
+static unsigned int cached_num_sign_bit_copies (const_rtx, scalar_int_mode,
+ const_rtx, machine_mode,
unsigned int);
-static unsigned int num_sign_bit_copies1 (const_rtx, machine_mode, const_rtx,
- machine_mode, unsigned int);
+static unsigned int num_sign_bit_copies1 (const_rtx, scalar_int_mode,
+ const_rtx, machine_mode,
+ unsigned int);
rtx_subrtx_bound_info rtx_all_subrtx_bounds[NUM_RTX_CODE];
rtx_subrtx_bound_info rtx_nonconst_subrtx_bounds[NUM_RTX_CODE];
FROM and TO for the current function, as it was at the start
of the routine. */
-static HOST_WIDE_INT
+static poly_int64
get_initial_register_offset (int from, int to)
{
static const struct elim_table_t
const int from;
const int to;
} table[] = ELIMINABLE_REGS;
- HOST_WIDE_INT offset1, offset2;
+ poly_int64 offset1, offset2;
unsigned int i, j;
if (to == from)
references on strict alignment machines. */
static int
-rtx_addr_can_trap_p_1 (const_rtx x, HOST_WIDE_INT offset, HOST_WIDE_INT size,
+rtx_addr_can_trap_p_1 (const_rtx x, poly_int64 offset, poly_int64 size,
machine_mode mode, bool unaligned_mems)
{
enum rtx_code code = GET_CODE (x);
+ gcc_checking_assert (mode == BLKmode || known_size_p (size));
/* The offset must be a multiple of the mode size if we are considering
unaligned memory references on strict alignment machines. */
- if (STRICT_ALIGNMENT && unaligned_mems && GET_MODE_SIZE (mode) != 0)
+ if (STRICT_ALIGNMENT && unaligned_mems && mode != BLKmode)
{
- HOST_WIDE_INT actual_offset = offset;
+ poly_int64 actual_offset = offset;
#ifdef SPARC_STACK_BOUNDARY_HACK
/* ??? The SPARC port may claim a STACK_BOUNDARY higher than
actual_offset -= STACK_POINTER_OFFSET;
#endif
- if (actual_offset % GET_MODE_SIZE (mode) != 0)
+ if (!multiple_p (actual_offset, GET_MODE_SIZE (mode)))
return 1;
}
case SYMBOL_REF:
if (SYMBOL_REF_WEAK (x))
return 1;
- if (!CONSTANT_POOL_ADDRESS_P (x))
+ if (!CONSTANT_POOL_ADDRESS_P (x) && !SYMBOL_REF_FUNCTION_P (x))
{
tree decl;
- HOST_WIDE_INT decl_size;
+ poly_int64 decl_size;
- if (offset < 0)
+ if (maybe_lt (offset, 0))
return 1;
- if (size == 0)
- size = GET_MODE_SIZE (mode);
- if (size == 0)
- return offset != 0;
+ if (!known_size_p (size))
+ return maybe_ne (offset, 0);
/* If the size of the access or of the symbol is unknown,
assume the worst. */
if (!decl)
decl_size = -1;
else if (DECL_P (decl) && DECL_SIZE_UNIT (decl))
- decl_size = (tree_fits_shwi_p (DECL_SIZE_UNIT (decl))
- ? tree_to_shwi (DECL_SIZE_UNIT (decl))
- : -1);
+ {
+ if (!poly_int_tree_p (DECL_SIZE_UNIT (decl), &decl_size))
+ decl_size = -1;
+ }
else if (TREE_CODE (decl) == STRING_CST)
decl_size = TREE_STRING_LENGTH (decl);
else if (TYPE_SIZE_UNIT (TREE_TYPE (decl)))
else
decl_size = -1;
- return (decl_size <= 0 ? offset != 0 : offset + size > decl_size);
+ return (!known_size_p (decl_size) || known_eq (decl_size, 0)
+ ? maybe_ne (offset, 0)
+ : maybe_gt (offset + size, decl_size));
}
return 0;
|| (x == arg_pointer_rtx && fixed_regs[ARG_POINTER_REGNUM]))
{
#ifdef RED_ZONE_SIZE
- HOST_WIDE_INT red_zone_size = RED_ZONE_SIZE;
+ poly_int64 red_zone_size = RED_ZONE_SIZE;
#else
- HOST_WIDE_INT red_zone_size = 0;
+ poly_int64 red_zone_size = 0;
#endif
- HOST_WIDE_INT stack_boundary = PREFERRED_STACK_BOUNDARY
- / BITS_PER_UNIT;
- HOST_WIDE_INT low_bound, high_bound;
+ poly_int64 stack_boundary = PREFERRED_STACK_BOUNDARY / BITS_PER_UNIT;
+ poly_int64 low_bound, high_bound;
- if (size == 0)
- size = GET_MODE_SIZE (mode);
- if (size == 0)
+ if (!known_size_p (size))
return 1;
if (x == frame_pointer_rtx)
{
if (FRAME_GROWS_DOWNWARD)
{
- high_bound = STARTING_FRAME_OFFSET;
+ high_bound = targetm.starting_frame_offset ();
low_bound = high_bound - get_frame_size ();
}
else
{
- low_bound = STARTING_FRAME_OFFSET;
+ low_bound = targetm.starting_frame_offset ();
high_bound = low_bound + get_frame_size ();
}
}
else if (x == hard_frame_pointer_rtx)
{
- HOST_WIDE_INT sp_offset
+ poly_int64 sp_offset
= get_initial_register_offset (STACK_POINTER_REGNUM,
HARD_FRAME_POINTER_REGNUM);
- HOST_WIDE_INT ap_offset
+ poly_int64 ap_offset
= get_initial_register_offset (ARG_POINTER_REGNUM,
HARD_FRAME_POINTER_REGNUM);
}
else if (x == stack_pointer_rtx)
{
- HOST_WIDE_INT ap_offset
+ poly_int64 ap_offset
= get_initial_register_offset (ARG_POINTER_REGNUM,
STACK_POINTER_REGNUM);
#endif
}
- if (offset >= low_bound && offset <= high_bound - size)
+ if (known_ge (offset, low_bound)
+ && known_le (offset, high_bound - size))
return 0;
return 1;
}
case PLUS:
/* An address is assumed not to trap if:
- - it is the pic register plus a constant. */
- if (XEXP (x, 0) == pic_offset_table_rtx && CONSTANT_P (XEXP (x, 1)))
+ - it is the pic register plus a const unspec without offset. */
+ if (XEXP (x, 0) == pic_offset_table_rtx
+ && GET_CODE (XEXP (x, 1)) == CONST
+ && GET_CODE (XEXP (XEXP (x, 1), 0)) == UNSPEC
+ && known_eq (offset, 0))
return 0;
/* - or it is an address that can't trap plus a constant integer. */
int
rtx_addr_can_trap_p (const_rtx x)
{
- return rtx_addr_can_trap_p_1 (x, 0, 0, VOIDmode, false);
+ return rtx_addr_can_trap_p_1 (x, 0, -1, BLKmode, false);
}
/* Return true if X contains a MEM subrtx. */
*base_out = x;
*offset_out = const0_rtx;
}
+
+/* Express integer value X as some value Y plus a polynomial offset,
+ where Y is either const0_rtx, X or something within X (as opposed
+ to a new rtx). Return the Y and store the offset in *OFFSET_OUT. */
+
+rtx
+strip_offset (rtx x, poly_int64_pod *offset_out)
+{
+ rtx base = const0_rtx;
+ rtx test = x;
+ if (GET_CODE (test) == CONST)
+ test = XEXP (test, 0);
+ if (GET_CODE (test) == PLUS)
+ {
+ base = XEXP (test, 0);
+ test = XEXP (test, 1);
+ }
+ if (poly_int_rtx_p (test, offset_out))
