public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r14-3746] _BitInt lowering support [PR102989]
@ 2023-09-06 15:57 Jakub Jelinek
  0 siblings, 0 replies; only message in thread
From: Jakub Jelinek @ 2023-09-06 15:57 UTC (permalink / raw)
  To: gcc-cvs

https://gcc.gnu.org/g:a9d6c7fbeb374365058ffe2b9815d2b4b7193d38

commit r14-3746-ga9d6c7fbeb374365058ffe2b9815d2b4b7193d38
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Wed Sep 6 17:27:41 2023 +0200

    _BitInt lowering support [PR102989]
    
    The following patch adds a new bitintlower lowering pass which lowers most
    operations on medium _BitInt into operations on corresponding integer types,
    large _BitInt into straight line code operating on 2 or more limbs and
    finally huge _BitInt into a loop plus optional straight line code.
    
    As the only supported architecture is little-endian, the lowering only
    supports little-endian for now, because it would be impossible to test it
    all for big-endian.  Rest is written with any endian support in mind, but
    of course only little-endian has been actually tested.
    I hope it is ok to add big-endian support to the lowering pass incrementally
    later when first big-endian target shows with the backend support.
    There are 2 possibilities of adding such support, one would be minimal one,
    just tweak limb_access function and perhaps one or two other spots and
    transform there the indexes from little endian (index 0 is least significant)
    to big endian for just the memory access.  Advantage is I think maintainance
    costs, disadvantage is that the loops will still iterate from 0 to some number
    of limbs and we'd rely on IVOPTs or something similar changing it later if
    needed.  Or we could make those indexes endian related everywhere, though
    I'm afraid that would be several hundreds of changes.
    
    For switches indexed by large/huge _BitInt the patch invokes what the switch
    lowering pass does (but only on those specific switches, not all of them);
    the switch lowering breaks the switches into clusters and none of the clusters
    can have a range which doesn't fit into 64-bit UWHI, everything else will be
    turned into a tree of comparisons.  For clusters normally emitted as smaller
    switches, because we already have a guarantee that the low .. high range is
    at most 64 bits, the patch forces subtraction of the low and turns it into
    a 64-bit switch.  This is done before the actual pass starts.
    Similarly, we cancel lowering of certain constructs like ABS_EXPR, ABSU_EXPR,
    MIN_EXPR, MAX_EXPR and COND_EXPR and turn those back to simpler comparisons
    etc., so that fewer operations need to be lowered later.
    
    2023-09-06  Jakub Jelinek  <jakub@redhat.com>
    
            PR c/102989
            * Makefile.in (OBJS): Add gimple-lower-bitint.o.
            * passes.def: Add pass_lower_bitint after pass_lower_complex and
            pass_lower_bitint_O0 after pass_lower_complex_O0.
            * tree-pass.h (PROP_gimple_lbitint): Define.
            (make_pass_lower_bitint_O0, make_pass_lower_bitint): Declare.
            * gimple-lower-bitint.h: New file.
            * tree-ssa-live.h (struct _var_map): Add bitint member.
            (init_var_map): Adjust declaration.
            (region_contains_p): Handle map->bitint like map->outofssa_p.
            * tree-ssa-live.cc (init_var_map): Add BITINT argument, initialize
            map->bitint and set map->outofssa_p to false if it is non-NULL.
            * tree-ssa-coalesce.cc: Include gimple-lower-bitint.h.
            (build_ssa_conflict_graph): Call build_bitint_stmt_ssa_conflicts if
            map->bitint.
            (create_coalesce_list_for_region): For map->bitint ignore SSA_NAMEs
            not in that bitmap, and allow res without default def.
            (compute_optimized_partition_bases): In map->bitint mode try hard to
            coalesce any SSA_NAMEs with the same size.
            (coalesce_bitint): New function.
            (coalesce_ssa_name): In map->bitint mode, or map->bitmap into
            used_in_copies and call coalesce_bitint.
            * gimple-lower-bitint.cc: New file.

Diff:
---
 gcc/Makefile.in            |    1 +
 gcc/gimple-lower-bitint.cc | 6074 ++++++++++++++++++++++++++++++++++++++++++++
 gcc/gimple-lower-bitint.h  |   31 +
 gcc/passes.def             |    3 +
 gcc/tree-pass.h            |    3 +
 gcc/tree-ssa-coalesce.cc   |  148 +-
 gcc/tree-ssa-live.cc       |    8 +-
 gcc/tree-ssa-live.h        |    8 +-
 8 files changed, 6270 insertions(+), 6 deletions(-)