+ return base;
+ *offset_out = 0;
+ return x;
+}
+
+/* Return the argument size in REG_ARGS_SIZE note X. */
+
+poly_int64
+get_args_size (const_rtx x)
+{
+ gcc_checking_assert (REG_NOTE_KIND (x) == REG_ARGS_SIZE);
+ return rtx_to_poly_int64 (XEXP (x, 0));
+}
\f
/* Return the number of places FIND appears within X. If COUNT_DEST is
zero, we do not count occurrences inside the destination of a SET. */
&& !REG_P (SET_DEST (body))
&& ! (GET_CODE (SET_DEST (body)) == SUBREG
&& REG_P (SUBREG_REG (SET_DEST (body)))
- && (((GET_MODE_SIZE (GET_MODE (SUBREG_REG (SET_DEST (body))))
- + (UNITS_PER_WORD - 1)) / UNITS_PER_WORD)
- == ((GET_MODE_SIZE (GET_MODE (SET_DEST (body)))
- + (UNITS_PER_WORD - 1)) / UNITS_PER_WORD)))
+ && !read_modify_subreg_p (SET_DEST (body)))
&& reg_overlap_mentioned_p (x, SET_DEST (body)))
return 1;
return 0;
return 0;
}
+
+/* Return true if X is a SUBREG and if storing a value to X would
+ preserve some of its SUBREG_REG. For example, on a normal 32-bit
+ target, using a SUBREG to store to one half of a DImode REG would
+ preserve the other half. */
+
+bool
+read_modify_subreg_p (const_rtx x)
+{
+ if (GET_CODE (x) != SUBREG)
+ return false;
+ poly_uint64 isize = GET_MODE_SIZE (GET_MODE (SUBREG_REG (x)));
+ poly_uint64 osize = GET_MODE_SIZE (GET_MODE (x));
+ poly_uint64 regsize = REGMODE_NATURAL_SIZE (GET_MODE (SUBREG_REG (x)));
+ /* The inner and outer modes of a subreg must be ordered, so that we
+ can tell whether they're paradoxical or partial. */
+ gcc_checking_assert (ordered_p (isize, osize));
+ return (maybe_gt (isize, osize) && maybe_gt (isize, regsize));
+}
\f
/* Helper function for set_of. */
struct set_of_data
if (GET_CODE (src) == SUBREG && GET_CODE (dst) == SUBREG)
{
- if (SUBREG_BYTE (src) != SUBREG_BYTE (dst))
+ if (maybe_ne (SUBREG_BYTE (src), SUBREG_BYTE (dst)))
return 0;
src = SUBREG_REG (src);
dst = SUBREG_REG (dst);
return 1;
}
-/* Return TRUE iff DEST is a register or subreg of a register and
- doesn't change the number of words of the inner register, and any
- part of the register is TEST_REGNO. */
+/* Return TRUE iff DEST is a register or subreg of a register, is a
+ complete rather than read-modify-write destination, and contains
+ register TEST_REGNO. */
static bool
covers_regno_no_parallel_p (const_rtx dest, unsigned int test_regno)
{
unsigned int regno, endregno;
- if (GET_CODE (dest) == SUBREG
- && (((GET_MODE_SIZE (GET_MODE (dest))
- + UNITS_PER_WORD - 1) / UNITS_PER_WORD)
- == ((GET_MODE_SIZE (GET_MODE (SUBREG_REG (dest)))
- + UNITS_PER_WORD - 1) / UNITS_PER_WORD)))
+ if (GET_CODE (dest) == SUBREG && !read_modify_subreg_p (dest))
dest = SUBREG_REG (dest);
if (!REG_P (dest))
if (GET_CODE (pattern) == COND_EXEC)
return 0;
- if (GET_CODE (pattern) == SET)
+ if (GET_CODE (pattern) == SET || GET_CODE (pattern) == CLOBBER)
return covers_regno_p (SET_DEST (pattern), test_regno);
else if (GET_CODE (pattern) == PARALLEL)
{
datum, REG_NOTES (insn));
}
+/* Add a REG_ARGS_SIZE note to INSN with value VALUE. */
+
+void
+add_args_size_note (rtx_insn *insn, poly_int64 value)
+{
+ gcc_checking_assert (!find_reg_note (insn, REG_ARGS_SIZE, NULL_RTX));
+ add_reg_note (insn, REG_ARGS_SIZE, gen_int_mode (value, Pmode));
+}
+
/* Add a register note like NOTE to INSN. */
void
code_changed
|| !MEM_NOTRAP_P (x))
{
- HOST_WIDE_INT size = MEM_SIZE_KNOWN_P (x) ? MEM_SIZE (x) : 0;
+ poly_int64 size = MEM_SIZE_KNOWN_P (x) ? MEM_SIZE (x) : -1;
return rtx_addr_can_trap_p_1 (XEXP (x, 0), 0, size,
GET_MODE (x), code_changed);
}
case PRE_INC:
case POST_INC:
{
- int size = GET_MODE_SIZE (GET_MODE (mem));
+ poly_int64 size = GET_MODE_SIZE (GET_MODE (mem));
rtx r1 = XEXP (x, 0);
rtx c = gen_int_mode (size, GET_MODE (r1));
return fn (mem, x, r1, r1, c, data);
case PRE_DEC:
case POST_DEC:
{
- int size = GET_MODE_SIZE (GET_MODE (mem));
+ poly_int64 size = GET_MODE_SIZE (GET_MODE (mem));
rtx r1 = XEXP (x, 0);
rtx c = gen_int_mode (-size, GET_MODE (r1));
return fn (mem, x, r1, r1, c, data);
/* Constants always become the second operand. Prefer "nice" constants. */
if (code == CONST_INT)
- return -8;
+ return -10;
if (code == CONST_WIDE_INT)
- return -7;
+ return -9;
+ if (code == CONST_POLY_INT)
+ return -8;
if (code == CONST_DOUBLE)
- return -7;
+ return -8;
if (code == CONST_FIXED)
- return -7;
+ return -8;
op = avoid_constant_pool_reference (op);
code = GET_CODE (op);
{
case RTX_CONST_OBJ:
if (code == CONST_INT)
- return -6;
+ return -7;
if (code == CONST_WIDE_INT)
- return -6;
+ return -6;
+ if (code == CONST_POLY_INT)
+ return -5;
if (code == CONST_DOUBLE)
- return -5;
+ return -5;
if (code == CONST_FIXED)
- return -5;
+ return -5;
return -4;
case RTX_EXTRA:
and SUBREG_BYTE, return the bit offset where the subreg begins
(counting from the least significant bit of the operand). */
-unsigned int
+poly_uint64
subreg_lsb_1 (machine_mode outer_mode,
machine_mode inner_mode,
- unsigned int subreg_byte)
+ poly_uint64 subreg_byte)
{
- unsigned int bitpos;
- unsigned int byte;
- unsigned int word;
+ poly_uint64 subreg_end, trailing_bytes, byte_pos;
/* A paradoxical subreg begins at bit position 0. */
- if (GET_MODE_PRECISION (outer_mode) > GET_MODE_PRECISION (inner_mode))
+ if (paradoxical_subreg_p (outer_mode, inner_mode))
return 0;
- if (WORDS_BIG_ENDIAN != BYTES_BIG_ENDIAN)
- /* If the subreg crosses a word boundary ensure that
- it also begins and ends on a word boundary. */
- gcc_assert (!((subreg_byte % UNITS_PER_WORD
- + GET_MODE_SIZE (outer_mode)) > UNITS_PER_WORD
- && (subreg_byte % UNITS_PER_WORD
- || GET_MODE_SIZE (outer_mode) % UNITS_PER_WORD)));
-
- if (WORDS_BIG_ENDIAN)
- word = (GET_MODE_SIZE (inner_mode)