diff --git a/gcc/Makefile.in b/gcc/Makefile.in
index 5930b52462aa..6d608db4dd24 100644
--- a/gcc/Makefile.in
+++ b/gcc/Makefile.in
@@ -1457,6 +1457,7 @@ OBJS = \
 	gimple-loop-jam.o \
 	gimple-loop-versioning.o \
 	gimple-low.o \
+	gimple-lower-bitint.o \
 	gimple-predicate-analysis.o \
 	gimple-pretty-print.o \
 	gimple-range.o \
diff --git a/gcc/gimple-lower-bitint.cc b/gcc/gimple-lower-bitint.cc
new file mode 100644
index 000000000000..3906c00bf3fa
--- /dev/null
+++ b/gcc/gimple-lower-bitint.cc
@@ -0,0 +1,6074 @@
+/* Lower _BitInt(N) operations to scalar operations.
+   Copyright (C) 2023 Free Software Foundation, Inc.
+   Contributed by Jakub Jelinek <jakub@redhat.com>.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published by the
+Free Software Foundation; either version 3, or (at your option) any
+later version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT
+ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3.  If not see
+<http://www.gnu.org/licenses/>.  */
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "backend.h"
+#include "rtl.h"
+#include "tree.h"
+#include "gimple.h"
+#include "cfghooks.h"
+#include "tree-pass.h"
+#include "ssa.h"
+#include "fold-const.h"
+#include "gimplify.h"
+#include "gimple-iterator.h"
+#include "tree-cfg.h"
+#include "tree-dfa.h"
+#include "cfgloop.h"
+#include "cfganal.h"
+#include "target.h"
+#include "tree-ssa-live.h"
+#include "tree-ssa-coalesce.h"
+#include "domwalk.h"
+#include "memmodel.h"
+#include "optabs.h"
+#include "varasm.h"
+#include "gimple-range.h"
+#include "value-range.h"
+#include "langhooks.h"
+#include "gimplify-me.h"
+#include "diagnostic-core.h"
+#include "tree-eh.h"
+#include "tree-pretty-print.h"
+#include "alloc-pool.h"
+#include "tree-into-ssa.h"
+#include "tree-cfgcleanup.h"
+#include "tree-switch-conversion.h"
+#include "ubsan.h"
+#include "gimple-lower-bitint.h"
+
+/* Split BITINT_TYPE precisions in 4 categories.  Small _BitInt, where
+   target hook says it is a single limb, middle _BitInt which per ABI
+   does not, but there is some INTEGER_TYPE in which arithmetics can be
+   performed (operations on such _BitInt are lowered to casts to that
+   arithmetic type and cast back; e.g. on x86_64 limb is DImode, but
+   target supports TImode, so _BitInt(65) to _BitInt(128) are middle
+   ones), large _BitInt which should by straight line code and
+   finally huge _BitInt which should be handled by loops over the limbs.  */
+
+enum bitint_prec_kind {
+  bitint_prec_small,
+  bitint_prec_middle,
+  bitint_prec_large,
+  bitint_prec_huge
+};
+
+/* Caches to speed up bitint_precision_kind.  */
+
+static int small_max_prec, mid_min_prec, large_min_prec, huge_min_prec;
+static int limb_prec;
+
+/* Categorize _BitInt(PREC) as small, middle, large or huge.  */
+
+static bitint_prec_kind
+bitint_precision_kind (int prec)
+{
+  if (prec <= small_max_prec)
+    return bitint_prec_small;
+  if (huge_min_prec && prec >= huge_min_prec)
+    return bitint_prec_huge;
+  if (large_min_prec && prec >= large_min_prec)
+    return bitint_prec_large;
+  if (mid_min_prec && prec >= mid_min_prec)
+    return bitint_prec_middle;
+
+  struct bitint_info info;
+  gcc_assert (targetm.c.bitint_type_info (prec, &info));
+  scalar_int_mode limb_mode = as_a <scalar_int_mode> (info.limb_mode);
+  if (prec <= GET_MODE_PRECISION (limb_mode))
+    {
+      small_max_prec = prec;
+      return bitint_prec_small;
+    }
+  scalar_int_mode arith_mode = (targetm.scalar_mode_supported_p (TImode)
+				? TImode : DImode);
+  if (!large_min_prec
+      && GET_MODE_PRECISION (arith_mode) > GET_MODE_PRECISION (limb_mode))
+    large_min_prec = GET_MODE_PRECISION (arith_mode) + 1;
+  if (!limb_prec)
+    limb_prec = GET_MODE_PRECISION (limb_mode);
+  if (!huge_min_prec)
+    {
+      if (4 * limb_prec >= GET_MODE_PRECISION (arith_mode))
+	huge_min_prec = 4 * limb_prec;
+      else
+	huge_min_prec = GET_MODE_PRECISION (arith_mode) + 1;
+    }
+  if (prec <= GET_MODE_PRECISION (arith_mode))
+    {
+      if (!mid_min_prec || prec < mid_min_prec)
+	mid_min_prec = prec;
+      return bitint_prec_middle;
+    }
+  if (large_min_prec && prec <= large_min_prec)
+    return bitint_prec_large;
+  return bitint_prec_huge;
+}
+
+/* Same for a TYPE.  */
+
+static bitint_prec_kind
+bitint_precision_kind (tree type)
+{
+  return bitint_precision_kind (TYPE_PRECISION (type));
+}
+
+/* Return minimum precision needed to describe INTEGER_CST
+   CST.  All bits above that precision up to precision of
+   TREE_TYPE (CST) are cleared if EXT is set to 0, or set
+   if EXT is set to -1.  */
+
+static unsigned
+bitint_min_cst_precision (tree cst, int &ext)
+{
+  ext = tree_int_cst_sgn (cst) < 0 ? -1 : 0;
+  wide_int w = wi::to_wide (cst);
+  unsigned min_prec = wi::min_precision (w, TYPE_SIGN (TREE_TYPE (cst)));
+  /* For signed values, we don't need to count the sign bit,
+     we'll use constant 0 or -1 for the upper bits.  */
+  if (!TYPE_UNSIGNED (TREE_TYPE (cst)))
+    --min_prec;
+  else
+    {
+      /* For unsigned values, also try signed min_precision
+	 in case the constant has lots of most significant bits set.  */
+      unsigned min_prec2 = wi::min_precision (w, SIGNED) - 1;
+      if (min_prec2 < min_prec)
+	{
+	  ext = -1;
+	  return min_prec2;
+	}
+    }
+  return min_prec;
+}
+
+namespace {
+
+/* If OP is middle _BitInt, cast it to corresponding INTEGER_TYPE
+   cached in TYPE and return it.  */
+
+tree
+maybe_cast_middle_bitint (gimple_stmt_iterator *gsi, tree op, tree &type)
+{
+  if (op == NULL_TREE
+      || TREE_CODE (TREE_TYPE (op)) != BITINT_TYPE
+      || bitint_precision_kind (TREE_TYPE (op)) != bitint_prec_middle)
+    return op;
+
+  int prec = TYPE_PRECISION (TREE_TYPE (op));
+  int uns = TYPE_UNSIGNED (TREE_TYPE (op));
+  if (type == NULL_TREE
+      || TYPE_PRECISION (type) != prec
+      || TYPE_UNSIGNED (type) != uns)
+    type = build_nonstandard_integer_type (prec, uns);
+
+  if (TREE_CODE (op) != SSA_NAME)
+    {
+      tree nop = fold_convert (type, op);
+      if (is_gimple_val (nop))
+	return nop;
+    }
+
+  tree nop = make_ssa_name (type);
+  gimple *g = gimple_build_assign (nop, NOP_EXPR, op);
+  gsi_insert_before (gsi, g, GSI_SAME_STMT);
+  return nop;
+}
+
+/* Return true if STMT can be handled in a loop from least to most
+   significant limb together with its dependencies.  */
+
+bool
+mergeable_op (gimple *stmt)
+{
+  if (!is_gimple_assign (stmt))
+    return false;
+  switch (gimple_assign_rhs_code (stmt))
+    {
+    case PLUS_EXPR:
+    case MINUS_EXPR:
+    case NEGATE_EXPR:
+    case BIT_AND_EXPR:
+    case BIT_IOR_EXPR:
+    case BIT_XOR_EXPR:
+    case BIT_NOT_EXPR:
+    case SSA_NAME:
+    case INTEGER_CST:
+      return true;
+    case LSHIFT_EXPR:
+      {
+	tree cnt = gimple_assign_rhs2 (stmt);
+	if (tree_fits_uhwi_p (cnt)
+	    && tree_to_uhwi (cnt) < (unsigned HOST_WIDE_INT) limb_prec)
+	  return true;
+      }
+      break;
+    CASE_CONVERT:
+    case VIEW_CONVERT_EXPR:
+      {
+	tree lhs_type = TREE_TYPE (gimple_assign_lhs (stmt));
+	tree rhs_type = TREE_TYPE (gimple_assign_rhs1 (stmt));
+	if (TREE_CODE (gimple_assign_rhs1 (stmt)) == SSA_NAME
+	    && TREE_CODE (lhs_type) == BITINT_TYPE
+	    && TREE_CODE (rhs_type) == BITINT_TYPE
+	    && bitint_precision_kind (lhs_type) >= bitint_prec_large
+	    && bitint_precision_kind (rhs_type) >= bitint_prec_large
+	    && tree_int_cst_equal (TYPE_SIZE (lhs_type), TYPE_SIZE (rhs_type)))
+	  {
+	    if (TYPE_PRECISION (rhs_type) >= TYPE_PRECISION (lhs_type))
+	      return true;
+	    if ((unsigned) TYPE_PRECISION (lhs_type) % (2 * limb_prec) != 0)
+	      return true;
+	    if (bitint_precision_kind (lhs_type) == bitint_prec_large)
+	      return true;
+	  }
+	break;
+      }
+    default:
+      break;
+    }
+  return false;
+}
+
+/* Return non-zero if stmt is .{ADD,SUB,MUL}_OVERFLOW call with
+   _Complex large/huge _BitInt lhs which has at most two immediate uses,
+   at most one use in REALPART_EXPR stmt in the same bb and exactly one
+   IMAGPART_EXPR use in the same bb with a single use which casts it to
+   non-BITINT_TYPE integral type.  If there is a REALPART_EXPR use,
+   return 2.  Such cases (most common uses of those builtins) can be
+   optimized by marking their lhs and lhs of IMAGPART_EXPR and maybe lhs
+   of REALPART_EXPR as not needed to be backed up by a stack variable.
+   For .UBSAN_CHECK_{ADD,SUB,MUL} return 3.  */
+
+int
+optimizable_arith_overflow (gimple *stmt)
+{
+  bool is_ubsan = false;
+  if (!is_gimple_call (stmt) || !gimple_call_internal_p (stmt))
+    return false;
+  switch (gimple_call_internal_fn (stmt))
+    {
+    case IFN_ADD_OVERFLOW:
+    case IFN_SUB_OVERFLOW:
+    case IFN_MUL_OVERFLOW:
+      break;
+    case IFN_UBSAN_CHECK_ADD:
+    case IFN_UBSAN_CHECK_SUB:
+    case IFN_UBSAN_CHECK_MUL:
+      is_ubsan = true;
+      break;
+    default:
+      return 0;
+    }
+  tree lhs = gimple_call_lhs (stmt);
+  if (!lhs)
+    return 0;
+  if (SSA_NAME_OCCURS_IN_ABNORMAL_PHI (lhs))
+    return 0;
+  tree type = is_ubsan ? TREE_TYPE (lhs) : TREE_TYPE (TREE_TYPE (lhs));
+  if (TREE_CODE (type) != BITINT_TYPE
+      || bitint_precision_kind (type) < bitint_prec_large)
+    return 0;
+
+  if (is_ubsan)
+    {
+      use_operand_p use_p;
+      gimple *use_stmt;
+      if (!single_imm_use (lhs, &use_p, &use_stmt)
+	  || gimple_bb (use_stmt) != gimple_bb (stmt)
+	  || !gimple_store_p (use_stmt)
+	  || !is_gimple_assign (use_stmt)
+	  || gimple_has_volatile_ops (use_stmt)
+	  || stmt_ends_bb_p (use_stmt))
+	return 0;
+      return 3;
+    }
+
+  imm_use_iterator ui;
+  use_operand_p use_p;
+  int seen = 0;
+  FOR_EACH_IMM_USE_FAST (use_p, ui, lhs)
+    {
+      gimple *g = USE_STMT (use_p);
+      if (is_gimple_debug (g))
+	continue;
+      if (!is_gimple_assign (g) || gimple_bb (g) != gimple_bb (stmt))
+	return 0;
+      if (gimple_assign_rhs_code (g) == REALPART_EXPR)
+	{
+	  if ((seen & 1) != 0)
+	    return 0;
+	  seen |= 1;
+	}
+      else if (gimple_assign_rhs_code (g) == IMAGPART_EXPR)
+	{
+	  if ((seen & 2) != 0)
+	    return 0;
+	  seen |= 2;
+
+	  use_operand_p use2_p;
+	  gimple *use_stmt;
+	  tree lhs2 = gimple_assign_lhs (g);
+	  if (SSA_NAME_OCCURS_IN_ABNORMAL_PHI (lhs2))
+	    return 0;
+	  if (!single_imm_use (lhs2, &use2_p, &use_stmt)
+	      || gimple_bb (use_stmt) != gimple_bb (stmt)
+	      || !gimple_assign_cast_p (use_stmt))
+	    return 0;
+
+	  lhs2 = gimple_assign_lhs (use_stmt);
+	  if (!INTEGRAL_TYPE_P (TREE_TYPE (lhs2))
+	      || TREE_CODE (TREE_TYPE (lhs2)) == BITINT_TYPE)
+	    return 0;
+	}
+      else
+	return 0;
+    }
+  if ((seen & 2) == 0)
+    return 0;
+  return seen == 3 ? 2 : 1;
+}
+
+/* If STMT is some kind of comparison (GIMPLE_COND, comparison assignment)
+   comparing large/huge _BitInt types, return the comparison code and if
+   non-NULL fill in the comparison operands to *POP1 and *POP2.  */
+
+tree_code
+comparison_op (gimple *stmt, tree *pop1, tree *pop2)
+{
+  tree op1 = NULL_TREE, op2 = NULL_TREE;
+  tree_code code = ERROR_MARK;
+  if (gimple_code (stmt) == GIMPLE_COND)
+    {
+      code = gimple_cond_code (stmt);
+      op1 = gimple_cond_lhs (stmt);
+      op2 = gimple_cond_rhs (stmt);
+    }
+  else if (is_gimple_assign (stmt))
+    {
+      code = gimple_assign_rhs_code (stmt);
+      op1 = gimple_assign_rhs1 (stmt);
+      if (TREE_CODE_CLASS (code) == tcc_comparison
+	  || TREE_CODE_CLASS (code) == tcc_binary)
+	op2 = gimple_assign_rhs2 (stmt);
+    }
+  if (TREE_CODE_CLASS (code) != tcc_comparison)
+    return ERROR_MARK;
+  tree type = TREE_TYPE (op1);
+  if (TREE_CODE (type) != BITINT_TYPE
+      || bitint_precision_kind (type) < bitint_prec_large)
+    return ERROR_MARK;
+  if (pop1)
+    {
+      *pop1 = op1;
+      *pop2 = op2;
+    }
+  return code;
+}
+
+/* Class used during large/huge _BitInt lowering containing all the
+   state for the methods.  */
+
+struct bitint_large_huge
+{
+  bitint_large_huge ()
+    : m_names (NULL), m_loads (NULL), m_preserved (NULL),
+      m_single_use_names (NULL), m_map (NULL), m_vars (NULL),
+      m_limb_type (NULL_TREE), m_data (vNULL) {}
+
+  ~bitint_large_huge ();
+
+  void insert_before (gimple *);
+  tree limb_access_type (tree, tree);
+  tree limb_access (tree, tree, tree, bool);
+  void if_then (gimple *, profile_probability, edge &, edge &);
+  void if_then_else (gimple *, profile_probability, edge &, edge &);
+  void if_then_if_then_else (gimple *g, gimple *,
+			     profile_probability, profile_probability,
+			     edge &, edge &, edge &);
+  tree handle_operand (tree, tree);
+  tree prepare_data_in_out (tree, tree, tree *);
+  tree add_cast (tree, tree);
+  tree handle_plus_minus (tree_code, tree, tree, tree);
+  tree handle_lshift (tree, tree, tree);
+  tree handle_cast (tree, tree, tree);
+  tree handle_load (gimple *, tree);
+  tree handle_stmt (gimple *, tree);
+  tree handle_operand_addr (tree, gimple *, int *, int *);
+  tree create_loop (tree, tree *);
+  tree lower_mergeable_stmt (gimple *, tree_code &, tree, tree);
+  tree lower_comparison_stmt (gimple *, tree_code &, tree, tree);
+  void lower_shift_stmt (tree, gimple *);
+  void lower_muldiv_stmt (tree, gimple *);
+  void lower_float_conv_stmt (tree, gimple *);
+  tree arith_overflow_extract_bits (unsigned int, unsigned int, tree,
+				    unsigned int, bool);
+  void finish_arith_overflow (tree, tree, tree, tree, tree, tree, gimple *,
+			      tree_code);
+  void lower_addsub_overflow (tree, gimple *);
+  void lower_mul_overflow (tree, gimple *);
+  void lower_cplxpart_stmt (tree, gimple *);
+  void lower_complexexpr_stmt (gimple *);
+  void lower_call (tree, gimple *);
+  void lower_asm (gimple *);
+  void lower_stmt (gimple *);
+
+  /* Bitmap of large/huge _BitInt SSA_NAMEs except those can be
+     merged with their uses.  */
+  bitmap m_names;
+  /* Subset of those for lhs of load statements.  These will be
+     cleared in m_names if the loads will be mergeable with all
+     their uses.  */
+  bitmap m_loads;
+  /* Bitmap of large/huge _BitInt SSA_NAMEs that should survive
+     to later passes (arguments or return values of calls).  */
+  bitmap m_preserved;
+  /* Subset of m_names which have a single use.  As the lowering
+     can replace various original statements with their lowered
+     form even before it is done iterating over all basic blocks,
+     testing has_single_use for the purpose of emitting clobbers
+     doesn't work properly.  */
+  bitmap m_single_use_names;
+  /* Used for coalescing/partitioning of large/huge _BitInt SSA_NAMEs
+     set in m_names.  */
+  var_map m_map;
+  /* Mapping of the partitions to corresponding decls.  */
+  tree *m_vars;
+  /* Unsigned integer type with limb precision.  */
+  tree m_limb_type;
+  /* Its TYPE_SIZE_UNIT.  */
+  unsigned HOST_WIDE_INT m_limb_size;
+  /* Location of a gimple stmt which is being currently lowered.  */
+  location_t m_loc;
+  /* Current stmt iterator where code is being lowered currently.  */
+  gimple_stmt_iterator m_gsi;
+  /* Statement after which any clobbers should be added if non-NULL.  */
+  gimple *m_after_stmt;
+  /* Set when creating loops to the loop header bb and its preheader.  */
+  basic_block m_bb, m_preheader_bb;
+  /* Stmt iterator after which initialization statements should be emitted.  */
+  gimple_stmt_iterator m_init_gsi;
+  /* Decl into which a mergeable statement stores result.  */
+  tree m_lhs;
+  /* handle_operand/handle_stmt can be invoked in various ways.
+
+     lower_mergeable_stmt for large _BitInt calls those with constant
+     idx only, expanding to straight line code, for huge _BitInt
+     emits a loop from least significant limb upwards, where each loop
+     iteration handles 2 limbs, plus there can be up to one full limb
+     and one partial limb processed after the loop, where handle_operand
+     and/or handle_stmt are called with constant idx.  m_upwards_2limb
+     is set for this case, false otherwise.  m_upwards is true if it
+     is either large or huge _BitInt handled by lower_mergeable_stmt,
+     i.e. indexes always increase.
+
+     Another way is used by lower_comparison_stmt, which walks limbs
+     from most significant to least significant, partial limb if any
+     processed first with constant idx and then loop processing a single
+     limb per iteration with non-constant idx.
+
+     Another way is used in lower_shift_stmt, where for LSHIFT_EXPR
+     destination limbs are processed from most significant to least
+     significant or for RSHIFT_EXPR the other way around, in loops or
+     straight line code, but idx usually is non-constant (so from
+     handle_operand/handle_stmt POV random access).  The LSHIFT_EXPR
+     handling there can access even partial limbs using non-constant
+     idx (then m_var_msb should be true, for all the other cases
+     including lower_mergeable_stmt/lower_comparison_stmt that is
+     not the case and so m_var_msb should be false.
+
+     m_first should be set the first time handle_operand/handle_stmt
+     is called and clear when it is called for some other limb with
+     the same argument.  If the lowering of an operand (e.g. INTEGER_CST)
+     or statement (e.g. +/-/<< with < limb_prec constant) needs some
+     state between the different calls, when m_first is true it should
+     push some trees to m_data vector and also make sure m_data_cnt is
+     incremented by how many trees were pushed, and when m_first is
+     false, it can use the m_data[m_data_cnt] etc. data or update them,
+     just needs to bump m_data_cnt by the same amount as when it was
+     called with m_first set.  The toplevel calls to
+     handle_operand/handle_stmt should set m_data_cnt to 0 and truncate
+     m_data vector when setting m_first to true.
+
+     m_cast_conditional and m_bitfld_load are used when handling a
+     bit-field load inside of a widening cast.  handle_cast sometimes
+     needs to do runtime comparisons and handle_operand only conditionally
+     or even in two separate conditional blocks for one idx (once with
+     constant index after comparing the runtime one for equality with the
+     constant).  In these cases, m_cast_conditional is set to true and
+     the bit-field load then communicates its m_data_cnt to handle_cast
+     using m_bitfld_load.  */
+  bool m_first;
+  bool m_var_msb;
+  unsigned m_upwards_2limb;
+  bool m_upwards;
+  bool m_cast_conditional;
+  unsigned m_bitfld_load;
+  vec<tree> m_data;
+  unsigned int m_data_cnt;
+};
+
+bitint_large_huge::~bitint_large_huge ()
+{
+  BITMAP_FREE (m_names);
+  BITMAP_FREE (m_loads);
+  BITMAP_FREE (m_preserved);
+  BITMAP_FREE (m_single_use_names);
+  if (m_map)
+    delete_var_map (m_map);
+  XDELETEVEC (m_vars);
+  m_data.release ();
+}
+
+/* Insert gimple statement G before current location
+   and set its gimple_location.  */
+
+void
+bitint_large_huge::insert_before (gimple *g)
+{
+  gimple_set_location (g, m_loc);
+  gsi_insert_before (&m_gsi, g, GSI_SAME_STMT);
+}
+
+/* Return type for accessing limb IDX of BITINT_TYPE TYPE.
+   This is normally m_limb_type, except for a partial most
+   significant limb if any.  */
+
+tree
+bitint_large_huge::limb_access_type (tree type, tree idx)
+{
+  if (type == NULL_TREE)
+    return m_limb_type;
+  unsigned HOST_WIDE_INT i = tree_to_uhwi (idx);
+  unsigned int prec = TYPE_PRECISION (type);
+  gcc_assert (i * limb_prec < prec);
+  if ((i + 1) * limb_prec <= prec)
+    return m_limb_type;
+  else
+    return build_nonstandard_integer_type (prec % limb_prec,
+					   TYPE_UNSIGNED (type));
+}
+
+/* Return a tree how to access limb IDX of VAR corresponding to BITINT_TYPE
+   TYPE.  If WRITE_P is true, it will be a store, otherwise a read.  */
+
+tree
+bitint_large_huge::limb_access (tree type, tree var, tree idx, bool write_p)
+{
+  tree atype = (tree_fits_uhwi_p (idx)
+		? limb_access_type (type, idx) : m_limb_type);
+  tree ret;
+  if (DECL_P (var) && tree_fits_uhwi_p (idx))
+    {
+      tree ptype = build_pointer_type (strip_array_types (TREE_TYPE (var)));
+      unsigned HOST_WIDE_INT off = tree_to_uhwi (idx) * m_limb_size;
+      ret = build2 (MEM_REF, m_limb_type,
+		    build_fold_addr_expr (var),
+		    build_int_cst (ptype, off));
+      TREE_THIS_VOLATILE (ret) = TREE_THIS_VOLATILE (var);
+      TREE_SIDE_EFFECTS (ret) = TREE_SIDE_EFFECTS (var);
+    }
+  else if (TREE_CODE (var) == MEM_REF && tree_fits_uhwi_p (idx))
+    {
+      ret
+	= build2 (MEM_REF, m_limb_type, TREE_OPERAND (var, 0),
+		  size_binop (PLUS_EXPR, TREE_OPERAND (var, 1),
+			      build_int_cst (TREE_TYPE (TREE_OPERAND (var, 1)),
+					     tree_to_uhwi (idx)
+					     * m_limb_size)));
+      TREE_THIS_VOLATILE (ret) = TREE_THIS_VOLATILE (var);
+      TREE_SIDE_EFFECTS (ret) = TREE_SIDE_EFFECTS (var);
+      TREE_THIS_NOTRAP (ret) = TREE_THIS_NOTRAP (var);
+    }
+  else
+    {
+      var = unshare_expr (var);
+      if (TREE_CODE (TREE_TYPE (var)) != ARRAY_TYPE
+	  || !useless_type_conversion_p (m_limb_type,
+					 TREE_TYPE (TREE_TYPE (var))))
+	{
+	  unsigned HOST_WIDE_INT nelts
+	    = CEIL (tree_to_uhwi (TYPE_SIZE (type)), limb_prec);
+	  tree atype = build_array_type_nelts (m_limb_type, nelts);
+	  var = build1 (VIEW_CONVERT_EXPR, atype, var);
+	}
+      ret = build4 (ARRAY_REF, m_limb_type, var, idx, NULL_TREE, NULL_TREE);
+    }
+  if (!write_p && !useless_type_conversion_p (atype, m_limb_type))
+    {
+      gimple *g = gimple_build_assign (make_ssa_name (m_limb_type), ret);
+      insert_before (g);
+      ret = gimple_assign_lhs (g);
+      ret = build1 (NOP_EXPR, atype, ret);
+    }
+  return ret;
+}
+
+/* Emit a half diamond,
+   if (COND)
+     |\
+     | \
+     |  \
+     | new_bb1
+     |  /
+     | /
+     |/
+   or if (COND) new_bb1;
+   PROB is the probability that the condition is true.
+   Updates m_gsi to start of new_bb1.
+   Sets EDGE_TRUE to edge from new_bb1 to successor and
+   EDGE_FALSE to the EDGE_FALSE_VALUE edge from if (COND) bb.  */
+
+void
+bitint_large_huge::if_then (gimple *cond, profile_probability prob,
+			    edge &edge_true, edge &edge_false)
+{
+  insert_before (cond);
+  edge e1 = split_block (gsi_bb (m_gsi), cond);
+  edge e2 = split_block (e1->dest, (gimple *) NULL);
+  edge e3 = make_edge (e1->src, e2->dest, EDGE_FALSE_VALUE);
+  e1->flags = EDGE_TRUE_VALUE;
+  e1->probability = prob;
+  e3->probability = prob.invert ();
+  set_immediate_dominator (CDI_DOMINATORS, e2->dest, e1->src);
+  edge_true = e2;
+  edge_false = e3;
+  m_gsi = gsi_after_labels (e1->dest);
+}
+
+/* Emit a full diamond,
+       if (COND)
+	 /\
+	/  \
+       /    \
+   new_bb1 new_bb2
+       \    /
+	\  /
+	 \/
+   or if (COND) new_bb2; else new_bb1;
+   PROB is the probability that the condition is true.
+   Updates m_gsi to start of new_bb2.
+   Sets EDGE_TRUE to edge from new_bb1 to successor and
+   EDGE_FALSE to the EDGE_FALSE_VALUE edge from if (COND) bb.  */
+
+void
+bitint_large_huge::if_then_else (gimple *cond, profile_probability prob,
+				 edge &edge_true, edge &edge_false)
+{
+  insert_before (cond);
+  edge e1 = split_block (gsi_bb (m_gsi), cond);
+  edge e2 = split_block (e1->dest, (gimple *) NULL);
+  basic_block bb = create_empty_bb (e1->dest);
+  add_bb_to_loop (bb, e1->dest->loop_father);
+  edge e3 = make_edge (e1->src, bb, EDGE_TRUE_VALUE);
+  e1->flags = EDGE_FALSE_VALUE;
+  e3->probability = prob;
+  e1->probability = prob.invert ();
+  set_immediate_dominator (CDI_DOMINATORS, bb, e1->src);
+  set_immediate_dominator (CDI_DOMINATORS, e2->dest, e1->src);
+  edge_true = make_edge (bb, e2->dest, EDGE_FALLTHRU);
+  edge_false = e2;
+  m_gsi = gsi_after_labels (bb);
+}
+
+/* Emit a half diamond with full diamond in it
+   if (COND1)
+     |\
+     | \
+     |  \
+     | if (COND2)
+     |    /  \
+     |   /    \
+     |new_bb1 new_bb2
+     |   |    /
+     \   |   /
+      \  |  /
+       \ | /
+	\|/
+   or if (COND1) { if (COND2) new_bb2; else new_bb1; }
+   PROB1 is the probability that the condition 1 is true.
+   PROB2 is the probability that the condition 2 is true.
+   Updates m_gsi to start of new_bb1.
+   Sets EDGE_TRUE_TRUE to edge from new_bb2 to successor,
+   EDGE_TRUE_FALSE to edge from new_bb1 to successor and
+   EDGE_FALSE to the EDGE_FALSE_VALUE edge from if (COND1) bb.
+   If COND2 is NULL, this is equivalent to
+   if_then (COND1, PROB1, EDGE_TRUE_FALSE, EDGE_FALSE);
+   EDGE_TRUE_TRUE = NULL;  */
+
+void
+bitint_large_huge::if_then_if_then_else (gimple *cond1, gimple *cond2,
+					 profile_probability prob1,
+					 profile_probability prob2,
+					 edge &edge_true_true,
+					 edge &edge_true_false,
+					 edge &edge_false)
+{
+  edge e2, e3, e4 = NULL;
+  if_then (cond1, prob1, e2, e3);
+  if (cond2 == NULL)
+    {
+      edge_true_true = NULL;
+      edge_true_false = e2;
+      edge_false = e3;
+      return;
+    }
+  insert_before (cond2);
+  e2 = split_block (gsi_bb (m_gsi), cond2);
+  basic_block bb = create_empty_bb (e2->dest);
+  add_bb_to_loop (bb, e2->dest->loop_father);
+  e4 = make_edge (e2->src, bb, EDGE_TRUE_VALUE);
+  set_immediate_dominator (CDI_DOMINATORS, bb, e2->src);
+  e4->probability = prob2;
+  e2->flags = EDGE_FALSE_VALUE;
+  e2->probability = prob2.invert ();
+  e4 = make_edge (bb, e3->dest, EDGE_FALLTHRU);
+  e2 = find_edge (e2->dest, e3->dest);
+  edge_true_true = e4;
+  edge_true_false = e2;
+  edge_false = e3;
+  m_gsi = gsi_after_labels (e2->src);
+}
+
+/* Emit code to access limb IDX from OP.  */
+
+tree
+bitint_large_huge::handle_operand (tree op, tree idx)
+{
+  switch (TREE_CODE (op))
+    {
+    case SSA_NAME:
+      if (m_names == NULL
+	  || !bitmap_bit_p (m_names, SSA_NAME_VERSION (op)))
+	{
+	  if (SSA_NAME_IS_DEFAULT_DEF (op))
+	    {
+	      if (m_first)
+		{
+		  tree v = create_tmp_reg (m_limb_type);
+		  if (SSA_NAME_VAR (op) && VAR_P (SSA_NAME_VAR (op)))
+		    {
+		      DECL_NAME (v) = DECL_NAME (SSA_NAME_VAR (op));
+		      DECL_SOURCE_LOCATION (v)
+			= DECL_SOURCE_LOCATION (SSA_NAME_VAR (op));
+		    }
+		  v = get_or_create_ssa_default_def (cfun, v);
+		  m_data.safe_push (v);
+		}
+	      tree ret = m_data[m_data_cnt];
+	      m_data_cnt++;
+	      if (tree_fits_uhwi_p (idx))
+		{
+		  tree type = limb_access_type (TREE_TYPE (op), idx);
+		  ret = add_cast (type, ret);
+		}
+	      return ret;
+	    }
+	  location_t loc_save = m_loc;
+	  m_loc = gimple_location (SSA_NAME_DEF_STMT (op));
+	  tree ret = handle_stmt (SSA_NAME_DEF_STMT (op), idx);
+	  m_loc = loc_save;
+	  return ret;
+	}
+      int p;
+      gimple *g;
+      tree t;
+      p = var_to_partition (m_map, op);
+      gcc_assert (m_vars[p] != NULL_TREE);
+      t = limb_access (TREE_TYPE (op), m_vars[p], idx, false);
+      g = gimple_build_assign (make_ssa_name (TREE_TYPE (t)), t);
+      insert_before (g);
+      t = gimple_assign_lhs (g);
+      if (m_first
+	  && m_single_use_names
+	  && m_vars[p] != m_lhs
+	  && m_after_stmt
+	  && bitmap_bit_p (m_single_use_names, SSA_NAME_VERSION (op)))
+	{
+	  tree clobber = build_clobber (TREE_TYPE (m_vars[p]), CLOBBER_EOL);
+	  g = gimple_build_assign (m_vars[p], clobber);
+	  gimple_stmt_iterator gsi = gsi_for_stmt (m_after_stmt);
+	  gsi_insert_after (&gsi, g, GSI_SAME_STMT);
+	}
+      return t;
+    case INTEGER_CST:
+      if (tree_fits_uhwi_p (idx))
+	{
+	  tree c, type = limb_access_type (TREE_TYPE (op), idx);
+	  unsigned HOST_WIDE_INT i = tree_to_uhwi (idx);
+	  if (m_first)
+	    {
+	      m_data.safe_push (NULL_TREE);
+	      m_data.safe_push (NULL_TREE);
+	    }
+	  if (limb_prec != HOST_BITS_PER_WIDE_INT)
+	    {
+	      wide_int w = wi::rshift (wi::to_wide (op), i * limb_prec,
+				       TYPE_SIGN (TREE_TYPE (op)));
+	      c = wide_int_to_tree (type,
+				    wide_int::from (w, TYPE_PRECISION (type),
+						    UNSIGNED));
+	    }
+	  else if (i >= TREE_INT_CST_EXT_NUNITS (op))
+	    c = build_int_cst (type,
+			       tree_int_cst_sgn (op) < 0 ? -1 : 0);
+	  else
+	    c = build_int_cst (type, TREE_INT_CST_ELT (op, i));
+	  m_data_cnt += 2;
+	  return c;
+	}
+      if (m_first
+	  || (m_data[m_data_cnt] == NULL_TREE
+	      && m_data[m_data_cnt + 1] == NULL_TREE))
+	{
+	  unsigned int prec = TYPE_PRECISION (TREE_TYPE (op));
+	  unsigned int rem = prec % (2 * limb_prec);
+	  int ext;
+	  unsigned min_prec = bitint_min_cst_precision (op, ext);
+	  if (m_first)
+	    {
+	      m_data.safe_push (NULL_TREE);
+	      m_data.safe_push (NULL_TREE);
+	    }
+	  if (integer_zerop (op))
+	    {
+	      tree c = build_zero_cst (m_limb_type);
+	      m_data[m_data_cnt] = c;
+	      m_data[m_data_cnt + 1] = c;
+	    }
+	  else if (integer_all_onesp (op))
+	    {
+	      tree c = build_all_ones_cst (m_limb_type);
+	      m_data[m_data_cnt] = c;
+	      m_data[m_data_cnt + 1] = c;
+	    }
+	  else if (m_upwards_2limb && min_prec <= (unsigned) limb_prec)
+	    {
+	      /* Single limb constant.  Use a phi with that limb from
+		 the preheader edge and 0 or -1 constant from the other edge
+		 and for the second limb in the loop.  */
+	      tree out;
+	      gcc_assert (m_first);
+	      m_data.pop ();
+	      m_data.pop ();
+	      prepare_data_in_out (fold_convert (m_limb_type, op), idx, &out);
+	      g = gimple_build_assign (m_data[m_data_cnt + 1],
+				       build_int_cst (m_limb_type, ext));
+	      insert_before (g);
+	      m_data[m_data_cnt + 1] = gimple_assign_rhs1 (g);
+	    }
+	  else if (min_prec > prec - rem - 2 * limb_prec)
+	    {
+	      /* Constant which has enough significant bits that it isn't
+		 worth trying to save .rodata space by extending from smaller
+		 number.  */
+	      tree type;
+	      if (m_var_msb)
+		type = TREE_TYPE (op);
+	      else
+		/* If we have a guarantee the most significant partial limb
+		   (if any) will be only accessed through handle_operand
+		   with INTEGER_CST idx, we don't need to include the partial
+		   limb in .rodata.  */
+		type = build_bitint_type (prec - rem, 1);
+	      tree c = tree_output_constant_def (fold_convert (type, op));
+	      m_data[m_data_cnt] = c;
+	      m_data[m_data_cnt + 1] = NULL_TREE;
+	    }
+	  else if (m_upwards_2limb)
+	    {
+	      /* Constant with smaller number of bits.  Trade conditional
+		 code for .rodata space by extending from smaller number.  */
+	      min_prec = CEIL (min_prec, 2 * limb_prec) * (2 * limb_prec);
+	      tree type = build_bitint_type (min_prec, 1);
+	      tree c = tree_output_constant_def (fold_convert (type, op));
+	      tree idx2 = make_ssa_name (sizetype);
+	      g = gimple_build_assign (idx2, PLUS_EXPR, idx, size_one_node);
+	      insert_before (g);
+	      g = gimple_build_cond (LT_EXPR, idx,
+				     size_int (min_prec / limb_prec),
+				     NULL_TREE, NULL_TREE);
+	      edge edge_true, edge_false;
+	      if_then (g, (min_prec >= (prec - rem) / 2
+			   ? profile_probability::likely ()
+			   : profile_probability::unlikely ()),
+		       edge_true, edge_false);
+	      tree c1 = limb_access (TREE_TYPE (op), c, idx, false);
+	      g = gimple_build_assign (make_ssa_name (TREE_TYPE (c1)), c1);
+	      insert_before (g);
+	      c1 = gimple_assign_lhs (g);
+	      tree c2 = limb_access (TREE_TYPE (op), c, idx2, false);
+	      g = gimple_build_assign (make_ssa_name (TREE_TYPE (c2)), c2);
+	      insert_before (g);
+	      c2 = gimple_assign_lhs (g);
+	      tree c3 = build_int_cst (m_limb_type, ext);
+	      m_gsi = gsi_after_labels (edge_true->dest);
+	      m_data[m_data_cnt] = make_ssa_name (m_limb_type);
+	      m_data[m_data_cnt + 1] = make_ssa_name (m_limb_type);
+	      gphi *phi = create_phi_node (m_data[m_data_cnt],
+					   edge_true->dest);
+	      add_phi_arg (phi, c1, edge_true, UNKNOWN_LOCATION);
+	      add_phi_arg (phi, c3, edge_false, UNKNOWN_LOCATION);
+	      phi = create_phi_node (m_data[m_data_cnt + 1], edge_true->dest);
+	      add_phi_arg (phi, c2, edge_true, UNKNOWN_LOCATION);
+	      add_phi_arg (phi, c3, edge_false, UNKNOWN_LOCATION);
+	    }
+	  else
+	    {
+	      /* Constant with smaller number of bits.  Trade conditional
+		 code for .rodata space by extending from smaller number.
+		 Version for loops with random access to the limbs or
+		 downwards loops.  */
+	      min_prec = CEIL (min_prec, limb_prec) * limb_prec;
+	      tree c;
+	      if (min_prec <= (unsigned) limb_prec)
+		c = fold_convert (m_limb_type, op);
+	      else
+		{
+		  tree type = build_bitint_type (min_prec, 1);
+		  c = tree_output_constant_def (fold_convert (type, op));
+		}
+	      m_data[m_data_cnt] = c;
+	      m_data[m_data_cnt + 1] = integer_type_node;
+	    }
+	  t = m_data[m_data_cnt];
+	  if (m_data[m_data_cnt + 1] == NULL_TREE)
+	    {
+	      t = limb_access (TREE_TYPE (op), t, idx, false);
+	      g = gimple_build_assign (make_ssa_name (TREE_TYPE (t)), t);
+	      insert_before (g);
+	      t = gimple_assign_lhs (g);
+	    }
+	}
+      else if (m_data[m_data_cnt + 1] == NULL_TREE)
+	{
+	  t = limb_access (TREE_TYPE (op), m_data[m_data_cnt], idx, false);
+	  g = gimple_build_assign (make_ssa_name (TREE_TYPE (t)), t);
+	  insert_before (g);
+	  t = gimple_assign_lhs (g);
+	}
+      else
+	t = m_data[m_data_cnt + 1];
+      if (m_data[m_data_cnt + 1] == integer_type_node)
+	{
+	  unsigned int prec = TYPE_PRECISION (TREE_TYPE (op));
+	  unsigned rem = prec % (2 * limb_prec);
+	  int ext = tree_int_cst_sgn (op) < 0 ? -1 : 0;
+	  tree c = m_data[m_data_cnt];
+	  unsigned min_prec = TYPE_PRECISION (TREE_TYPE (c));
+	  g = gimple_build_cond (LT_EXPR, idx,
+				 size_int (min_prec / limb_prec),
+				 NULL_TREE, NULL_TREE);
+	  edge edge_true, edge_false;
+	  if_then (g, (min_prec >= (prec - rem) / 2
+		       ? profile_probability::likely ()
+		       : profile_probability::unlikely ()),
+		   edge_true, edge_false);
+	  if (min_prec > (unsigned) limb_prec)
+	    {
+	      c = limb_access (TREE_TYPE (op), c, idx, false);
+	      g = gimple_build_assign (make_ssa_name (TREE_TYPE (c)), c);
+	      insert_before (g);
+	      c = gimple_assign_lhs (g);
+	    }
+	  tree c2 = build_int_cst (m_limb_type, ext);
+	  m_gsi = gsi_after_labels (edge_true->dest);
+	  t = make_ssa_name (m_limb_type);
+	  gphi *phi = create_phi_node (t, edge_true->dest);
+	  add_phi_arg (phi, c, edge_true, UNKNOWN_LOCATION);
+	  add_phi_arg (phi, c2, edge_false, UNKNOWN_LOCATION);
+	}
+      m_data_cnt += 2;
+      return t;
+    default:
+      gcc_unreachable ();
+    }
+}
+
+/* Helper method, add a PHI node with VAL from preheader edge if
+   inside of a loop and m_first.  Keep state in a pair of m_data
+   elements.  */
+
+tree
+bitint_large_huge::prepare_data_in_out (tree val, tree idx, tree *data_out)
+{
+  if (!m_first)
+    {
+      *data_out = tree_fits_uhwi_p (idx) ? NULL_TREE : m_data[m_data_cnt + 1];
+      return m_data[m_data_cnt];
+    }
+
+  *data_out = NULL_TREE;
+  if (tree_fits_uhwi_p (idx))
+    {
+      m_data.safe_push (val);
+      m_data.safe_push (NULL_TREE);
+      return val;
+    }
+
+  tree in = make_ssa_name (TREE_TYPE (val));
+  gphi *phi = create_phi_node (in, m_bb);
+  edge e1 = find_edge (m_preheader_bb, m_bb);
+  edge e2 = EDGE_PRED (m_bb, 0);
+  if (e1 == e2)
+    e2 = EDGE_PRED (m_bb, 1);
+  add_phi_arg (phi, val, e1, UNKNOWN_LOCATION);
+  tree out = make_ssa_name (TREE_TYPE (val));
+  add_phi_arg (phi, out, e2, UNKNOWN_LOCATION);
+  m_data.safe_push (in);
+  m_data.safe_push (out);
+  return in;
+}
+
+/* Return VAL cast to TYPE.  If VAL is INTEGER_CST, just
+   convert it without emitting any code, otherwise emit
+   the conversion statement before the current location.  */
+
+tree
+bitint_large_huge::add_cast (tree type, tree val)
+{
+  if (TREE_CODE (val) == INTEGER_CST)
+    return fold_convert (type, val);
+
+  tree lhs = make_ssa_name (type);
+  gimple *g = gimple_build_assign (lhs, NOP_EXPR, val);
+  insert_before (g);
+  return lhs;
+}
+
+/* Helper of handle_stmt method, handle PLUS_EXPR or MINUS_EXPR.  */
+
+tree
+bitint_large_huge::handle_plus_minus (tree_code code, tree rhs1, tree rhs2,
+				      tree idx)
+{
+  tree lhs, data_out, ctype;
+  tree rhs1_type = TREE_TYPE (rhs1);
+  gimple *g;
+  tree data_in = prepare_data_in_out (build_zero_cst (m_limb_type), idx,
+				      &data_out);
+
+  if (optab_handler (code == PLUS_EXPR ? uaddc5_optab : usubc5_optab,
+		     TYPE_MODE (m_limb_type)) != CODE_FOR_nothing)
+    {
+      ctype = build_complex_type (m_limb_type);
+      if (!types_compatible_p (rhs1_type, m_limb_type))
+	{
+	  if (!TYPE_UNSIGNED (rhs1_type))
+	    {
+	      tree type = unsigned_type_for (rhs1_type);
+	      rhs1 = add_cast (type, rhs1);
+	      rhs2 = add_cast (type, rhs2);
+	    }
+	  rhs1 = add_cast (m_limb_type, rhs1);
+	  rhs2 = add_cast (m_limb_type, rhs2);
+	}
+      lhs = make_ssa_name (ctype);
+      g = gimple_build_call_internal (code == PLUS_EXPR
+				      ? IFN_UADDC : IFN_USUBC,
+				      3, rhs1, rhs2, data_in);
+      gimple_call_set_lhs (g, lhs);
+      insert_before (g);
+      if (data_out == NULL_TREE)
+	data_out = make_ssa_name (m_limb_type);
+      g = gimple_build_assign (data_out, IMAGPART_EXPR,
+			       build1 (IMAGPART_EXPR, m_limb_type, lhs));
+      insert_before (g);
+    }
+  else if (types_compatible_p (rhs1_type, m_limb_type))
+    {
+      ctype = build_complex_type (m_limb_type);
+      lhs = make_ssa_name (ctype);
+      g = gimple_build_call_internal (code == PLUS_EXPR
+				      ? IFN_ADD_OVERFLOW : IFN_SUB_OVERFLOW,
+				      2, rhs1, rhs2);
+      gimple_call_set_lhs (g, lhs);
+      insert_before (g);
+      if (data_out == NULL_TREE)
+	data_out = make_ssa_name (m_limb_type);
+      if (!integer_zerop (data_in))
+	{
+	  rhs1 = make_ssa_name (m_limb_type);
+	  g = gimple_build_assign (rhs1, REALPART_EXPR,
+				   build1 (REALPART_EXPR, m_limb_type, lhs));
+	  insert_before (g);
+	  rhs2 = make_ssa_name (m_limb_type);
+	  g = gimple_build_assign (rhs2, IMAGPART_EXPR,
+				   build1 (IMAGPART_EXPR, m_limb_type, lhs));
+	  insert_before (g);
+	  lhs = make_ssa_name (ctype);
+	  g = gimple_build_call_internal (code == PLUS_EXPR
+					  ? IFN_ADD_OVERFLOW
+					  : IFN_SUB_OVERFLOW,
+					  2, rhs1, data_in);
+	  gimple_call_set_lhs (g, lhs);
+	  insert_before (g);
+	  data_in = make_ssa_name (m_limb_type);
+	  g = gimple_build_assign (data_in, IMAGPART_EXPR,
+				   build1 (IMAGPART_EXPR, m_limb_type, lhs));
+	  insert_before (g);
+	  g = gimple_build_assign (data_out, PLUS_EXPR, rhs2, data_in);
+	  insert_before (g);
+	}
+      else
+	{
+	  g = gimple_build_assign (data_out, IMAGPART_EXPR,
+				   build1 (IMAGPART_EXPR, m_limb_type, lhs));
+	  insert_before (g);
+	}
+    }
+  else
+    {
+      tree in = add_cast (rhs1_type, data_in);
+      lhs = make_ssa_name (rhs1_type);
+      g = gimple_build_assign (lhs, code, rhs1, rhs2);
+      insert_before (g);
+      rhs1 = make_ssa_name (rhs1_type);
+      g = gimple_build_assign (rhs1, code, lhs, in);
+      insert_before (g);
+      m_data[m_data_cnt] = NULL_TREE;
+      m_data_cnt += 2;
+      return rhs1;
+    }
+  rhs1 = make_ssa_name (m_limb_type);
+  g = gimple_build_assign (rhs1, REALPART_EXPR,
+			   build1 (REALPART_EXPR, m_limb_type, lhs));
+  insert_before (g);
+  if (!types_compatible_p (rhs1_type, m_limb_type))
+    rhs1 = add_cast (rhs1_type, rhs1);
+  m_data[m_data_cnt] = data_out;
+  m_data_cnt += 2;
+  return rhs1;
+}
+
+/* Helper function for handle_stmt method, handle LSHIFT_EXPR by
+   count in [0, limb_prec - 1] range.  */
+
+tree
+bitint_large_huge::handle_lshift (tree rhs1, tree rhs2, tree idx)
+{
+  unsigned HOST_WIDE_INT cnt = tree_to_uhwi (rhs2);
+  gcc_checking_assert (cnt < (unsigned) limb_prec);
+  if (cnt == 0)
+    return rhs1;
+
+  tree lhs, data_out, rhs1_type = TREE_TYPE (rhs1);
+  gimple *g;
+  tree data_in = prepare_data_in_out (build_zero_cst (m_limb_type), idx,
+				      &data_out);
+
+  if (!integer_zerop (data_in))
+    {
+      lhs = make_ssa_name (m_limb_type);
+      g = gimple_build_assign (lhs, RSHIFT_EXPR, data_in,
+			       build_int_cst (unsigned_type_node,
+					      limb_prec - cnt));
+      insert_before (g);
+      if (!types_compatible_p (rhs1_type, m_limb_type))
+	lhs = add_cast (rhs1_type, lhs);
+      data_in = lhs;
+    }
+  if (types_compatible_p (rhs1_type, m_limb_type))
+    {
+      if (data_out == NULL_TREE)
+	data_out = make_ssa_name (m_limb_type);
+      g = gimple_build_assign (data_out, rhs1);
+      insert_before (g);
+    }
+  if (cnt < (unsigned) TYPE_PRECISION (rhs1_type))
+    {
+      lhs = make_ssa_name (rhs1_type);
+      g = gimple_build_assign (lhs, LSHIFT_EXPR, rhs1, rhs2);
+      insert_before (g);
+      if (!integer_zerop (data_in))
+	{
+	  rhs1 = lhs;
+	  lhs = make_ssa_name (rhs1_type);
+	  g = gimple_build_assign (lhs, BIT_IOR_EXPR, rhs1, data_in);
+	  insert_before (g);
+	}
+    }
+  else
+    lhs = data_in;
+  m_data[m_data_cnt] = data_out;
+  m_data_cnt += 2;
+  return lhs;
+}
+
+/* Helper function for handle_stmt method, handle an integral
+   to integral conversion.  */
+
+tree
+bitint_large_huge::handle_cast (tree lhs_type, tree rhs1, tree idx)
+{
+  tree rhs_type = TREE_TYPE (rhs1);
+  gimple *g;
+  if (TREE_CODE (rhs1) == SSA_NAME
+      && TREE_CODE (lhs_type) == BITINT_TYPE
+      && TREE_CODE (rhs_type) == BITINT_TYPE
+      && bitint_precision_kind (lhs_type) >= bitint_prec_large
+      && bitint_precision_kind (rhs_type) >= bitint_prec_large)
+    {
+      if (TYPE_PRECISION (rhs_type) >= TYPE_PRECISION (lhs_type)
+	  /* If lhs has bigger precision than rhs, we can use
+	     the simple case only if there is a guarantee that
+	     the most significant limb is handled in straight
+	     line code.  If m_var_msb (on left shifts) or
+	     if m_upwards_2limb * limb_prec is equal to
+	     lhs precision that is not the case.  */
+	  || (!m_var_msb
+	      && tree_int_cst_equal (TYPE_SIZE (rhs_type),
+				     TYPE_SIZE (lhs_type))
+	      && (!m_upwards_2limb
+		  || (m_upwards_2limb * limb_prec
+		      < TYPE_PRECISION (lhs_type)))))
+	{
+	  rhs1 = handle_operand (rhs1, idx);
+	  if (tree_fits_uhwi_p (idx))
+	    {
+	      tree type = limb_access_type (lhs_type, idx);
+	      if (!types_compatible_p (type, TREE_TYPE (rhs1)))
+		rhs1 = add_cast (type, rhs1);
+	    }
+	  return rhs1;
+	}
+      tree t;
+      /* Indexes lower than this don't need any special processing.  */
+      unsigned low = ((unsigned) TYPE_PRECISION (rhs_type)
+		      - !TYPE_UNSIGNED (rhs_type)) / limb_prec;
+      /* Indexes >= than this always contain an extension.  */
+      unsigned high = CEIL ((unsigned) TYPE_PRECISION (rhs_type), limb_prec);
+      bool save_first = m_first;
+      if (m_first)
+	{
+	  m_data.safe_push (NULL_TREE);
+	  m_data.safe_push (NULL_TREE);
+	  m_data.safe_push (NULL_TREE);
+	  if (TYPE_UNSIGNED (rhs_type))
+	    /* No need to keep state between iterations.  */
+	    ;
+	  else if (m_upwards && !m_upwards_2limb)
+	    /* We need to keep state between iterations, but
+	       not within any loop, everything is straight line
+	       code with only increasing indexes.  */
+	    ;
+	  else if (!m_upwards_2limb)
+	    {
+	      unsigned save_data_cnt = m_data_cnt;
+	      gimple_stmt_iterator save_gsi = m_gsi;
+	      m_gsi = m_init_gsi;
+	      if (gsi_end_p (m_gsi))
+		m_gsi = gsi_after_labels (gsi_bb (m_gsi));
+	      else
+		gsi_next (&m_gsi);
+	      m_data_cnt = save_data_cnt + 3;
+	      t = handle_operand (rhs1, size_int (low));
+	      m_first = false;
+	      m_data[save_data_cnt + 2]
+		= build_int_cst (NULL_TREE, m_data_cnt);
+	      m_data_cnt = save_data_cnt;
+	      t = add_cast (signed_type_for (m_limb_type), t);
+	      tree lpm1 = build_int_cst (unsigned_type_node, limb_prec - 1);
+	      tree n = make_ssa_name (TREE_TYPE (t));
+	      g = gimple_build_assign (n, RSHIFT_EXPR, t, lpm1);
+	      insert_before (g);
+	      m_data[save_data_cnt + 1] = add_cast (m_limb_type, n);
+	      m_gsi = save_gsi;
+	    }
+	  else if (m_upwards_2limb * limb_prec < TYPE_PRECISION (rhs_type))
+	    /* We need to keep state between iterations, but
+	       fortunately not within the loop, only afterwards.  */
+	    ;
+	  else
+	    {
+	      tree out;
+	      m_data.truncate (m_data_cnt);
+	      prepare_data_in_out (build_zero_cst (m_limb_type), idx, &out);
+	      m_data.safe_push (NULL_TREE);
+	    }
+	}
+
+      unsigned save_data_cnt = m_data_cnt;
+      m_data_cnt += 3;
+      if (!tree_fits_uhwi_p (idx))
+	{
+	  if (m_upwards_2limb
+	      && (m_upwards_2limb * limb_prec
+		  <= ((unsigned) TYPE_PRECISION (rhs_type)
+		      - !TYPE_UNSIGNED (rhs_type))))
+	    {
+	      rhs1 = handle_operand (rhs1, idx);
+	      if (m_first)
+		m_data[save_data_cnt + 2]
+		  = build_int_cst (NULL_TREE, m_data_cnt);
+	      m_first = save_first;
+	      return rhs1;
+	    }
+	  bool single_comparison
+	    = low == high || (m_upwards_2limb && (low & 1) == m_first);
+	  g = gimple_build_cond (single_comparison ? LT_EXPR : LE_EXPR,
+				 idx, size_int (low), NULL_TREE, NULL_TREE);
+	  edge edge_true_true, edge_true_false, edge_false;
+	  if_then_if_then_else (g, (single_comparison ? NULL
+				    : gimple_build_cond (EQ_EXPR, idx,
+							 size_int (low),
+							 NULL_TREE,
+							 NULL_TREE)),
+				profile_probability::likely (),
+				profile_probability::unlikely (),
+				edge_true_true, edge_true_false, edge_false);
+	  bool save_cast_conditional = m_cast_conditional;
+	  m_cast_conditional = true;
+	  m_bitfld_load = 0;
+	  tree t1 = handle_operand (rhs1, idx), t2 = NULL_TREE;
+	  if (m_first)
+	    m_data[save_data_cnt + 2]
+	      = build_int_cst (NULL_TREE, m_data_cnt);
+	  tree ext = NULL_TREE;
+	  tree bitfld = NULL_TREE;
+	  if (!single_comparison)
+	    {
+	      m_gsi = gsi_after_labels (edge_true_true->src);
+	      m_first = false;
+	      m_data_cnt = save_data_cnt + 3;
+	      if (m_bitfld_load)
+		{
+		  bitfld = m_data[m_bitfld_load];
+		  m_data[m_bitfld_load] = m_data[m_bitfld_load + 2];
+		  m_bitfld_load = 0;
+		}
+	      t2 = handle_operand (rhs1, size_int (low));
+	      if (!useless_type_conversion_p (m_limb_type, TREE_TYPE (t2)))
+		t2 = add_cast (m_limb_type, t2);
+	      if (!TYPE_UNSIGNED (rhs_type) && m_upwards_2limb)
+		{
+		  ext = add_cast (signed_type_for (m_limb_type), t2);
+		  tree lpm1 = build_int_cst (unsigned_type_node,
+					     limb_prec - 1);
+		  tree n = make_ssa_name (TREE_TYPE (ext));
+		  g = gimple_build_assign (n, RSHIFT_EXPR, ext, lpm1);
+		  insert_before (g);
+		  ext = add_cast (m_limb_type, n);
+		}
+	    }
+	  tree t3;
+	  if (TYPE_UNSIGNED (rhs_type))
+	    t3 = build_zero_cst (m_limb_type);
+	  else if (m_upwards_2limb && (save_first || ext != NULL_TREE))
+	    t3 = m_data[save_data_cnt];
+	  else
+	    t3 = m_data[save_data_cnt + 1];
+	  m_gsi = gsi_after_labels (edge_true_false->dest);
+	  t = make_ssa_name (m_limb_type);
+	  gphi *phi = create_phi_node (t, edge_true_false->dest);
+	  add_phi_arg (phi, t1, edge_true_false, UNKNOWN_LOCATION);
+	  add_phi_arg (phi, t3, edge_false, UNKNOWN_LOCATION);
+	  if (edge_true_true)
+	    add_phi_arg (phi, t2, edge_true_true, UNKNOWN_LOCATION);
+	  if (ext)
+	    {
+	      tree t4 = make_ssa_name (m_limb_type);
+	      phi = create_phi_node (t4, edge_true_false->dest);
+	      add_phi_arg (phi, build_zero_cst (m_limb_type), edge_true_false,
+			   UNKNOWN_LOCATION);
+	      add_phi_arg (phi, m_data[save_data_cnt], edge_false,
+			   UNKNOWN_LOCATION);
+	      add_phi_arg (phi, ext, edge_true_true, UNKNOWN_LOCATION);
+	      g = gimple_build_assign (m_data[save_data_cnt + 1], t4);
+	      insert_before (g);
+	    }
+	  if (m_bitfld_load)
+	    {
+	      tree t4;
+	      if (!m_first)
+		t4 = m_data[m_bitfld_load + 1];
+	      else
+		t4 = make_ssa_name (m_limb_type);
+	      phi = create_phi_node (t4, edge_true_false->dest);
+	      add_phi_arg (phi,
+			   edge_true_true ? bitfld : m_data[m_bitfld_load],
+			   edge_true_false, UNKNOWN_LOCATION);
+	      add_phi_arg (phi, m_data[m_bitfld_load + 2],
+			   edge_false, UNKNOWN_LOCATION);
+	      if (edge_true_true)
+		add_phi_arg (phi, m_data[m_bitfld_load], edge_true_true,
+			     UNKNOWN_LOCATION);
+	      m_data[m_bitfld_load] = t4;
+	      m_data[m_bitfld_load + 2] = t4;
+	      m_bitfld_load = 0;
+	    }
+	  m_cast_conditional = save_cast_conditional;
+	  m_first = save_first;
+	  return t;
+	}
+      else
+	{
+	  if (tree_to_uhwi (idx) < low)
+	    {
+	      t = handle_operand (rhs1, idx);
+	      if (m_first)
+		m_data[save_data_cnt + 2]
+		  = build_int_cst (NULL_TREE, m_data_cnt);
+	    }
+	  else if (tree_to_uhwi (idx) < high)
+	    {
+	      t = handle_operand (rhs1, size_int (low));
+	      if (m_first)
+		m_data[save_data_cnt + 2]
+		  = build_int_cst (NULL_TREE, m_data_cnt);
+	      if (!useless_type_conversion_p (m_limb_type, TREE_TYPE (t)))
+		t = add_cast (m_limb_type, t);
+	      tree ext = NULL_TREE;
+	      if (!TYPE_UNSIGNED (rhs_type) && m_upwards)
+		{
+		  ext = add_cast (signed_type_for (m_limb_type), t);
+		  tree lpm1 = build_int_cst (unsigned_type_node,
+					     limb_prec - 1);
+		  tree n = make_ssa_name (TREE_TYPE (ext));
+		  g = gimple_build_assign (n, RSHIFT_EXPR, ext, lpm1);
+		  insert_before (g);
+		  ext = add_cast (m_limb_type, n);
+		  m_data[save_data_cnt + 1] = ext;
+		}
+	    }
+	  else
+	    {
+	      if (TYPE_UNSIGNED (rhs_type) && m_first)
+		{
+		  handle_operand (rhs1, size_zero_node);
+		  m_data[save_data_cnt + 2]
+		    = build_int_cst (NULL_TREE, m_data_cnt);
+		}
+	      else
+		m_data_cnt = tree_to_uhwi (m_data[save_data_cnt + 2]);
+	      if (TYPE_UNSIGNED (rhs_type))
+		t = build_zero_cst (m_limb_type);
+	      else
+		t = m_data[save_data_cnt + 1];
+	    }
+	  tree type = limb_access_type (lhs_type, idx);
+	  if (!useless_type_conversion_p (type, m_limb_type))
+	    t = add_cast (type, t);
+	  m_first = save_first;
+	  return t;
+	}
+    }
+  else if (TREE_CODE (lhs_type) == BITINT_TYPE
+	   && bitint_precision_kind (lhs_type) >= bitint_prec_large
+	   && INTEGRAL_TYPE_P (rhs_type))