- - (subreg_byte + GET_MODE_SIZE (outer_mode))) / UNITS_PER_WORD;
- else
- word = subreg_byte / UNITS_PER_WORD;
- bitpos = word * BITS_PER_WORD;
-
- if (BYTES_BIG_ENDIAN)
- byte = (GET_MODE_SIZE (inner_mode)
- - (subreg_byte + GET_MODE_SIZE (outer_mode))) % UNITS_PER_WORD;
+ subreg_end = subreg_byte + GET_MODE_SIZE (outer_mode);
+ trailing_bytes = GET_MODE_SIZE (inner_mode) - subreg_end;
+ if (WORDS_BIG_ENDIAN && BYTES_BIG_ENDIAN)
+ byte_pos = trailing_bytes;
+ else if (!WORDS_BIG_ENDIAN && !BYTES_BIG_ENDIAN)
+ byte_pos = subreg_byte;
else
- byte = subreg_byte % UNITS_PER_WORD;
- bitpos += byte * BITS_PER_UNIT;
+ {
+ /* When bytes and words have opposite endianness, we must be able
+ to split offsets into words and bytes at compile time. */
+ poly_uint64 leading_word_part
+ = force_align_down (subreg_byte, UNITS_PER_WORD);
+ poly_uint64 trailing_word_part
+ = force_align_down (trailing_bytes, UNITS_PER_WORD);
+ /* If the subreg crosses a word boundary ensure that
+ it also begins and ends on a word boundary. */
+ gcc_assert (known_le (subreg_end - leading_word_part,
+ (unsigned int) UNITS_PER_WORD)
+ || (known_eq (leading_word_part, subreg_byte)
+ && known_eq (trailing_word_part, trailing_bytes)));
+ if (WORDS_BIG_ENDIAN)
+ byte_pos = trailing_word_part + (subreg_byte - leading_word_part);
+ else
+ byte_pos = leading_word_part + (trailing_bytes - trailing_word_part);
+ }
- return bitpos;
+ return byte_pos * BITS_PER_UNIT;
}
/* Given a subreg X, return the bit offset where the subreg begins
(counting from the least significant bit of the reg). */
-unsigned int
+poly_uint64
subreg_lsb (const_rtx x)
{
return subreg_lsb_1 (GET_MODE (x), GET_MODE (SUBREG_REG (x)),
lsb of the inner value. This is the inverse of the calculation
performed by subreg_lsb_1 (which converts byte offsets to bit shifts). */
-unsigned int
-subreg_size_offset_from_lsb (unsigned int outer_bytes,
- unsigned int inner_bytes,
- unsigned int lsb_shift)
+poly_uint64
+subreg_size_offset_from_lsb (poly_uint64 outer_bytes, poly_uint64 inner_bytes,
+ poly_uint64 lsb_shift)
{
/* A paradoxical subreg begins at bit position 0. */
- if (outer_bytes > inner_bytes)
+ gcc_checking_assert (ordered_p (outer_bytes, inner_bytes));
+ if (maybe_gt (outer_bytes, inner_bytes))
{
- gcc_checking_assert (lsb_shift == 0);
+ gcc_checking_assert (known_eq (lsb_shift, 0U));
return 0;
}
- gcc_assert (lsb_shift % BITS_PER_UNIT == 0);
- unsigned int lower_bytes = lsb_shift / BITS_PER_UNIT;
- unsigned int upper_bytes = inner_bytes - (lower_bytes + outer_bytes);
+ poly_uint64 lower_bytes = exact_div (lsb_shift, BITS_PER_UNIT);
+ poly_uint64 upper_bytes = inner_bytes - (lower_bytes + outer_bytes);
if (WORDS_BIG_ENDIAN && BYTES_BIG_ENDIAN)
return upper_bytes;
else if (!WORDS_BIG_ENDIAN && !BYTES_BIG_ENDIAN)
return lower_bytes;
else
{
- unsigned int lower_word_part = lower_bytes & -UNITS_PER_WORD;
- unsigned int upper_word_part = upper_bytes & -UNITS_PER_WORD;
+ /* When bytes and words have opposite endianness, we must be able
+ to split offsets into words and bytes at compile time. */
+ poly_uint64 lower_word_part = force_align_down (lower_bytes,
+ UNITS_PER_WORD);
+ poly_uint64 upper_word_part = force_align_down (upper_bytes,
+ UNITS_PER_WORD);
if (WORDS_BIG_ENDIAN)
return upper_word_part + (lower_bytes - lower_word_part);
else
function does not check whether adding INFO->offset to XREGNO gives
a valid hard register; even if INFO->offset + XREGNO is out of range,
there might be another register of the same type that is in range.
- Likewise it doesn't check whether HARD_REGNO_MODE_OK accepts the new
- register, since that can depend on things like whether the final
+ Likewise it doesn't check whether targetm.hard_regno_mode_ok accepts
+ the new register, since that can depend on things like whether the final
register number is even or odd. Callers that want to check whether
this particular subreg can be replaced by a simple (reg ...) should
use simplify_subreg_regno. */
void
subreg_get_info (unsigned int xregno, machine_mode xmode,
- unsigned int offset, machine_mode ymode,
+ poly_uint64 offset, machine_mode ymode,
struct subreg_info *info)
{
unsigned int nregs_xmode, nregs_ymode;
gcc_assert (xregno < FIRST_PSEUDO_REGISTER);
- unsigned int xsize = GET_MODE_SIZE (xmode);
- unsigned int ysize = GET_MODE_SIZE (ymode);
+ poly_uint64 xsize = GET_MODE_SIZE (xmode);
+ poly_uint64 ysize = GET_MODE_SIZE (ymode);
+
bool rknown = false;
/* If the register representation of a non-scalar mode has holes in it,
at least one register. */
if (HARD_REGNO_NREGS_HAS_PADDING (xregno, xmode))
{
+ /* As a consequence, we must be dealing with a constant number of
+ scalars, and thus a constant offset and number of units. */
+ HOST_WIDE_INT coffset = offset.to_constant ();
+ HOST_WIDE_INT cysize = ysize.to_constant ();
nregs_xmode = HARD_REGNO_NREGS_WITH_PADDING (xregno, xmode);
- unsigned int nunits = GET_MODE_NUNITS (xmode);
- machine_mode xmode_unit = GET_MODE_INNER (xmode);
+ unsigned int nunits = GET_MODE_NUNITS (xmode).to_constant ();
+ scalar_mode xmode_unit = GET_MODE_INNER (xmode);
gcc_assert (HARD_REGNO_NREGS_HAS_PADDING (xregno, xmode_unit));
gcc_assert (nregs_xmode
== (nunits
* HARD_REGNO_NREGS_WITH_PADDING (xregno, xmode_unit)));
- gcc_assert (hard_regno_nregs[xregno][xmode]
- == hard_regno_nregs[xregno][xmode_unit] * nunits);
+ gcc_assert (hard_regno_nregs (xregno, xmode)
+ == hard_regno_nregs (xregno, xmode_unit) * nunits);
/* You can only ask for a SUBREG of a value with holes in the middle
if you don't cross the holes. (Such a SUBREG should be done by
3 for each part, but in memory it's two 128-bit parts.
Padding is assumed to be at the end (not necessarily the 'high part')
of each unit. */
- if ((offset / GET_MODE_SIZE (xmode_unit) + 1 < nunits)
- && (offset / GET_MODE_SIZE (xmode_unit)
- != ((offset + ysize - 1) / GET_MODE_SIZE (xmode_unit))))
+ if ((coffset / GET_MODE_SIZE (xmode_unit) + 1 < nunits)
+ && (coffset / GET_MODE_SIZE (xmode_unit)
+ != ((coffset + cysize - 1) / GET_MODE_SIZE (xmode_unit))))
{
info->representable_p = false;
rknown = true;
}
}
else
- nregs_xmode = hard_regno_nregs[xregno][xmode];
+ nregs_xmode = hard_regno_nregs (xregno, xmode);
- nregs_ymode = hard_regno_nregs[xregno][ymode];
+ nregs_ymode = hard_regno_nregs (xregno, ymode);
+
+ /* Subreg sizes must be ordered, so that we can tell whether they are
+ partial, paradoxical or complete. */
+ gcc_checking_assert (ordered_p (xsize, ysize));
/* Paradoxical subregs are otherwise valid. */
- if (!rknown && offset == 0 && ysize > xsize)
+ if (!rknown && known_eq (offset, 0U) && maybe_gt (ysize, xsize))
{
info->representable_p = true;
/* If this is a big endian paradoxical subreg, which uses more
/* If registers store different numbers of bits in the different
modes, we cannot generally form this subreg. */
+ poly_uint64 regsize_xmode, regsize_ymode;
if (!HARD_REGNO_NREGS_HAS_PADDING (xregno, xmode)
&& !HARD_REGNO_NREGS_HAS_PADDING (xregno, ymode)
- && (xsize % nregs_xmode) == 0
- && (ysize % nregs_ymode) == 0)
+ && multiple_p (xsize, nregs_xmode, ®size_xmode)
+ && multiple_p (ysize, nregs_ymode, ®size_ymode))
{
- int regsize_xmode = xsize / nregs_xmode;
- int regsize_ymode = ysize / nregs_ymode;
if (!rknown
- && ((nregs_ymode > 1 && regsize_xmode > regsize_ymode)
- || (nregs_xmode > 1 && regsize_ymode > regsize_xmode)))
+ && ((nregs_ymode > 1 && maybe_gt (regsize_xmode, regsize_ymode))
+ || (nregs_xmode > 1 && maybe_gt (regsize_ymode, regsize_xmode))))
{
info->representable_p = false;
- info->nregs = CEIL (ysize, regsize_xmode);
- info->offset = offset / regsize_xmode;
+ if (!can_div_away_from_zero_p (ysize, regsize_xmode, &info->nregs)
+ || !can_div_trunc_p (offset, regsize_xmode, &info->offset))
+ /* Checked by validate_subreg. We must know at compile time
+ which inner registers are being accessed. */
+ gcc_unreachable ();
return;
}
/* It's not valid to extract a subreg of mode YMODE at OFFSET that
would go outside of XMODE. */
- if (!rknown && ysize + offset > xsize)
+ if (!rknown && maybe_gt (ysize + offset, xsize))
{
info->representable_p = false;
info->nregs = nregs_ymode;
- info->offset = offset / regsize_xmode;
+ if (!can_div_trunc_p (offset, regsize_xmode, &info->offset))
+ /* Checked by validate_subreg. We must know at compile time
+ which inner registers are being accessed. */
+ gcc_unreachable ();
return;
}
/* Quick exit for the simple and common case of extracting whole
/* ??? It would be better to integrate this into the code below,
if we can generalize the concept enough and figure out how
odd-sized modes can coexist with the other weird cases we support. */
+ HOST_WIDE_INT count;
if (!rknown
&& WORDS_BIG_ENDIAN == REG_WORDS_BIG_ENDIAN
- && regsize_xmode == regsize_ymode
- && (offset % regsize_ymode) == 0)
+ && known_eq (regsize_xmode, regsize_ymode)
+ && constant_multiple_p (offset, regsize_ymode, &count))
{
info->representable_p = true;
info->nregs = nregs_ymode;
- info->offset = offset / regsize_ymode;
+ info->offset = count;
gcc_assert (info->offset + info->nregs <= (int) nregs_xmode);
return;
}
}
/* Lowpart subregs are otherwise valid. */
- if (!rknown && offset == subreg_lowpart_offset (ymode, xmode))
+ if (!rknown && known_eq (offset, subreg_lowpart_offset (ymode, xmode)))
{
info->representable_p = true;
rknown = true;
- if (offset == 0 || nregs_xmode == nregs_ymode)
+ if (known_eq (offset, 0U) || nregs_xmode == nregs_ymode)
{
info->offset = 0;
info->nregs = nregs_ymode;
be exact, otherwise we don't know how to verify the constraint.
These conditions may be relaxed but subreg_regno_offset would
need to be redesigned. */
- gcc_assert ((xsize % num_blocks) == 0);
- unsigned int bytes_per_block = xsize / num_blocks;
+ poly_uint64 bytes_per_block = exact_div (xsize, num_blocks);
/* Get the number of the first block that contains the subreg and the byte
offset of the subreg from the start of that block. */
- unsigned int block_number = offset / bytes_per_block;
- unsigned int subblock_offset = offset % bytes_per_block;
+ unsigned int block_number;
+ poly_uint64 subblock_offset;
+ if (!can_div_trunc_p (offset, bytes_per_block, &block_number,
+ &subblock_offset))
+ /* Checked by validate_subreg. We must know at compile time which
+ inner registers are being accessed. */
+ gcc_unreachable ();
if (!rknown)
{
/* Only the lowpart of each block is representable. */
info->representable_p
- = (subblock_offset
- == subreg_size_lowpart_offset (ysize, bytes_per_block));
+ = known_eq (subblock_offset,
+ subreg_size_lowpart_offset (ysize, bytes_per_block));
rknown = true;
}
RETURN - The regno offset which would be used. */
unsigned int
subreg_regno_offset (unsigned int xregno, machine_mode xmode,
- unsigned int offset, machine_mode ymode)
+ poly_uint64 offset, machine_mode ymode)
{
struct subreg_info info;
subreg_get_info (xregno, xmode, offset, ymode, &info);
RETURN - Whether the offset is representable. */
bool
subreg_offset_representable_p (unsigned int xregno, machine_mode xmode,
- unsigned int offset, machine_mode ymode)
+ poly_uint64 offset, machine_mode ymode)
{
struct subreg_info info;
subreg_get_info (xregno, xmode, offset, ymode, &info);
int
simplify_subreg_regno (unsigned int xregno, machine_mode xmode,
- unsigned int offset, machine_mode ymode)
+ poly_uint64 offset, machine_mode ymode)
{
struct subreg_info info;
unsigned int yregno;
-#ifdef CANNOT_CHANGE_MODE_CLASS
/* Give the backend a chance to disallow the mode change. */
if (GET_MODE_CLASS (xmode) != MODE_COMPLEX_INT
&& GET_MODE_CLASS (xmode) != MODE_COMPLEX_FLOAT
- && REG_CANNOT_CHANGE_MODE_P (xregno, xmode, ymode)
+ && !REG_CAN_CHANGE_MODE_P (xregno, xmode, ymode)
/* We can use mode change in LRA for some transformations. */
&& ! lra_in_progress)
return -1;
-#endif
/* We shouldn't simplify stack-related registers. */
if ((!reload_completed || frame_pointer_needed)
??? We allow invalid registers if (reg:XMODE XREGNO) is also invalid.
This is a kludge to work around how complex FP arguments are passed
on IA-64 and should be fixed. See PR target/49226. */
- if (!HARD_REGNO_MODE_OK (yregno, ymode)
- && HARD_REGNO_MODE_OK (xregno, xmode))
+ if (!targetm.hard_regno_mode_ok (yregno, ymode)
+ && targetm.hard_regno_mode_ok (xregno, xmode))
return -1;
return (int) yregno;
/* A size N times larger than UNITS_PER_WORD likely needs N times as
many insns, taking N times as long. */