+    {
+      /* Add support for 3 or more limbs filled in from normal integral
+	 type if this assert fails.  If no target chooses limb mode smaller
+	 than half of largest supported normal integral type, this will not
+	 be needed.  */
+      gcc_assert (TYPE_PRECISION (rhs_type) <= 2 * limb_prec);
+      tree r1 = NULL_TREE, r2 = NULL_TREE, rext = NULL_TREE;
+      if (m_first)
+	{
+	  gimple_stmt_iterator save_gsi = m_gsi;
+	  m_gsi = m_init_gsi;
+	  if (gsi_end_p (m_gsi))
+	    m_gsi = gsi_after_labels (gsi_bb (m_gsi));
+	  else
+	    gsi_next (&m_gsi);
+	  if (TREE_CODE (rhs_type) == BITINT_TYPE
+	      && bitint_precision_kind (rhs_type) == bitint_prec_middle)
+	    {
+	      tree type = NULL_TREE;
+	      rhs1 = maybe_cast_middle_bitint (&m_gsi, rhs1, type);
+	      rhs_type = TREE_TYPE (rhs1);
+	    }
+	  r1 = rhs1;
+	  if (!useless_type_conversion_p (m_limb_type, TREE_TYPE (rhs1)))
+	    r1 = add_cast (m_limb_type, rhs1);
+	  if (TYPE_PRECISION (rhs_type) > limb_prec)
+	    {
+	      g = gimple_build_assign (make_ssa_name (rhs_type),
+				       RSHIFT_EXPR, rhs1,
+				       build_int_cst (unsigned_type_node,
+						      limb_prec));
+	      insert_before (g);
+	      r2 = add_cast (m_limb_type, gimple_assign_lhs (g));
+	    }
+	  if (TYPE_UNSIGNED (rhs_type))
+	    rext = build_zero_cst (m_limb_type);
+	  else
+	    {
+	      rext = add_cast (signed_type_for (m_limb_type), r2 ? r2 : r1);
+	      g = gimple_build_assign (make_ssa_name (TREE_TYPE (rext)),
+				       RSHIFT_EXPR, rext,
+				       build_int_cst (unsigned_type_node,
+						      limb_prec - 1));
+	      insert_before (g);
+	      rext = add_cast (m_limb_type, gimple_assign_lhs (g));
+	    }
+	  m_gsi = save_gsi;
+	}
+      tree t;
+      if (m_upwards_2limb)
+	{
+	  if (m_first)
+	    {
+	      tree out1, out2;
+	      prepare_data_in_out (r1, idx, &out1);
+	      g = gimple_build_assign (m_data[m_data_cnt + 1], rext);
+	      insert_before (g);
+	      if (TYPE_PRECISION (rhs_type) > limb_prec)
+		{
+		  prepare_data_in_out (r2, idx, &out2);
+		  g = gimple_build_assign (m_data[m_data_cnt + 3], rext);
+		  insert_before (g);
+		  m_data.pop ();
+		  t = m_data.pop ();
+		  m_data[m_data_cnt + 1] = t;
+		}
+	      else
+		m_data[m_data_cnt + 1] = rext;
+	      m_data.safe_push (rext);
+	      t = m_data[m_data_cnt];
+	    }
+	  else if (!tree_fits_uhwi_p (idx))
+	    t = m_data[m_data_cnt + 1];
+	  else
+	    {
+	      tree type = limb_access_type (lhs_type, idx);
+	      t = m_data[m_data_cnt + 2];
+	      if (!useless_type_conversion_p (type, m_limb_type))
+		t = add_cast (type, t);
+	    }
+	  m_data_cnt += 3;
+	  return t;
+	}
+      else if (m_first)
+	{
+	  m_data.safe_push (r1);
+	  m_data.safe_push (r2);
+	  m_data.safe_push (rext);
+	}
+      if (tree_fits_uhwi_p (idx))
+	{
+	  tree type = limb_access_type (lhs_type, idx);
+	  if (integer_zerop (idx))
+	    t = m_data[m_data_cnt];
+	  else if (TYPE_PRECISION (rhs_type) > limb_prec
+		   && integer_onep (idx))
+	    t = m_data[m_data_cnt + 1];
+	  else
+	    t = m_data[m_data_cnt + 2];
+	  if (!useless_type_conversion_p (type, m_limb_type))
+	    t = add_cast (type, t);
+	  m_data_cnt += 3;
+	  return t;
+	}
+      g = gimple_build_cond (NE_EXPR, idx, size_zero_node,
+			     NULL_TREE, NULL_TREE);
+      edge e2, e3, e4 = NULL;
+      if_then (g, profile_probability::likely (), e2, e3);
+      if (m_data[m_data_cnt + 1])
+	{
+	  g = gimple_build_cond (EQ_EXPR, idx, size_one_node,
+				 NULL_TREE, NULL_TREE);
+	  insert_before (g);
+	  edge e5 = split_block (gsi_bb (m_gsi), g);
+	  e4 = make_edge (e5->src, e2->dest, EDGE_TRUE_VALUE);
+	  e2 = find_edge (e5->dest, e2->dest);
+	  e4->probability = profile_probability::unlikely ();
+	  e5->flags = EDGE_FALSE_VALUE;
+	  e5->probability = e4->probability.invert ();
+	}
+      m_gsi = gsi_after_labels (e2->dest);
+      t = make_ssa_name (m_limb_type);
+      gphi *phi = create_phi_node (t, e2->dest);
+      add_phi_arg (phi, m_data[m_data_cnt + 2], e2, UNKNOWN_LOCATION);
+      add_phi_arg (phi, m_data[m_data_cnt], e3, UNKNOWN_LOCATION);
+      if (e4)
+	add_phi_arg (phi, m_data[m_data_cnt + 1], e4, UNKNOWN_LOCATION);
+      m_data_cnt += 3;
+      return t;
+    }
+  return NULL_TREE;
+}
+
+/* Helper function for handle_stmt method, handle a load from memory.  */
+
+tree
+bitint_large_huge::handle_load (gimple *stmt, tree idx)
+{
+  tree rhs1 = gimple_assign_rhs1 (stmt);
+  tree rhs_type = TREE_TYPE (rhs1);
+  bool eh = stmt_ends_bb_p (stmt);
+  edge eh_edge = NULL;
+  gimple *g;
+
+  if (eh)
+    {
+      edge_iterator ei;
+      basic_block bb = gimple_bb (stmt);
+
+      FOR_EACH_EDGE (eh_edge, ei, bb->succs)
+	if (eh_edge->flags & EDGE_EH)
+	    break;
+    }
+
+  if (TREE_CODE (rhs1) == COMPONENT_REF
+      && DECL_BIT_FIELD_TYPE (TREE_OPERAND (rhs1, 1)))
+    {
+      tree fld = TREE_OPERAND (rhs1, 1);
+      /* For little-endian, we can allow as inputs bit-fields
+	 which start at a limb boundary.  */
+      gcc_assert (tree_fits_uhwi_p (DECL_FIELD_BIT_OFFSET (fld)));
+      if (DECL_OFFSET_ALIGN (fld) >= TYPE_ALIGN (TREE_TYPE (rhs1))
+	  && (tree_to_uhwi (DECL_FIELD_BIT_OFFSET (fld)) % limb_prec) == 0)
+	goto normal_load;
+      /* Even if DECL_FIELD_BIT_OFFSET (fld) is a multiple of UNITS_PER_BIT,
+	 handle it normally for now.  */
+      if ((tree_to_uhwi (DECL_FIELD_BIT_OFFSET (fld)) % BITS_PER_UNIT) == 0)
+	goto normal_load;
+      tree repr = DECL_BIT_FIELD_REPRESENTATIVE (fld);
+      poly_int64 bitoffset;
+      poly_uint64 field_offset, repr_offset;
+      bool var_field_off = false;
+      if (poly_int_tree_p (DECL_FIELD_OFFSET (fld), &field_offset)
+	  && poly_int_tree_p (DECL_FIELD_OFFSET (repr), &repr_offset))
+	bitoffset = (field_offset - repr_offset) * BITS_PER_UNIT;
+      else
+	{
+	  bitoffset = 0;
+	  var_field_off = true;
+	}
+      bitoffset += (tree_to_uhwi (DECL_FIELD_BIT_OFFSET (fld))
+		    - tree_to_uhwi (DECL_FIELD_BIT_OFFSET (repr)));
+      tree nrhs1 = build3 (COMPONENT_REF, TREE_TYPE (repr),
+			   TREE_OPERAND (rhs1, 0), repr,
+			   var_field_off ? TREE_OPERAND (rhs1, 2) : NULL_TREE);
+      HOST_WIDE_INT bo = bitoffset.to_constant ();
+      unsigned bo_idx = (unsigned HOST_WIDE_INT) bo / limb_prec;
+      unsigned bo_bit = (unsigned HOST_WIDE_INT) bo % limb_prec;
+      if (m_first)
+	{
+	  if (m_upwards)
+	    {
+	      gimple_stmt_iterator save_gsi = m_gsi;
+	      m_gsi = m_init_gsi;
+	      if (gsi_end_p (m_gsi))
+		m_gsi = gsi_after_labels (gsi_bb (m_gsi));
+	      else
+		gsi_next (&m_gsi);
+	      tree t = limb_access (rhs_type, nrhs1, size_int (bo_idx), true);
+	      tree iv = make_ssa_name (m_limb_type);
+	      g = gimple_build_assign (iv, t);
+	      insert_before (g);
+	      if (eh)
+		{
+		  maybe_duplicate_eh_stmt (g, stmt);
+		  if (eh_edge)
+		    {
+		      edge e = split_block (gsi_bb (m_gsi), g);
+		      make_edge (e->src, eh_edge->dest, EDGE_EH)->probability
+			= profile_probability::very_unlikely ();
+		      m_init_gsi.bb = e->dest;
+		    }
+		}
+	      m_gsi = save_gsi;
+	      tree out;
+	      prepare_data_in_out (iv, idx, &out);
+	      out = m_data[m_data_cnt];
+	      m_data.safe_push (out);
+	    }
+	  else
+	    {
+	      m_data.safe_push (NULL_TREE);
+	      m_data.safe_push (NULL_TREE);
+	      m_data.safe_push (NULL_TREE);
+	    }
+	}
+
+      tree nidx0 = NULL_TREE, nidx1;
+      tree iv = m_data[m_data_cnt];
+      if (m_cast_conditional && iv)
+	{
+	  gcc_assert (!m_bitfld_load);
+	  m_bitfld_load = m_data_cnt;
+	}
+      if (tree_fits_uhwi_p (idx))
+	{
+	  unsigned prec = TYPE_PRECISION (rhs_type);
+	  unsigned HOST_WIDE_INT i = tree_to_uhwi (idx);
+	  gcc_assert (i * limb_prec < prec);
+	  nidx1 = size_int (i + bo_idx + 1);
+	  if ((i + 1) * limb_prec > prec)
+	    {
+	      prec %= limb_prec;
+	      if (prec + bo_bit <= (unsigned) limb_prec)
+		nidx1 = NULL_TREE;
+	    }
+	  if (!iv)
+	    nidx0 = size_int (i + bo_idx);
+	}
+      else
+	{
+	  if (!iv)
+	    {
+	      if (bo_idx == 0)
+		nidx0 = idx;
+	      else
+		{
+		  nidx0 = make_ssa_name (sizetype);
+		  g = gimple_build_assign (nidx0, PLUS_EXPR, idx,
+					   size_int (bo_idx));
+		  insert_before (g);
+		}
+	    }
+	  nidx1 = make_ssa_name (sizetype);
+	  g = gimple_build_assign (nidx1, PLUS_EXPR, idx,
+				   size_int (bo_idx + 1));
+	  insert_before (g);
+	}
+
+      tree iv2 = NULL_TREE;
+      if (nidx0)
+	{
+	  tree t = limb_access (rhs_type, nrhs1, nidx0, true);
+	  iv = make_ssa_name (m_limb_type);
+	  g = gimple_build_assign (iv, t);
+	  insert_before (g);
+	  gcc_assert (!eh);
+	}
+      if (nidx1)
+	{
+	  bool conditional = m_var_msb && !tree_fits_uhwi_p (idx);
+	  unsigned prec = TYPE_PRECISION (rhs_type);
+	  if (conditional)
+	    {
+	      if ((prec % limb_prec) == 0
+		  || ((prec % limb_prec) + bo_bit > (unsigned) limb_prec))
+		conditional = false;
+	    }
+	  edge edge_true = NULL, edge_false = NULL;
+	  if (conditional)
+	    {
+	      g = gimple_build_cond (NE_EXPR, idx,
+				     size_int (prec / limb_prec),
+				     NULL_TREE, NULL_TREE);
+	      if_then (g, profile_probability::likely (),
+		       edge_true, edge_false);
+	    }
+	  tree t = limb_access (rhs_type, nrhs1, nidx1, true);
+	  if (m_upwards_2limb
+	      && !m_first
+	      && !m_bitfld_load
+	      && !tree_fits_uhwi_p (idx))
+	    iv2 = m_data[m_data_cnt + 1];
+	  else
+	    iv2 = make_ssa_name (m_limb_type);
+	  g = gimple_build_assign (iv2, t);
+	  insert_before (g);
+	  if (eh)
+	    {
+	      maybe_duplicate_eh_stmt (g, stmt);
+	      if (eh_edge)
+		{
+		  edge e = split_block (gsi_bb (m_gsi), g);
+		  m_gsi = gsi_after_labels (e->dest);
+		  make_edge (e->src, eh_edge->dest, EDGE_EH)->probability
+		    = profile_probability::very_unlikely ();
+		}
+	    }
+	  if (conditional)
+	    {
+	      tree iv3 = make_ssa_name (m_limb_type);
+	      if (eh)
+		edge_true = find_edge (gsi_bb (m_gsi), edge_false->dest);
+	      gphi *phi = create_phi_node (iv3, edge_true->dest);
+	      add_phi_arg (phi, iv2, edge_true, UNKNOWN_LOCATION);
+	      add_phi_arg (phi, build_zero_cst (m_limb_type),
+			   edge_false, UNKNOWN_LOCATION);
+	      m_gsi = gsi_after_labels (edge_true->dest);
+	    }
+	}
+      g = gimple_build_assign (make_ssa_name (m_limb_type), RSHIFT_EXPR,
+			       iv, build_int_cst (unsigned_type_node, bo_bit));
+      insert_before (g);
+      iv = gimple_assign_lhs (g);
+      if (iv2)
+	{
+	  g = gimple_build_assign (make_ssa_name (m_limb_type), LSHIFT_EXPR,
+				   iv2, build_int_cst (unsigned_type_node,
+						       limb_prec - bo_bit));
+	  insert_before (g);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type), BIT_IOR_EXPR,
+				   gimple_assign_lhs (g), iv);
+	  insert_before (g);
+	  iv = gimple_assign_lhs (g);
+	  if (m_data[m_data_cnt])
+	    m_data[m_data_cnt] = iv2;
+	}
+      if (tree_fits_uhwi_p (idx))
+	{
+	  tree atype = limb_access_type (rhs_type, idx);
+	  if (!useless_type_conversion_p (atype, TREE_TYPE (iv)))
+	    iv = add_cast (atype, iv);
+	}
+      m_data_cnt += 3;
+      return iv;
+    }
+
+normal_load:
+  /* Use write_p = true for loads with EH edges to make
+     sure limb_access doesn't add a cast as separate
+     statement after it.  */
+  rhs1 = limb_access (rhs_type, rhs1, idx, eh);
+  tree ret = make_ssa_name (TREE_TYPE (rhs1));
+  g = gimple_build_assign (ret, rhs1);
+  insert_before (g);
+  if (eh)
+    {
+      maybe_duplicate_eh_stmt (g, stmt);
+      if (eh_edge)
+	{
+	  edge e = split_block (gsi_bb (m_gsi), g);
+	  m_gsi = gsi_after_labels (e->dest);
+	  make_edge (e->src, eh_edge->dest, EDGE_EH)->probability
+	    = profile_probability::very_unlikely ();
+	}
+      if (tree_fits_uhwi_p (idx))
+	{
+	  tree atype = limb_access_type (rhs_type, idx);
+	  if (!useless_type_conversion_p (atype, TREE_TYPE (rhs1)))
+	    ret = add_cast (atype, ret);
+	}
+    }
+  return ret;
+}
+
+/* Return a limb IDX from a mergeable statement STMT.  */
+
+tree
+bitint_large_huge::handle_stmt (gimple *stmt, tree idx)
+{
+  tree lhs, rhs1, rhs2 = NULL_TREE;
+  gimple *g;
+  switch (gimple_code (stmt))
+    {
+    case GIMPLE_ASSIGN:
+      if (gimple_assign_load_p (stmt))
+	return handle_load (stmt, idx);
+      switch (gimple_assign_rhs_code (stmt))
+	{
+	case BIT_AND_EXPR:
+	case BIT_IOR_EXPR:
+	case BIT_XOR_EXPR:
+	  rhs2 = handle_operand (gimple_assign_rhs2 (stmt), idx);
+	  /* FALLTHRU */
+	case BIT_NOT_EXPR:
+	  rhs1 = handle_operand (gimple_assign_rhs1 (stmt), idx);
+	  lhs = make_ssa_name (TREE_TYPE (rhs1));
+	  g = gimple_build_assign (lhs, gimple_assign_rhs_code (stmt),
+				   rhs1, rhs2);
+	  insert_before (g);
+	  return lhs;
+	case PLUS_EXPR:
+	case MINUS_EXPR:
+	  rhs1 = handle_operand (gimple_assign_rhs1 (stmt), idx);
+	  rhs2 = handle_operand (gimple_assign_rhs2 (stmt), idx);
+	  return handle_plus_minus (gimple_assign_rhs_code (stmt),
+				    rhs1, rhs2, idx);
+	case NEGATE_EXPR:
+	  rhs2 = handle_operand (gimple_assign_rhs1 (stmt), idx);
+	  rhs1 = build_zero_cst (TREE_TYPE (rhs2));
+	  return handle_plus_minus (MINUS_EXPR, rhs1, rhs2, idx);
+	case LSHIFT_EXPR:
+	  return handle_lshift (handle_operand (gimple_assign_rhs1 (stmt),
+						idx),
+				gimple_assign_rhs2 (stmt), idx);
+	case SSA_NAME:
+	case INTEGER_CST:
+	  return handle_operand (gimple_assign_rhs1 (stmt), idx);
+	CASE_CONVERT:
+	case VIEW_CONVERT_EXPR:
+	  return handle_cast (TREE_TYPE (gimple_assign_lhs (stmt)),
+			      gimple_assign_rhs1 (stmt), idx);
+	default:
+	  break;
+	}
+      break;
+    default:
+      break;
+    }
+  gcc_unreachable ();
+}
+
+/* Return minimum precision of OP at STMT.
+   Positive value is minimum precision above which all bits
+   are zero, negative means all bits above negation of the
+   value are copies of the sign bit.  */
+
+static int
+range_to_prec (tree op, gimple *stmt)
+{
+  int_range_max r;
+  wide_int w;
+  tree type = TREE_TYPE (op);
+  unsigned int prec = TYPE_PRECISION (type);
+
+  if (!optimize
+      || !get_range_query (cfun)->range_of_expr (r, op, stmt))
+    {
+      if (TYPE_UNSIGNED (type))
+	return prec;
+      else
+	return -prec;
+    }
+
+  if (!TYPE_UNSIGNED (TREE_TYPE (op)))
+    {
+      w = r.lower_bound ();
+      if (wi::neg_p (w))
+	{
+	  int min_prec1 = wi::min_precision (w, SIGNED);
+	  w = r.upper_bound ();
+	  int min_prec2 = wi::min_precision (w, SIGNED);
+	  int min_prec = MAX (min_prec1, min_prec2);
+	  return MIN (-min_prec, -2);
+	}
+    }
+
+  w = r.upper_bound ();
+  int min_prec = wi::min_precision (w, UNSIGNED);
+  return MAX (min_prec, 1);
+}
+
+/* Return address of the first limb of OP and write into *PREC
+   its precision.  If positive, the operand is zero extended
+   from that precision, if it is negative, the operand is sign-extended
+   from -*PREC.  If PREC_STORED is NULL, it is the toplevel call,
+   otherwise *PREC_STORED is prec from the innermost call without
+   range optimizations.  */
+
+tree
+bitint_large_huge::handle_operand_addr (tree op, gimple *stmt,
+					int *prec_stored, int *prec)
+{
+  wide_int w;
+  location_t loc_save = m_loc;
+  if ((TREE_CODE (TREE_TYPE (op)) != BITINT_TYPE
+       || bitint_precision_kind (TREE_TYPE (op)) < bitint_prec_large)
+      && TREE_CODE (op) != INTEGER_CST)
+    {
+    do_int:
+      *prec = range_to_prec (op, stmt);
+      bitint_prec_kind kind = bitint_prec_small;
+      gcc_assert (INTEGRAL_TYPE_P (TREE_TYPE (op)));
+      if (TREE_CODE (TREE_TYPE (op)) == BITINT_TYPE)
+	kind = bitint_precision_kind (TREE_TYPE (op));
+      if (kind == bitint_prec_middle)
+	{
+	  tree type = NULL_TREE;
+	  op = maybe_cast_middle_bitint (&m_gsi, op, type);
+	}
+      tree op_type = TREE_TYPE (op);
+      unsigned HOST_WIDE_INT nelts
+	= CEIL (TYPE_PRECISION (op_type), limb_prec);
+      /* Add support for 3 or more limbs filled in from normal
+	 integral type if this assert fails.  If no target chooses
+	 limb mode smaller than half of largest supported normal
+	 integral type, this will not be needed.  */
+      gcc_assert (nelts <= 2);
+      if (prec_stored)
+	*prec_stored = (TYPE_UNSIGNED (op_type)
+			? TYPE_PRECISION (op_type)
+			: -TYPE_PRECISION (op_type));
+      if (*prec <= limb_prec && *prec >= -limb_prec)
+	{
+	  nelts = 1;
+	  if (prec_stored)
+	    {
+	      if (TYPE_UNSIGNED (op_type))
+		{
+		  if (*prec_stored > limb_prec)
+		    *prec_stored = limb_prec;
+		}
+	      else if (*prec_stored < -limb_prec)
+		*prec_stored = -limb_prec;
+	    }
+	}
+      tree atype = build_array_type_nelts (m_limb_type, nelts);
+      tree var = create_tmp_var (atype);
+      tree t1 = op;
+      if (!useless_type_conversion_p (m_limb_type, op_type))
+	t1 = add_cast (m_limb_type, t1);
+      tree v = build4 (ARRAY_REF, m_limb_type, var, size_zero_node,
+		       NULL_TREE, NULL_TREE);
+      gimple *g = gimple_build_assign (v, t1);
+      insert_before (g);
+      if (nelts > 1)
+	{
+	  tree lp = build_int_cst (unsigned_type_node, limb_prec);
+	  g = gimple_build_assign (make_ssa_name (op_type),
+				   RSHIFT_EXPR, op, lp);
+	  insert_before (g);
+	  tree t2 = gimple_assign_lhs (g);
+	  t2 = add_cast (m_limb_type, t2);
+	  v = build4 (ARRAY_REF, m_limb_type, var, size_one_node,
+		      NULL_TREE, NULL_TREE);
+	  g = gimple_build_assign (v, t2);
+	  insert_before (g);
+	}
+      tree ret = build_fold_addr_expr (var);
+      if (!stmt_ends_bb_p (gsi_stmt (m_gsi)))
+	{
+	  tree clobber = build_clobber (atype, CLOBBER_EOL);
+	  g = gimple_build_assign (var, clobber);
+	  gsi_insert_after (&m_gsi, g, GSI_SAME_STMT);
+	}
+      m_loc = loc_save;
+      return ret;
+    }
+  switch (TREE_CODE (op))
+    {
+    case SSA_NAME:
+      if (m_names == NULL
+	  || !bitmap_bit_p (m_names, SSA_NAME_VERSION (op)))
+	{
+	  gimple *g = SSA_NAME_DEF_STMT (op);
+	  tree ret;
+	  m_loc = gimple_location (g);
+	  if (gimple_assign_load_p (g))
+	    {
+	      *prec = range_to_prec (op, NULL);
+	      if (prec_stored)
+		*prec_stored = (TYPE_UNSIGNED (TREE_TYPE (op))
+				? TYPE_PRECISION (TREE_TYPE (op))
+				: -TYPE_PRECISION (TREE_TYPE (op)));
+	      ret = build_fold_addr_expr (gimple_assign_rhs1 (g));
+	      ret = force_gimple_operand_gsi (&m_gsi, ret, true,
+					      NULL_TREE, true, GSI_SAME_STMT);
+	    }
+	  else if (gimple_code (g) == GIMPLE_NOP)
+	    {
+	      tree var = create_tmp_var (m_limb_type);
+	      TREE_ADDRESSABLE (var) = 1;
+	      ret = build_fold_addr_expr (var);
+	      if (!stmt_ends_bb_p (gsi_stmt (m_gsi)))
+		{
+		  tree clobber = build_clobber (m_limb_type, CLOBBER_EOL);
+		  g = gimple_build_assign (var, clobber);
+		  gsi_insert_after (&m_gsi, g, GSI_SAME_STMT);
+		}
+	    }
+	  else
+	    {
+	      gcc_assert (gimple_assign_cast_p (g));
+	      tree rhs1 = gimple_assign_rhs1 (g);
+	      bitint_prec_kind kind = bitint_prec_small;
+	      gcc_assert (INTEGRAL_TYPE_P (TREE_TYPE (rhs1)));
+	      if (TREE_CODE (TREE_TYPE (rhs1)) == BITINT_TYPE)
+		kind = bitint_precision_kind (TREE_TYPE (rhs1));
+	      if (kind >= bitint_prec_large)
+		{
+		  tree lhs_type = TREE_TYPE (op);
+		  tree rhs_type = TREE_TYPE (rhs1);
+		  int prec_stored_val = 0;
+		  ret = handle_operand_addr (rhs1, g, &prec_stored_val, prec);
+		  if (TYPE_PRECISION (lhs_type) > TYPE_PRECISION (rhs_type))
+		    {
+		      if (TYPE_UNSIGNED (lhs_type)
+			  && !TYPE_UNSIGNED (rhs_type))
+			gcc_assert (*prec >= 0 || prec_stored == NULL);
+		    }
+		  else
+		    {
+		      if (*prec > 0 && *prec < TYPE_PRECISION (lhs_type))
+			;
+		      else if (TYPE_UNSIGNED (lhs_type))
+			{
+			  gcc_assert (*prec > 0
+				      || prec_stored_val > 0
+				      || (-prec_stored_val
+					  >= TYPE_PRECISION (lhs_type)));
+			  *prec = TYPE_PRECISION (lhs_type);
+			}
+		      else if (*prec < 0 && -*prec < TYPE_PRECISION (lhs_type))
+			;
+		      else
+			*prec = -TYPE_PRECISION (lhs_type);
+		    }
+		}
+	      else
+		{
+		  op = rhs1;
+		  stmt = g;
+		  goto do_int;
+		}
+	    }
+	  m_loc = loc_save;
+	  return ret;
+	}
+      else
+	{
+	  int p = var_to_partition (m_map, op);
+	  gcc_assert (m_vars[p] != NULL_TREE);
+	  *prec = range_to_prec (op, stmt);
+	  if (prec_stored)
+	    *prec_stored = (TYPE_UNSIGNED (TREE_TYPE (op))
+			    ? TYPE_PRECISION (TREE_TYPE (op))
+			    : -TYPE_PRECISION (TREE_TYPE (op)));
+	  return build_fold_addr_expr (m_vars[p]);
+	}
+    case INTEGER_CST:
+      unsigned int min_prec, mp;
+      tree type;
+      w = wi::to_wide (op);
+      if (tree_int_cst_sgn (op) >= 0)
+	{
+	  min_prec = wi::min_precision (w, UNSIGNED);
+	  *prec = MAX (min_prec, 1);
+	}
+      else
+	{
+	  min_prec = wi::min_precision (w, SIGNED);
+	  *prec = MIN ((int) -min_prec, -2);
+	}
+      mp = CEIL (min_prec, limb_prec) * limb_prec;
+      if (mp >= (unsigned) TYPE_PRECISION (TREE_TYPE (op)))
+	type = TREE_TYPE (op);
+      else
+	type = build_bitint_type (mp, 1);
+      if (TREE_CODE (type) != BITINT_TYPE
+	  || bitint_precision_kind (type) == bitint_prec_small)
+	{
+	  if (TYPE_PRECISION (type) <= limb_prec)
+	    type = m_limb_type;
+	  else
+	    /* This case is for targets which e.g. have 64-bit
+	       limb but categorize up to 128-bits _BitInts as
+	       small.  We could use type of m_limb_type[2] and
+	       similar instead to save space.  */
+	    type = build_bitint_type (mid_min_prec, 1);
+	}
+      if (prec_stored)
+	{
+	  if (tree_int_cst_sgn (op) >= 0)
+	    *prec_stored = MAX (TYPE_PRECISION (type), 1);
+	  else
+	    *prec_stored = MIN ((int) -TYPE_PRECISION (type), -2);
+	}
+      op = tree_output_constant_def (fold_convert (type, op));
+      return build_fold_addr_expr (op);
+    default:
+      gcc_unreachable ();
+    }
+}
+
+/* Helper function, create a loop before the current location,
+   start with sizetype INIT value from the preheader edge.  Return
+   a PHI result and set *IDX_NEXT to SSA_NAME it creates and uses
+   from the latch edge.  */
+
+tree
+bitint_large_huge::create_loop (tree init, tree *idx_next)
+{
+  if (!gsi_end_p (m_gsi))
+    gsi_prev (&m_gsi);
+  else
+    m_gsi = gsi_last_bb (gsi_bb (m_gsi));
+  edge e1 = split_block (gsi_bb (m_gsi), gsi_stmt (m_gsi));
+  edge e2 = split_block (e1->dest, (gimple *) NULL);
+  edge e3 = make_edge (e1->dest, e1->dest, EDGE_TRUE_VALUE);
+  e3->probability = profile_probability::very_unlikely ();
+  e2->flags = EDGE_FALSE_VALUE;
+  e2->probability = e3->probability.invert ();
+  tree idx = make_ssa_name (sizetype);
+  gphi *phi = create_phi_node (idx, e1->dest);
+  add_phi_arg (phi, init, e1, UNKNOWN_LOCATION);
+  *idx_next = make_ssa_name (sizetype);
+  add_phi_arg (phi, *idx_next, e3, UNKNOWN_LOCATION);
+  m_gsi = gsi_after_labels (e1->dest);
+  m_bb = e1->dest;
+  m_preheader_bb = e1->src;
+  class loop *loop = alloc_loop ();
+  loop->header = e1->dest;
+  add_loop (loop, e1->src->loop_father);
+  return idx;
+}
+
+/* Lower large/huge _BitInt statement mergeable or similar STMT which can be
+   lowered using iteration from the least significant limb up to the most
+   significant limb.  For large _BitInt it is emitted as straight line code
+   before current location, for huge _BitInt as a loop handling two limbs
+   at once, followed by handling up to limbs in straight line code (at most
+   one full and one partial limb).  It can also handle EQ_EXPR/NE_EXPR
+   comparisons, in that case CMP_CODE should be the comparison code and
+   CMP_OP1/CMP_OP2 the comparison operands.  */
+
+tree
+bitint_large_huge::lower_mergeable_stmt (gimple *stmt, tree_code &cmp_code,
+					 tree cmp_op1, tree cmp_op2)
+{
+  bool eq_p = cmp_code != ERROR_MARK;
+  tree type;
+  if (eq_p)
+    type = TREE_TYPE (cmp_op1);
+  else
+    type = TREE_TYPE (gimple_assign_lhs (stmt));
+  gcc_assert (TREE_CODE (type) == BITINT_TYPE);
+  bitint_prec_kind kind = bitint_precision_kind (type);
+  gcc_assert (kind >= bitint_prec_large);
+  gimple *g;
+  tree lhs = gimple_get_lhs (stmt);
+  tree rhs1, lhs_type = lhs ? TREE_TYPE (lhs) : NULL_TREE;
+  if (lhs
+      && TREE_CODE (lhs) == SSA_NAME
+      && TREE_CODE (TREE_TYPE (lhs)) == BITINT_TYPE
+      && bitint_precision_kind (TREE_TYPE (lhs)) >= bitint_prec_large)
+    {
+      int p = var_to_partition (m_map, lhs);
+      gcc_assert (m_vars[p] != NULL_TREE);
+      m_lhs = lhs = m_vars[p];
+    }
+  unsigned cnt, rem = 0, end = 0, prec = TYPE_PRECISION (type);
+  bool sext = false;
+  tree ext = NULL_TREE, store_operand = NULL_TREE;
+  bool eh = false;
+  basic_block eh_pad = NULL;
+  tree nlhs = NULL_TREE;
+  unsigned HOST_WIDE_INT bo_idx = 0;
+  unsigned HOST_WIDE_INT bo_bit = 0;
+  tree bf_cur = NULL_TREE, bf_next = NULL_TREE;
+  if (gimple_store_p (stmt))
+    {
+      store_operand = gimple_assign_rhs1 (stmt);
+      eh = stmt_ends_bb_p (stmt);
+      if (eh)
+	{
+	  edge e;
+	  edge_iterator ei;
+	  basic_block bb = gimple_bb (stmt);
+
+	  FOR_EACH_EDGE (e, ei, bb->succs)
+	    if (e->flags & EDGE_EH)
+	      {
+		eh_pad = e->dest;
+		break;
+	      }
+	}
+      if (TREE_CODE (lhs) == COMPONENT_REF
+	  && DECL_BIT_FIELD_TYPE (TREE_OPERAND (lhs, 1)))
+	{
+	  tree fld = TREE_OPERAND (lhs, 1);
+	  gcc_assert (tree_fits_uhwi_p (DECL_FIELD_BIT_OFFSET (fld)));
+	  tree repr = DECL_BIT_FIELD_REPRESENTATIVE (fld);
+	  poly_int64 bitoffset;
+	  poly_uint64 field_offset, repr_offset;
+	  if ((tree_to_uhwi (DECL_FIELD_BIT_OFFSET (fld)) % BITS_PER_UNIT) == 0)
+	    nlhs = lhs;
+	  else
+	    {
+	      bool var_field_off = false;
+	      if (poly_int_tree_p (DECL_FIELD_OFFSET (fld), &field_offset)
+		  && poly_int_tree_p (DECL_FIELD_OFFSET (repr), &repr_offset))
+		bitoffset = (field_offset - repr_offset) * BITS_PER_UNIT;
+	      else
+		{
+		  bitoffset = 0;
+		  var_field_off = true;
+		}
+	      bitoffset += (tree_to_uhwi (DECL_FIELD_BIT_OFFSET (fld))
+			    - tree_to_uhwi (DECL_FIELD_BIT_OFFSET (repr)));
+	      nlhs = build3 (COMPONENT_REF, TREE_TYPE (repr),
+			     TREE_OPERAND (lhs, 0), repr,
+			     var_field_off
+			     ? TREE_OPERAND (lhs, 2) : NULL_TREE);
+	      HOST_WIDE_INT bo = bitoffset.to_constant ();
+	      bo_idx = (unsigned HOST_WIDE_INT) bo / limb_prec;
+	      bo_bit = (unsigned HOST_WIDE_INT) bo % limb_prec;
+	    }
+	}
+    }
+  if ((store_operand
+       && TREE_CODE (store_operand) == SSA_NAME
+       && (m_names == NULL
+	   || !bitmap_bit_p (m_names, SSA_NAME_VERSION (store_operand)))
+       && gimple_assign_cast_p (SSA_NAME_DEF_STMT (store_operand)))
+      || gimple_assign_cast_p (stmt))
+    {
+      rhs1 = gimple_assign_rhs1 (store_operand
+				 ? SSA_NAME_DEF_STMT (store_operand)
+				 : stmt);
+      /* Optimize mergeable ops ending with widening cast to _BitInt
+	 (or followed by store).  We can lower just the limbs of the
+	 cast operand and widen afterwards.  */
+      if (TREE_CODE (rhs1) == SSA_NAME
+	  && (m_names == NULL
+	      || !bitmap_bit_p (m_names, SSA_NAME_VERSION (rhs1)))
+	  && TREE_CODE (TREE_TYPE (rhs1)) == BITINT_TYPE
+	  && bitint_precision_kind (TREE_TYPE (rhs1)) >= bitint_prec_large
+	  && (CEIL ((unsigned) TYPE_PRECISION (TREE_TYPE (rhs1)),
+		    limb_prec) < CEIL (prec, limb_prec)
+	      || (kind == bitint_prec_huge
+		  && TYPE_PRECISION (TREE_TYPE (rhs1)) < prec)))
+	{
+	  store_operand = rhs1;
+	  prec = TYPE_PRECISION (TREE_TYPE (rhs1));
+	  kind = bitint_precision_kind (TREE_TYPE (rhs1));
+	  if (!TYPE_UNSIGNED (TREE_TYPE (rhs1)))
+	    sext = true;
+	}
+    }
+  tree idx = NULL_TREE, idx_first = NULL_TREE, idx_next = NULL_TREE;
+  if (kind == bitint_prec_large)
+    cnt = CEIL (prec, limb_prec);
+  else
+    {
+      rem = (prec % (2 * limb_prec));
+      end = (prec - rem) / limb_prec;
+      cnt = 2 + CEIL (rem, limb_prec);
+      idx = idx_first = create_loop (size_zero_node, &idx_next);
+    }
+
+  basic_block edge_bb = NULL;
+  if (eq_p)
+    {
+      gimple_stmt_iterator gsi = gsi_for_stmt (stmt);
+      gsi_prev (&gsi);
+      edge e = split_block (gsi_bb (gsi), gsi_stmt (gsi));
+      edge_bb = e->src;
+      if (kind == bitint_prec_large)
+	{
+	  m_gsi = gsi_last_bb (edge_bb);
+	  if (!gsi_end_p (m_gsi))
+	    gsi_next (&m_gsi);
+	}
+    }
+  else
+    m_after_stmt = stmt;
+  if (kind != bitint_prec_large)
+    m_upwards_2limb = end;
+  m_upwards = true;
+
+  bool separate_ext
+    = (prec != (unsigned) TYPE_PRECISION (type)
+       && (CEIL ((unsigned) TYPE_PRECISION (type), limb_prec)
+	   > CEIL (prec, limb_prec)));
+
+  for (unsigned i = 0; i < cnt; i++)
+    {
+      m_data_cnt = 0;
+      if (kind == bitint_prec_large)
+	idx = size_int (i);
+      else if (i >= 2)
+	idx = size_int (end + (i > 2));
+      if (eq_p)
+	{
+	  rhs1 = handle_operand (cmp_op1, idx);
+	  tree rhs2 = handle_operand (cmp_op2, idx);
+	  g = gimple_build_cond (NE_EXPR, rhs1, rhs2, NULL_TREE, NULL_TREE);
+	  insert_before (g);
+	  edge e1 = split_block (gsi_bb (m_gsi), g);
+	  e1->flags = EDGE_FALSE_VALUE;
+	  edge e2 = make_edge (e1->src, gimple_bb (stmt), EDGE_TRUE_VALUE);
+	  e1->probability = profile_probability::unlikely ();
+	  e2->probability = e1->probability.invert ();
+	  if (i == 0)
+	    set_immediate_dominator (CDI_DOMINATORS, e2->dest, e2->src);
+	  m_gsi = gsi_after_labels (e1->dest);
+	}
+      else
+	{
+	  if (store_operand)
+	    rhs1 = handle_operand (store_operand, idx);
+	  else
+	    rhs1 = handle_stmt (stmt, idx);
+	  if (!useless_type_conversion_p (m_limb_type, TREE_TYPE (rhs1)))
+	    rhs1 = add_cast (m_limb_type, rhs1);
+	  if (sext && i == cnt - 1)
+	    ext = rhs1;
+	  tree nidx = idx;
+	  if (bo_idx)
+	    {
+	      if (tree_fits_uhwi_p (idx))
+		nidx = size_int (tree_to_uhwi (idx) + bo_idx);
+	      else
+		{
+		  nidx = make_ssa_name (sizetype);
+		  g = gimple_build_assign (nidx, PLUS_EXPR, idx,
+					   size_int (bo_idx));
+		  insert_before (g);
+		}
+	    }
+	  bool done = false;
+	  basic_block new_bb = NULL;
+	  /* Handle stores into bit-fields.  */
+	  if (bo_bit)
+	    {
+	      if (i == 0)
+		{
+		  edge e2 = NULL;
+		  if (kind != bitint_prec_large)
+		    {
+		      prepare_data_in_out (build_zero_cst (m_limb_type),
+					   idx, &bf_next);
+		      bf_next = m_data.pop ();
+		      bf_cur = m_data.pop ();
+		      g = gimple_build_cond (EQ_EXPR, idx, size_zero_node,
+					     NULL_TREE, NULL_TREE);
+		      edge edge_true;
+		      if_then_else (g, profile_probability::unlikely (),
+				    edge_true, e2);
+		      new_bb = e2->dest;
+		    }
+		  tree ftype
+		    = build_nonstandard_integer_type (limb_prec - bo_bit, 1);
+		  tree bfr = build3 (BIT_FIELD_REF, ftype, unshare_expr (nlhs),
+				     bitsize_int (limb_prec - bo_bit),
+				     bitsize_int (bo_idx * limb_prec + bo_bit));
+		  tree t = add_cast (ftype, rhs1);
+		  g = gimple_build_assign (bfr, t);
+		  insert_before (g);
+		  if (eh)
+		    {
+		      maybe_duplicate_eh_stmt (g, stmt);
+		      if (eh_pad)
+			{
+			  edge e = split_block (gsi_bb (m_gsi), g);
+			  m_gsi = gsi_after_labels (e->dest);
+			  make_edge (e->src, eh_pad, EDGE_EH)->probability
+			    = profile_probability::very_unlikely ();
+			}
+		    }
+		  if (kind == bitint_prec_large)
+		    {
+		      bf_cur = rhs1;
+		      done = true;
+		    }
+		  else if (e2)
+		    m_gsi = gsi_after_labels (e2->src);
+		}
+	      if (!done)
+		{
+		  tree t1 = make_ssa_name (m_limb_type);
+		  tree t2 = make_ssa_name (m_limb_type);
+		  tree t3 = make_ssa_name (m_limb_type);
+		  g = gimple_build_assign (t1, RSHIFT_EXPR, bf_cur,
+					   build_int_cst (unsigned_type_node,
+							  limb_prec - bo_bit));
+		  insert_before (g);
+		  g = gimple_build_assign (t2, LSHIFT_EXPR, rhs1,
+					   build_int_cst (unsigned_type_node,
+							  bo_bit));
+		  insert_before (g);
+		  bf_cur = rhs1;
+		  g = gimple_build_assign (t3, BIT_IOR_EXPR, t1, t2);
+		  insert_before (g);
+		  rhs1 = t3;
+		  if (bf_next && i == 1)
+		    {
+		      g = gimple_build_assign (bf_next, bf_cur);
+		      insert_before (g);
+		    }
+		}
+	    }
+	  if (!done)
+	    {
+	      /* Handle bit-field access to partial last limb if needed.  */
+	      if (nlhs
+		  && i == cnt - 1
+		  && !separate_ext
+		  && tree_fits_uhwi_p (idx))
+		{
+		  unsigned int tprec = TYPE_PRECISION (type);
+		  unsigned int rprec = tprec % limb_prec;
+		  if (rprec + bo_bit < (unsigned) limb_prec)
+		    {
+		      tree ftype
+			= build_nonstandard_integer_type (rprec + bo_bit, 1);
+		      tree bfr = build3 (BIT_FIELD_REF, ftype,
+					 unshare_expr (nlhs),
+					 bitsize_int (rprec + bo_bit),
+					 bitsize_int ((bo_idx
+						       + tprec / limb_prec)
+						      * limb_prec));
+		      tree t = add_cast (ftype, rhs1);
+		      g = gimple_build_assign (bfr, t);
+		      done = true;
+		      bf_cur = NULL_TREE;
+		    }
+		  else if (rprec + bo_bit == (unsigned) limb_prec)
+		    bf_cur = NULL_TREE;
+		}
+	      /* Otherwise, stores to any other lhs.  */
+	      if (!done)
+		{
+		  tree l = limb_access (lhs_type, nlhs ? nlhs : lhs,
+					nidx, true);
+		  g = gimple_build_assign (l, rhs1);
+		}
+	      insert_before (g);
+	      if (eh)
+		{
+		  maybe_duplicate_eh_stmt (g, stmt);
+		  if (eh_pad)
+		    {
+		      edge e = split_block (gsi_bb (m_gsi), g);
+		      m_gsi = gsi_after_labels (e->dest);
+		      make_edge (e->src, eh_pad, EDGE_EH)->probability
+			= profile_probability::very_unlikely ();
+		    }
+		}
+	      if (new_bb)
+		m_gsi = gsi_after_labels (new_bb);
+	    }
+	}
+      m_first = false;
+      if (kind == bitint_prec_huge && i <= 1)
+	{
+	  if (i == 0)
+	    {
+	      idx = make_ssa_name (sizetype);
+	      g = gimple_build_assign (idx, PLUS_EXPR, idx_first,
+				       size_one_node);
+	      insert_before (g);
+	    }
+	  else
+	    {
+	      g = gimple_build_assign (idx_next, PLUS_EXPR, idx_first,
+				       size_int (2));
+	      insert_before (g);
+	      g = gimple_build_cond (NE_EXPR, idx_next, size_int (end),
+				     NULL_TREE, NULL_TREE);
+	      insert_before (g);
+	      if (eq_p)
+		m_gsi = gsi_after_labels (edge_bb);
+	      else
+		m_gsi = gsi_for_stmt (stmt);
+	    }
+	}
+    }
+
+  if (separate_ext)
+    {
+      if (sext)
+	{
+	  ext = add_cast (signed_type_for (m_limb_type), ext);
+	  tree lpm1 = build_int_cst (unsigned_type_node,
+				     limb_prec - 1);
+	  tree n = make_ssa_name (TREE_TYPE (ext));
+	  g = gimple_build_assign (n, RSHIFT_EXPR, ext, lpm1);
+	  insert_before (g);
+	  ext = add_cast (m_limb_type, n);
+	}
+      else
+	ext = build_zero_cst (m_limb_type);
+      kind = bitint_precision_kind (type);
+      unsigned start = CEIL (prec, limb_prec);
+      prec = TYPE_PRECISION (type);
+      idx = idx_first = idx_next = NULL_TREE;
+      if (prec <= (start + 2 + (bo_bit != 0)) * limb_prec)
+	kind = bitint_prec_large;
+      if (kind == bitint_prec_large)
+	cnt = CEIL (prec, limb_prec) - start;
+      else
+	{
+	  rem = prec % limb_prec;
+	  end = (prec - rem) / limb_prec;
+	  cnt = (bo_bit != 0) + 1 + (rem != 0);
+	}
+      for (unsigned i = 0; i < cnt; i++)
+	{
+	  if (kind == bitint_prec_large || (i == 0 && bo_bit != 0))
+	    idx = size_int (start + i);
+	  else if (i == cnt - 1)
+	    idx = size_int (end);
+	  else if (i == (bo_bit != 0))
+	    idx = create_loop (size_int (start + i), &idx_next);
+	  rhs1 = ext;
+	  if (bf_cur != NULL_TREE && bf_cur != ext)
+	    {
+	      tree t1 = make_ssa_name (m_limb_type);
+	      g = gimple_build_assign (t1, RSHIFT_EXPR, bf_cur,
+				       build_int_cst (unsigned_type_node,
+						      limb_prec - bo_bit));
+	      insert_before (g);
+	      if (integer_zerop (ext))
+		rhs1 = t1;
+	      else
+		{
+		  tree t2 = make_ssa_name (m_limb_type);
+		  rhs1 = make_ssa_name (m_limb_type);
+		  g = gimple_build_assign (t2, LSHIFT_EXPR, ext,
+					   build_int_cst (unsigned_type_node,
+							  bo_bit));
+		  insert_before (g);
+		  g = gimple_build_assign (rhs1, BIT_IOR_EXPR, t1, t2);
+		  insert_before (g);
+		}
+	      bf_cur = ext;
+	    }
+	  tree nidx = idx;
+	  if (bo_idx)
+	    {
+	      if (tree_fits_uhwi_p (idx))
+		nidx = size_int (tree_to_uhwi (idx) + bo_idx);
+	      else
+		{
+		  nidx = make_ssa_name (sizetype);
+		  g = gimple_build_assign (nidx, PLUS_EXPR, idx,
+					   size_int (bo_idx));
+		  insert_before (g);
+		}
+	    }
+	  bool done = false;
+	  /* Handle bit-field access to partial last limb if needed.  */
+	  if (nlhs && i == cnt - 1)
+	    {
+	      unsigned int tprec = TYPE_PRECISION (type);
+	      unsigned int rprec = tprec % limb_prec;
+	      if (rprec + bo_bit < (unsigned) limb_prec)
+		{
+		  tree ftype
+		    = build_nonstandard_integer_type (rprec + bo_bit, 1);
+		  tree bfr = build3 (BIT_FIELD_REF, ftype,
+				     unshare_expr (nlhs),
+				     bitsize_int (rprec + bo_bit),
+				     bitsize_int ((bo_idx + tprec / limb_prec)
+						  * limb_prec));
+		  tree t = add_cast (ftype, rhs1);
+		  g = gimple_build_assign (bfr, t);
+		  done = true;
+		  bf_cur = NULL_TREE;
+		}
+	      else if (rprec + bo_bit == (unsigned) limb_prec)
+		bf_cur = NULL_TREE;
+	    }
+	  /* Otherwise, stores to any other lhs.  */
+	  if (!done)
+	    {
+	      tree l = limb_access (lhs_type, nlhs ? nlhs : lhs, nidx, true);
+	      g = gimple_build_assign (l, rhs1);
+	    }
+	  insert_before (g);
+	  if (eh)
+	    {
+	      maybe_duplicate_eh_stmt (g, stmt);
+	      if (eh_pad)
+		{
+		  edge e = split_block (gsi_bb (m_gsi), g);
+		  m_gsi = gsi_after_labels (e->dest);
+		  make_edge (e->src, eh_pad, EDGE_EH)->probability
+		    = profile_probability::very_unlikely ();
+		}
+	    }
+	  if (kind == bitint_prec_huge && i == (bo_bit != 0))
+	    {
+	      g = gimple_build_assign (idx_next, PLUS_EXPR, idx,
+				       size_one_node);
+	      insert_before (g);
+	      g = gimple_build_cond (NE_EXPR, idx_next, size_int (end),
+				     NULL_TREE, NULL_TREE);
+	      insert_before (g);
+	      m_gsi = gsi_for_stmt (stmt);
+	    }
+	}
+    }
+  if (bf_cur != NULL_TREE)
+    {
+      unsigned int tprec = TYPE_PRECISION (type);
+      unsigned int rprec = tprec % limb_prec;
+      tree ftype = build_nonstandard_integer_type (rprec + bo_bit, 1);
+      tree bfr = build3 (BIT_FIELD_REF, ftype, unshare_expr (nlhs),
+			 bitsize_int (rprec + bo_bit),
+			 bitsize_int ((bo_idx + tprec / limb_prec)
+				      * limb_prec));
+      rhs1 = bf_cur;
+      if (bf_cur != ext)
+	{
+	  rhs1 = make_ssa_name (TREE_TYPE (rhs1));
+	  g = gimple_build_assign (rhs1, RSHIFT_EXPR, bf_cur,
+				   build_int_cst (unsigned_type_node,
+						  limb_prec - bo_bit));
+	  insert_before (g);
+	}
+      rhs1 = add_cast (ftype, rhs1);
+      g = gimple_build_assign (bfr, rhs1);
+      insert_before (g);
+      if (eh)
+	{
+	  maybe_duplicate_eh_stmt (g, stmt);
+	  if (eh_pad)
+	    {
+	      edge e = split_block (gsi_bb (m_gsi), g);
+	      m_gsi = gsi_after_labels (e->dest);
+	      make_edge (e->src, eh_pad, EDGE_EH)->probability
+		= profile_probability::very_unlikely ();
+	    }
+	}
+    }
+
+  if (gimple_store_p (stmt))
+    {
+      unlink_stmt_vdef (stmt);
+      release_ssa_name (gimple_vdef (stmt));
+      gsi_remove (&m_gsi, true);
+    }
+  if (eq_p)
+    {
+      lhs = make_ssa_name (boolean_type_node);
+      basic_block bb = gimple_bb (stmt);
+      gphi *phi = create_phi_node (lhs, bb);
+      edge e = find_edge (gsi_bb (m_gsi), bb);
+      unsigned int n = EDGE_COUNT (bb->preds);
+      for (unsigned int i = 0; i < n; i++)
+	{
+	  edge e2 = EDGE_PRED (bb, i);
+	  add_phi_arg (phi, e == e2 ? boolean_true_node : boolean_false_node,
+		       e2, UNKNOWN_LOCATION);
+	}
+      cmp_code = cmp_code == EQ_EXPR ? NE_EXPR : EQ_EXPR;
+      return lhs;
+    }
+  else
+    return NULL_TREE;
+}
+
+/* Handle a large/huge _BitInt comparison statement STMT other than
+   EQ_EXPR/NE_EXPR.  CMP_CODE, CMP_OP1 and CMP_OP2 meaning is like in
+   lower_mergeable_stmt.  The {GT,GE,LT,LE}_EXPR comparisons are
+   lowered by iteration from the most significant limb downwards to
+   the least significant one, for large _BitInt in straight line code,
+   otherwise with most significant limb handled in
+   straight line code followed by a loop handling one limb at a time.
+   Comparisons with unsigned huge _BitInt with precisions which are
+   multiples of limb precision can use just the loop and don't need to
+   handle most significant limb before the loop.  The loop or straight
+   line code jumps to final basic block if a particular pair of limbs
+   is not equal.  */
+
+tree
+bitint_large_huge::lower_comparison_stmt (gimple *stmt, tree_code &cmp_code,
+					  tree cmp_op1, tree cmp_op2)
+{
+  tree type = TREE_TYPE (cmp_op1);
+  gcc_assert (TREE_CODE (type) == BITINT_TYPE);
+  bitint_prec_kind kind = bitint_precision_kind (type);
+  gcc_assert (kind >= bitint_prec_large);
+  gimple *g;
+  if (!TYPE_UNSIGNED (type)
+      && integer_zerop (cmp_op2)
+      && (cmp_code == GE_EXPR || cmp_code == LT_EXPR))
+    {
+      unsigned end = CEIL ((unsigned) TYPE_PRECISION (type), limb_prec) - 1;
+      tree idx = size_int (end);
+      m_data_cnt = 0;
+      tree rhs1 = handle_operand (cmp_op1, idx);
+      if (TYPE_UNSIGNED (TREE_TYPE (rhs1)))
+	{
+	  tree stype = signed_type_for (TREE_TYPE (rhs1));
+	  rhs1 = add_cast (stype, rhs1);
+	}
+      tree lhs = make_ssa_name (boolean_type_node);
+      g = gimple_build_assign (lhs, cmp_code, rhs1,
+			       build_zero_cst (TREE_TYPE (rhs1)));
+      insert_before (g);
+      cmp_code = NE_EXPR;
+      return lhs;
+    }
+
+  unsigned cnt, rem = 0, end = 0;
+  tree idx = NULL_TREE, idx_next = NULL_TREE;
+  if (kind == bitint_prec_large)
+    cnt = CEIL ((unsigned) TYPE_PRECISION (type), limb_prec);
+  else
+    {
+      rem = ((unsigned) TYPE_PRECISION (type) % limb_prec);
+      if (rem == 0 && !TYPE_UNSIGNED (type))
+	rem = limb_prec;
+      end = ((unsigned) TYPE_PRECISION (type) - rem) / limb_prec;
+      cnt = 1 + (rem != 0);
+    }
+
+  basic_block edge_bb = NULL;
+  gimple_stmt_iterator gsi = gsi_for_stmt (stmt);
+  gsi_prev (&gsi);
+  edge e = split_block (gsi_bb (gsi), gsi_stmt (gsi));
+  edge_bb = e->src;
+  m_gsi = gsi_last_bb (edge_bb);
+  if (!gsi_end_p (m_gsi))
+    gsi_next (&m_gsi);
+
+  edge *edges = XALLOCAVEC (edge, cnt * 2);
+  for (unsigned i = 0; i < cnt; i++)
+    {
+      m_data_cnt = 0;
+      if (kind == bitint_prec_large)
+	idx = size_int (cnt - i - 1);
+      else if (i == cnt - 1)
+	idx = create_loop (size_int (end - 1), &idx_next);
+      else
+	idx = size_int (end);
+      tree rhs1 = handle_operand (cmp_op1, idx);
+      tree rhs2 = handle_operand (cmp_op2, idx);
+      if (i == 0
+	  && !TYPE_UNSIGNED (type)
+	  && TYPE_UNSIGNED (TREE_TYPE (rhs1)))
+	{
+	  tree stype = signed_type_for (TREE_TYPE (rhs1));
+	  rhs1 = add_cast (stype, rhs1);
+	  rhs2 = add_cast (stype, rhs2);
+	}
+      g = gimple_build_cond (GT_EXPR, rhs1, rhs2, NULL_TREE, NULL_TREE);
+      insert_before (g);
+      edge e1 = split_block (gsi_bb (m_gsi), g);
+      e1->flags = EDGE_FALSE_VALUE;
+      edge e2 = make_edge (e1->src, gimple_bb (stmt), EDGE_TRUE_VALUE);
+      e1->probability = profile_probability::likely ();
+      e2->probability = e1->probability.invert ();
+      if (i == 0)
+	set_immediate_dominator (CDI_DOMINATORS, e2->dest, e2->src);
+      m_gsi = gsi_after_labels (e1->dest);
+      edges[2 * i] = e2;
+      g = gimple_build_cond (LT_EXPR, rhs1, rhs2, NULL_TREE, NULL_TREE);
+      insert_before (g);
+      e1 = split_block (gsi_bb (m_gsi), g);
+      e1->flags = EDGE_FALSE_VALUE;
+      e2 = make_edge (e1->src, gimple_bb (stmt), EDGE_TRUE_VALUE);
+      e1->probability = profile_probability::unlikely ();
+      e2->probability = e1->probability.invert ();
+      m_gsi = gsi_after_labels (e1->dest);
+      edges[2 * i + 1] = e2;
+      m_first = false;
+      if (kind == bitint_prec_huge && i == cnt - 1)
+	{
+	  g = gimple_build_assign (idx_next, PLUS_EXPR, idx, size_int (-1));
+	  insert_before (g);
+	  g = gimple_build_cond (NE_EXPR, idx, size_zero_node,
+				 NULL_TREE, NULL_TREE);
+	  insert_before (g);
+	  edge true_edge, false_edge;
+	  extract_true_false_edges_from_block (gsi_bb (m_gsi),
+					       &true_edge, &false_edge);
+	  m_gsi = gsi_after_labels (false_edge->dest);
+	}
+    }
+
+  tree lhs = make_ssa_name (boolean_type_node);
+  basic_block bb = gimple_bb (stmt);
+  gphi *phi = create_phi_node (lhs, bb);
+  for (unsigned int i = 0; i < cnt * 2; i++)
+    {
+      tree val = ((cmp_code == GT_EXPR || cmp_code == GE_EXPR)
+		  ^ (i & 1)) ? boolean_true_node : boolean_false_node;
+      add_phi_arg (phi, val, edges[i], UNKNOWN_LOCATION);
+    }
+  add_phi_arg (phi, (cmp_code == GE_EXPR || cmp_code == LE_EXPR)
+		    ? boolean_true_node : boolean_false_node,
+	       find_edge (gsi_bb (m_gsi), bb), UNKNOWN_LOCATION);
+  cmp_code = NE_EXPR;
+  return lhs;
+}
+
+/* Lower large/huge _BitInt left and right shift except for left
+   shift by < limb_prec constant.  */
+
+void
+bitint_large_huge::lower_shift_stmt (tree obj, gimple *stmt)
+{
+  tree rhs1 = gimple_assign_rhs1 (stmt);
+  tree lhs = gimple_assign_lhs (stmt);
+  tree_code rhs_code = gimple_assign_rhs_code (stmt);
+  tree type = TREE_TYPE (rhs1);
+  gimple *final_stmt = gsi_stmt (m_gsi);
+  gcc_assert (TREE_CODE (type) == BITINT_TYPE
+	      && bitint_precision_kind (type) >= bitint_prec_large);
+  int prec = TYPE_PRECISION (type);
+  tree n = gimple_assign_rhs2 (stmt), n1, n2, n3, n4;
+  gimple *g;
+  if (obj == NULL_TREE)
+    {
+      int part = var_to_partition (m_map, lhs);
+      gcc_assert (m_vars[part] != NULL_TREE);
+      obj = m_vars[part];
+    }
+  /* Preparation code common for both left and right shifts.
+     unsigned n1 = n % limb_prec;
+     size_t n2 = n / limb_prec;
+     size_t n3 = n1 != 0;
+     unsigned n4 = (limb_prec - n1) % limb_prec;
+     (for power of 2 limb_prec n4 can be -n1 & (limb_prec)).  */
+  if (TREE_CODE (n) == INTEGER_CST)
+    {
+      tree lp = build_int_cst (TREE_TYPE (n), limb_prec);
+      n1 = int_const_binop (TRUNC_MOD_EXPR, n, lp);
+      n2 = fold_convert (sizetype, int_const_binop (TRUNC_DIV_EXPR, n, lp));
+      n3 = size_int (!integer_zerop (n1));
+      n4 = int_const_binop (TRUNC_MOD_EXPR,
+			    int_const_binop (MINUS_EXPR, lp, n1), lp);
+    }
+  else
+    {
+      n1 = make_ssa_name (TREE_TYPE (n));
+      n2 = make_ssa_name (sizetype);
+      n3 = make_ssa_name (sizetype);
+      n4 = make_ssa_name (TREE_TYPE (n));
+      if (pow2p_hwi (limb_prec))
+	{
+	  tree lpm1 = build_int_cst (TREE_TYPE (n), limb_prec - 1);
+	  g = gimple_build_assign (n1, BIT_AND_EXPR, n, lpm1);
+	  insert_before (g);
+	  g = gimple_build_assign (useless_type_conversion_p (sizetype,
+							      TREE_TYPE (n))
+				   ? n2 : make_ssa_name (TREE_TYPE (n)),
+				   RSHIFT_EXPR, n,
+				   build_int_cst (TREE_TYPE (n),
+						  exact_log2 (limb_prec)));
+	  insert_before (g);
+	  if (gimple_assign_lhs (g) != n2)
+	    {
+	      g = gimple_build_assign (n2, NOP_EXPR, gimple_assign_lhs (g));
+	      insert_before (g);
+	    }
+	  g = gimple_build_assign (make_ssa_name (TREE_TYPE (n)),
+				   NEGATE_EXPR, n1);
+	  insert_before (g);
+	  g = gimple_build_assign (n4, BIT_AND_EXPR, gimple_assign_lhs (g),
+				   lpm1);
+	  insert_before (g);
+	}
+      else
+	{
+	  tree lp = build_int_cst (TREE_TYPE (n), limb_prec);
+	  g = gimple_build_assign (n1, TRUNC_MOD_EXPR, n, lp);
+	  insert_before (g);
+	  g = gimple_build_assign (useless_type_conversion_p (sizetype,
+							      TREE_TYPE (n))
+				   ? n2 : make_ssa_name (TREE_TYPE (n)),
+				   TRUNC_DIV_EXPR, n, lp);
+	  insert_before (g);
+	  if (gimple_assign_lhs (g) != n2)
+	    {
+	      g = gimple_build_assign (n2, NOP_EXPR, gimple_assign_lhs (g));
+	      insert_before (g);
+	    }
+	  g = gimple_build_assign (make_ssa_name (TREE_TYPE (n)),
+				   MINUS_EXPR, lp, n1);
+	  insert_before (g);
+	  g = gimple_build_assign (n4, TRUNC_MOD_EXPR, gimple_assign_lhs (g),
+				   lp);
+	  insert_before (g);
+	}
+      g = gimple_build_assign (make_ssa_name (boolean_type_node), NE_EXPR, n1,
+			       build_zero_cst (TREE_TYPE (n)));
+      insert_before (g);
+      g = gimple_build_assign (n3, NOP_EXPR, gimple_assign_lhs (g));
+      insert_before (g);
+    }
+  tree p = build_int_cst (sizetype,
+			  prec / limb_prec - (prec % limb_prec == 0));
+  if (rhs_code == RSHIFT_EXPR)
+    {
+      /* Lower
+	   dst = src >> n;
+	 as
+	   unsigned n1 = n % limb_prec;
+	   size_t n2 = n / limb_prec;
+	   size_t n3 = n1 != 0;
+	   unsigned n4 = (limb_prec - n1) % limb_prec;
+	   size_t idx;
+	   size_t p = prec / limb_prec - (prec % limb_prec == 0);
+	   int signed_p = (typeof (src) -1) < 0;
+	   for (idx = n2; idx < ((!signed_p && (prec % limb_prec == 0))
+				 ? p : p - n3); ++idx)
+	     dst[idx - n2] = (src[idx] >> n1) | (src[idx + n3] << n4);
+	   limb_type ext;
+	   if (prec % limb_prec == 0)
+	     ext = src[p];
+	   else if (signed_p)
+	     ext = ((signed limb_type) (src[p] << (limb_prec
+						   - (prec % limb_prec))))
+		   >> (limb_prec - (prec % limb_prec));
+	   else
+	     ext = src[p] & (((limb_type) 1 << (prec % limb_prec)) - 1);
+	   if (!signed_p && (prec % limb_prec == 0))
+	     ;
+	   else if (idx < prec / 64)
+	     {
+	       dst[idx - n2] = (src[idx] >> n1) | (ext << n4);
+	       ++idx;
+	     }
+	   idx -= n2;
+	   if (signed_p)
+	     {
+	       dst[idx] = ((signed limb_type) ext) >> n1;
+	       ext = ((signed limb_type) ext) >> (limb_prec - 1);
+	     }
+	   else
+	     {
+	       dst[idx] = ext >> n1;
+	       ext = 0;
+	     }
+	   for (++idx; idx <= p; ++idx)
+	     dst[idx] = ext;  */
+      tree pmn3;
+      if (TYPE_UNSIGNED (type) && prec % limb_prec == 0)
+	pmn3 = p;
+      else if (TREE_CODE (n3) == INTEGER_CST)
+	pmn3 = int_const_binop (MINUS_EXPR, p, n3);
+      else
+	{
+	  pmn3 = make_ssa_name (sizetype);
+	  g = gimple_build_assign (pmn3, MINUS_EXPR, p, n3);
+	  insert_before (g);
+	}
+      g = gimple_build_cond (LT_EXPR, n2, pmn3, NULL_TREE, NULL_TREE);
+      edge edge_true, edge_false;
+      if_then (g, profile_probability::likely (), edge_true, edge_false);
+      tree idx_next;
+      tree idx = create_loop (n2, &idx_next);