- factor = GET_MODE_SIZE (mode) / UNITS_PER_WORD;
+ factor = estimated_poly_value (GET_MODE_SIZE (mode)) / UNITS_PER_WORD;
if (factor == 0)
factor = 1;
/* A SET doesn't have a mode, so let's look at the SET_DEST to get
the mode for the factor. */
mode = GET_MODE (SET_DEST (x));
- factor = GET_MODE_SIZE (mode) / UNITS_PER_WORD;
+ factor = estimated_poly_value (GET_MODE_SIZE (mode)) / UNITS_PER_WORD;
if (factor == 0)
factor = 1;
/* FALLTHRU */
total = 0;
/* If we can't tie these modes, make this expensive. The larger
the mode, the more expensive it is. */
- if (! MODES_TIEABLE_P (mode, GET_MODE (SUBREG_REG (x))))
+ if (!targetm.modes_tieable_p (mode, GET_MODE (SUBREG_REG (x))))
return COSTS_N_INSNS (2 + factor);
break;
case TRUNCATE:
- /* If we can tie these modes, make this cheap. */
- if (MODES_TIEABLE_P (mode, GET_MODE (SUBREG_REG (x))))
+ if (targetm.modes_tieable_p (mode, GET_MODE (XEXP (x, 0))))
{
total = 0;
break;
unsigned HOST_WIDE_INT
nonzero_bits (const_rtx x, machine_mode mode)
{
- return cached_nonzero_bits (x, mode, NULL_RTX, VOIDmode, 0);
+ if (mode == VOIDmode)
+ mode = GET_MODE (x);
+ scalar_int_mode int_mode;
+ if (!is_a <scalar_int_mode> (mode, &int_mode))
+ return GET_MODE_MASK (mode);
+ return cached_nonzero_bits (x, int_mode, NULL_RTX, VOIDmode, 0);
}
unsigned int
num_sign_bit_copies (const_rtx x, machine_mode mode)
{
- return cached_num_sign_bit_copies (x, mode, NULL_RTX, VOIDmode, 0);
+ if (mode == VOIDmode)
+ mode = GET_MODE (x);
+ scalar_int_mode int_mode;
+ if (!is_a <scalar_int_mode> (mode, &int_mode))
+ return 1;
+ return cached_num_sign_bit_copies (x, int_mode, NULL_RTX, VOIDmode, 0);
}
/* Return true if nonzero_bits1 might recurse into both operands
identical subexpressions on the first or the second level. */
static unsigned HOST_WIDE_INT
-cached_nonzero_bits (const_rtx x, machine_mode mode, const_rtx known_x,
+cached_nonzero_bits (const_rtx x, scalar_int_mode mode, const_rtx known_x,
machine_mode known_mode,
unsigned HOST_WIDE_INT known_ret)
{
an arithmetic operation, we can do better. */
static unsigned HOST_WIDE_INT
-nonzero_bits1 (const_rtx x, machine_mode mode, const_rtx known_x,
+nonzero_bits1 (const_rtx x, scalar_int_mode mode, const_rtx known_x,
machine_mode known_mode,
unsigned HOST_WIDE_INT known_ret)
{
unsigned HOST_WIDE_INT nonzero = GET_MODE_MASK (mode);
unsigned HOST_WIDE_INT inner_nz;
- enum rtx_code code;
+ enum rtx_code code = GET_CODE (x);
machine_mode inner_mode;
+ unsigned int inner_width;
+ scalar_int_mode xmode;
+
unsigned int mode_width = GET_MODE_PRECISION (mode);
- /* For floating-point and vector values, assume all bits are needed. */
- if (FLOAT_MODE_P (GET_MODE (x)) || FLOAT_MODE_P (mode)
- || VECTOR_MODE_P (GET_MODE (x)) || VECTOR_MODE_P (mode))
+ if (CONST_INT_P (x))
+ {
+ if (SHORT_IMMEDIATES_SIGN_EXTEND
+ && INTVAL (x) > 0
+ && mode_width < BITS_PER_WORD
+ && (UINTVAL (x) & (HOST_WIDE_INT_1U << (mode_width - 1))) != 0)
+ return UINTVAL (x) | (HOST_WIDE_INT_M1U << mode_width);
+
+ return UINTVAL (x);
+ }
+
+ if (!is_a <scalar_int_mode> (GET_MODE (x), &xmode))
return nonzero;
+ unsigned int xmode_width = GET_MODE_PRECISION (xmode);
/* If X is wider than MODE, use its mode instead. */
- if (GET_MODE_PRECISION (GET_MODE (x)) > mode_width)
+ if (xmode_width > mode_width)
{
- mode = GET_MODE (x);
+ mode = xmode;
nonzero = GET_MODE_MASK (mode);
- mode_width = GET_MODE_PRECISION (mode);
+ mode_width = xmode_width;
}
if (mode_width > HOST_BITS_PER_WIDE_INT)
return nonzero;
/* If MODE is wider than X, but both are a single word for both the host
- and target machines, we can compute this from which bits of the
- object might be nonzero in its own mode, taking into account the fact
- that on many CISC machines, accessing an object in a wider mode
- causes the high-order bits to become undefined. So they are
- not known to be zero. */
-
- if (!WORD_REGISTER_OPERATIONS
- && GET_MODE (x) != VOIDmode
- && GET_MODE (x) != mode
- && GET_MODE_PRECISION (GET_MODE (x)) <= BITS_PER_WORD
- && GET_MODE_PRECISION (GET_MODE (x)) <= HOST_BITS_PER_WIDE_INT
- && GET_MODE_PRECISION (mode) > GET_MODE_PRECISION (GET_MODE (x)))
+ and target machines, we can compute this from which bits of the object
+ might be nonzero in its own mode, taking into account the fact that, on
+ CISC machines, accessing an object in a wider mode generally causes the
+ high-order bits to become undefined, so they are not known to be zero.
+ We extend this reasoning to RISC machines for rotate operations since the
+ semantics of the operations in the larger mode is not well defined. */
+ if (mode_width > xmode_width
+ && xmode_width <= BITS_PER_WORD
+ && xmode_width <= HOST_BITS_PER_WIDE_INT
+ && (!WORD_REGISTER_OPERATIONS || code == ROTATE || code == ROTATERT))
{
- nonzero &= cached_nonzero_bits (x, GET_MODE (x),
+ nonzero &= cached_nonzero_bits (x, xmode,
known_x, known_mode, known_ret);
- nonzero |= GET_MODE_MASK (mode) & ~GET_MODE_MASK (GET_MODE (x));
+ nonzero |= GET_MODE_MASK (mode) & ~GET_MODE_MASK (xmode);
return nonzero;
}
/* Please keep nonzero_bits_binary_arith_p above in sync with
the code in the switch below. */
- code = GET_CODE (x);
switch (code)
{
case REG:
we can do this only if the target does not support different pointer
or address modes depending on the address space. */
if (target_default_pointer_address_modes_p ()
- && POINTERS_EXTEND_UNSIGNED && GET_MODE (x) == Pmode
+ && POINTERS_EXTEND_UNSIGNED
+ && xmode == Pmode
&& REG_POINTER (x)
&& !targetm.have_ptr_extend ())
nonzero &= GET_MODE_MASK (ptr_mode);
stack to be momentarily aligned only to that amount,
so we pick the least alignment. */
if (x == stack_pointer_rtx && PUSH_ARGS)
- alignment = MIN ((unsigned HOST_WIDE_INT) PUSH_ROUNDING (1),
- alignment);
+ {
+ poly_uint64 rounded_1 = PUSH_ROUNDING (poly_int64 (1));
+ alignment = MIN (known_alignment (rounded_1), alignment);
+ }
#endif
nonzero &= ~(alignment - 1);
{
unsigned HOST_WIDE_INT nonzero_for_hook = nonzero;
- rtx new_rtx = rtl_hooks.reg_nonzero_bits (x, mode, known_x,
- known_mode, known_ret,
- &nonzero_for_hook);
+ rtx new_rtx = rtl_hooks.reg_nonzero_bits (x, xmode, mode,
+ &nonzero_for_hook);
if (new_rtx)
nonzero_for_hook &= cached_nonzero_bits (new_rtx, mode, known_x,
return nonzero_for_hook;
}
- case CONST_INT:
- /* If X is negative in MODE, sign-extend the value. */
- if (SHORT_IMMEDIATES_SIGN_EXTEND && INTVAL (x) > 0
- && mode_width < BITS_PER_WORD
- && (UINTVAL (x) & (HOST_WIDE_INT_1U << (mode_width - 1)))
- != 0)
- return UINTVAL (x) | (HOST_WIDE_INT_M1U << mode_width);
-
- return UINTVAL (x);
-
case MEM:
/* In many, if not most, RISC machines, reading a byte from memory
zeros the rest of the register. Noticing that fact saves a lot