+      tree idxmn2 = make_ssa_name (sizetype);
+      tree idxpn3 = make_ssa_name (sizetype);
+      g = gimple_build_assign (idxmn2, MINUS_EXPR, idx, n2);
+      insert_before (g);
+      g = gimple_build_assign (idxpn3, PLUS_EXPR, idx, n3);
+      insert_before (g);
+      m_data_cnt = 0;
+      tree t1 = handle_operand (rhs1, idx);
+      m_first = false;
+      g = gimple_build_assign (make_ssa_name (m_limb_type),
+			       RSHIFT_EXPR, t1, n1);
+      insert_before (g);
+      t1 = gimple_assign_lhs (g);
+      if (!integer_zerop (n3))
+	{
+	  m_data_cnt = 0;
+	  tree t2 = handle_operand (rhs1, idxpn3);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   LSHIFT_EXPR, t2, n4);
+	  insert_before (g);
+	  t2 = gimple_assign_lhs (g);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   BIT_IOR_EXPR, t1, t2);
+	  insert_before (g);
+	  t1 = gimple_assign_lhs (g);
+	}
+      tree l = limb_access (TREE_TYPE (lhs), obj, idxmn2, true);
+      g = gimple_build_assign (l, t1);
+      insert_before (g);
+      g = gimple_build_assign (idx_next, PLUS_EXPR, idx, size_one_node);
+      insert_before (g);
+      g = gimple_build_cond (LT_EXPR, idx_next, pmn3, NULL_TREE, NULL_TREE);
+      insert_before (g);
+      idx = make_ssa_name (sizetype);
+      m_gsi = gsi_for_stmt (final_stmt);
+      gphi *phi = create_phi_node (idx, gsi_bb (m_gsi));
+      edge_false = find_edge (edge_false->src, gsi_bb (m_gsi));
+      edge_true = EDGE_PRED (gsi_bb (m_gsi),
+			     EDGE_PRED (gsi_bb (m_gsi), 0) == edge_false);
+      add_phi_arg (phi, n2, edge_false, UNKNOWN_LOCATION);
+      add_phi_arg (phi, idx_next, edge_true, UNKNOWN_LOCATION);
+      m_data_cnt = 0;
+      tree ms = handle_operand (rhs1, p);
+      tree ext = ms;
+      if (!types_compatible_p (TREE_TYPE (ms), m_limb_type))
+	ext = add_cast (m_limb_type, ms);
+      if (!(TYPE_UNSIGNED (type) && prec % limb_prec == 0)
+	  && !integer_zerop (n3))
+	{
+	  g = gimple_build_cond (LT_EXPR, idx, p, NULL_TREE, NULL_TREE);
+	  if_then (g, profile_probability::likely (), edge_true, edge_false);
+	  m_data_cnt = 0;
+	  t1 = handle_operand (rhs1, idx);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   RSHIFT_EXPR, t1, n1);
+	  insert_before (g);
+	  t1 = gimple_assign_lhs (g);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   LSHIFT_EXPR, ext, n4);
+	  insert_before (g);
+	  tree t2 = gimple_assign_lhs (g);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   BIT_IOR_EXPR, t1, t2);
+	  insert_before (g);
+	  t1 = gimple_assign_lhs (g);
+	  idxmn2 = make_ssa_name (sizetype);
+	  g = gimple_build_assign (idxmn2, MINUS_EXPR, idx, n2);
+	  insert_before (g);
+	  l = limb_access (TREE_TYPE (lhs), obj, idxmn2, true);
+	  g = gimple_build_assign (l, t1);
+	  insert_before (g);
+	  idx_next = make_ssa_name (sizetype);
+	  g = gimple_build_assign (idx_next, PLUS_EXPR, idx, size_one_node);
+	  insert_before (g);
+	  m_gsi = gsi_for_stmt (final_stmt);
+	  tree nidx = make_ssa_name (sizetype);
+	  phi = create_phi_node (nidx, gsi_bb (m_gsi));
+	  edge_false = find_edge (edge_false->src, gsi_bb (m_gsi));
+	  edge_true = EDGE_PRED (gsi_bb (m_gsi),
+				 EDGE_PRED (gsi_bb (m_gsi), 0) == edge_false);
+	  add_phi_arg (phi, idx, edge_false, UNKNOWN_LOCATION);
+	  add_phi_arg (phi, idx_next, edge_true, UNKNOWN_LOCATION);
+	  idx = nidx;
+	}
+      g = gimple_build_assign (make_ssa_name (sizetype), MINUS_EXPR, idx, n2);
+      insert_before (g);
+      idx = gimple_assign_lhs (g);
+      tree sext = ext;
+      if (!TYPE_UNSIGNED (type))
+	sext = add_cast (signed_type_for (m_limb_type), ext);
+      g = gimple_build_assign (make_ssa_name (TREE_TYPE (sext)),
+			       RSHIFT_EXPR, sext, n1);
+      insert_before (g);
+      t1 = gimple_assign_lhs (g);
+      if (!TYPE_UNSIGNED (type))
+	{
+	  t1 = add_cast (m_limb_type, t1);
+	  g = gimple_build_assign (make_ssa_name (TREE_TYPE (sext)),
+				   RSHIFT_EXPR, sext,
+				   build_int_cst (TREE_TYPE (n),
+						  limb_prec - 1));
+	  insert_before (g);
+	  ext = add_cast (m_limb_type, gimple_assign_lhs (g));
+	}
+      else
+	ext = build_zero_cst (m_limb_type);
+      l = limb_access (TREE_TYPE (lhs), obj, idx, true);
+      g = gimple_build_assign (l, t1);
+      insert_before (g);
+      g = gimple_build_assign (make_ssa_name (sizetype), PLUS_EXPR, idx,
+			       size_one_node);
+      insert_before (g);
+      idx = gimple_assign_lhs (g);
+      g = gimple_build_cond (LE_EXPR, idx, p, NULL_TREE, NULL_TREE);
+      if_then (g, profile_probability::likely (), edge_true, edge_false);
+      idx = create_loop (idx, &idx_next);
+      l = limb_access (TREE_TYPE (lhs), obj, idx, true);
+      g = gimple_build_assign (l, ext);
+      insert_before (g);
+      g = gimple_build_assign (idx_next, PLUS_EXPR, idx, size_one_node);
+      insert_before (g);
+      g = gimple_build_cond (LE_EXPR, idx_next, p, NULL_TREE, NULL_TREE);
+      insert_before (g);
+    }
+  else
+    {
+      /* Lower
+	   dst = src << n;
+	 as
+	   unsigned n1 = n % limb_prec;
+	   size_t n2 = n / limb_prec;
+	   size_t n3 = n1 != 0;
+	   unsigned n4 = (limb_prec - n1) % limb_prec;
+	   size_t idx;
+	   size_t p = prec / limb_prec - (prec % limb_prec == 0);
+	   for (idx = p; (ssize_t) idx >= (ssize_t) (n2 + n3); --idx)
+	     dst[idx] = (src[idx - n2] << n1) | (src[idx - n2 - n3] >> n4);
+	   if (n1)
+	     {
+	       dst[idx] = src[idx - n2] << n1;
+	       --idx;
+	     }
+	   for (; (ssize_t) idx >= 0; --idx)
+	     dst[idx] = 0;  */
+      tree n2pn3;
+      if (TREE_CODE (n2) == INTEGER_CST && TREE_CODE (n3) == INTEGER_CST)
+	n2pn3 = int_const_binop (PLUS_EXPR, n2, n3);
+      else
+	{
+	  n2pn3 = make_ssa_name (sizetype);
+	  g = gimple_build_assign (n2pn3, PLUS_EXPR, n2, n3);
+	  insert_before (g);
+	}
+      /* For LSHIFT_EXPR, we can use handle_operand with non-INTEGER_CST
+	 idx even to access the most significant partial limb.  */
+      m_var_msb = true;
+      if (integer_zerop (n3))
+	/* For n3 == 0 p >= n2 + n3 is always true for all valid shift
+	   counts.  Emit if (true) condition that can be optimized later.  */
+	g = gimple_build_cond (NE_EXPR, boolean_true_node, boolean_false_node,
+			       NULL_TREE, NULL_TREE);
+      else
+	g = gimple_build_cond (LE_EXPR, n2pn3, p, NULL_TREE, NULL_TREE);
+      edge edge_true, edge_false;
+      if_then (g, profile_probability::likely (), edge_true, edge_false);
+      tree idx_next;
+      tree idx = create_loop (p, &idx_next);
+      tree idxmn2 = make_ssa_name (sizetype);
+      tree idxmn2mn3 = make_ssa_name (sizetype);
+      g = gimple_build_assign (idxmn2, MINUS_EXPR, idx, n2);
+      insert_before (g);
+      g = gimple_build_assign (idxmn2mn3, MINUS_EXPR, idxmn2, n3);
+      insert_before (g);
+      m_data_cnt = 0;
+      tree t1 = handle_operand (rhs1, idxmn2);
+      m_first = false;
+      g = gimple_build_assign (make_ssa_name (m_limb_type),
+			       LSHIFT_EXPR, t1, n1);
+      insert_before (g);
+      t1 = gimple_assign_lhs (g);
+      if (!integer_zerop (n3))
+	{
+	  m_data_cnt = 0;
+	  tree t2 = handle_operand (rhs1, idxmn2mn3);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   RSHIFT_EXPR, t2, n4);
+	  insert_before (g);
+	  t2 = gimple_assign_lhs (g);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   BIT_IOR_EXPR, t1, t2);
+	  insert_before (g);
+	  t1 = gimple_assign_lhs (g);
+	}
+      tree l = limb_access (TREE_TYPE (lhs), obj, idx, true);
+      g = gimple_build_assign (l, t1);
+      insert_before (g);
+      g = gimple_build_assign (idx_next, PLUS_EXPR, idx, size_int (-1));
+      insert_before (g);
+      tree sn2pn3 = add_cast (ssizetype, n2pn3);
+      g = gimple_build_cond (GE_EXPR, add_cast (ssizetype, idx_next), sn2pn3,
+			     NULL_TREE, NULL_TREE);
+      insert_before (g);
+      idx = make_ssa_name (sizetype);
+      m_gsi = gsi_for_stmt (final_stmt);
+      gphi *phi = create_phi_node (idx, gsi_bb (m_gsi));
+      edge_false = find_edge (edge_false->src, gsi_bb (m_gsi));
+      edge_true = EDGE_PRED (gsi_bb (m_gsi),
+			     EDGE_PRED (gsi_bb (m_gsi), 0) == edge_false);
+      add_phi_arg (phi, p, edge_false, UNKNOWN_LOCATION);
+      add_phi_arg (phi, idx_next, edge_true, UNKNOWN_LOCATION);
+      m_data_cnt = 0;
+      if (!integer_zerop (n3))
+	{
+	  g = gimple_build_cond (NE_EXPR, n3, size_zero_node,
+				 NULL_TREE, NULL_TREE);
+	  if_then (g, profile_probability::likely (), edge_true, edge_false);
+	  idxmn2 = make_ssa_name (sizetype);
+	  g = gimple_build_assign (idxmn2, MINUS_EXPR, idx, n2);
+	  insert_before (g);
+	  m_data_cnt = 0;
+	  t1 = handle_operand (rhs1, idxmn2);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   LSHIFT_EXPR, t1, n1);
+	  insert_before (g);
+	  t1 = gimple_assign_lhs (g);
+	  l = limb_access (TREE_TYPE (lhs), obj, idx, true);
+	  g = gimple_build_assign (l, t1);
+	  insert_before (g);
+	  idx_next = make_ssa_name (sizetype);
+	  g = gimple_build_assign (idx_next, PLUS_EXPR, idx, size_int (-1));
+	  insert_before (g);
+	  m_gsi = gsi_for_stmt (final_stmt);
+	  tree nidx = make_ssa_name (sizetype);
+	  phi = create_phi_node (nidx, gsi_bb (m_gsi));
+	  edge_false = find_edge (edge_false->src, gsi_bb (m_gsi));
+	  edge_true = EDGE_PRED (gsi_bb (m_gsi),
+				 EDGE_PRED (gsi_bb (m_gsi), 0) == edge_false);
+	  add_phi_arg (phi, idx, edge_false, UNKNOWN_LOCATION);
+	  add_phi_arg (phi, idx_next, edge_true, UNKNOWN_LOCATION);
+	  idx = nidx;
+	}
+      g = gimple_build_cond (GE_EXPR, add_cast (ssizetype, idx),
+			     ssize_int (0), NULL_TREE, NULL_TREE);
+      if_then (g, profile_probability::likely (), edge_true, edge_false);
+      idx = create_loop (idx, &idx_next);
+      l = limb_access (TREE_TYPE (lhs), obj, idx, true);
+      g = gimple_build_assign (l, build_zero_cst (m_limb_type));
+      insert_before (g);
+      g = gimple_build_assign (idx_next, PLUS_EXPR, idx, size_int (-1));
+      insert_before (g);
+      g = gimple_build_cond (GE_EXPR, add_cast (ssizetype, idx_next),
+			     ssize_int (0), NULL_TREE, NULL_TREE);
+      insert_before (g);
+    }
+}
+
+/* Lower large/huge _BitInt multiplication or division.  */
+
+void
+bitint_large_huge::lower_muldiv_stmt (tree obj, gimple *stmt)
+{
+  tree rhs1 = gimple_assign_rhs1 (stmt);
+  tree rhs2 = gimple_assign_rhs2 (stmt);
+  tree lhs = gimple_assign_lhs (stmt);
+  tree_code rhs_code = gimple_assign_rhs_code (stmt);
+  tree type = TREE_TYPE (rhs1);
+  gcc_assert (TREE_CODE (type) == BITINT_TYPE
+	      && bitint_precision_kind (type) >= bitint_prec_large);
+  int prec = TYPE_PRECISION (type), prec1, prec2;
+  rhs1 = handle_operand_addr (rhs1, stmt, NULL, &prec1);
+  rhs2 = handle_operand_addr (rhs2, stmt, NULL, &prec2);
+  if (obj == NULL_TREE)
+    {
+      int part = var_to_partition (m_map, lhs);
+      gcc_assert (m_vars[part] != NULL_TREE);
+      obj = m_vars[part];
+      lhs = build_fold_addr_expr (obj);
+    }
+  else
+    {
+      lhs = build_fold_addr_expr (obj);
+      lhs = force_gimple_operand_gsi (&m_gsi, lhs, true,
+				      NULL_TREE, true, GSI_SAME_STMT);
+    }
+  tree sitype = lang_hooks.types.type_for_mode (SImode, 0);
+  gimple *g;
+  switch (rhs_code)
+    {
+    case MULT_EXPR:
+      g = gimple_build_call_internal (IFN_MULBITINT, 6,
+				      lhs, build_int_cst (sitype, prec),
+				      rhs1, build_int_cst (sitype, prec1),
+				      rhs2, build_int_cst (sitype, prec2));
+      insert_before (g);
+      break;
+    case TRUNC_DIV_EXPR:
+      g = gimple_build_call_internal (IFN_DIVMODBITINT, 8,
+				      lhs, build_int_cst (sitype, prec),
+				      null_pointer_node,
+				      build_int_cst (sitype, 0),
+				      rhs1, build_int_cst (sitype, prec1),
+				      rhs2, build_int_cst (sitype, prec2));
+      if (!stmt_ends_bb_p (stmt))
+	gimple_call_set_nothrow (as_a <gcall *> (g), true);
+      insert_before (g);
+      break;
+    case TRUNC_MOD_EXPR:
+      g = gimple_build_call_internal (IFN_DIVMODBITINT, 8, null_pointer_node,
+				      build_int_cst (sitype, 0),
+				      lhs, build_int_cst (sitype, prec),
+				      rhs1, build_int_cst (sitype, prec1),
+				      rhs2, build_int_cst (sitype, prec2));
+      if (!stmt_ends_bb_p (stmt))
+	gimple_call_set_nothrow (as_a <gcall *> (g), true);
+      insert_before (g);
+      break;
+    default:
+      gcc_unreachable ();
+    }
+  if (stmt_ends_bb_p (stmt))
+    {
+      maybe_duplicate_eh_stmt (g, stmt);
+      edge e1;
+      edge_iterator ei;
+      basic_block bb = gimple_bb (stmt);
+
+      FOR_EACH_EDGE (e1, ei, bb->succs)
+	if (e1->flags & EDGE_EH)
+	  break;
+      if (e1)
+	{
+	  edge e2 = split_block (gsi_bb (m_gsi), g);
+	  m_gsi = gsi_after_labels (e2->dest);
+	  make_edge (e2->src, e1->dest, EDGE_EH)->probability
+	    = profile_probability::very_unlikely ();
+	}
+    }
+}
+
+/* Lower large/huge _BitInt conversion to/from floating point.  */
+
+void
+bitint_large_huge::lower_float_conv_stmt (tree obj, gimple *stmt)
+{
+  tree rhs1 = gimple_assign_rhs1 (stmt);
+  tree lhs = gimple_assign_lhs (stmt);
+  tree_code rhs_code = gimple_assign_rhs_code (stmt);
+  tree sitype = lang_hooks.types.type_for_mode (SImode, 0);
+  gimple *g;
+  if (rhs_code == FIX_TRUNC_EXPR)
+    {
+      int prec = TYPE_PRECISION (TREE_TYPE (lhs));
+      if (!TYPE_UNSIGNED (TREE_TYPE (lhs)))
+	prec = -prec;
+      if (obj == NULL_TREE)
+	{
+	  int part = var_to_partition (m_map, lhs);
+	  gcc_assert (m_vars[part] != NULL_TREE);
+	  obj = m_vars[part];
+	  lhs = build_fold_addr_expr (obj);
+	}
+      else
+	{
+	  lhs = build_fold_addr_expr (obj);
+	  lhs = force_gimple_operand_gsi (&m_gsi, lhs, true,
+					  NULL_TREE, true, GSI_SAME_STMT);
+	}
+      scalar_mode from_mode
+	= as_a <scalar_mode> (TYPE_MODE (TREE_TYPE (rhs1)));
+#ifdef HAVE_SFmode
+      /* IEEE single is a full superset of both IEEE half and
+	 bfloat formats, convert to float first and then to _BitInt
+	 to avoid the need of another 2 library routines.  */
+      if ((REAL_MODE_FORMAT (from_mode) == &arm_bfloat_half_format
+	   || REAL_MODE_FORMAT (from_mode) == &ieee_half_format)
+	  && REAL_MODE_FORMAT (SFmode) == &ieee_single_format)
+	{
+	  tree type = lang_hooks.types.type_for_mode (SFmode, 0);
+	  if (type)
+	    rhs1 = add_cast (type, rhs1);
+	}
+#endif
+      g = gimple_build_call_internal (IFN_FLOATTOBITINT, 3,
+				      lhs, build_int_cst (sitype, prec),
+				      rhs1);
+      insert_before (g);
+    }
+  else
+    {
+      int prec;
+      rhs1 = handle_operand_addr (rhs1, stmt, NULL, &prec);
+      g = gimple_build_call_internal (IFN_BITINTTOFLOAT, 2,
+				      rhs1, build_int_cst (sitype, prec));
+      gimple_call_set_lhs (g, lhs);
+      if (!stmt_ends_bb_p (stmt))
+	gimple_call_set_nothrow (as_a <gcall *> (g), true);
+      gsi_replace (&m_gsi, g, true);
+    }
+}
+
+/* Helper method for lower_addsub_overflow and lower_mul_overflow.
+   If check_zero is true, caller wants to check if all bits in [start, end)
+   are zero, otherwise if bits in [start, end) are either all zero or
+   all ones.  L is the limb with index LIMB, START and END are measured
+   in bits.  */
+
+tree
+bitint_large_huge::arith_overflow_extract_bits (unsigned int start,
+						unsigned int end, tree l,
+						unsigned int limb,
+						bool check_zero)
+{
+  unsigned startlimb = start / limb_prec;
+  unsigned endlimb = (end - 1) / limb_prec;
+  gimple *g;
+
+  if ((start % limb_prec) == 0 && (end % limb_prec) == 0)
+    return l;
+  if (startlimb == endlimb && limb == startlimb)
+    {
+      if (check_zero)
+	{
+	  wide_int w = wi::shifted_mask (start % limb_prec,
+					 end - start, false, limb_prec);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   BIT_AND_EXPR, l,
+				   wide_int_to_tree (m_limb_type, w));
+	  insert_before (g);
+	  return gimple_assign_lhs (g);
+	}
+      unsigned int shift = start % limb_prec;
+      if ((end % limb_prec) != 0)
+	{
+	  unsigned int lshift = (-end) % limb_prec;
+	  shift += lshift;
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   LSHIFT_EXPR, l,
+				   build_int_cst (unsigned_type_node,
+						  lshift));
+	  insert_before (g);
+	  l = gimple_assign_lhs (g);
+	}
+      l = add_cast (signed_type_for (m_limb_type), l);
+      g = gimple_build_assign (make_ssa_name (TREE_TYPE (l)),
+			       RSHIFT_EXPR, l,
+			       build_int_cst (unsigned_type_node, shift));
+      insert_before (g);
+      return add_cast (m_limb_type, gimple_assign_lhs (g));
+    }
+  else if (limb == startlimb)
+    {
+      if ((start % limb_prec) == 0)
+	return l;
+      if (!check_zero)
+	l = add_cast (signed_type_for (m_limb_type), l);
+      g = gimple_build_assign (make_ssa_name (TREE_TYPE (l)),
+			       RSHIFT_EXPR, l,
+			       build_int_cst (unsigned_type_node,
+					      start % limb_prec));
+      insert_before (g);
+      l = gimple_assign_lhs (g);
+      if (!check_zero)
+	l = add_cast (m_limb_type, l);
+      return l;
+    }
+  else if (limb == endlimb)
+    {
+      if ((end % limb_prec) == 0)
+	return l;
+      if (check_zero)
+	{
+	  wide_int w = wi::mask (end % limb_prec, false, limb_prec);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type),
+				   BIT_AND_EXPR, l,
+				   wide_int_to_tree (m_limb_type, w));
+	  insert_before (g);
+	  return gimple_assign_lhs (g);
+	}
+      unsigned int shift = (-end) % limb_prec;
+      g = gimple_build_assign (make_ssa_name (m_limb_type),
+			       LSHIFT_EXPR, l,
+			       build_int_cst (unsigned_type_node, shift));
+      insert_before (g);
+      l = add_cast (signed_type_for (m_limb_type), gimple_assign_lhs (g));
+      g = gimple_build_assign (make_ssa_name (TREE_TYPE (l)),
+			       RSHIFT_EXPR, l,
+			       build_int_cst (unsigned_type_node, shift));
+      insert_before (g);
+      return add_cast (m_limb_type, gimple_assign_lhs (g));
+    }
+  return l;
+}
+
+/* Helper method for lower_addsub_overflow and lower_mul_overflow.  Store
+   result including overflow flag into the right locations.  */
+
+void
+bitint_large_huge::finish_arith_overflow (tree var, tree obj, tree type,
+					  tree ovf, tree lhs, tree orig_obj,
+					  gimple *stmt, tree_code code)
+{
+  gimple *g;
+
+  if (obj == NULL_TREE
+      && (TREE_CODE (type) != BITINT_TYPE
+	  || bitint_precision_kind (type) < bitint_prec_large))
+    {
+      /* Add support for 3 or more limbs filled in from normal integral
+	 type if this assert fails.  If no target chooses limb mode smaller
+	 than half of largest supported normal integral type, this will not
+	 be needed.  */
+      gcc_assert (TYPE_PRECISION (type) <= 2 * limb_prec);
+      tree lhs_type = type;
+      if (TREE_CODE (type) == BITINT_TYPE
+	  && bitint_precision_kind (type) == bitint_prec_middle)
+	lhs_type = build_nonstandard_integer_type (TYPE_PRECISION (type),
+						   TYPE_UNSIGNED (type));
+      tree r1 = limb_access (NULL_TREE, var, size_int (0), true);
+      g = gimple_build_assign (make_ssa_name (m_limb_type), r1);
+      insert_before (g);
+      r1 = gimple_assign_lhs (g);
+      if (!useless_type_conversion_p (lhs_type, TREE_TYPE (r1)))
+	r1 = add_cast (lhs_type, r1);
+      if (TYPE_PRECISION (lhs_type) > limb_prec)
+	{
+	  tree r2 = limb_access (NULL_TREE, var, size_int (1), true);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type), r2);
+	  insert_before (g);
+	  r2 = gimple_assign_lhs (g);
+	  r2 = add_cast (lhs_type, r2);
+	  g = gimple_build_assign (make_ssa_name (lhs_type), LSHIFT_EXPR, r2,
+				   build_int_cst (unsigned_type_node,
+						  limb_prec));
+	  insert_before (g);
+	  g = gimple_build_assign (make_ssa_name (lhs_type), BIT_IOR_EXPR, r1,
+				   gimple_assign_lhs (g));
+	  insert_before (g);
+	  r1 = gimple_assign_lhs (g);
+	}
+      if (lhs_type != type)
+	r1 = add_cast (type, r1);
+      ovf = add_cast (lhs_type, ovf);
+      if (lhs_type != type)
+	ovf = add_cast (type, ovf);
+      g = gimple_build_assign (lhs, COMPLEX_EXPR, r1, ovf);
+      m_gsi = gsi_for_stmt (stmt);
+      gsi_replace (&m_gsi, g, true);
+    }
+  else
+    {
+      unsigned HOST_WIDE_INT nelts = 0;
+      tree atype = NULL_TREE;
+      if (obj)
+	{
+	  nelts = tree_to_uhwi (TYPE_SIZE (TREE_TYPE (obj))) / limb_prec;
+	  if (orig_obj == NULL_TREE)
+	    nelts >>= 1;
+	  atype = build_array_type_nelts (m_limb_type, nelts);
+	}
+      if (var && obj)
+	{
+	  tree v1, v2;
+	  tree zero;
+	  if (orig_obj == NULL_TREE)
+	    {
+	      zero = build_zero_cst (build_pointer_type (TREE_TYPE (obj)));
+	      v1 = build2 (MEM_REF, atype,
+			   build_fold_addr_expr (unshare_expr (obj)), zero);
+	    }
+	  else if (!useless_type_conversion_p (atype, TREE_TYPE (obj)))
+	    v1 = build1 (VIEW_CONVERT_EXPR, atype, unshare_expr (obj));
+	  else
+	    v1 = unshare_expr (obj);
+	  zero = build_zero_cst (build_pointer_type (TREE_TYPE (var)));
+	  v2 = build2 (MEM_REF, atype, build_fold_addr_expr (var), zero);
+	  g = gimple_build_assign (v1, v2);
+	  insert_before (g);
+	}
+      if (orig_obj == NULL_TREE && obj)
+	{
+	  ovf = add_cast (m_limb_type, ovf);
+	  tree l = limb_access (NULL_TREE, obj, size_int (nelts), true);
+	  g = gimple_build_assign (l, ovf);
+	  insert_before (g);
+	  if (nelts > 1)
+	    {
+	      atype = build_array_type_nelts (m_limb_type, nelts - 1);
+	      tree off = build_int_cst (build_pointer_type (TREE_TYPE (obj)),
+					(nelts + 1) * m_limb_size);
+	      tree v1 = build2 (MEM_REF, atype,
+				build_fold_addr_expr (unshare_expr (obj)),
+				off);
+	      g = gimple_build_assign (v1, build_zero_cst (atype));
+	      insert_before (g);
+	    }
+	}
+      else if (TREE_CODE (TREE_TYPE (lhs)) == COMPLEX_TYPE)
+	{
+	  imm_use_iterator ui;
+	  use_operand_p use_p;
+	  FOR_EACH_IMM_USE_FAST (use_p, ui, lhs)
+	    {
+	      g = USE_STMT (use_p);
+	      if (!is_gimple_assign (g)
+		  || gimple_assign_rhs_code (g) != IMAGPART_EXPR)
+		continue;
+	      tree lhs2 = gimple_assign_lhs (g);
+	      gimple *use_stmt;
+	      single_imm_use (lhs2, &use_p, &use_stmt);
+	      lhs2 = gimple_assign_lhs (use_stmt);
+	      gimple_stmt_iterator gsi = gsi_for_stmt (use_stmt);
+	      if (useless_type_conversion_p (TREE_TYPE (lhs2), TREE_TYPE (ovf)))
+		g = gimple_build_assign (lhs2, ovf);
+	      else
+		g = gimple_build_assign (lhs2, NOP_EXPR, ovf);
+	      gsi_replace (&gsi, g, true);
+	      break;
+	    }
+	}
+      else if (ovf != boolean_false_node)
+	{
+	  g = gimple_build_cond (NE_EXPR, ovf, boolean_false_node,
+				 NULL_TREE, NULL_TREE);
+	  edge edge_true, edge_false;
+	  if_then (g, profile_probability::very_unlikely (),
+		   edge_true, edge_false);
+	  tree zero = build_zero_cst (TREE_TYPE (lhs));
+	  tree fn = ubsan_build_overflow_builtin (code, m_loc,
+						  TREE_TYPE (lhs),
+						  zero, zero, NULL);
+	  force_gimple_operand_gsi (&m_gsi, fn, true, NULL_TREE,
+				    true, GSI_SAME_STMT);
+	  m_gsi = gsi_after_labels (edge_true->dest);
+	}
+    }
+  if (var)
+    {
+      tree clobber = build_clobber (TREE_TYPE (var), CLOBBER_EOL);
+      g = gimple_build_assign (var, clobber);
+      gsi_insert_after (&m_gsi, g, GSI_SAME_STMT);
+    }
+}
+
+/* Helper function for lower_addsub_overflow and lower_mul_overflow.
+   Given precisions of result TYPE (PREC), argument 0 precision PREC0,
+   argument 1 precision PREC1 and minimum precision for the result
+   PREC2, compute *START, *END, *CHECK_ZERO and return OVF.  */
+
+static tree
+arith_overflow (tree_code code, tree type, int prec, int prec0, int prec1,
+		int prec2, unsigned *start, unsigned *end, bool *check_zero)
+{
+  *start = 0;
+  *end = 0;
+  *check_zero = true;
+  /* Ignore this special rule for subtraction, even if both
+     prec0 >= 0 and prec1 >= 0, their subtraction can be negative
+     in infinite precision.  */
+  if (code != MINUS_EXPR && prec0 >= 0 && prec1 >= 0)
+    {
+      /* Result in [0, prec2) is unsigned, if prec > prec2,
+	 all bits above it will be zero.  */
+      if ((prec - !TYPE_UNSIGNED (type)) >= prec2)
+	return boolean_false_node;
+      else
+	{
+	  /* ovf if any of bits in [start, end) is non-zero.  */
+	  *start = prec - !TYPE_UNSIGNED (type);
+	  *end = prec2;
+	}
+    }
+  else if (TYPE_UNSIGNED (type))
+    {
+      /* If result in [0, prec2) is signed and if prec > prec2,
+	 all bits above it will be sign bit copies.  */
+      if (prec >= prec2)
+	{
+	  /* ovf if bit prec - 1 is non-zero.  */
+	  *start = prec - 1;
+	  *end = prec;
+	}
+      else
+	{
+	  /* ovf if any of bits in [start, end) is non-zero.  */
+	  *start = prec;
+	  *end = prec2;
+	}
+    }
+  else if (prec >= prec2)
+    return boolean_false_node;
+  else
+    {
+      /* ovf if [start, end) bits aren't all zeros or all ones.  */
+      *start = prec - 1;
+      *end = prec2;
+      *check_zero = false;
+    }
+  return NULL_TREE;
+}
+
+/* Lower a .{ADD,SUB}_OVERFLOW call with at least one large/huge _BitInt
+   argument or return type _Complex large/huge _BitInt.  */
+
+void
+bitint_large_huge::lower_addsub_overflow (tree obj, gimple *stmt)
+{
+  tree arg0 = gimple_call_arg (stmt, 0);
+  tree arg1 = gimple_call_arg (stmt, 1);
+  tree lhs = gimple_call_lhs (stmt);
+  gimple *g;
+
+  if (!lhs)
+    {
+      gimple_stmt_iterator gsi = gsi_for_stmt (stmt);
+      gsi_remove (&gsi, true);
+      return;
+    }
+  gimple *final_stmt = gsi_stmt (m_gsi);
+  tree type = TREE_TYPE (lhs);
+  if (TREE_CODE (type) == COMPLEX_TYPE)
+    type = TREE_TYPE (type);
+  int prec = TYPE_PRECISION (type);
+  int prec0 = range_to_prec (arg0, stmt);
+  int prec1 = range_to_prec (arg1, stmt);
+  int prec2 = ((prec0 < 0) == (prec1 < 0)
+	       ? MAX (prec0 < 0 ? -prec0 : prec0,
+		      prec1 < 0 ? -prec1 : prec1) + 1
+	       : MAX (prec0 < 0 ? -prec0 : prec0 + 1,
+		      prec1 < 0 ? -prec1 : prec1 + 1) + 1);
+  int prec3 = MAX (prec0 < 0 ? -prec0 : prec0,
+		   prec1 < 0 ? -prec1 : prec1);
+  prec3 = MAX (prec3, prec);
+  tree var = NULL_TREE;
+  tree orig_obj = obj;
+  if (obj == NULL_TREE
+      && TREE_CODE (type) == BITINT_TYPE
+      && bitint_precision_kind (type) >= bitint_prec_large
+      && m_names
+      && bitmap_bit_p (m_names, SSA_NAME_VERSION (lhs)))
+    {
+      int part = var_to_partition (m_map, lhs);
+      gcc_assert (m_vars[part] != NULL_TREE);
+      obj = m_vars[part];
+      if (TREE_TYPE (lhs) == type)
+	orig_obj = obj;
+    }
+  if (TREE_CODE (type) != BITINT_TYPE
+      || bitint_precision_kind (type) < bitint_prec_large)
+    {
+      unsigned HOST_WIDE_INT nelts = CEIL (prec, limb_prec);
+      tree atype = build_array_type_nelts (m_limb_type, nelts);
+      var = create_tmp_var (atype);
+    }
+
+  enum tree_code code;
+  switch (gimple_call_internal_fn (stmt))
+    {
+    case IFN_ADD_OVERFLOW:
+    case IFN_UBSAN_CHECK_ADD:
+      code = PLUS_EXPR;
+      break;
+    case IFN_SUB_OVERFLOW:
+    case IFN_UBSAN_CHECK_SUB:
+      code = MINUS_EXPR;
+      break;
+    default:
+      gcc_unreachable ();
+    }
+  unsigned start, end;
+  bool check_zero;
+  tree ovf = arith_overflow (code, type, prec, prec0, prec1, prec2,
+			     &start, &end, &check_zero);
+
+  unsigned startlimb, endlimb;
+  if (ovf)
+    {
+      startlimb = ~0U;
+      endlimb = ~0U;
+    }
+  else
+    {
+      startlimb = start / limb_prec;
+      endlimb = (end - 1) / limb_prec;
+    }
+
+  int prec4 = ovf != NULL_TREE ? prec : prec3;
+  bitint_prec_kind kind = bitint_precision_kind (prec4);
+  unsigned cnt, rem = 0, fin = 0;
+  tree idx = NULL_TREE, idx_first = NULL_TREE, idx_next = NULL_TREE;
+  bool last_ovf = (ovf == NULL_TREE
+		   && CEIL (prec2, limb_prec) > CEIL (prec3, limb_prec));
+  if (kind != bitint_prec_huge)
+    cnt = CEIL (prec4, limb_prec) + last_ovf;
+  else
+    {
+      rem = (prec4 % (2 * limb_prec));
+      fin = (prec4 - rem) / limb_prec;
+      cnt = 2 + CEIL (rem, limb_prec) + last_ovf;
+      idx = idx_first = create_loop (size_zero_node, &idx_next);
+    }
+
+  if (kind == bitint_prec_huge)
+    m_upwards_2limb = fin;
+  m_upwards = true;
+
+  tree type0 = TREE_TYPE (arg0);
+  tree type1 = TREE_TYPE (arg1);
+  if (TYPE_PRECISION (type0) < prec3)
+    {
+      type0 = build_bitint_type (prec3, TYPE_UNSIGNED (type0));
+      if (TREE_CODE (arg0) == INTEGER_CST)
+	arg0 = fold_convert (type0, arg0);
+    }
+  if (TYPE_PRECISION (type1) < prec3)
+    {
+      type1 = build_bitint_type (prec3, TYPE_UNSIGNED (type1));
+      if (TREE_CODE (arg1) == INTEGER_CST)
+	arg1 = fold_convert (type1, arg1);
+    }
+  unsigned int data_cnt = 0;
+  tree last_rhs1 = NULL_TREE, last_rhs2 = NULL_TREE;
+  tree cmp = build_zero_cst (m_limb_type);
+  unsigned prec_limbs = CEIL ((unsigned) prec, limb_prec);
+  tree ovf_out = NULL_TREE, cmp_out = NULL_TREE;
+  for (unsigned i = 0; i < cnt; i++)
+    {
+      m_data_cnt = 0;
+      tree rhs1, rhs2;
+      if (kind != bitint_prec_huge)
+	idx = size_int (i);
+      else if (i >= 2)
+	idx = size_int (fin + (i > 2));
+      if (!last_ovf || i < cnt - 1)
+	{
+	  if (type0 != TREE_TYPE (arg0))
+	    rhs1 = handle_cast (type0, arg0, idx);
+	  else
+	    rhs1 = handle_operand (arg0, idx);
+	  if (type1 != TREE_TYPE (arg1))
+	    rhs2 = handle_cast (type1, arg1, idx);
+	  else
+	    rhs2 = handle_operand (arg1, idx);
+	  if (i == 0)
+	    data_cnt = m_data_cnt;
+	  if (!useless_type_conversion_p (m_limb_type, TREE_TYPE (rhs1)))
+	    rhs1 = add_cast (m_limb_type, rhs1);
+	  if (!useless_type_conversion_p (m_limb_type, TREE_TYPE (rhs2)))
+	    rhs2 = add_cast (m_limb_type, rhs2);
+	  last_rhs1 = rhs1;
+	  last_rhs2 = rhs2;
+	}
+      else
+	{
+	  m_data_cnt = data_cnt;
+	  if (TYPE_UNSIGNED (type0))
+	    rhs1 = build_zero_cst (m_limb_type);
+	  else
+	    {
+	      rhs1 = add_cast (signed_type_for (m_limb_type), last_rhs1);
+	      if (TREE_CODE (rhs1) == INTEGER_CST)
+		rhs1 = build_int_cst (m_limb_type,
+				      tree_int_cst_sgn (rhs1) < 0 ? -1 : 0);
+	      else
+		{
+		  tree lpm1 = build_int_cst (unsigned_type_node,
+					     limb_prec - 1);
+		  g = gimple_build_assign (make_ssa_name (TREE_TYPE (rhs1)),
+					   RSHIFT_EXPR, rhs1, lpm1);
+		  insert_before (g);
+		  rhs1 = add_cast (m_limb_type, gimple_assign_lhs (g));
+		}
+	    }
+	  if (TYPE_UNSIGNED (type1))
+	    rhs2 = build_zero_cst (m_limb_type);
+	  else
+	    {
+	      rhs2 = add_cast (signed_type_for (m_limb_type), last_rhs2);
+	      if (TREE_CODE (rhs2) == INTEGER_CST)
+		rhs2 = build_int_cst (m_limb_type,
+				      tree_int_cst_sgn (rhs2) < 0 ? -1 : 0);
+	      else
+		{
+		  tree lpm1 = build_int_cst (unsigned_type_node,
+					     limb_prec - 1);
+		  g = gimple_build_assign (make_ssa_name (TREE_TYPE (rhs2)),
+					   RSHIFT_EXPR, rhs2, lpm1);
+		  insert_before (g);
+		  rhs2 = add_cast (m_limb_type, gimple_assign_lhs (g));
+		}
+	    }
+	}
+      tree rhs = handle_plus_minus (code, rhs1, rhs2, idx);
+      if (ovf != boolean_false_node)
+	{
+	  if (tree_fits_uhwi_p (idx))
+	    {
+	      unsigned limb = tree_to_uhwi (idx);
+	      if (limb >= startlimb && limb <= endlimb)
+		{
+		  tree l = arith_overflow_extract_bits (start, end, rhs,
+							limb, check_zero);
+		  tree this_ovf = make_ssa_name (boolean_type_node);
+		  if (ovf == NULL_TREE && !check_zero)
+		    {
+		      cmp = l;
+		      g = gimple_build_assign (make_ssa_name (m_limb_type),
+					       PLUS_EXPR, l,
+					       build_int_cst (m_limb_type, 1));
+		      insert_before (g);
+		      g = gimple_build_assign (this_ovf, GT_EXPR,
+					       gimple_assign_lhs (g),
+					       build_int_cst (m_limb_type, 1));
+		    }
+		  else
+		    g = gimple_build_assign (this_ovf, NE_EXPR, l, cmp);
+		  insert_before (g);
+		  if (ovf == NULL_TREE)
+		    ovf = this_ovf;
+		  else
+		    {
+		      tree b = make_ssa_name (boolean_type_node);
+		      g = gimple_build_assign (b, BIT_IOR_EXPR, ovf, this_ovf);
+		      insert_before (g);
+		      ovf = b;
+		    }
+		}
+	    }
+	  else if (startlimb < fin)
+	    {
+	      if (m_first && startlimb + 2 < fin)
+		{
+		  tree data_out;
+		  ovf = prepare_data_in_out (boolean_false_node, idx, &data_out);
+		  ovf_out = m_data.pop ();
+		  m_data.pop ();
+		  if (!check_zero)
+		    {
+		      cmp = prepare_data_in_out (cmp, idx, &data_out);
+		      cmp_out = m_data.pop ();
+		      m_data.pop ();
+		    }
+		}
+	      if (i != 0 || startlimb != fin - 1)
+		{
+		  tree_code cmp_code;
+		  bool single_comparison
+		    = (startlimb + 2 >= fin || (startlimb & 1) != (i & 1));
+		  if (!single_comparison)
+		    {
+		      cmp_code = GE_EXPR;
+		      if (!check_zero && (start % limb_prec) == 0)
+			single_comparison = true;
+		    }
+		  else if ((startlimb & 1) == (i & 1))
+		    cmp_code = EQ_EXPR;
+		  else
+		    cmp_code = GT_EXPR;
+		  g = gimple_build_cond (cmp_code, idx, size_int (startlimb),
+					 NULL_TREE, NULL_TREE);
+		  edge edge_true_true, edge_true_false, edge_false;
+		  gimple *g2 = NULL;
+		  if (!single_comparison)
+		    g2 = gimple_build_cond (EQ_EXPR, idx,
+					    size_int (startlimb), NULL_TREE,
+					    NULL_TREE);
+		  if_then_if_then_else (g, g2, profile_probability::likely (),
+					profile_probability::unlikely (),
+					edge_true_true, edge_true_false,
+					edge_false);
+		  unsigned tidx = startlimb + (cmp_code == GT_EXPR);
+		  tree l = arith_overflow_extract_bits (start, end, rhs, tidx,
+							check_zero);
+		  tree this_ovf = make_ssa_name (boolean_type_node);
+		  if (cmp_code != GT_EXPR && !check_zero)
+		    {
+		      g = gimple_build_assign (make_ssa_name (m_limb_type),
+					       PLUS_EXPR, l,
+					       build_int_cst (m_limb_type, 1));
+		      insert_before (g);
+		      g = gimple_build_assign (this_ovf, GT_EXPR,
+					       gimple_assign_lhs (g),
+					       build_int_cst (m_limb_type, 1));
+		    }
+		  else
+		    g = gimple_build_assign (this_ovf, NE_EXPR, l, cmp);
+		  insert_before (g);
+		  if (cmp_code == GT_EXPR)
+		    {
+		      tree t = make_ssa_name (boolean_type_node);
+		      g = gimple_build_assign (t, BIT_IOR_EXPR, ovf, this_ovf);
+		      insert_before (g);
+		      this_ovf = t;
+		    }
+		  tree this_ovf2 = NULL_TREE;
+		  if (!single_comparison)
+		    {
+		      m_gsi = gsi_after_labels (edge_true_true->src);
+		      tree t = make_ssa_name (boolean_type_node);
+		      g = gimple_build_assign (t, NE_EXPR, rhs, cmp);
+		      insert_before (g);
+		      this_ovf2 = make_ssa_name (boolean_type_node);
+		      g = gimple_build_assign (this_ovf2, BIT_IOR_EXPR,
+					       ovf, t);
+		      insert_before (g);
+		    }
+		  m_gsi = gsi_after_labels (edge_true_false->dest);
+		  tree t;
+		  if (i == 1 && ovf_out)
+		    t = ovf_out;
+		  else
+		    t = make_ssa_name (boolean_type_node);
+		  gphi *phi = create_phi_node (t, edge_true_false->dest);
+		  add_phi_arg (phi, this_ovf, edge_true_false,
+			       UNKNOWN_LOCATION);
+		  add_phi_arg (phi, ovf ? ovf
+				    : boolean_false_node, edge_false,
+			       UNKNOWN_LOCATION);
+		  if (edge_true_true)
+		    add_phi_arg (phi, this_ovf2, edge_true_true,
+				 UNKNOWN_LOCATION);
+		  ovf = t;
+		  if (!check_zero && cmp_code != GT_EXPR)
+		    {
+		      t = cmp_out ? cmp_out : make_ssa_name (m_limb_type);
+		      phi = create_phi_node (t, edge_true_false->dest);
+		      add_phi_arg (phi, l, edge_true_false, UNKNOWN_LOCATION);
+		      add_phi_arg (phi, cmp, edge_false, UNKNOWN_LOCATION);
+		      if (edge_true_true)
+			add_phi_arg (phi, cmp, edge_true_true,
+				     UNKNOWN_LOCATION);
+		      cmp = t;
+		    }
+		}
+	    }
+	}
+
+      if (var || obj)
+	{
+	  if (tree_fits_uhwi_p (idx) && tree_to_uhwi (idx) >= prec_limbs)
+	    ;
+	  else if (!tree_fits_uhwi_p (idx)
+		   && (unsigned) prec < (fin - (i == 0)) * limb_prec)
+	    {
+	      bool single_comparison
+		= (((unsigned) prec % limb_prec) == 0
+		   || prec_limbs + 1 >= fin
+		   || (prec_limbs & 1) == (i & 1));
+	      g = gimple_build_cond (LE_EXPR, idx, size_int (prec_limbs - 1),
+				     NULL_TREE, NULL_TREE);
+	      gimple *g2 = NULL;
+	      if (!single_comparison)
+		g2 = gimple_build_cond (LT_EXPR, idx,
+					size_int (prec_limbs - 1),
+					NULL_TREE, NULL_TREE);
+	      edge edge_true_true, edge_true_false, edge_false;
+	      if_then_if_then_else (g, g2, profile_probability::likely (),
+				    profile_probability::likely (),
+				    edge_true_true, edge_true_false,
+				    edge_false);
+	      tree l = limb_access (type, var ? var : obj, idx, true);
+	      g = gimple_build_assign (l, rhs);
+	      insert_before (g);
+	      if (!single_comparison)
+		{
+		  m_gsi = gsi_after_labels (edge_true_true->src);
+		  l = limb_access (type, var ? var : obj,
+				   size_int (prec_limbs - 1), true);
+		  if (!useless_type_conversion_p (TREE_TYPE (l),
+						  TREE_TYPE (rhs)))
+		    rhs = add_cast (TREE_TYPE (l), rhs);
+		  g = gimple_build_assign (l, rhs);
+		  insert_before (g);
+		}
+	      m_gsi = gsi_after_labels (edge_true_false->dest);
+	    }
+	  else
+	    {
+	      tree l = limb_access (type, var ? var : obj, idx, true);
+	      if (!useless_type_conversion_p (TREE_TYPE (l), TREE_TYPE (rhs)))
+		rhs = add_cast (TREE_TYPE (l), rhs);
+	      g = gimple_build_assign (l, rhs);
+	      insert_before (g);
+	    }
+	}
+      m_first = false;
+      if (kind == bitint_prec_huge && i <= 1)
+	{
+	  if (i == 0)
+	    {
+	      idx = make_ssa_name (sizetype);
+	      g = gimple_build_assign (idx, PLUS_EXPR, idx_first,
+				       size_one_node);
+	      insert_before (g);
+	    }
+	  else
+	    {
+	      g = gimple_build_assign (idx_next, PLUS_EXPR, idx_first,
+				       size_int (2));
+	      insert_before (g);
+	      g = gimple_build_cond (NE_EXPR, idx_next, size_int (fin),
+				     NULL_TREE, NULL_TREE);
+	      insert_before (g);
+	      m_gsi = gsi_for_stmt (final_stmt);
+	    }
+	}
+    }
+
+  finish_arith_overflow (var, obj, type, ovf, lhs, orig_obj, stmt, code);
+}
+
+/* Lower a .MUL_OVERFLOW call with at least one large/huge _BitInt
+   argument or return type _Complex large/huge _BitInt.  */
+
+void
+bitint_large_huge::lower_mul_overflow (tree obj, gimple *stmt)
+{
+  tree arg0 = gimple_call_arg (stmt, 0);
+  tree arg1 = gimple_call_arg (stmt, 1);
+  tree lhs = gimple_call_lhs (stmt);
+  if (!lhs)
+    {
+      gimple_stmt_iterator gsi = gsi_for_stmt (stmt);
+      gsi_remove (&gsi, true);
+      return;
+    }
+  gimple *final_stmt = gsi_stmt (m_gsi);
+  tree type = TREE_TYPE (lhs);
+  if (TREE_CODE (type) == COMPLEX_TYPE)
+    type = TREE_TYPE (type);
+  int prec = TYPE_PRECISION (type), prec0, prec1;
+  arg0 = handle_operand_addr (arg0, stmt, NULL, &prec0);
+  arg1 = handle_operand_addr (arg1, stmt, NULL, &prec1);
+  int prec2 = ((prec0 < 0 ? -prec0 : prec0)
+	       + (prec1 < 0 ? -prec1 : prec1)
+	       + ((prec0 < 0) != (prec1 < 0)));
+  tree var = NULL_TREE;
+  tree orig_obj = obj;
+  bool force_var = false;
+  if (obj == NULL_TREE
+      && TREE_CODE (type) == BITINT_TYPE
+      && bitint_precision_kind (type) >= bitint_prec_large
+      && m_names
+      && bitmap_bit_p (m_names, SSA_NAME_VERSION (lhs)))
+    {
+      int part = var_to_partition (m_map, lhs);
+      gcc_assert (m_vars[part] != NULL_TREE);
+      obj = m_vars[part];
+      if (TREE_TYPE (lhs) == type)
+	orig_obj = obj;
+    }
+  else if (obj != NULL_TREE && DECL_P (obj))
+    {
+      for (int i = 0; i < 2; ++i)
+	{
+	  tree arg = i ? arg1 : arg0;
+	  if (TREE_CODE (arg) == ADDR_EXPR)
+	    arg = TREE_OPERAND (arg, 0);
+	  if (get_base_address (arg) == obj)
+	    {
+	      force_var = true;
+	      break;
+	    }
+	}
+    }
+  if (obj == NULL_TREE
+      || force_var
+      || TREE_CODE (type) != BITINT_TYPE
+      || bitint_precision_kind (type) < bitint_prec_large
+      || prec2 > (CEIL (prec, limb_prec) * limb_prec * (orig_obj ? 1 : 2)))
+    {
+      unsigned HOST_WIDE_INT nelts = CEIL (MAX (prec, prec2), limb_prec);
+      tree atype = build_array_type_nelts (m_limb_type, nelts);
+      var = create_tmp_var (atype);
+    }
+  tree addr = build_fold_addr_expr (var ? var : obj);
+  addr = force_gimple_operand_gsi (&m_gsi, addr, true,
+				   NULL_TREE, true, GSI_SAME_STMT);
+  tree sitype = lang_hooks.types.type_for_mode (SImode, 0);
+  gimple *g
+    = gimple_build_call_internal (IFN_MULBITINT, 6,
+				  addr, build_int_cst (sitype,
+						       MAX (prec2, prec)),
+				  arg0, build_int_cst (sitype, prec0),
+				  arg1, build_int_cst (sitype, prec1));
+  insert_before (g);
+
+  unsigned start, end;
+  bool check_zero;
+  tree ovf = arith_overflow (MULT_EXPR, type, prec, prec0, prec1, prec2,
+			     &start, &end, &check_zero);
+  if (ovf == NULL_TREE)
+    {
+      unsigned startlimb = start / limb_prec;
+      unsigned endlimb = (end - 1) / limb_prec;
+      unsigned cnt;
+      bool use_loop = false;
+      if (startlimb == endlimb)
+	cnt = 1;
+      else if (startlimb + 1 == endlimb)
+	cnt = 2;
+      else if ((end % limb_prec) == 0)
+	{
+	  cnt = 2;
+	  use_loop = true;
+	}
+      else
+	{
+	  cnt = 3;
+	  use_loop = startlimb + 2 < endlimb;
+	}
+      if (cnt == 1)
+	{
+	  tree l = limb_access (NULL_TREE, var ? var : obj,
+				size_int (startlimb), true);
+	  g = gimple_build_assign (make_ssa_name (m_limb_type), l);
+	  insert_before (g);
+	  l = arith_overflow_extract_bits (start, end, gimple_assign_lhs (g),
+					   startlimb, check_zero);
+	  ovf = make_ssa_name (boolean_type_node);
+	  if (check_zero)
+	    g = gimple_build_assign (ovf, NE_EXPR, l,
+				     build_zero_cst (m_limb_type));
+	  else
+	    {
+	      g = gimple_build_assign (make_ssa_name (m_limb_type),
+				       PLUS_EXPR, l,
+				       build_int_cst (m_limb_type, 1));
+	      insert_before (g);
+	      g = gimple_build_assign (ovf, GT_EXPR, gimple_assign_lhs (g),
+				       build_int_cst (m_limb_type, 1));
+	    }
+	  insert_before (g);
+	}
+      else
+	{
+	  basic_block edge_bb = NULL;
+	  gimple_stmt_iterator gsi = m_gsi;
+	  gsi_prev (&gsi);
+	  edge e = split_block (gsi_bb (gsi), gsi_stmt (gsi));
+	  edge_bb = e->src;
+	  m_gsi = gsi_last_bb (edge_bb);
+	  if (!gsi_end_p (m_gsi))
+	    gsi_next (&m_gsi);
+
+	  tree cmp = build_zero_cst (m_limb_type);
+	  for (unsigned i = 0; i < cnt; i++)
+	    {
+	      tree idx, idx_next = NULL_TREE;
+	      if (i == 0)
+		idx = size_int (startlimb);
+	      else if (i == 2)
+		idx = size_int (endlimb);
+	      else if (use_loop)
+		idx = create_loop (size_int (startlimb + 1), &idx_next);
+	      else
+		idx = size_int (startlimb + 1);
+	      tree l = limb_access (NULL_TREE, var ? var : obj, idx, true);
+	      g = gimple_build_assign (make_ssa_name (m_limb_type), l);
+	      insert_before (g);
+	      l = gimple_assign_lhs (g);
+	      if (i == 0 || i == 2)
+		l = arith_overflow_extract_bits (start, end, l,
+						 tree_to_uhwi (idx),
+						 check_zero);
+	      if (i == 0 && !check_zero)
+		{
+		  cmp = l;
+		  g = gimple_build_assign (make_ssa_name (m_limb_type),
+					   PLUS_EXPR, l,
+					   build_int_cst (m_limb_type, 1));
+		  insert_before (g);
+		  g = gimple_build_cond (GT_EXPR, gimple_assign_lhs (g),
+					 build_int_cst (m_limb_type, 1),
+					 NULL_TREE, NULL_TREE);
+		}
+	      else
+		g = gimple_build_cond (NE_EXPR, l, cmp, NULL_TREE, NULL_TREE);
+	      insert_before (g);
+	      edge e1 = split_block (gsi_bb (m_gsi), g);
+	      e1->flags = EDGE_FALSE_VALUE;
+	      edge e2 = make_edge (e1->src, gimple_bb (final_stmt),
+				   EDGE_TRUE_VALUE);
+	      e1->probability = profile_probability::likely ();
+	      e2->probability = e1->probability.invert ();
+	      if (i == 0)
+		set_immediate_dominator (CDI_DOMINATORS, e2->dest, e2->src);
+	      m_gsi = gsi_after_labels (e1->dest);
+	      if (i == 1 && use_loop)
+		{
+		  g = gimple_build_assign (idx_next, PLUS_EXPR, idx,
+					   size_one_node);
+		  insert_before (g);
+		  g = gimple_build_cond (NE_EXPR, idx_next,
+					 size_int (endlimb + (cnt == 1)),
+					 NULL_TREE, NULL_TREE);
+		  insert_before (g);
+		  edge true_edge, false_edge;
+		  extract_true_false_edges_from_block (gsi_bb (m_gsi),
+						       &true_edge,
+						       &false_edge);
+		  m_gsi = gsi_after_labels (false_edge->dest);
+		}
+	    }
+
+	  ovf = make_ssa_name (boolean_type_node);
+	  basic_block bb = gimple_bb (final_stmt);
+	  gphi *phi = create_phi_node (ovf, bb);
+	  edge e1 = find_edge (gsi_bb (m_gsi), bb);
+	  edge_iterator ei;
+	  FOR_EACH_EDGE (e, ei, bb->preds)
+	    {
+	      tree val = e == e1 ? boolean_false_node : boolean_true_node;
+	      add_phi_arg (phi, val, e, UNKNOWN_LOCATION);
+	    }
+	  m_gsi = gsi_for_stmt (final_stmt);
+	}
+    }
+
+  finish_arith_overflow (var, obj, type, ovf, lhs, orig_obj, stmt, MULT_EXPR);
+}
+
+/* Lower REALPART_EXPR or IMAGPART_EXPR stmt extracting part of result from
+   .{ADD,SUB,MUL}_OVERFLOW call.  */
+
+void
+bitint_large_huge::lower_cplxpart_stmt (tree obj, gimple *stmt)
+{
+  tree rhs1 = gimple_assign_rhs1 (stmt);
+  rhs1 = TREE_OPERAND (rhs1, 0);
+  if (obj == NULL_TREE)
+    {
+      int part = var_to_partition (m_map, gimple_assign_lhs (stmt));
+      gcc_assert (m_vars[part] != NULL_TREE);
+      obj = m_vars[part];
+    }
+  if (TREE_CODE (rhs1) == SSA_NAME
+      && (m_names == NULL
+	  || !bitmap_bit_p (m_names, SSA_NAME_VERSION (rhs1))))
+    {
+      lower_call (obj, SSA_NAME_DEF_STMT (rhs1));
+      return;
+    }
+  int part = var_to_partition (m_map, rhs1);
+  gcc_assert (m_vars[part] != NULL_TREE);
+  tree var = m_vars[part];
+  unsigned HOST_WIDE_INT nelts
+    = tree_to_uhwi (TYPE_SIZE (TREE_TYPE (obj))) / limb_prec;
+  tree atype = build_array_type_nelts (m_limb_type, nelts);
+  if (!useless_type_conversion_p (atype, TREE_TYPE (obj)))
+    obj = build1 (VIEW_CONVERT_EXPR, atype, obj);
+  tree off = build_int_cst (build_pointer_type (TREE_TYPE (var)),
+			    gimple_assign_rhs_code (stmt) == REALPART_EXPR
+			    ? 0 : nelts * m_limb_size);
+  tree v2 = build2 (MEM_REF, atype, build_fold_addr_expr (var), off);
+  gimple *g = gimple_build_assign (obj, v2);
+  insert_before (g);
+}
+
+/* Lower COMPLEX_EXPR stmt.  */
+
+void
+bitint_large_huge::lower_complexexpr_stmt (gimple *stmt)
+{
+  tree lhs = gimple_assign_lhs (stmt);
+  tree rhs1 = gimple_assign_rhs1 (stmt);
+  tree rhs2 = gimple_assign_rhs2 (stmt);
+  int part = var_to_partition (m_map, lhs);
+  gcc_assert (m_vars[part] != NULL_TREE);
+  lhs = m_vars[part];
+  unsigned HOST_WIDE_INT nelts
+    = tree_to_uhwi (TYPE_SIZE (TREE_TYPE (rhs1))) / limb_prec;
+  tree atype = build_array_type_nelts (m_limb_type, nelts);
+  tree zero = build_zero_cst (build_pointer_type (TREE_TYPE (lhs)));
+  tree v1 = build2 (MEM_REF, atype, build_fold_addr_expr (lhs), zero);
+  tree v2;
+  if (TREE_CODE (rhs1) == SSA_NAME)
+    {
+      part = var_to_partition (m_map, rhs1);
+      gcc_assert (m_vars[part] != NULL_TREE);
+      v2 = m_vars[part];
+    }
+  else if (integer_zerop (rhs1))
+    v2 = build_zero_cst (atype);
+  else
+    v2 = tree_output_constant_def (rhs1);
+  if (!useless_type_conversion_p (atype, TREE_TYPE (v2)))
+    v2 = build1 (VIEW_CONVERT_EXPR, atype, v2);
+  gimple *g = gimple_build_assign (v1, v2);
+  insert_before (g);
+  tree off = fold_convert (build_pointer_type (TREE_TYPE (lhs)),
+			   TYPE_SIZE_UNIT (atype));
+  v1 = build2 (MEM_REF, atype, build_fold_addr_expr (lhs), off);
+  if (TREE_CODE (rhs2) == SSA_NAME)
+    {
+      part = var_to_partition (m_map, rhs2);
+      gcc_assert (m_vars[part] != NULL_TREE);
+      v2 = m_vars[part];
+    }
+  else if (integer_zerop (rhs2))
+    v2 = build_zero_cst (atype);
+  else
+    v2 = tree_output_constant_def (rhs2);
+  if (!useless_type_conversion_p (atype, TREE_TYPE (v2)))
+    v2 = build1 (VIEW_CONVERT_EXPR, atype, v2);
+  g = gimple_build_assign (v1, v2);
+  insert_before (g);
+}
+
+/* Lower a call statement with one or more large/huge _BitInt
+   arguments or large/huge _BitInt return value.  */
+
+void
+bitint_large_huge::lower_call (tree obj, gimple *stmt)
+{
+  gimple_stmt_iterator gsi = gsi_for_stmt (stmt);
+  unsigned int nargs = gimple_call_num_args (stmt);
+  if (gimple_call_internal_p (stmt))
+    switch (gimple_call_internal_fn (stmt))
+      {
+      case IFN_ADD_OVERFLOW:
+      case IFN_SUB_OVERFLOW:
+      case IFN_UBSAN_CHECK_ADD:
+      case IFN_UBSAN_CHECK_SUB:
+	lower_addsub_overflow (obj, stmt);
+	return;
+      case IFN_MUL_OVERFLOW:
+      case IFN_UBSAN_CHECK_MUL:
+	lower_mul_overflow (obj, stmt);
+	return;
+      default:
+	break;
+      }
+  for (unsigned int i = 0; i < nargs; ++i)
+    {
+      tree arg = gimple_call_arg (stmt, i);
+      if (TREE_CODE (arg) != SSA_NAME
+	  || TREE_CODE (TREE_TYPE (arg)) != BITINT_TYPE
+	  || bitint_precision_kind (TREE_TYPE (arg)) <= bitint_prec_middle)
+	continue;
+      int p = var_to_partition (m_map, arg);
+      tree v = m_vars[p];
+      gcc_assert (v != NULL_TREE);
+      if (!types_compatible_p (TREE_TYPE (arg), TREE_TYPE (v)))
+	v = build1 (VIEW_CONVERT_EXPR, TREE_TYPE (arg), v);
+      arg = make_ssa_name (TREE_TYPE (arg));
+      gimple *g = gimple_build_assign (arg, v);
+      gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+      gimple_call_set_arg (stmt, i, arg);
+      if (m_preserved == NULL)
+	m_preserved = BITMAP_ALLOC (NULL);
+      bitmap_set_bit (m_preserved, SSA_NAME_VERSION (arg));
+    }
+  tree lhs = gimple_call_lhs (stmt);
+  if (lhs
+      && TREE_CODE (lhs) == SSA_NAME
+      && TREE_CODE (TREE_TYPE (lhs)) == BITINT_TYPE
+      && bitint_precision_kind (TREE_TYPE (lhs)) >= bitint_prec_large)
+    {
+      int p = var_to_partition (m_map, lhs);
+      tree v = m_vars[p];
+      gcc_assert (v != NULL_TREE);
+      if (!types_compatible_p (TREE_TYPE (lhs), TREE_TYPE (v)))
+	v = build1 (VIEW_CONVERT_EXPR, TREE_TYPE (lhs), v);
+      gimple_call_set_lhs (stmt, v);
+      SSA_NAME_DEF_STMT (lhs) = gimple_build_nop ();
+    }
+  update_stmt (stmt);
+}
+
+/* Lower __asm STMT which involves large/huge _BitInt values.  */
+
+void
+bitint_large_huge::lower_asm (gimple *stmt)
+{
+  gasm *g = as_a <gasm *> (stmt);
+  unsigned noutputs = gimple_asm_noutputs (g);
+  unsigned ninputs = gimple_asm_ninputs (g);
+
+  for (unsigned i = 0; i < noutputs; ++i)
+    {
+      tree t = gimple_asm_output_op (g, i);
+      tree s = TREE_VALUE (t);
+      if (TREE_CODE (s) == SSA_NAME
+	  && TREE_CODE (TREE_TYPE (s)) == BITINT_TYPE
+	  && bitint_precision_kind (TREE_TYPE (s)) >= bitint_prec_large)
+	{
+	  int part = var_to_partition (m_map, s);
+	  gcc_assert (m_vars[part] != NULL_TREE);
+	  TREE_VALUE (t) = m_vars[part];
+	}
+    }
+  for (unsigned i = 0; i < ninputs; ++i)
+    {
+      tree t = gimple_asm_input_op (g, i);
+      tree s = TREE_VALUE (t);
+      if (TREE_CODE (s) == SSA_NAME
+	  && TREE_CODE (TREE_TYPE (s)) == BITINT_TYPE
+	  && bitint_precision_kind (TREE_TYPE (s)) >= bitint_prec_large)
+	{
+	  int part = var_to_partition (m_map, s);
+	  gcc_assert (m_vars[part] != NULL_TREE);
+	  TREE_VALUE (t) = m_vars[part];
+	}
+    }
+  update_stmt (stmt);
+}
+
+/* Lower statement STMT which involves large/huge _BitInt values
+   into code accessing individual limbs.  */
+
+void
+bitint_large_huge::lower_stmt (gimple *stmt)
+{
+  m_first = true;
+  m_lhs = NULL_TREE;
+  m_data.truncate (0);
+  m_data_cnt = 0;
+  m_gsi = gsi_for_stmt (stmt);
+  m_after_stmt = NULL;
+  m_bb = NULL;
+  m_init_gsi = m_gsi;
+  gsi_prev (&m_init_gsi);
+  m_preheader_bb = NULL;
+  m_upwards_2limb = 0;
+  m_upwards = false;
+  m_var_msb = false;
+  m_cast_conditional = false;
+  m_bitfld_load = 0;
+  m_loc = gimple_location (stmt);
+  if (is_gimple_call (stmt))
+    {
+      lower_call (NULL_TREE, stmt);
+      return;
+    }
+  if (gimple_code (stmt) == GIMPLE_ASM)
+    {
+      lower_asm (stmt);
+      return;
+    }
+  tree lhs = NULL_TREE, cmp_op1 = NULL_TREE, cmp_op2 = NULL_TREE;
+  tree_code cmp_code = comparison_op (stmt, &cmp_op1, &cmp_op2);
+  bool eq_p = (cmp_code == EQ_EXPR || cmp_code == NE_EXPR);
+  bool mergeable_cast_p = false;
+  bool final_cast_p = false;
+  if (gimple_assign_cast_p (stmt))
+    {
+      lhs = gimple_assign_lhs (stmt);
+      tree rhs1 = gimple_assign_rhs1 (stmt);
+      if (TREE_CODE (TREE_TYPE (lhs)) == BITINT_TYPE
+	  && bitint_precision_kind (TREE_TYPE (lhs)) >= bitint_prec_large
+	  && INTEGRAL_TYPE_P (TREE_TYPE (rhs1)))
+	mergeable_cast_p = true;
+      else if (TREE_CODE (TREE_TYPE (rhs1)) == BITINT_TYPE
+	       && bitint_precision_kind (TREE_TYPE (rhs1)) >= bitint_prec_large
+	       && INTEGRAL_TYPE_P (TREE_TYPE (lhs)))
+	{
+	  final_cast_p = true;
+	  if (TREE_CODE (rhs1) == SSA_NAME
+	      && (m_names == NULL
+		  || !bitmap_bit_p (m_names, SSA_NAME_VERSION (rhs1))))
+	    {
+	      gimple *g = SSA_NAME_DEF_STMT (rhs1);
+	      if (is_gimple_assign (g)
+		  && gimple_assign_rhs_code (g) == IMAGPART_EXPR)
+		{
+		  tree rhs2 = TREE_OPERAND (gimple_assign_rhs1 (g), 0);
+		  if (TREE_CODE (rhs2) == SSA_NAME
+		      && (m_names == NULL
+			  || !bitmap_bit_p (m_names, SSA_NAME_VERSION (rhs2))))
+		    {
+		      g = SSA_NAME_DEF_STMT (rhs2);
+		      int ovf = optimizable_arith_overflow (g);
+		      if (ovf == 2)
+			/* If .{ADD,SUB,MUL}_OVERFLOW has both REALPART_EXPR
+			   and IMAGPART_EXPR uses, where the latter is cast to
+			   non-_BitInt, it will be optimized when handling
+			   the REALPART_EXPR.  */
+			return;
+		      if (ovf == 1)
+			{
+			  lower_call (NULL_TREE, g);
+			  return;
+			}
+		    }
+		}
+	    }
+	}
+    }
+  if (gimple_store_p (stmt))
+    {
+      tree rhs1 = gimple_assign_rhs1 (stmt);
+      if (TREE_CODE (rhs1) == SSA_NAME
+	  && (m_names == NULL
+	      || !bitmap_bit_p (m_names, SSA_NAME_VERSION (rhs1))))
+	{
+	  gimple *g = SSA_NAME_DEF_STMT (rhs1);
+	  m_loc = gimple_location (g);
+	  lhs = gimple_assign_lhs (stmt);
+	  if (is_gimple_assign (g) && !mergeable_op (g))
+	    switch (gimple_assign_rhs_code (g))
+	      {
+	      case LSHIFT_EXPR:
+	      case RSHIFT_EXPR:
+		lower_shift_stmt (lhs, g);
+	      handled:
+		m_gsi = gsi_for_stmt (stmt);
+		unlink_stmt_vdef (stmt);
+		release_ssa_name (gimple_vdef (stmt));
+		gsi_remove (&m_gsi, true);
+		return;
+	      case MULT_EXPR:
+	      case TRUNC_DIV_EXPR:
+	      case TRUNC_MOD_EXPR:
+		lower_muldiv_stmt (lhs, g);
+		goto handled;
+	      case FIX_TRUNC_EXPR:
+		lower_float_conv_stmt (lhs, g);
+		goto handled;
+	      case REALPART_EXPR:
+	      case IMAGPART_EXPR:
+		lower_cplxpart_stmt (lhs, g);
+		goto handled;
+	      default:
+		break;
+	      }
+	  else if (optimizable_arith_overflow (g) == 3)
+	    {
+	      lower_call (lhs, g);
+	      goto handled;
+	    }
+	  m_loc = gimple_location (stmt);
+	}
+    }
+  if (mergeable_op (stmt)
+      || gimple_store_p (stmt)
+      || gimple_assign_load_p (stmt)
+      || eq_p
+      || mergeable_cast_p)
+    {
+      lhs = lower_mergeable_stmt (stmt, cmp_code, cmp_op1, cmp_op2);
+      if (!eq_p)
+	return;
+    }
+  else if (cmp_code != ERROR_MARK)
+    lhs = lower_comparison_stmt (stmt, cmp_code, cmp_op1, cmp_op2);
+  if (cmp_code != ERROR_MARK)
+    {
+      if (gimple_code (stmt) == GIMPLE_COND)
+	{
+	  gcond *cstmt = as_a <gcond *> (stmt);
+	  gimple_cond_set_lhs (cstmt, lhs);
+	  gimple_cond_set_rhs (cstmt, boolean_false_node);
+	  gimple_cond_set_code (cstmt, cmp_code);
+	  update_stmt (stmt);
+	  return;
+	}
+      if (gimple_assign_rhs_code (stmt) == COND_EXPR)
+	{
+	  tree cond = build2 (cmp_code, boolean_type_node, lhs,
+			      boolean_false_node);
+	  gimple_assign_set_rhs1 (stmt, cond);
+	  lhs = gimple_assign_lhs (stmt);
+	  gcc_assert (TREE_CODE (TREE_TYPE (lhs)) != BITINT_TYPE
+		      || (bitint_precision_kind (TREE_TYPE (lhs))
+			  <= bitint_prec_middle));
+	  update_stmt (stmt);
+	  return;
+	}
+      gimple_assign_set_rhs1 (stmt, lhs);
+      gimple_assign_set_rhs2 (stmt, boolean_false_node);
+      gimple_assign_set_rhs_code (stmt, cmp_code);
+      update_stmt (stmt);
+      return;
+    }
+  if (final_cast_p)
+    {
+      tree lhs_type = TREE_TYPE (lhs);
+      /* Add support for 3 or more limbs filled in from normal integral
+	 type if this assert fails.  If no target chooses limb mode smaller
+	 than half of largest supported normal integral type, this will not
+	 be needed.  */
+      gcc_assert (TYPE_PRECISION (lhs_type) <= 2 * limb_prec);
+      gimple *g;
+      if (TREE_CODE (lhs_type) == BITINT_TYPE
+	  && bitint_precision_kind (lhs_type) == bitint_prec_middle)
+	lhs_type = build_nonstandard_integer_type (TYPE_PRECISION (lhs_type),
+						   TYPE_UNSIGNED (lhs_type));
+      m_data_cnt = 0;
+      tree rhs1 = gimple_assign_rhs1 (stmt);
+      tree r1 = handle_operand (rhs1, size_int (0));
+      if (!useless_type_conversion_p (lhs_type, TREE_TYPE (r1)))
+	r1 = add_cast (lhs_type, r1);
+      if (TYPE_PRECISION (lhs_type) > limb_prec)
+	{
+	  m_data_cnt = 0;
+	  m_first = false;
+	  tree r2 = handle_operand (rhs1, size_int (1));
+	  r2 = add_cast (lhs_type, r2);
+	  g = gimple_build_assign (make_ssa_name (lhs_type), LSHIFT_EXPR, r2,
+				   build_int_cst (unsigned_type_node,
+						  limb_prec));
+	  insert_before (g);
+	  g = gimple_build_assign (make_ssa_name (lhs_type), BIT_IOR_EXPR, r1,
+				   gimple_assign_lhs (g));
+	  insert_before (g);
+	  r1 = gimple_assign_lhs (g);
+	}
+      if (lhs_type != TREE_TYPE (lhs))
+	g = gimple_build_assign (lhs, NOP_EXPR, r1);
+      else
+	g = gimple_build_assign (lhs, r1);
+      gsi_replace (&m_gsi, g, true);
+      return;
+    }
+  if (is_gimple_assign (stmt))
+    switch (gimple_assign_rhs_code (stmt))
+      {
+      case LSHIFT_EXPR:
+      case RSHIFT_EXPR:
+	lower_shift_stmt (NULL_TREE, stmt);
+	return;
+      case MULT_EXPR:
+      case TRUNC_DIV_EXPR:
+      case TRUNC_MOD_EXPR:
+	lower_muldiv_stmt (NULL_TREE, stmt);
+	return;
+      case FIX_TRUNC_EXPR:
+      case FLOAT_EXPR:
+	lower_float_conv_stmt (NULL_TREE, stmt);
+	return;
+      case REALPART_EXPR:
+      case IMAGPART_EXPR:
+	lower_cplxpart_stmt (NULL_TREE, stmt);
+	return;
+      case COMPLEX_EXPR:
+	lower_complexexpr_stmt (stmt);
+	return;
+      default:
+	break;
+      }
+  gcc_unreachable ();
+}
+
+/* Helper for walk_non_aliased_vuses.  Determine if we arrived at
+   the desired memory state.  */
+
+void *
+vuse_eq (ao_ref *, tree vuse1, void *data)
+{
+  tree vuse2 = (tree) data;
+  if (vuse1 == vuse2)
+    return data;
+
+  return NULL;
+}
+
+/* Return true if STMT uses a library function and needs to take
+   address of its inputs.  We need to avoid bit-fields in those
+   cases.  */
+
+bool
+stmt_needs_operand_addr (gimple *stmt)
+{
+  if (is_gimple_assign (stmt))
+    switch (gimple_assign_rhs_code (stmt))
+      {
+      case MULT_EXPR:
+      case TRUNC_DIV_EXPR:
+      case TRUNC_MOD_EXPR:
+      case FLOAT_EXPR:
+	return true;
+      default:
+	break;
+      }
+  else if (gimple_call_internal_p (stmt, IFN_MUL_OVERFLOW)
+	   || gimple_call_internal_p (stmt, IFN_UBSAN_CHECK_MUL))
+    return true;
+  return false;
+}
+
+/* Dominator walker used to discover which large/huge _BitInt
+   loads could be sunk into all their uses.  */
+
+class bitint_dom_walker : public dom_walker
+{
+public:
+  bitint_dom_walker (bitmap names, bitmap loads)
+    : dom_walker (CDI_DOMINATORS), m_names (names), m_loads (loads) {}
+
+  edge before_dom_children (basic_block) final override;
+
+private:
+  bitmap m_names, m_loads;
+};
+
+edge
+bitint_dom_walker::before_dom_children (basic_block bb)
+{
+  gphi *phi = get_virtual_phi (bb);
+  tree vop;
+  if (phi)
+    vop = gimple_phi_result (phi);
+  else if (bb == ENTRY_BLOCK_PTR_FOR_FN (cfun))
+    vop = NULL_TREE;
+  else
+    vop = (tree) get_immediate_dominator (CDI_DOMINATORS, bb)->aux;
+
+  auto_vec<tree, 16> worklist;
+  for (gimple_stmt_iterator gsi = gsi_start_bb (bb);
+       !gsi_end_p (gsi); gsi_next (&gsi))
+    {
+      gimple *stmt = gsi_stmt (gsi);
+      if (is_gimple_debug (stmt))
+	continue;
+
+      if (!vop && gimple_vuse (stmt))
+	vop = gimple_vuse (stmt);
+
+      tree cvop = vop;
+      if (gimple_vdef (stmt))
+	vop = gimple_vdef (stmt);
+
+      tree lhs = gimple_get_lhs (stmt);
+      if (lhs
+	  && TREE_CODE (lhs) == SSA_NAME
+	  && TREE_CODE (TREE_TYPE (lhs)) != BITINT_TYPE
+	  && bitint_precision_kind (TREE_TYPE (lhs)) >= bitint_prec_large
+	  && !bitmap_bit_p (m_names, SSA_NAME_VERSION (lhs)))
+	/* If lhs of stmt is large/huge _BitInt SSA_NAME not in m_names,
+	   it means it will be handled in a loop or straight line code
+	   at the location of its (ultimate) immediate use, so for
+	   vop checking purposes check these only at the ultimate
+	   immediate use.  */
+	continue;
+
+      ssa_op_iter oi;
+      use_operand_p use_p;
+      FOR_EACH_SSA_USE_OPERAND (use_p, stmt, oi, SSA_OP_USE)
+	{
+	  tree s = USE_FROM_PTR (use_p);
+	  if (TREE_CODE (TREE_TYPE (s)) == BITINT_TYPE
+	      && bitint_precision_kind (TREE_TYPE (s)) >= bitint_prec_large)
+	    worklist.safe_push (s);
+	}
+
+      bool needs_operand_addr = stmt_needs_operand_addr (stmt);
+      while (worklist.length () > 0)
+	{
+	  tree s = worklist.pop ();
+
+	  if (!bitmap_bit_p (m_names, SSA_NAME_VERSION (s)))
+	    {
+	      gimple *g = SSA_NAME_DEF_STMT (s);
+	      needs_operand_addr |= stmt_needs_operand_addr (g);
+	      FOR_EACH_SSA_USE_OPERAND (use_p, g, oi, SSA_OP_USE)
+		{
+		  tree s2 = USE_FROM_PTR (use_p);
+		  if (TREE_CODE (TREE_TYPE (s2)) == BITINT_TYPE
+		      && (bitint_precision_kind (TREE_TYPE (s2))
+			  >= bitint_prec_large))
+		    worklist.safe_push (s2);
+		}
+	      continue;
+	    }
+	  if (!SSA_NAME_OCCURS_IN_ABNORMAL_PHI (s)
+	      && gimple_assign_cast_p (SSA_NAME_DEF_STMT (s)))
+	    {
+	      tree rhs = gimple_assign_rhs1 (SSA_NAME_DEF_STMT (s));
+	      if (TREE_CODE (rhs) == SSA_NAME
+		  && bitmap_bit_p (m_loads, SSA_NAME_VERSION (rhs)))
+		s = rhs;
+	      else
+		continue;
+	    }
+	  else if (!bitmap_bit_p (m_loads, SSA_NAME_VERSION (s)))
+	    continue;
+
+	  tree rhs1 = gimple_assign_rhs1 (SSA_NAME_DEF_STMT (s));
+	  if (needs_operand_addr
+	      && TREE_CODE (rhs1) == COMPONENT_REF
+	      && DECL_BIT_FIELD_TYPE (TREE_OPERAND (rhs1, 1)))
+	    {
+	      tree fld = TREE_OPERAND (rhs1, 1);
+	      /* For little-endian, we can allow as inputs bit-fields
+		 which start at a limb boundary.  */
+	      if (DECL_OFFSET_ALIGN (fld) >= TYPE_ALIGN (TREE_TYPE (rhs1))
+		  && tree_fits_uhwi_p (DECL_FIELD_BIT_OFFSET (fld))
+		  && (tree_to_uhwi (DECL_FIELD_BIT_OFFSET (fld))
+		      % limb_prec) == 0)
+		;
+	      else
+		{
+		  bitmap_clear_bit (m_loads, SSA_NAME_VERSION (s));
+		  continue;
+		}
+	    }
+
+	  ao_ref ref;
+	  ao_ref_init (&ref, rhs1);
+	  tree lvop = gimple_vuse (SSA_NAME_DEF_STMT (s));
+	  unsigned limit = 64;
+	  tree vuse = cvop;
+	  if (vop != cvop
+	      && is_gimple_assign (stmt)
+	      && gimple_store_p (stmt)
+	      && !operand_equal_p (lhs,
+				   gimple_assign_rhs1 (SSA_NAME_DEF_STMT (s)),
+				   0))
+	    vuse = vop;
+	  if (vuse != lvop
+	      && walk_non_aliased_vuses (&ref, vuse, false, vuse_eq,
+					 NULL, NULL, limit, lvop) == NULL)
+	    bitmap_clear_bit (m_loads, SSA_NAME_VERSION (s));
+	}
+    }
+
+  bb->aux = (void *) vop;
+  return NULL;
+}
+
+}
+
+/* Replacement for normal processing of STMT in tree-ssa-coalesce.cc
+   build_ssa_conflict_graph.
+   The differences are:
+   1) don't process assignments with large/huge _BitInt lhs not in NAMES
+   2) for large/huge _BitInt multiplication/division/modulo process def
+      only after processing uses rather than before to make uses conflict
+      with the definition
+   3) for large/huge _BitInt uses not in NAMES mark the uses of their
+      SSA_NAME_DEF_STMT (recursively), because those uses will be sunk into
+      the final statement.  */
+
+void
+build_bitint_stmt_ssa_conflicts (gimple *stmt, live_track *live,
+				 ssa_conflicts *graph, bitmap names,
+				 void (*def) (live_track *, tree,
+					      ssa_conflicts *),
+				 void (*use) (live_track *, tree))
+{
+  bool muldiv_p = false;
+  tree lhs = NULL_TREE;
+  if (is_gimple_assign (stmt))
+    {
+      lhs = gimple_assign_lhs (stmt);
+      if (TREE_CODE (lhs) == SSA_NAME
+	  && TREE_CODE (TREE_TYPE (lhs)) == BITINT_TYPE
+	  && bitint_precision_kind (TREE_TYPE (lhs)) >= bitint_prec_large)
+	{
+	  if (!bitmap_bit_p (names, SSA_NAME_VERSION (lhs)))
+	    return;
+	  switch (gimple_assign_rhs_code (stmt))
+	    {
+	    case MULT_EXPR:
+	    case TRUNC_DIV_EXPR:
+	    case TRUNC_MOD_EXPR:
+	      muldiv_p = true;
+	    default:
+	      break;
+	    }
+	}
+    }
+
+  ssa_op_iter iter;
+  tree var;
+  if (!muldiv_p)
+    {
+      /* For stmts with more than one SSA_NAME definition pretend all the
+	 SSA_NAME outputs but the first one are live at this point, so
+	 that conflicts are added in between all those even when they are
+	 actually not really live after the asm, because expansion might
+	 copy those into pseudos after the asm and if multiple outputs
+	 share the same partition, it might overwrite those that should
+	 be live.  E.g.
+	 asm volatile (".." : "=r" (a) : "=r" (b) : "0" (a), "1" (a));
+	 return a;
+	 See PR70593.  */
+      bool first = true;
+      FOR_EACH_SSA_TREE_OPERAND (var, stmt, iter, SSA_OP_DEF)
+	if (first)
+	  first = false;
+	else
+	  use (live, var);
+
+      FOR_EACH_SSA_TREE_OPERAND (var, stmt, iter, SSA_OP_DEF)
+	def (live, var, graph);
+    }
+
+  auto_vec<tree, 16> worklist;
+  FOR_EACH_SSA_TREE_OPERAND (var, stmt, iter, SSA_OP_USE)
+    if (TREE_CODE (TREE_TYPE (var)) == BITINT_TYPE
+	&& bitint_precision_kind (TREE_TYPE (var)) >= bitint_prec_large)
+      {
+	if (bitmap_bit_p (names, SSA_NAME_VERSION (var)))
+	  use (live, var);
+	else
+	  worklist.safe_push (var);
+      }
+
+  while (worklist.length () > 0)
+    {
+      tree s = worklist.pop ();
+      FOR_EACH_SSA_TREE_OPERAND (var, SSA_NAME_DEF_STMT (s), iter, SSA_OP_USE)
+	if (TREE_CODE (TREE_TYPE (var)) == BITINT_TYPE
+	    && bitint_precision_kind (TREE_TYPE (var)) >= bitint_prec_large)
+	  {
+	    if (bitmap_bit_p (names, SSA_NAME_VERSION (var)))
+	      use (live, var);
+	    else
+	      worklist.safe_push (var);
+	  }
+    }
+
+  if (muldiv_p)
+    def (live, lhs, graph);
+}
+
+/* Entry point for _BitInt(N) operation lowering during optimization.  */
+
+static unsigned int
+gimple_lower_bitint (void)
+{
+  small_max_prec = mid_min_prec = large_min_prec = huge_min_prec = 0;
+  limb_prec = 0;
+
+  unsigned int i;
+  for (i = 0; i < num_ssa_names; ++i)
+    {
+      tree s = ssa_name (i);
+      if (s == NULL)
+	continue;
+      tree type = TREE_TYPE (s);
+      if (TREE_CODE (type) == COMPLEX_TYPE)
+	type = TREE_TYPE (type);
+      if (TREE_CODE (type) == BITINT_TYPE
+	  && bitint_precision_kind (type) != bitint_prec_small)
+	break;
+      /* We need to also rewrite stores of large/huge _BitInt INTEGER_CSTs
+	 into memory.  Such functions could have no large/huge SSA_NAMEs.  */
+      if (SSA_NAME_IS_VIRTUAL_OPERAND (s))
+	{
+	  gimple *g = SSA_NAME_DEF_STMT (s);
+	  if (is_gimple_assign (g) && gimple_store_p (g))
+	    {
+	      tree t = gimple_assign_rhs1 (g);
+	      if (TREE_CODE (TREE_TYPE (t)) == BITINT_TYPE
+		  && (bitint_precision_kind (TREE_TYPE (t))
+		      >= bitint_prec_large))
+		break;
+	    }
+	}
+    }
+  if (i == num_ssa_names)
+    return 0;
+
+  basic_block bb;
+  auto_vec<gimple *, 4> switch_statements;
+  FOR_EACH_BB_FN (bb, cfun)
+    {
+      if (gswitch *swtch = safe_dyn_cast <gswitch *> (*gsi_last_bb (bb)))
+	{
+	  tree idx = gimple_switch_index (swtch);
+	  if (TREE_CODE (TREE_TYPE (idx)) != BITINT_TYPE
+	      || bitint_precision_kind (TREE_TYPE (idx)) < bitint_prec_large)
+	    continue;
+
+	  if (optimize)
+	    group_case_labels_stmt (swtch);
+	  switch_statements.safe_push (swtch);
+	}
+    }
+
+  if (!switch_statements.is_empty ())
+    {
+      bool expanded = false;
+      gimple *stmt;
+      unsigned int j;
+      i = 0;
+      FOR_EACH_VEC_ELT (switch_statements, j, stmt)
+	{
+	  gswitch *swtch = as_a<gswitch *> (stmt);
+	  tree_switch_conversion::switch_decision_tree dt (swtch);
+	  expanded |= dt.analyze_switch_statement ();
+	}
+
+      if (expanded)
+	{
+	  free_dominance_info (CDI_DOMINATORS);
+	  free_dominance_info (CDI_POST_DOMINATORS);
+	  mark_virtual_operands_for_renaming (cfun);
+	  cleanup_tree_cfg (TODO_update_ssa);
+	}
+    }
+
+  struct bitint_large_huge large_huge;
+  bool has_large_huge_parm_result = false;
+  bool has_large_huge = false;
+  unsigned int ret = 0, first_large_huge = ~0U;
+  bool edge_insertions = false;
+  for (; i < num_ssa_names; ++i)
+    {
+      tree s = ssa_name (i);
+      if (s == NULL)
+	continue;
+      tree type = TREE_TYPE (s);
+      if (TREE_CODE (type) == COMPLEX_TYPE)
+	type = TREE_TYPE (type);
+      if (TREE_CODE (type) == BITINT_TYPE
+	  && bitint_precision_kind (type) >= bitint_prec_large)
+	{
+	  if (first_large_huge == ~0U)
+	    first_large_huge = i;
+	  gimple *stmt = SSA_NAME_DEF_STMT (s), *g;
+	  gimple_stmt_iterator gsi;
+	  tree_code rhs_code;
+	  /* Unoptimize certain constructs to simpler alternatives to
+	     avoid having to lower all of them.  */
+	  if (is_gimple_assign (stmt))
+	    switch (rhs_code = gimple_assign_rhs_code (stmt))
+	      {
+	      default:
+		break;
+	      case LROTATE_EXPR:
+	      case RROTATE_EXPR:
+		{
+		  first_large_huge = 0;
+		  location_t loc = gimple_location (stmt);
+		  gsi = gsi_for_stmt (stmt);
+		  tree rhs1 = gimple_assign_rhs1 (stmt);
+		  tree type = TREE_TYPE (rhs1);
+		  tree n = gimple_assign_rhs2 (stmt), m;
+		  tree p = build_int_cst (TREE_TYPE (n),
+					  TYPE_PRECISION (type));
+		  if (TREE_CODE (n) == INTEGER_CST)
+		    m = fold_build2 (MINUS_EXPR, TREE_TYPE (n), p, n);
+		  else
+		    {
+		      m = make_ssa_name (TREE_TYPE (n));
+		      g = gimple_build_assign (m, MINUS_EXPR, p, n);
+		      gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+		      gimple_set_location (g, loc);
+		    }
+		  if (!TYPE_UNSIGNED (type))
+		    {
+		      tree utype = build_bitint_type (TYPE_PRECISION (type),
+						      1);
+		      if (TREE_CODE (rhs1) == INTEGER_CST)
+			rhs1 = fold_convert (utype, rhs1);
+		      else
+			{
+			  tree t = make_ssa_name (type);
+			  g = gimple_build_assign (t, NOP_EXPR, rhs1);
+			  gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+			  gimple_set_location (g, loc);
+			}
+		    }
+		  g = gimple_build_assign (make_ssa_name (TREE_TYPE (rhs1)),
+					   rhs_code == LROTATE_EXPR
+					   ? LSHIFT_EXPR : RSHIFT_EXPR,
+					   rhs1, n);
+		  gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+		  gimple_set_location (g, loc);
+		  tree op1 = gimple_assign_lhs (g);
+		  g = gimple_build_assign (make_ssa_name (TREE_TYPE (rhs1)),
+					   rhs_code == LROTATE_EXPR
+					   ? RSHIFT_EXPR : LSHIFT_EXPR,
+					   rhs1, m);
+		  gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+		  gimple_set_location (g, loc);
+		  tree op2 = gimple_assign_lhs (g);
+		  tree lhs = gimple_assign_lhs (stmt);
+		  if (!TYPE_UNSIGNED (type))
+		    {
+		      g = gimple_build_assign (make_ssa_name (TREE_TYPE (op1)),
+					       BIT_IOR_EXPR, op1, op2);
+		      gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+		      gimple_set_location (g, loc);
+		      g = gimple_build_assign (lhs, NOP_EXPR,
+					       gimple_assign_lhs (g));
+		    }
+		  else
+		    g = gimple_build_assign (lhs, BIT_IOR_EXPR, op1, op2);
+		  gsi_replace (&gsi, g, true);
+		  gimple_set_location (g, loc);
+		}
+		break;
+	      case ABS_EXPR:
+	      case ABSU_EXPR:
+	      case MIN_EXPR:
+	      case MAX_EXPR:
+	      case COND_EXPR:
+		first_large_huge = 0;
+		gsi = gsi_for_stmt (stmt);
+		tree lhs = gimple_assign_lhs (stmt);
+		tree rhs1 = gimple_assign_rhs1 (stmt), rhs2 = NULL_TREE;
+		location_t loc = gimple_location (stmt);
+		if (rhs_code == ABS_EXPR)
+		  g = gimple_build_cond (LT_EXPR, rhs1,
+		  			 build_zero_cst (TREE_TYPE (rhs1)),
+					 NULL_TREE, NULL_TREE);
+		else if (rhs_code == ABSU_EXPR)
+		  {
+		    rhs2 = make_ssa_name (TREE_TYPE (lhs));
+		    g = gimple_build_assign (rhs2, NOP_EXPR, rhs1);
+		    gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+		    gimple_set_location (g, loc);
+		    g = gimple_build_cond (LT_EXPR, rhs1,
+					   build_zero_cst (TREE_TYPE (rhs1)),
+					   NULL_TREE, NULL_TREE);
+		    rhs1 = rhs2;
+		  }
+		else if (rhs_code == MIN_EXPR || rhs_code == MAX_EXPR)
+		  {
+		    rhs2 = gimple_assign_rhs2 (stmt);
+		    if (TREE_CODE (rhs1) == INTEGER_CST)
+		      std::swap (rhs1, rhs2);
+		    g = gimple_build_cond (LT_EXPR, rhs1, rhs2,
+					   NULL_TREE, NULL_TREE);
+		    if (rhs_code == MAX_EXPR)
+		      std::swap (rhs1, rhs2);
+		  }
+		else
+		  {
+		    g = gimple_build_cond (NE_EXPR, rhs1,
+					   build_zero_cst (TREE_TYPE (rhs1)),
+					   NULL_TREE, NULL_TREE);
+		    rhs1 = gimple_assign_rhs2 (stmt);
+		    rhs2 = gimple_assign_rhs3 (stmt);
+		  }
+		gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+		gimple_set_location (g, loc);
+		edge e1 = split_block (gsi_bb (gsi), g);
+		edge e2 = split_block (e1->dest, (gimple *) NULL);
+		edge e3 = make_edge (e1->src, e2->dest, EDGE_FALSE_VALUE);
+		e3->probability = profile_probability::even ();
+		e1->flags = EDGE_TRUE_VALUE;
+		e1->probability = e3->probability.invert ();
+		if (dom_info_available_p (CDI_DOMINATORS))
+		  set_immediate_dominator (CDI_DOMINATORS, e2->dest, e1->src);
+		if (rhs_code == ABS_EXPR || rhs_code == ABSU_EXPR)
+		  {
+		    gsi = gsi_after_labels (e1->dest);
+		    g = gimple_build_assign (make_ssa_name (TREE_TYPE (rhs1)),
+					     NEGATE_EXPR, rhs1);
+		    gsi_insert_before (&gsi, g, GSI_SAME_STMT);
+		    gimple_set_location (g, loc);
+		    rhs2 = gimple_assign_lhs (g);
+		    std::swap (rhs1, rhs2);
+		  }
+		gsi = gsi_for_stmt (stmt);
+		gsi_remove (&gsi, true);
+		gphi *phi = create_phi_node (lhs, e2->dest);
+		add_phi_arg (phi, rhs1, e2, UNKNOWN_LOCATION);
+		add_phi_arg (phi, rhs2, e3, UNKNOWN_LOCATION);
+		break;
+	      }
+	}
+      /* We need to also rewrite stores of large/huge _BitInt INTEGER_CSTs
+	 into memory.  Such functions could have no large/huge SSA_NAMEs.  */
+      else if (SSA_NAME_IS_VIRTUAL_OPERAND (s))
+	{
+	  gimple *g = SSA_NAME_DEF_STMT (s);
+	  if (is_gimple_assign (g) && gimple_store_p (g))
+	    {
+	      tree t = gimple_assign_rhs1 (g);
+	      if (TREE_CODE (TREE_TYPE (t)) == BITINT_TYPE
+		  && (bitint_precision_kind (TREE_TYPE (t))
+		      >= bitint_prec_large))
+		has_large_huge = true;
+	    }
+	}
+    }
+  for (i = first_large_huge; i < num_ssa_names; ++i)
+    {
+      tree s = ssa_name (i);
+      if (s == NULL)
+	continue;
+      tree type = TREE_TYPE (s);
+      if (TREE_CODE (type) == COMPLEX_TYPE)
+	type = TREE_TYPE (type);
+      if (TREE_CODE (type) == BITINT_TYPE
+	  && bitint_precision_kind (type) >= bitint_prec_large)
+	{
+	  use_operand_p use_p;
+	  gimple *use_stmt;
+	  has_large_huge = true;
+	  if (optimize
+	      && optimizable_arith_overflow (SSA_NAME_DEF_STMT (s)))
+	    continue;
+	  /* Ignore large/huge _BitInt SSA_NAMEs which have single use in
+	     the same bb and could be handled in the same loop with the
+	     immediate use.  */
+	  if (optimize
+	      && !SSA_NAME_OCCURS_IN_ABNORMAL_PHI (s)
+	      && single_imm_use (s, &use_p, &use_stmt)
+	      && gimple_bb (SSA_NAME_DEF_STMT (s)) == gimple_bb (use_stmt))
+	    {
+	      if (mergeable_op (SSA_NAME_DEF_STMT (s)))
+		{
+		  if (mergeable_op (use_stmt))
+		    continue;
+		  tree_code cmp_code = comparison_op (use_stmt, NULL, NULL);
+		  if (cmp_code == EQ_EXPR || cmp_code == NE_EXPR)
+		    continue;
+		  if (gimple_assign_cast_p (use_stmt))
+		    {
+		      tree lhs = gimple_assign_lhs (use_stmt);
+		      if (INTEGRAL_TYPE_P (TREE_TYPE (lhs)))
+			continue;
+		    }
+		  else if (gimple_store_p (use_stmt)
+			   && is_gimple_assign (use_stmt)
+			   && !gimple_has_volatile_ops (use_stmt)
+			   && !stmt_ends_bb_p (use_stmt))
+		    continue;
+		}
+	      if (gimple_assign_cast_p (SSA_NAME_DEF_STMT (s)))
+		{
+		  tree rhs1 = gimple_assign_rhs1 (SSA_NAME_DEF_STMT (s));
+		  if (INTEGRAL_TYPE_P (TREE_TYPE (rhs1))
+		      && ((is_gimple_assign (use_stmt)
+			   && (gimple_assign_rhs_code (use_stmt)
+			       != COMPLEX_EXPR))
+			  || gimple_code (use_stmt) == GIMPLE_COND)
+		      && (!gimple_store_p (use_stmt)
+			  || (is_gimple_assign (use_stmt)
+			      && !gimple_has_volatile_ops (use_stmt)
+			      && !stmt_ends_bb_p (use_stmt)))
+		      && (TREE_CODE (rhs1) != SSA_NAME
+			  || !SSA_NAME_OCCURS_IN_ABNORMAL_PHI (rhs1)))
+		    {
+		      if (TREE_CODE (TREE_TYPE (rhs1)) != BITINT_TYPE
+			  || (bitint_precision_kind (TREE_TYPE (rhs1))
+			      < bitint_prec_large)
+			  || (TYPE_PRECISION (TREE_TYPE (rhs1))
+			      >= TYPE_PRECISION (TREE_TYPE (s)))
+			  || mergeable_op (SSA_NAME_DEF_STMT (s)))
+			continue;
+		      /* Prevent merging a widening non-mergeable cast
+			 on result of some narrower mergeable op
+			 together with later mergeable operations.  E.g.
+			 result of _BitInt(223) addition shouldn't be
+			 sign-extended to _BitInt(513) and have another
+			 _BitInt(513) added to it, as handle_plus_minus
+			 with its PHI node handling inside of handle_cast
+			 will not work correctly.  An exception is if
+			 use_stmt is a store, this is handled directly
+			 in lower_mergeable_stmt.  */
+		      if (TREE_CODE (rhs1) != SSA_NAME
+			  || !has_single_use (rhs1)
+			  || (gimple_bb (SSA_NAME_DEF_STMT (rhs1))
+			      != gimple_bb (SSA_NAME_DEF_STMT (s)))
+			  || !mergeable_op (SSA_NAME_DEF_STMT (rhs1))
+			  || gimple_store_p (use_stmt))
+			continue;
+		      if (gimple_assign_cast_p (SSA_NAME_DEF_STMT (rhs1)))
+			{
+			  /* Another exception is if the widening cast is
+			     from mergeable same precision cast from something
+			     not mergeable.  */
+			  tree rhs2
+			    = gimple_assign_rhs1 (SSA_NAME_DEF_STMT (rhs1));
+			  if (TREE_CODE (TREE_TYPE (rhs2)) == BITINT_TYPE
+			      && (TYPE_PRECISION (TREE_TYPE (rhs1))
+				  == TYPE_PRECISION (TREE_TYPE (rhs2))))
+			    {
+			      if (TREE_CODE (rhs2) != SSA_NAME
+				  || !has_single_use (rhs2)
+				  || (gimple_bb (SSA_NAME_DEF_STMT (rhs2))
+				      != gimple_bb (SSA_NAME_DEF_STMT (s)))
+				  || !mergeable_op (SSA_NAME_DEF_STMT (rhs2)))
+				continue;
+			    }
+			}
+		    }
+		}
+	      if (is_gimple_assign (SSA_NAME_DEF_STMT (s)))
+		switch (gimple_assign_rhs_code (SSA_NAME_DEF_STMT (s)))
+		  {
+		  case IMAGPART_EXPR:
+		    {
+		      tree rhs1 = gimple_assign_rhs1 (SSA_NAME_DEF_STMT (s));
+		      rhs1 = TREE_OPERAND (rhs1, 0);
+		      if (TREE_CODE (rhs1) == SSA_NAME)
+			{
+			  gimple *g = SSA_NAME_DEF_STMT (rhs1);
+			  if (optimizable_arith_overflow (g))
+			    continue;
+			}
+		    }
+		    /* FALLTHRU */
+		  case LSHIFT_EXPR:
+		  case RSHIFT_EXPR:
+		  case MULT_EXPR:
+		  case TRUNC_DIV_EXPR:
+		  case TRUNC_MOD_EXPR:
+		  case FIX_TRUNC_EXPR:
+		  case REALPART_EXPR:
+		    if (gimple_store_p (use_stmt)
+			&& is_gimple_assign (use_stmt)
+			&& !gimple_has_volatile_ops (use_stmt)
+			&& !stmt_ends_bb_p (use_stmt))
+		      {
+			tree lhs = gimple_assign_lhs (use_stmt);
+			/* As multiply/division passes address of the lhs
+			   to library function and that assumes it can extend
+			   it to whole number of limbs, avoid merging those
+			   with bit-field stores.  Don't allow it for
+			   shifts etc. either, so that the bit-field store
+			   handling doesn't have to be done everywhere.  */
+			if (TREE_CODE (lhs) == COMPONENT_REF
+			    && DECL_BIT_FIELD_TYPE (TREE_OPERAND (lhs, 1)))
+			  break;
+			continue;
+		      }
+		    break;
+		  default:
+		    break;
+		  }
+	    }
+
+	  /* Also ignore uninitialized uses.  */
+	  if (SSA_NAME_IS_DEFAULT_DEF (s)
+	      && (!SSA_NAME_VAR (s) || VAR_P (SSA_NAME_VAR (s))))
+	    continue;
+
+	  if (!large_huge.m_names)
+	    large_huge.m_names = BITMAP_ALLOC (NULL);
+	  bitmap_set_bit (large_huge.m_names, SSA_NAME_VERSION (s));
+	  if (has_single_use (s))
+	    {
+	      if (!large_huge.m_single_use_names)
+		large_huge.m_single_use_names = BITMAP_ALLOC (NULL);
+	      bitmap_set_bit (large_huge.m_single_use_names,
+			      SSA_NAME_VERSION (s));
+	    }
+	  if (SSA_NAME_VAR (s)
+	      && ((TREE_CODE (SSA_NAME_VAR (s)) == PARM_DECL
+		   && SSA_NAME_IS_DEFAULT_DEF (s))
+		  || TREE_CODE (SSA_NAME_VAR (s)) == RESULT_DECL))
+	    has_large_huge_parm_result = true;
+	  if (optimize
+	      && !SSA_NAME_OCCURS_IN_ABNORMAL_PHI (s)
+	      && gimple_assign_load_p (SSA_NAME_DEF_STMT (s))
+	      && !gimple_has_volatile_ops (SSA_NAME_DEF_STMT (s))
+	      && !stmt_ends_bb_p (SSA_NAME_DEF_STMT (s)))
+	    {
+	      use_operand_p use_p;
+	      imm_use_iterator iter;
+	      bool optimizable_load = true;
+	      FOR_EACH_IMM_USE_FAST (use_p, iter, s)
+		{
+		  gimple *use_stmt = USE_STMT (use_p);
+		  if (is_gimple_debug (use_stmt))
+		    continue;
+		  if (gimple_code (use_stmt) == GIMPLE_PHI
+		      || is_gimple_call (use_stmt))
+		    {
+		      optimizable_load = false;
+		      break;
+		    }
+		}
+
+	      ssa_op_iter oi;
+	      FOR_EACH_SSA_USE_OPERAND (use_p, SSA_NAME_DEF_STMT (s),
+					oi, SSA_OP_USE)
+		{
+		  tree s2 = USE_FROM_PTR (use_p);
+		  if (SSA_NAME_OCCURS_IN_ABNORMAL_PHI (s2))
+		    {
+		      optimizable_load = false;
+		      break;
+		    }
+		}
+
+	      if (optimizable_load && !stmt_ends_bb_p (SSA_NAME_DEF_STMT (s)))
+		{
+		  if (!large_huge.m_loads)
+		    large_huge.m_loads = BITMAP_ALLOC (NULL);
+		  bitmap_set_bit (large_huge.m_loads, SSA_NAME_VERSION (s));
+		}
+	    }
+	}
+      /* We need to also rewrite stores of large/huge _BitInt INTEGER_CSTs
+	 into memory.  Such functions could have no large/huge SSA_NAMEs.  */
+      else if (SSA_NAME_IS_VIRTUAL_OPERAND (s))
+	{
+	  gimple *g = SSA_NAME_DEF_STMT (s);
+	  if (is_gimple_assign (g) && gimple_store_p (g))
+	    {
+	      tree t = gimple_assign_rhs1 (g);
+	      if (TREE_CODE (TREE_TYPE (t)) == BITINT_TYPE
+		  && bitint_precision_kind (TREE_TYPE (t)) >= bitint_prec_large)
+		has_large_huge = true;
+	    }
+	}
+    }
+
+  if (large_huge.m_names || has_large_huge)
+    {
+      ret = TODO_update_ssa_only_virtuals | TODO_cleanup_cfg;
+      calculate_dominance_info (CDI_DOMINATORS);
+      if (optimize)
+	enable_ranger (cfun);
+      if (large_huge.m_loads)
+	{
+	  basic_block entry = ENTRY_BLOCK_PTR_FOR_FN (cfun);
+	  entry->aux = NULL;
+	  bitint_dom_walker (large_huge.m_names,
+			     large_huge.m_loads).walk (entry);
+	  bitmap_and_compl_into (large_huge.m_names, large_huge.m_loads);
+	  clear_aux_for_blocks ();
+	  BITMAP_FREE (large_huge.m_loads);
+	}
+      large_huge.m_limb_type = build_nonstandard_integer_type (limb_prec, 1);
+      large_huge.m_limb_size
+	= tree_to_uhwi (TYPE_SIZE_UNIT (large_huge.m_limb_type));
+    }
+  if (large_huge.m_names)
+    {
+      large_huge.m_map
+	= init_var_map (num_ssa_names, NULL, large_huge.m_names);
+      coalesce_ssa_name (large_huge.m_map);
+      partition_view_normal (large_huge.m_map);
+      if (dump_file && (dump_flags & TDF_DETAILS))
+	{
+	  fprintf (dump_file, "After Coalescing:\n");
+	  dump_var_map (dump_file, large_huge.m_map);
+	}
+      large_huge.m_vars
+	= XCNEWVEC (tree, num_var_partitions (large_huge.m_map));
+      bitmap_iterator bi;
+      if (has_large_huge_parm_result)
+	EXECUTE_IF_SET_IN_BITMAP (large_huge.m_names, 0, i, bi)
+	  {
+	    tree s = ssa_name (i);
+	    if (SSA_NAME_VAR (s)
+		&& ((TREE_CODE (SSA_NAME_VAR (s)) == PARM_DECL
+		     && SSA_NAME_IS_DEFAULT_DEF (s))
+		    || TREE_CODE (SSA_NAME_VAR (s)) == RESULT_DECL))
+	      {
+		int p = var_to_partition (large_huge.m_map, s);
+		if (large_huge.m_vars[p] == NULL_TREE)
+		  {
+		    large_huge.m_vars[p] = SSA_NAME_VAR (s);
+		    mark_addressable (SSA_NAME_VAR (s));
+		  }
+	      }
+	  }
+      tree atype = NULL_TREE;
+      EXECUTE_IF_SET_IN_BITMAP (large_huge.m_names, 0, i, bi)
+	{
+	  tree s = ssa_name (i);
+	  int p = var_to_partition (large_huge.m_map, s);
+	  if (large_huge.m_vars[p] != NULL_TREE)
+	    continue;
+	  if (atype == NULL_TREE
+	      || !tree_int_cst_equal (TYPE_SIZE (atype),
+				      TYPE_SIZE (TREE_TYPE (s))))
+	    {
+	      unsigned HOST_WIDE_INT nelts
+		= tree_to_uhwi (TYPE_SIZE (TREE_TYPE (s))) / limb_prec;
+	      atype = build_array_type_nelts (large_huge.m_limb_type, nelts);
+	    }
+	  large_huge.m_vars[p] = create_tmp_var (atype, "bitint");
+	  mark_addressable (large_huge.m_vars[p]);
+	}
+    }
+
+  FOR_EACH_BB_REVERSE_FN (bb, cfun)
+    {
+      gimple_stmt_iterator prev;
+      for (gimple_stmt_iterator gsi = gsi_last_bb (bb); !gsi_end_p (gsi);
+	   gsi = prev)
+	{
+	  prev = gsi;
+	  gsi_prev (&prev);
+	  ssa_op_iter iter;
+	  gimple *stmt = gsi_stmt (gsi);
+	  if (is_gimple_debug (stmt))
+	    continue;
+	  bitint_prec_kind kind = bitint_prec_small;
+	  tree t;
+	  FOR_EACH_SSA_TREE_OPERAND (t, stmt, iter, SSA_OP_ALL_OPERANDS)
+	    if (TREE_CODE (TREE_TYPE (t)) == BITINT_TYPE)
+	      {
+		bitint_prec_kind this_kind
+		  = bitint_precision_kind (TREE_TYPE (t));
+		if (this_kind > kind)
+		  kind = this_kind;
+	      }
+	  if (is_gimple_assign (stmt) && gimple_store_p (stmt))
+	    {
+	      t = gimple_assign_rhs1 (stmt);
+	      if (TREE_CODE (TREE_TYPE (t)) == BITINT_TYPE)
+		{
+		  bitint_prec_kind this_kind
+		    = bitint_precision_kind (TREE_TYPE (t));
+		  if (this_kind > kind)
+		    kind = this_kind;
+		}
+	    }
+	  if (is_gimple_call (stmt))
+	    {
+	      t = gimple_call_lhs (stmt);
+	      if (t
+		  && TREE_CODE (TREE_TYPE (t)) == COMPLEX_TYPE
+		  && TREE_CODE (TREE_TYPE (TREE_TYPE (t))) == BITINT_TYPE)
+		{
+		  bitint_prec_kind this_kind
+		    = bitint_precision_kind (TREE_TYPE (TREE_TYPE (t)));
+		  if (this_kind > kind)
+		    kind = this_kind;
+		}
+	    }
+	  if (kind == bitint_prec_small)
+	    continue;
+	  switch (gimple_code (stmt))
+	    {
+	    case GIMPLE_CALL:
+	      /* For now.  We'll need to handle some internal functions and
+		 perhaps some builtins.  */
+	      if (kind == bitint_prec_middle)
+		continue;
+	      break;
+	    case GIMPLE_ASM:
+	      if (kind == bitint_prec_middle)
+		continue;
+	      break;
+	    case GIMPLE_RETURN:
+	      continue;
+	    case GIMPLE_ASSIGN:
+	      if (gimple_clobber_p (stmt))
+		continue;
+	      if (kind >= bitint_prec_large)
+		break;
+	      if (gimple_assign_single_p (stmt))
+		/* No need to lower copies, loads or stores.  */
+		continue;
+	      if (gimple_assign_cast_p (stmt))
+		{
+		  tree lhs = gimple_assign_lhs (stmt);
+		  tree rhs = gimple_assign_rhs1 (stmt);
+		  if (INTEGRAL_TYPE_P (TREE_TYPE (lhs))
+		      && INTEGRAL_TYPE_P (TREE_TYPE (rhs))
+		      && (TYPE_PRECISION (TREE_TYPE (lhs))
+			  == TYPE_PRECISION (TREE_TYPE (rhs))))
+		    /* No need to lower casts to same precision.  */
+		    continue;
+		}
+	      break;
+	    default:
+	      break;
+	    }
+
+	  if (kind == bitint_prec_middle)
+	    {
+	      tree type = NULL_TREE;
+	      /* Middle _BitInt(N) is rewritten to casts to INTEGER_TYPEs
+		 with the same precision and back.  */
+	      if (tree lhs = gimple_get_lhs (stmt))
+		if (TREE_CODE (TREE_TYPE (lhs)) == BITINT_TYPE
+		    && (bitint_precision_kind (TREE_TYPE (lhs))
+			== bitint_prec_middle))
+		  {
+		    int prec = TYPE_PRECISION (TREE_TYPE (lhs));
+		    int uns = TYPE_UNSIGNED (TREE_TYPE (lhs));
+		    type = build_nonstandard_integer_type (prec, uns);
+		    tree lhs2 = make_ssa_name (type);
+		    gimple *g = gimple_build_assign (lhs, NOP_EXPR, lhs2);
+		    gsi_insert_after (&gsi, g, GSI_SAME_STMT);
+		    gimple_set_lhs (stmt, lhs2);
+		  }
+	      unsigned int nops = gimple_num_ops (stmt);
+	      for (unsigned int i = 0; i < nops; ++i)
+		if (tree op = gimple_op (stmt, i))
+		  {
+		    tree nop = maybe_cast_middle_bitint (&gsi, op, type);
+		    if (nop != op)
+		      gimple_set_op (stmt, i, nop);
+		    else if (COMPARISON_CLASS_P (op))
+		      {
+			TREE_OPERAND (op, 0)
+			  = maybe_cast_middle_bitint (&gsi,
+						      TREE_OPERAND (op, 0),
+						      type);
+			TREE_OPERAND (op, 1)
+			  = maybe_cast_middle_bitint (&gsi,
+						      TREE_OPERAND (op, 1),
+						      type);
+		      }
+		    else if (TREE_CODE (op) == CASE_LABEL_EXPR)
+		      {
+			CASE_LOW (op)
+			  = maybe_cast_middle_bitint (&gsi, CASE_LOW (op),
+						      type);
+			CASE_HIGH (op)
+			  = maybe_cast_middle_bitint (&gsi, CASE_HIGH (op),
+						      type);
+		      }
+		  }
+	      update_stmt (stmt);
+	      continue;
+	    }
+
+	  if (tree lhs = gimple_get_lhs (stmt))
+	    if (TREE_CODE (lhs) == SSA_NAME)
+	      {
+		tree type = TREE_TYPE (lhs);
+		if (TREE_CODE (type) == COMPLEX_TYPE)
+		  type = TREE_TYPE (type);
+		if (TREE_CODE (type) == BITINT_TYPE
+		    && bitint_precision_kind (type) >= bitint_prec_large
+		    && (large_huge.m_names == NULL
+			|| !bitmap_bit_p (large_huge.m_names,
+					  SSA_NAME_VERSION (lhs))))
+		  continue;
+	      }
+
+	  large_huge.lower_stmt (stmt);
+	}
+
+      tree atype = NULL_TREE;
+      for (gphi_iterator gsi = gsi_start_phis (bb); !gsi_end_p (gsi);
+	   gsi_next (&gsi))
+	{
+	  gphi *phi = gsi.phi ();
+	  tree lhs = gimple_phi_result (phi);
+	  if (TREE_CODE (TREE_TYPE (lhs)) != BITINT_TYPE
+	      || bitint_precision_kind (TREE_TYPE (lhs)) < bitint_prec_large)
+	    continue;
+	  int p1 = var_to_partition (large_huge.m_map, lhs);
+	  gcc_assert (large_huge.m_vars[p1] != NULL_TREE);
+	  tree v1 = large_huge.m_vars[p1];
+	  for (unsigned i = 0; i < gimple_phi_num_args (phi); ++i)
+	    {
+	      tree arg = gimple_phi_arg_def (phi, i);
+	      edge e = gimple_phi_arg_edge (phi, i);
+	      gimple *g;
+	      switch (TREE_CODE (arg))
+		{
+		case INTEGER_CST:
+		  if (integer_zerop (arg) && VAR_P (v1))
+		    {
+		      tree zero = build_zero_cst (TREE_TYPE (v1));
+		      g = gimple_build_assign (v1, zero);
+		      gsi_insert_on_edge (e, g);
+		      edge_insertions = true;
+		      break;
+		    }
+		  int ext;
+		  unsigned int min_prec, prec, rem;
+		  tree c;
+		  prec = TYPE_PRECISION (TREE_TYPE (arg));
+		  rem = prec % (2 * limb_prec);
+		  min_prec = bitint_min_cst_precision (arg, ext);
+		  if (min_prec > prec - rem - 2 * limb_prec
+		      && min_prec > (unsigned) limb_prec)
+		    /* Constant which has enough significant bits that it
+		       isn't worth trying to save .rodata space by extending
+		       from smaller number.  */
+		    min_prec = prec;
+		  else
+		    min_prec = CEIL (min_prec, limb_prec) * limb_prec;
+		  if (min_prec == 0)
+		    c = NULL_TREE;
+		  else if (min_prec == prec)
+		    c = tree_output_constant_def (arg);
+		  else if (min_prec == (unsigned) limb_prec)
+		    c = fold_convert (large_huge.m_limb_type, arg);
+		  else
+		    {
+		      tree ctype = build_bitint_type (min_prec, 1);
+		      c = tree_output_constant_def (fold_convert (ctype, arg));
+		    }
+		  if (c)
+		    {
+		      if (VAR_P (v1) && min_prec == prec)
+			{
+			  tree v2 = build1 (VIEW_CONVERT_EXPR,
+					    TREE_TYPE (v1), c);
+			  g = gimple_build_assign (v1, v2);
+			  gsi_insert_on_edge (e, g);
+			  edge_insertions = true;
+			  break;
+			}
+		      if (TREE_CODE (TREE_TYPE (c)) == INTEGER_TYPE)
+			g = gimple_build_assign (build1 (VIEW_CONVERT_EXPR,
+							 TREE_TYPE (c), v1),
+						 c);
+		      else
+			{
+			  unsigned HOST_WIDE_INT nelts
+			    = tree_to_uhwi (TYPE_SIZE (TREE_TYPE (c)))
+			      / limb_prec;
+			  tree vtype
+			    = build_array_type_nelts (large_huge.m_limb_type,
+						      nelts);
+			  g = gimple_build_assign (build1 (VIEW_CONVERT_EXPR,
+							   vtype, v1),
+						   build1 (VIEW_CONVERT_EXPR,
+							   vtype, c));
+			}
+		      gsi_insert_on_edge (e, g);
+		    }
+		  if (ext == 0)
+		    {
+		      unsigned HOST_WIDE_INT nelts
+			= (tree_to_uhwi (TYPE_SIZE (TREE_TYPE (v1)))
+			   - min_prec) / limb_prec;
+		      tree vtype
+			= build_array_type_nelts (large_huge.m_limb_type,
+						  nelts);
+		      tree ptype = build_pointer_type (TREE_TYPE (v1));
+		      tree off = fold_convert (ptype,
+					       TYPE_SIZE_UNIT (TREE_TYPE (c)));
+		      tree vd = build2 (MEM_REF, vtype,
+					build_fold_addr_expr (v1), off);
+		      g = gimple_build_assign (vd, build_zero_cst (vtype));
+		    }
+		  else
+		    {
+		      tree vd = v1;
+		      if (c)
+			{
+			  tree ptype = build_pointer_type (TREE_TYPE (v1));
+			  tree off
+			    = fold_convert (ptype,
+					    TYPE_SIZE_UNIT (TREE_TYPE (c)));
+			  vd = build2 (MEM_REF, large_huge.m_limb_type,
+				       build_fold_addr_expr (v1), off);
+			}
+		      vd = build_fold_addr_expr (vd);
+		      unsigned HOST_WIDE_INT nbytes
+			= tree_to_uhwi (TYPE_SIZE_UNIT (TREE_TYPE (v1)));
+		      if (c)
+			nbytes
+			  -= tree_to_uhwi (TYPE_SIZE_UNIT (TREE_TYPE (c)));
+		      tree fn = builtin_decl_implicit (BUILT_IN_MEMSET);
+		      g = gimple_build_call (fn, 3, vd,
+					     integer_minus_one_node,
+					     build_int_cst (sizetype,
+							    nbytes));
+		    }
+		  gsi_insert_on_edge (e, g);
+		  edge_insertions = true;
+		  break;
+		default:
+		  gcc_unreachable ();
+		case SSA_NAME:
+		  if (gimple_code (SSA_NAME_DEF_STMT (arg)) == GIMPLE_NOP)
+		    {
+		      if (large_huge.m_names == NULL
+			  || !bitmap_bit_p (large_huge.m_names,
+					    SSA_NAME_VERSION (arg)))
+			continue;
+		    }
+		  int p2 = var_to_partition (large_huge.m_map, arg);
+		  if (p1 == p2)
+		    continue;
+		  gcc_assert (large_huge.m_vars[p2] != NULL_TREE);
+		  tree v2 = large_huge.m_vars[p2];
+		  if (VAR_P (v1) && VAR_P (v2))
+		    g = gimple_build_assign (v1, v2);
+		  else if (VAR_P (v1))
+		    g = gimple_build_assign (v1, build1 (VIEW_CONVERT_EXPR,
+							 TREE_TYPE (v1), v2));
+		  else if (VAR_P (v2))
+		    g = gimple_build_assign (build1 (VIEW_CONVERT_EXPR,
+						     TREE_TYPE (v2), v1), v2);
+		  else
+		    {
+		      if (atype == NULL_TREE
+			  || !tree_int_cst_equal (TYPE_SIZE (atype),
+						  TYPE_SIZE (TREE_TYPE (lhs))))
+			{
+			  unsigned HOST_WIDE_INT nelts
+			    = tree_to_uhwi (TYPE_SIZE (TREE_TYPE (lhs)))
+			      / limb_prec;
+			  atype
+			    = build_array_type_nelts (large_huge.m_limb_type,
+						      nelts);
+			}
+		      g = gimple_build_assign (build1 (VIEW_CONVERT_EXPR,
+						       atype, v1),
+					       build1 (VIEW_CONVERT_EXPR,
+						       atype, v2));
+		    }
+		  gsi_insert_on_edge (e, g);
+		  edge_insertions = true;
+		  break;
+		}
+	    }
+	}
+    }
+
+  if (large_huge.m_names || has_large_huge)
+    {
+      gimple *nop = NULL;
+      for (i = 0; i < num_ssa_names; ++i)
+	{
+	  tree s = ssa_name (i);
+	  if (s == NULL_TREE)
+	    continue;
+	  tree type = TREE_TYPE (s);
+	  if (TREE_CODE (type) == COMPLEX_TYPE)
+	    type = TREE_TYPE (type);
+	  if (TREE_CODE (type) == BITINT_TYPE
+	      && bitint_precision_kind (type) >= bitint_prec_large)
+	    {
+	      if (large_huge.m_preserved
+		  && bitmap_bit_p (large_huge.m_preserved,
+				   SSA_NAME_VERSION (s)))
+		continue;
+	      gimple *g = SSA_NAME_DEF_STMT (s);
+	      if (gimple_code (g) == GIMPLE_NOP)
+		{
+		  if (SSA_NAME_VAR (s))
+		    set_ssa_default_def (cfun, SSA_NAME_VAR (s), NULL_TREE);
+		  release_ssa_name (s);
+		  continue;
+		}
+	      if (gimple_code (g) != GIMPLE_ASM)
+		{
+		  gimple_stmt_iterator gsi = gsi_for_stmt (g);
+		  bool save_vta = flag_var_tracking_assignments;
+		  flag_var_tracking_assignments = false;
+		  gsi_remove (&gsi, true);
+		  flag_var_tracking_assignments = save_vta;
+		}
+	      if (nop == NULL)
+		nop = gimple_build_nop ();
+	      SSA_NAME_DEF_STMT (s) = nop;
+	      release_ssa_name (s);
+	    }
+	}
+      if (optimize)
+	disable_ranger (cfun);
+    }
+
+  if (edge_insertions)
+    gsi_commit_edge_inserts ();
+
+  return ret;
+}
+
+namespace {
+
+const pass_data pass_data_lower_bitint =
+{
+  GIMPLE_PASS, /* type */
+  "bitintlower", /* name */
+  OPTGROUP_NONE, /* optinfo_flags */
+  TV_NONE, /* tv_id */
+  PROP_ssa, /* properties_required */
+  PROP_gimple_lbitint, /* properties_provided */
+  0, /* properties_destroyed */
+  0, /* todo_flags_start */
+  0, /* todo_flags_finish */
+};
+
+class pass_lower_bitint : public gimple_opt_pass
+{
+public:
+  pass_lower_bitint (gcc::context *ctxt)
+    : gimple_opt_pass (pass_data_lower_bitint, ctxt)
+  {}
+
+  /* opt_pass methods: */
+  opt_pass * clone () final override { return new pass_lower_bitint (m_ctxt); }
+  unsigned int execute (function *) final override
+  {
+    return gimple_lower_bitint ();
+  }
+
+}; // class pass_lower_bitint
+
+} // anon namespace
+
+gimple_opt_pass *
+make_pass_lower_bitint (gcc::context *ctxt)
+{
+  return new pass_lower_bitint (ctxt);
+}
+
+\f
+namespace {
+
+const pass_data pass_data_lower_bitint_O0 =
+{
+  GIMPLE_PASS, /* type */
+  "bitintlower0", /* name */
+  OPTGROUP_NONE, /* optinfo_flags */
+  TV_NONE, /* tv_id */
+  PROP_cfg, /* properties_required */
+  PROP_gimple_lbitint, /* properties_provided */
+  0, /* properties_destroyed */
+  0, /* todo_flags_start */
+  0, /* todo_flags_finish */
+};
+
+class pass_lower_bitint_O0 : public gimple_opt_pass
+{
+public:
+  pass_lower_bitint_O0 (gcc::context *ctxt)
+    : gimple_opt_pass (pass_data_lower_bitint_O0, ctxt)
+  {}
+
+  /* opt_pass methods: */
+  bool gate (function *fun) final override
+    {
+      /* With errors, normal optimization passes are not run.  If we don't
+	 lower bitint operations at all, rtl expansion will abort.  */
+      return !(fun->curr_properties & PROP_gimple_lbitint);
+    }
+
+  unsigned int execute (function *) final override
+  {
+    return gimple_lower_bitint ();
+  }
+
+}; // class pass_lower_bitint_O0
+
+} // anon namespace
+
+gimple_opt_pass *
+make_pass_lower_bitint_O0 (gcc::context *ctxt)
+{
+  return new pass_lower_bitint_O0 (ctxt);
+}
diff --git a/gcc/gimple-lower-bitint.h b/gcc/gimple-lower-bitint.h
new file mode 100644
index 000000000000..60b159873212
--- /dev/null
+++ b/gcc/gimple-lower-bitint.h
@@ -0,0 +1,31 @@
+/* Header file for gimple-lower-bitint.cc exports.
+   Copyright (C) 2023 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 3, or (at your option) any later
+version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3.  If not see
+<http://www.gnu.org/licenses/>.  */
+
+#ifndef GCC_GIMPLE_LOWER_BITINT_H
+#define GCC_GIMPLE_LOWER_BITINT_H
+
+class live_track;
+struct ssa_conflicts;
+extern void build_bitint_stmt_ssa_conflicts (gimple *, live_track *,
+					     ssa_conflicts *, bitmap,
+					     void (*) (live_track *, tree,
+						       ssa_conflicts *),
+					     void (*) (live_track *, tree));
+
+#endif /* GCC_GIMPLE_LOWER_BITINT_H */
diff --git a/gcc/passes.def b/gcc/passes.def
index ef5a21afe497..4110a472914f 100644
--- a/gcc/passes.def
+++ b/gcc/passes.def
@@ -237,6 +237,7 @@ along with GCC; see the file COPYING3.  If not see
       NEXT_PASS (pass_tail_recursion);
       NEXT_PASS (pass_ch);
       NEXT_PASS (pass_lower_complex);
+      NEXT_PASS (pass_lower_bitint);
       NEXT_PASS (pass_sra);
       /* The dom pass will also resolve all __builtin_constant_p calls
          that are still there to 0.  This has to be done after some
@@ -386,6 +387,7 @@ along with GCC; see the file COPYING3.  If not see
       NEXT_PASS (pass_strip_predict_hints, false /* early_p */);
       /* Lower remaining pieces of GIMPLE.  */
       NEXT_PASS (pass_lower_complex);
+      NEXT_PASS (pass_lower_bitint);
       NEXT_PASS (pass_lower_vector_ssa);
       NEXT_PASS (pass_lower_switch);
       /* Perform simple scalar cleanup which is constant/copy propagation.  */
@@ -429,6 +431,7 @@ along with GCC; see the file COPYING3.  If not see
   NEXT_PASS (pass_lower_vaarg);
   NEXT_PASS (pass_lower_vector);
   NEXT_PASS (pass_lower_complex_O0);
+  NEXT_PASS (pass_lower_bitint_O0);
   NEXT_PASS (pass_sancov_O0);
   NEXT_PASS (pass_lower_switch_O0);
   NEXT_PASS (pass_asan_O0);
diff --git a/gcc/tree-pass.h b/gcc/tree-pass.h
index 57865cdfc424..eba2d54ac766 100644
--- a/gcc/tree-pass.h
+++ b/gcc/tree-pass.h
@@ -229,6 +229,7 @@ protected:
 						   have completed.  */
 #define PROP_assumptions_done	(1 << 19)	/* Assume function kept
 						   around.  */
+#define PROP_gimple_lbitint	(1 << 20)       /* lowered large _BitInt */
 