of extra zero-extends. */
- if (load_extend_op (GET_MODE (x)) == ZERO_EXTEND)
- nonzero &= GET_MODE_MASK (GET_MODE (x));
+ if (load_extend_op (xmode) == ZERO_EXTEND)
+ nonzero &= GET_MODE_MASK (xmode);
break;
case EQ: case NE:
operation in, and not the actual operation mode. We can wind
up with (subreg:DI (gt:V4HI x y)), and we don't have anything
that describes the results of a vector compare. */
- if (GET_MODE_CLASS (GET_MODE (x)) == MODE_INT
+ if (GET_MODE_CLASS (xmode) == MODE_INT
&& mode_width <= HOST_BITS_PER_WIDE_INT)
nonzero = STORE_FLAG_VALUE;
break;
#if 0
/* Disabled to avoid exponential mutual recursion between nonzero_bits
and num_sign_bit_copies. */
- if (num_sign_bit_copies (XEXP (x, 0), GET_MODE (x))
- == GET_MODE_PRECISION (GET_MODE (x)))
+ if (num_sign_bit_copies (XEXP (x, 0), xmode) == xmode_width)
nonzero = 1;
#endif
- if (GET_MODE_PRECISION (GET_MODE (x)) < mode_width)
- nonzero |= (GET_MODE_MASK (mode) & ~GET_MODE_MASK (GET_MODE (x)));
+ if (xmode_width < mode_width)
+ nonzero |= (GET_MODE_MASK (mode) & ~GET_MODE_MASK (xmode));
break;
case ABS:
#if 0
/* Disabled to avoid exponential mutual recursion between nonzero_bits
and num_sign_bit_copies. */
- if (num_sign_bit_copies (XEXP (x, 0), GET_MODE (x))
- == GET_MODE_PRECISION (GET_MODE (x)))
+ if (num_sign_bit_copies (XEXP (x, 0), xmode) == xmode_width)
nonzero = 1;
#endif
break;
unsigned HOST_WIDE_INT nz1
= cached_nonzero_bits (XEXP (x, 1), mode,
known_x, known_mode, known_ret);
- int sign_index = GET_MODE_PRECISION (GET_MODE (x)) - 1;
+ int sign_index = xmode_width - 1;
int width0 = floor_log2 (nz0) + 1;
int width1 = floor_log2 (nz1) + 1;
int low0 = ctz_or_zero (nz0);
been zero-extended, we know that at least the high-order bits
are zero, though others might be too. */
if (SUBREG_PROMOTED_VAR_P (x) && SUBREG_PROMOTED_UNSIGNED_P (x))
- nonzero = GET_MODE_MASK (GET_MODE (x))
- & cached_nonzero_bits (SUBREG_REG (x), GET_MODE (x),
+ nonzero = GET_MODE_MASK (xmode)
+ & cached_nonzero_bits (SUBREG_REG (x), xmode,
known_x, known_mode, known_ret);
/* If the inner mode is a single word for both the host and target
machines, we can compute this from which bits of the inner
object might be nonzero. */
inner_mode = GET_MODE (SUBREG_REG (x));
- if (GET_MODE_PRECISION (inner_mode) <= BITS_PER_WORD
- && GET_MODE_PRECISION (inner_mode) <= HOST_BITS_PER_WIDE_INT)
+ if (GET_MODE_PRECISION (inner_mode).is_constant (&inner_width)
+ && inner_width <= BITS_PER_WORD
+ && inner_width <= HOST_BITS_PER_WIDE_INT)
{
nonzero &= cached_nonzero_bits (SUBREG_REG (x), mode,
known_x, known_mode, known_ret);
? val_signbit_known_set_p (inner_mode, nonzero)
: extend_op != ZERO_EXTEND)
|| (!MEM_P (SUBREG_REG (x)) && !REG_P (SUBREG_REG (x))))
- && GET_MODE_PRECISION (GET_MODE (x))
- > GET_MODE_PRECISION (inner_mode))
+ && xmode_width > inner_width)
nonzero
|= (GET_MODE_MASK (GET_MODE (x)) & ~GET_MODE_MASK (inner_mode));
}
break;
+ case ASHIFT:
case ASHIFTRT:
case LSHIFTRT:
- case ASHIFT:
case ROTATE:
+ case ROTATERT:
/* The nonzero bits are in two classes: any bits within MODE
- that aren't in GET_MODE (x) are always significant. The rest of the
+ that aren't in xmode are always significant. The rest of the
nonzero bits are those that are significant in the operand of
the shift when shifted the appropriate number of bits. This
shows that high-order bits are cleared by the right shift and
if (CONST_INT_P (XEXP (x, 1))
&& INTVAL (XEXP (x, 1)) >= 0
&& INTVAL (XEXP (x, 1)) < HOST_BITS_PER_WIDE_INT
- && INTVAL (XEXP (x, 1)) < GET_MODE_PRECISION (GET_MODE (x)))
+ && INTVAL (XEXP (x, 1)) < xmode_width)
{
- machine_mode inner_mode = GET_MODE (x);
- unsigned int width = GET_MODE_PRECISION (inner_mode);
int count = INTVAL (XEXP (x, 1));
- unsigned HOST_WIDE_INT mode_mask = GET_MODE_MASK (inner_mode);
+ unsigned HOST_WIDE_INT mode_mask = GET_MODE_MASK (xmode);
unsigned HOST_WIDE_INT op_nonzero
= cached_nonzero_bits (XEXP (x, 0), mode,
known_x, known_mode, known_ret);
unsigned HOST_WIDE_INT inner = op_nonzero & mode_mask;
unsigned HOST_WIDE_INT outer = 0;
- if (mode_width > width)
+ if (mode_width > xmode_width)
outer = (op_nonzero & nonzero & ~mode_mask);
- if (code == LSHIFTRT)
- inner >>= count;
- else if (code == ASHIFTRT)
+ switch (code)
{
+ case ASHIFT:
+ inner <<= count;
+ break;
+
+ case LSHIFTRT:
+ inner >>= count;
+ break;
+
+ case ASHIFTRT:
inner >>= count;
/* If the sign bit may have been nonzero before the shift, we
need to mark all the places it could have been copied to
by the shift as possibly nonzero. */
- if (inner & (HOST_WIDE_INT_1U << (width - 1 - count)))
- inner |= ((HOST_WIDE_INT_1U << count) - 1)
- << (width - count);
+ if (inner & (HOST_WIDE_INT_1U << (xmode_width - 1 - count)))
+ inner |= (((HOST_WIDE_INT_1U << count) - 1)
+ << (xmode_width - count));
+ break;
+
+ case ROTATE:
+ inner = (inner << (count % xmode_width)
+ | (inner >> (xmode_width - (count % xmode_width))))
+ & mode_mask;
+ break;
+
+ case ROTATERT:
+ inner = (inner >> (count % xmode_width)
+ | (inner << (xmode_width - (count % xmode_width))))
+ & mode_mask;
+ break;
+
+ default:
+ gcc_unreachable ();
}
- else if (code == ASHIFT)
- inner <<= count;
- else
- inner = ((inner << (count % width)
- | (inner >> (width - (count % width)))) & mode_mask);
nonzero &= (outer | inner);
}
first or the second level. */
static unsigned int
-cached_num_sign_bit_copies (const_rtx x, machine_mode mode, const_rtx known_x,
- machine_mode known_mode,
+cached_num_sign_bit_copies (const_rtx x, scalar_int_mode mode,
+ const_rtx known_x, machine_mode known_mode,
unsigned int known_ret)
{
if (x == known_x && mode == known_mode)
}
/* Return the number of bits at the high-order end of X that are known to
- be equal to the sign bit. X will be used in mode MODE; if MODE is
- VOIDmode, X will be used in its own mode. The returned value will always
- be between 1 and the number of bits in MODE. */
+ be equal to the sign bit. X will be used in mode MODE. The returned
+ value will always be between 1 and the number of bits in MODE. */
static unsigned int
-num_sign_bit_copies1 (const_rtx x, machine_mode mode, const_rtx known_x,
+num_sign_bit_copies1 (const_rtx x, scalar_int_mode mode, const_rtx known_x,
machine_mode known_mode,
unsigned int known_ret)
{
enum rtx_code code = GET_CODE (x);
- machine_mode inner_mode;
+ unsigned int bitwidth = GET_MODE_PRECISION (mode);
int num0, num1, result;
unsigned HOST_WIDE_INT nonzero;
- /* If we weren't given a mode, use the mode of X. If the mode is still
- VOIDmode, we don't know anything. Likewise if one of the modes is
- floating-point. */
-
- if (mode == VOIDmode)
- mode = GET_MODE (x);
+ if (CONST_INT_P (x))
+ {
+ /* If the constant is negative, take its 1's complement and remask.