 #define PROP_gimple \
   (PROP_gimple_any | PROP_gimple_lcf | PROP_gimple_leh | PROP_gimple_lomp)
@@ -420,6 +421,8 @@ extern gimple_opt_pass *make_pass_strip_predict_hints (gcc::context *ctxt);
 extern gimple_opt_pass *make_pass_rebuild_frequencies (gcc::context *ctxt);
 extern gimple_opt_pass *make_pass_lower_complex_O0 (gcc::context *ctxt);
 extern gimple_opt_pass *make_pass_lower_complex (gcc::context *ctxt);
+extern gimple_opt_pass *make_pass_lower_bitint_O0 (gcc::context *ctxt);
+extern gimple_opt_pass *make_pass_lower_bitint (gcc::context *ctxt);
 extern gimple_opt_pass *make_pass_lower_switch (gcc::context *ctxt);
 extern gimple_opt_pass *make_pass_lower_switch_O0 (gcc::context *ctxt);
 extern gimple_opt_pass *make_pass_lower_vector (gcc::context *ctxt);
diff --git a/gcc/tree-ssa-coalesce.cc b/gcc/tree-ssa-coalesce.cc
index e9de9d4d14a2..482764ea29c1 100644
--- a/gcc/tree-ssa-coalesce.cc
+++ b/gcc/tree-ssa-coalesce.cc
@@ -38,6 +38,7 @@ along with GCC; see the file COPYING3.  If not see
 #include "explow.h"
 #include "tree-dfa.h"
 #include "stor-layout.h"
+#include "gimple-lower-bitint.h"
 
 /* This set of routines implements a coalesce_list.  This is an object which
    is used to track pairs of ssa_names which are desirable to coalesce
@@ -914,6 +915,14 @@ build_ssa_conflict_graph (tree_live_info_p liveinfo)
 	  else if (is_gimple_debug (stmt))
 	    continue;
 
+	  if (map->bitint)
+	    {
+	      build_bitint_stmt_ssa_conflicts (stmt, live, graph, map->bitint,
+					       live_track_process_def,
+					       live_track_process_use);
+	      continue;
+	    }
+
 	  /* For stmts with more than one SSA_NAME definition pretend all the
 	     SSA_NAME outputs but the first one are live at this point, so
 	     that conflicts are added in between all those even when they are
@@ -1058,6 +1067,8 @@ create_coalesce_list_for_region (var_map map, bitmap used_in_copy)
 	  if (virtual_operand_p (res))
 	    continue;
 	  ver = SSA_NAME_VERSION (res);
+	  if (map->bitint && !bitmap_bit_p (map->bitint, ver))
+	    continue;
 
 	  /* Register ssa_names and coalesces between the args and the result
 	     of all PHI.  */
@@ -1106,6 +1117,8 @@ create_coalesce_list_for_region (var_map map, bitmap used_in_copy)
 		  {
 		    v1 = SSA_NAME_VERSION (lhs);
 		    v2 = SSA_NAME_VERSION (rhs1);
+		    if (map->bitint && !bitmap_bit_p (map->bitint, v1))
+		      break;
 		    cost = coalesce_cost_bb (bb);
 		    add_coalesce (cl, v1, v2, cost);
 		    bitmap_set_bit (used_in_copy, v1);
@@ -1124,12 +1137,16 @@ create_coalesce_list_for_region (var_map map, bitmap used_in_copy)
 		if (!rhs1)
 		  break;
 		tree lhs = ssa_default_def (cfun, res);
+		if (map->bitint && !lhs)
+		  break;
 		gcc_assert (lhs);
 		if (TREE_CODE (rhs1) == SSA_NAME
 		    && gimple_can_coalesce_p (lhs, rhs1))
 		  {
 		    v1 = SSA_NAME_VERSION (lhs);
 		    v2 = SSA_NAME_VERSION (rhs1);
+		    if (map->bitint && !bitmap_bit_p (map->bitint, v1))
+		      break;
 		    cost = coalesce_cost_bb (bb);
 		    add_coalesce (cl, v1, v2, cost);
 		    bitmap_set_bit (used_in_copy, v1);
@@ -1177,6 +1194,8 @@ create_coalesce_list_for_region (var_map map, bitmap used_in_copy)
 