+ Then see how many zero bits we have. */
+ nonzero = UINTVAL (x) & GET_MODE_MASK (mode);
+ if (bitwidth <= HOST_BITS_PER_WIDE_INT
+ && (nonzero & (HOST_WIDE_INT_1U << (bitwidth - 1))) != 0)
+ nonzero = (~nonzero) & GET_MODE_MASK (mode);
- gcc_checking_assert (mode != BLKmode);
+ return (nonzero == 0 ? bitwidth : bitwidth - floor_log2 (nonzero) - 1);
+ }
- if (mode == VOIDmode || FLOAT_MODE_P (mode) || FLOAT_MODE_P (GET_MODE (x))
- || VECTOR_MODE_P (GET_MODE (x)) || VECTOR_MODE_P (mode))
+ scalar_int_mode xmode, inner_mode;
+ if (!is_a <scalar_int_mode> (GET_MODE (x), &xmode))
return 1;
+ unsigned int xmode_width = GET_MODE_PRECISION (xmode);
+
/* For a smaller mode, just ignore the high bits. */
- unsigned int bitwidth = GET_MODE_PRECISION (mode);
- if (bitwidth < GET_MODE_PRECISION (GET_MODE (x)))
+ if (bitwidth < xmode_width)
{
- num0 = cached_num_sign_bit_copies (x, GET_MODE (x),
+ num0 = cached_num_sign_bit_copies (x, xmode,
known_x, known_mode, known_ret);
- return MAX (1,
- num0 - (int) (GET_MODE_PRECISION (GET_MODE (x)) - bitwidth));
+ return MAX (1, num0 - (int) (xmode_width - bitwidth));
}
- if (GET_MODE (x) != VOIDmode && bitwidth > GET_MODE_PRECISION (GET_MODE (x)))
+ if (bitwidth > xmode_width)
{
/* If this machine does not do all register operations on the entire
register and MODE is wider than the mode of X, we can say nothing
- at all about the high-order bits. */
- if (!WORD_REGISTER_OPERATIONS)
+ at all about the high-order bits. We extend this reasoning to every
+ machine for rotate operations since the semantics of the operations
+ in the larger mode is not well defined. */
+ if (!WORD_REGISTER_OPERATIONS || code == ROTATE || code == ROTATERT)
return 1;
/* Likewise on machines that do, if the mode of the object is smaller
than a word and loads of that size don't sign extend, we can say
nothing about the high order bits. */
- if (GET_MODE_PRECISION (GET_MODE (x)) < BITS_PER_WORD
- && load_extend_op (GET_MODE (x)) != SIGN_EXTEND)
+ if (xmode_width < BITS_PER_WORD
+ && load_extend_op (xmode) != SIGN_EXTEND)
return 1;
}
we can do this only if the target does not support different pointer
or address modes depending on the address space. */
if (target_default_pointer_address_modes_p ()
- && ! POINTERS_EXTEND_UNSIGNED && GET_MODE (x) == Pmode
+ && ! POINTERS_EXTEND_UNSIGNED && xmode == Pmode
&& mode == Pmode && REG_POINTER (x)
&& !targetm.have_ptr_extend ())
return GET_MODE_PRECISION (Pmode) - GET_MODE_PRECISION (ptr_mode) + 1;
{
unsigned int copies_for_hook = 1, copies = 1;
- rtx new_rtx = rtl_hooks.reg_num_sign_bit_copies (x, mode, known_x,
- known_mode, known_ret,
- &copies_for_hook);
+ rtx new_rtx = rtl_hooks.reg_num_sign_bit_copies (x, xmode, mode,
+ &copies_for_hook);
if (new_rtx)
copies = cached_num_sign_bit_copies (new_rtx, mode, known_x,
case MEM:
/* Some RISC machines sign-extend all loads of smaller than a word. */
- if (load_extend_op (GET_MODE (x)) == SIGN_EXTEND)
- return MAX (1, ((int) bitwidth
- - (int) GET_MODE_PRECISION (GET_MODE (x)) + 1));
+ if (load_extend_op (xmode) == SIGN_EXTEND)
+ return MAX (1, ((int) bitwidth - (int) xmode_width + 1));
break;
- case CONST_INT:
- /* If the constant is negative, take its 1's complement and remask.
- Then see how many zero bits we have. */
- nonzero = UINTVAL (x) & GET_MODE_MASK (mode);
- if (bitwidth <= HOST_BITS_PER_WIDE_INT
- && (nonzero & (HOST_WIDE_INT_1U << (bitwidth - 1))) != 0)
- nonzero = (~nonzero) & GET_MODE_MASK (mode);
-
- return (nonzero == 0 ? bitwidth : bitwidth - floor_log2 (nonzero) - 1);
-
case SUBREG:
/* If this is a SUBREG for a promoted object that is sign-extended
and we are looking at it in a wider mode, we know that at least the
{
num0 = cached_num_sign_bit_copies (SUBREG_REG (x), mode,
known_x, known_mode, known_ret);
- return MAX ((int) bitwidth
- - (int) GET_MODE_PRECISION (GET_MODE (x)) + 1,
- num0);
+ return MAX ((int) bitwidth - (int) xmode_width + 1, num0);
}
- /* For a smaller object, just ignore the high bits. */
- inner_mode = GET_MODE (SUBREG_REG (x));
- if (bitwidth <= GET_MODE_PRECISION (inner_mode))
+ if (is_a <scalar_int_mode> (GET_MODE (SUBREG_REG (x)), &inner_mode))
{
- num0 = cached_num_sign_bit_copies (SUBREG_REG (x), VOIDmode,
- known_x, known_mode, known_ret);
- return
- MAX (1, num0 - (int) (GET_MODE_PRECISION (inner_mode) - bitwidth));
- }
+ /* For a smaller object, just ignore the high bits. */
+ if (bitwidth <= GET_MODE_PRECISION (inner_mode))
+ {
+ num0 = cached_num_sign_bit_copies (SUBREG_REG (x), inner_mode,
+ known_x, known_mode,
+ known_ret);
+ return MAX (1, num0 - (int) (GET_MODE_PRECISION (inner_mode)
+ - bitwidth));
+ }
- /* For paradoxical SUBREGs on machines where all register operations
- affect the entire register, just look inside. Note that we are
- passing MODE to the recursive call, so the number of sign bit copies
- will remain relative to that mode, not the inner mode. */
-
- /* This works only if loads sign extend. Otherwise, if we get a
- reload for the inner part, it may be loaded from the stack, and
- then we lose all sign bit copies that existed before the store
- to the stack. */
-
- if (WORD_REGISTER_OPERATIONS
- && load_extend_op (inner_mode) == SIGN_EXTEND
- && paradoxical_subreg_p (x)
- && (MEM_P (SUBREG_REG (x)) || REG_P (SUBREG_REG (x))))
- return cached_num_sign_bit_copies (SUBREG_REG (x), mode,
- known_x, known_mode, known_ret);
+ /* For paradoxical SUBREGs on machines where all register operations
+ affect the entire register, just look inside. Note that we are
+ passing MODE to the recursive call, so the number of sign bit
+ copies will remain relative to that mode, not the inner mode. */
+
+ /* This works only if loads sign extend. Otherwise, if we get a
+ reload for the inner part, it may be loaded from the stack, and
+ then we lose all sign bit copies that existed before the store
+ to the stack. */
+
+ if (WORD_REGISTER_OPERATIONS
+ && load_extend_op (inner_mode) == SIGN_EXTEND
+ && paradoxical_subreg_p (x)
+ && MEM_P (SUBREG_REG (x)))
+ return cached_num_sign_bit_copies (SUBREG_REG (x), mode,
+ known_x, known_mode, known_ret);
+ }
break;
case SIGN_EXTRACT:
break;
case SIGN_EXTEND:
- return (bitwidth - GET_MODE_PRECISION (GET_MODE (XEXP (x, 0)))
- + cached_num_sign_bit_copies (XEXP (x, 0), VOIDmode,
- known_x, known_mode, known_ret));
+ if (is_a <scalar_int_mode> (GET_MODE (XEXP (x, 0)), &inner_mode))
+ return (bitwidth - GET_MODE_PRECISION (inner_mode)
+ + cached_num_sign_bit_copies (XEXP (x, 0), inner_mode,
+ known_x, known_mode, known_ret));
+ break;
case TRUNCATE:
/* For a smaller object, just ignore the high bits. */
- num0 = cached_num_sign_bit_copies (XEXP (x, 0), VOIDmode,
+ inner_mode = as_a <scalar_int_mode> (GET_MODE (XEXP (x, 0)));
+ num0 = cached_num_sign_bit_copies (XEXP (x, 0), inner_mode,
known_x, known_mode, known_ret);
- return MAX (1, (num0 - (int) (GET_MODE_PRECISION (GET_MODE (XEXP (x, 0)))
+ return MAX (1, (num0 - (int) (GET_MODE_PRECISION (inner_mode)
- bitwidth)));
case NOT:
known_x, known_mode, known_ret);
if (CONST_INT_P (XEXP (x, 1))
&& INTVAL (XEXP (x, 1)) > 0
- && INTVAL (XEXP (x, 1)) < GET_MODE_PRECISION (GET_MODE (x)))
+ && INTVAL (XEXP (x, 1)) < xmode_width)
num0 = MIN ((int) bitwidth, num0 + INTVAL (XEXP (x, 1)));
return num0;
if (!CONST_INT_P (XEXP (x, 1))
|| INTVAL (XEXP (x, 1)) < 0
|| INTVAL (XEXP (x, 1)) >= (int) bitwidth
- || INTVAL (XEXP (x, 1)) >= GET_MODE_PRECISION (GET_MODE (x)))
+ || INTVAL (XEXP (x, 1)) >= xmode_width)
return 1;
num0 = cached_num_sign_bit_copies (XEXP (x, 0), mode,
? 1 : bitwidth - floor_log2 (nonzero) - 1;
}
-/* Calculate the rtx_cost of a single instruction. A return value of
+/* Calculate the rtx_cost of a single instruction pattern. A return value of
zero indicates an instruction pattern without a known cost. */
int
-insn_rtx_cost (rtx pat, bool speed)
+pattern_cost (rtx pat, bool speed)
{
int i, cost;
rtx set;
- /* Extract the single set rtx from the instruction pattern.