 		    v1 = SSA_NAME_VERSION (outputs[match]);
 		    v2 = SSA_NAME_VERSION (input);
+		    if (map->bitint && !bitmap_bit_p (map->bitint, v1))
+		      continue;
 
 		    if (gimple_can_coalesce_p (outputs[match], input))
 		      {
@@ -1651,6 +1670,33 @@ compute_optimized_partition_bases (var_map map, bitmap used_in_copies,
 	  }
     }
 
+  if (map->bitint
+      && flag_tree_coalesce_vars
+      && (optimize > 1 || parts < 500))
+    for (i = 0; i < (unsigned) parts; ++i)
+      {
+	tree s1 = partition_to_var (map, i);
+	int p1 = partition_find (tentative, i);
+	for (unsigned j = i + 1; j < (unsigned) parts; ++j)
+	  {
+	    tree s2 = partition_to_var (map, j);
+	    if (s1 == s2)
+	      continue;
+	    if (tree_int_cst_equal (TYPE_SIZE (TREE_TYPE (s1)),
+				    TYPE_SIZE (TREE_TYPE (s2))))
+	      {
+		int p2 = partition_find (tentative, j);
+
+		if (p1 == p2)
+		  continue;
+
+		partition_union (tentative, p1, p2);
+		if (partition_find (tentative, i) != p1)
+		  break;
+	      }
+	  }
+      }
+
   map->partition_to_base_index = XCNEWVEC (int, parts);
   auto_vec<unsigned int> index_map (parts);
   if (parts)
@@ -1692,6 +1738,101 @@ compute_optimized_partition_bases (var_map map, bitmap used_in_copies,
   partition_delete (tentative);
 }
 
+/* For the bitint lowering pass, try harder.  Partitions which contain
+   SSA_NAME default def of a PARM_DECL or have RESULT_DECL need to have
+   compatible types because they will use that RESULT_DECL or PARM_DECL.
+   Other partitions can have even incompatible _BitInt types, as long
+   as they have the same size - those will use VAR_DECLs which are just
+   arrays of the limbs.  */
+
+static void
+coalesce_bitint (var_map map, ssa_conflicts *graph)
+{
+  unsigned n = num_var_partitions (map);
+  if (optimize <= 1 && n > 500)
+    return;
+
+  bool try_same_size = false;
+  FILE *debug_file = (dump_flags & TDF_DETAILS) ? dump_file : NULL;
+  for (unsigned i = 0; i < n; ++i)
+    {
+      tree s1 = partition_to_var (map, i);
+      if ((unsigned) var_to_partition (map, s1) != i)
+	continue;
+      int v1 = SSA_NAME_VERSION (s1);
+      for (unsigned j = i + 1; j < n; ++j)
+	{
+	  tree s2 = partition_to_var (map, j);
+	  if (s1 == s2 || (unsigned) var_to_partition (map, s2) != j)
+	    continue;
+	  if (!types_compatible_p (TREE_TYPE (s1), TREE_TYPE (s2)))
+	    {
+	      if (!try_same_size
+		  && tree_int_cst_equal (TYPE_SIZE (TREE_TYPE (s1)),
+					 TYPE_SIZE (TREE_TYPE (s2))))
+		try_same_size = true;
+	      continue;
+	    }
+	  int v2 = SSA_NAME_VERSION (s2);
+	  if (attempt_coalesce (map, graph, v1, v2, debug_file)
+	      && partition_to_var (map, i) != s1)
+	    break;
+	}
+    }
+
+  if (!try_same_size)
+    return;
+
+  unsigned i;
+  bitmap_iterator bi;
+  bitmap same_type = NULL;
+
+  EXECUTE_IF_SET_IN_BITMAP (map->bitint, 0, i, bi)
+    {
+      tree s = ssa_name (i);
+      if (!SSA_NAME_VAR (s))
+	continue;
+      if (TREE_CODE (SSA_NAME_VAR (s)) != RESULT_DECL
+	  && (TREE_CODE (SSA_NAME_VAR (s)) != PARM_DECL
+	      || !SSA_NAME_IS_DEFAULT_DEF (s)))
+	continue;
+      if (same_type == NULL)
+	same_type = BITMAP_ALLOC (NULL);
+      int p = var_to_partition (map, s);
+      bitmap_set_bit (same_type, p);
+    }
+
+  for (i = 0; i < n; ++i)
+    {
+      if (same_type && bitmap_bit_p (same_type, i))
+	continue;
+      tree s1 = partition_to_var (map, i);
+      if ((unsigned) var_to_partition (map, s1) != i)
+	continue;
+      int v1 = SSA_NAME_VERSION (s1);
+      for (unsigned j = i + 1; j < n; ++j)
+	{
+	  if (same_type && bitmap_bit_p (same_type, j))
+	    continue;
+
+	  tree s2 = partition_to_var (map, j);
+	  if (s1 == s2 || (unsigned) var_to_partition (map, s2) != j)
+	    continue;
+
+	  if (!tree_int_cst_equal (TYPE_SIZE (TREE_TYPE (s1)),
+				   TYPE_SIZE (TREE_TYPE (s2))))
+	    continue;
+
+	  int v2 = SSA_NAME_VERSION (s2);
+	  if (attempt_coalesce (map, graph, v1, v2, debug_file)
+	      && partition_to_var (map, i) != s1)
+	    break;
+	}
+    }
+
+  BITMAP_FREE (same_type);
+}
+
 /* Given an initial var_map MAP, coalesce variables and return a partition map
    with the resulting coalesce.  Note that this function is called in either
    live range computation context or out-of-ssa context, indicated by MAP.  */
@@ -1709,6 +1850,8 @@ coalesce_ssa_name (var_map map)
   if (map->outofssa_p)
     populate_coalesce_list_for_outofssa (cl, used_in_copies);
   bitmap_list_view (used_in_copies);
+  if (map->bitint)
+    bitmap_ior_into (used_in_copies, map->bitint);
 
   if (dump_file && (dump_flags & TDF_DETAILS))
     dump_var_map (dump_file, map);
@@ -1756,6 +1899,9 @@ coalesce_ssa_name (var_map map)
 		       ((dump_flags & TDF_DETAILS) ? dump_file : NULL));
 
   delete_coalesce_list (cl);
+
+  if (map->bitint && flag_tree_coalesce_vars)
+    coalesce_bitint (map, graph);
+
   ssa_conflicts_delete (graph);
 }
-
diff --git a/gcc/tree-ssa-live.cc b/gcc/tree-ssa-live.cc
index 61789fae67d5..8d8a3189eada 100644
--- a/gcc/tree-ssa-live.cc
+++ b/gcc/tree-ssa-live.cc
@@ -77,10 +77,11 @@ var_map_base_fini (var_map map)
 }
 /* Create a variable partition map of SIZE for region, initialize and return
    it.  Region is a loop if LOOP is non-NULL, otherwise is the current
-   function.  */
+   function.  If BITINT is non-NULL, only SSA_NAMEs from that bitmap
+   will be coalesced.  */
 
 var_map
-init_var_map (int size, class loop *loop)
+init_var_map (int size, class loop *loop, bitmap bitint)
 {
   var_map map;
 
@@ -109,7 +110,8 @@ init_var_map (int size, class loop *loop)
   else
     {
       map->bmp_bbs = NULL;
-      map->outofssa_p = true;
+      map->outofssa_p = bitint == NULL;
+      map->bitint = bitint;
       basic_block bb;
       FOR_EACH_BB_FN (bb, cfun)
 	map->vec_bbs.safe_push (bb);
diff --git a/gcc/tree-ssa-live.h b/gcc/tree-ssa-live.h
index d175ad7247e2..73191dc434dc 100644
--- a/gcc/tree-ssa-live.h
+++ b/gcc/tree-ssa-live.h
@@ -70,6 +70,10 @@ typedef struct _var_map
   /* Vector of basic block in the region.  */
   vec<basic_block> vec_bbs;
 
+  /* If non-NULL, only coalesce SSA_NAMEs from this bitmap, and try harder
+     for those (for bitint lowering pass).  */
+  bitmap bitint;
+
   /* True if this map is for out-of-ssa, otherwise for live range
      computation.  When for out-of-ssa, it also means the var map is computed
      for whole current function.  */
@@ -80,7 +84,7 @@ typedef struct _var_map
 /* Value used to represent no partition number.  */
 #define NO_PARTITION		-1
 
-extern var_map init_var_map (int, class loop* = NULL);
+extern var_map init_var_map (int, class loop * = NULL, bitmap = NULL);
 extern void delete_var_map (var_map);
 extern int var_union (var_map, tree, tree);
 extern void partition_view_normal (var_map);
@@ -100,7 +104,7 @@ inline bool
 region_contains_p (var_map map, basic_block bb)
 {
   /* It's possible that the function is called with ENTRY_BLOCK/EXIT_BLOCK.  */
-  if (map->outofssa_p)
+  if (map->outofssa_p || map->bitint)
     return (bb->index != ENTRY_BLOCK && bb->index != EXIT_BLOCK);
 
   return bitmap_bit_p (map->bmp_bbs, bb->index);

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2023-09-06 15:57 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-09-06 15:57 [gcc r14-3746] _BitInt lowering support [PR102989] Jakub Jelinek

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).