- We can't use single_set since we only have the pattern. */
+ /* Extract the single set rtx from the instruction pattern. We
+ can't use single_set since we only have the pattern. We also
+ consider PARALLELs of a normal set and a single comparison. In
+ that case we use the cost of the non-comparison SET operation,
+ which is most-likely to be the real cost of this operation. */
if (GET_CODE (pat) == SET)
set = pat;
else if (GET_CODE (pat) == PARALLEL)
{
set = NULL_RTX;
+ rtx comparison = NULL_RTX;
+
for (i = 0; i < XVECLEN (pat, 0); i++)
{
rtx x = XVECEXP (pat, 0, i);
if (GET_CODE (x) == SET)
{
- if (set)
- return 0;
- set = x;
+ if (GET_CODE (SET_SRC (x)) == COMPARE)
+ {
+ if (comparison)
+ return 0;
+ comparison = x;
+ }
+ else
+ {
+ if (set)
+ return 0;
+ set = x;
+ }
}
}
+
+ if (!set && comparison)
+ set = comparison;
+
if (!set)
return 0;
}
return cost > 0 ? cost : COSTS_N_INSNS (1);
}
+/* Calculate the cost of a single instruction. A return value of zero
+ indicates an instruction pattern without a known cost. */
+
+int
+insn_cost (rtx_insn *insn, bool speed)
+{
+ if (targetm.insn_cost)
+ return targetm.insn_cost (insn, speed);
+
+ return pattern_cost (PATTERN (insn), speed);
+}
+
/* Returns estimate on cost of computing SEQ. */
unsigned
set = single_set (seq);
if (set)
cost += set_rtx_cost (set, speed);
- else
- cost++;
+ else if (NONDEBUG_INSN_P (seq))
+ {
+ int this_cost = insn_cost (CONST_CAST_RTX_INSN (seq), speed);
+ if (this_cost > 0)
+ cost += this_cost;
+ else
+ cost++;
+ }
}
return cost;
if we can do computations in the relevant mode and we do not
overflow. */
- if (GET_MODE_CLASS (GET_MODE (op0)) != MODE_CC
- && CONST_INT_P (op1)
- && GET_MODE (op0) != VOIDmode
- && GET_MODE_PRECISION (GET_MODE (op0)) <= HOST_BITS_PER_WIDE_INT)
+ scalar_int_mode op0_mode;
+ if (CONST_INT_P (op1)
+ && is_a <scalar_int_mode> (GET_MODE (op0), &op0_mode)
+ && GET_MODE_PRECISION (op0_mode) <= HOST_BITS_PER_WIDE_INT)
{
HOST_WIDE_INT const_val = INTVAL (op1);
unsigned HOST_WIDE_INT uconst_val = const_val;
unsigned HOST_WIDE_INT max_val
- = (unsigned HOST_WIDE_INT) GET_MODE_MASK (GET_MODE (op0));
+ = (unsigned HOST_WIDE_INT) GET_MODE_MASK (op0_mode);
switch (code)
{
case LE:
if ((unsigned HOST_WIDE_INT) const_val != max_val >> 1)
- code = LT, op1 = gen_int_mode (const_val + 1, GET_MODE (op0));
+ code = LT, op1 = gen_int_mode (const_val + 1, op0_mode);
break;
/* When cross-compiling, const_val might be sign-extended from
BITS_PER_WORD to HOST_BITS_PER_WIDE_INT */
case GE:
if ((const_val & max_val)
- != (HOST_WIDE_INT_1U
- << (GET_MODE_PRECISION (GET_MODE (op0)) - 1)))
- code = GT, op1 = gen_int_mode (const_val - 1, GET_MODE (op0));
+ != (HOST_WIDE_INT_1U << (GET_MODE_PRECISION (op0_mode) - 1)))
+ code = GT, op1 = gen_int_mode (const_val - 1, op0_mode);
break;
case LEU:
if (uconst_val < max_val)
- code = LTU, op1 = gen_int_mode (uconst_val + 1, GET_MODE (op0));
+ code = LTU, op1 = gen_int_mode (uconst_val + 1, op0_mode);
break;
case GEU:
if (uconst_val != 0)
- code = GTU, op1 = gen_int_mode (uconst_val - 1, GET_MODE (op0));
+ code = GTU, op1 = gen_int_mode (uconst_val - 1, op0_mode);
break;
default:
if (CC0_P (op0))
return 0;
- return gen_rtx_fmt_ee (code, VOIDmode, op0, op1);
+ /* We promised to return a comparison. */
+ rtx ret = gen_rtx_fmt_ee (code, VOIDmode, op0, op1);
+ if (COMPARISON_P (ret))
+ return ret;
+ return 0;
}
/* Given a jump insn JUMP, return the condition that will cause it to branch
static void
init_num_sign_bit_copies_in_rep (void)
{
- machine_mode mode, in_mode;
+ opt_scalar_int_mode in_mode_iter;
+ scalar_int_mode mode;
- for (in_mode = GET_CLASS_NARROWEST_MODE (MODE_INT); in_mode != VOIDmode;
- in_mode = GET_MODE_WIDER_MODE (mode))
- for (mode = GET_CLASS_NARROWEST_MODE (MODE_INT); mode != in_mode;
- mode = GET_MODE_WIDER_MODE (mode))
+ FOR_EACH_MODE_IN_CLASS (in_mode_iter, MODE_INT)
+ FOR_EACH_MODE_UNTIL (mode, in_mode_iter.require ())
{
- machine_mode i;
+ scalar_int_mode in_mode = in_mode_iter.require ();
+ scalar_int_mode i;
/* Currently, it is assumed that TARGET_MODE_REP_EXTENDED
extends to the next widest mode. */
gcc_assert (targetm.mode_rep_extended (mode, in_mode) == UNKNOWN
- || GET_MODE_WIDER_MODE (mode) == in_mode);
+ || GET_MODE_WIDER_MODE (mode).require () == in_mode);
/* We are in in_mode. Count how many bits outside of mode
have to be copies of the sign-bit. */
- for (i = mode; i != in_mode; i = GET_MODE_WIDER_MODE (i))
+ FOR_EACH_MODE (i, mode, in_mode)
{
- machine_mode wider = GET_MODE_WIDER_MODE (i);
+ /* This must always exist (for the last iteration it will be
+ IN_MODE). */
+ scalar_int_mode wider = GET_MODE_WIDER_MODE (i).require ();
if (targetm.mode_rep_extended (i, wider) == SIGN_EXTEND
/* We can only check sign-bit copies starting from the
{
if (mode != VOIDmode)
{
- if (GET_MODE_PRECISION (mode) > HOST_BITS_PER_WIDE_INT)
+ if (!HWI_COMPUTABLE_MODE_P (mode))
return -1;
m &= GET_MODE_MASK (mode);
}
/* Return the mode of MEM's address. */
-machine_mode
+scalar_int_mode
get_address_mode (rtx mem)
{
machine_mode mode;
gcc_assert (MEM_P (mem));
mode = GET_MODE (XEXP (mem, 0));
if (mode != VOIDmode)
- return mode;
+ return as_a <scalar_int_mode> (mode);
return targetm.addr_space.address_mode (MEM_ADDR_SPACE (mem));
}
\f
machine_mode mode = GET_MODE (XEXP (x, 0));
HOST_WIDE_INT len = INTVAL (XEXP (x, 1));
HOST_WIDE_INT pos = INTVAL (XEXP (x, 2));
+ poly_int64 remaining_bits = GET_MODE_PRECISION (mode) - len;
- return (pos == (BITS_BIG_ENDIAN ? GET_MODE_PRECISION (mode) - len : 0));
+ return known_eq (pos, BITS_BIG_ENDIAN ? remaining_bits : 0);
}
return false;
}