Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

1 files changed, 885 insertions, 0 deletions

diff --git a/include/math-emu/op-common.h b/include/math-emu/op-common.h
new file mode 100644
index 000000000..8ce066c03
--- /dev/null
+++ b/include/math-emu/op-common.h
@@ -0,0 +1,885 @@
+/* Software floating-point emulation. Common operations.
+   Copyright (C) 1997,1998,1999 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+   Contributed by Richard Henderson (rth@cygnus.com),
+		  Jakub Jelinek (jj@ultra.linux.cz),
+		  David S. Miller (davem@redhat.com) and
+		  Peter Maydell (pmaydell@chiark.greenend.org.uk).
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Library General Public License as
+   published by the Free Software Foundation; either version 2 of the
+   License, or (at your option) any later version.
+
+   The GNU C Library 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
+   Library General Public License for more details.
+
+   You should have received a copy of the GNU Library General Public
+   License along with the GNU C Library; see the file COPYING.LIB.  If
+   not, write to the Free Software Foundation, Inc.,
+   59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */
+
+#ifndef __MATH_EMU_OP_COMMON_H__
+#define __MATH_EMU_OP_COMMON_H__
+
+#define _FP_DECL(wc, X)			\
+  _FP_I_TYPE X##_c=0, X##_s=0, X##_e=0;	\
+  _FP_FRAC_DECL_##wc(X)
+
+/*
+ * Finish truly unpacking a native fp value by classifying the kind
+ * of fp value and normalizing both the exponent and the fraction.
+ */
+
+#define _FP_UNPACK_CANONICAL(fs, wc, X)					\
+do {									\
+  switch (X##_e)							\
+  {									\
+  default:								\
+    _FP_FRAC_HIGH_RAW_##fs(X) |= _FP_IMPLBIT_##fs;			\
+    _FP_FRAC_SLL_##wc(X, _FP_WORKBITS);					\
+    X##_e -= _FP_EXPBIAS_##fs;						\
+    X##_c = FP_CLS_NORMAL;						\
+    break;								\
+									\
+  case 0:								\
+    if (_FP_FRAC_ZEROP_##wc(X))						\
+      X##_c = FP_CLS_ZERO;						\
+    else								\
+      {									\
+	/* a denormalized number */					\
+	_FP_I_TYPE _shift;						\
+	_FP_FRAC_CLZ_##wc(_shift, X);					\
+	_shift -= _FP_FRACXBITS_##fs;					\
+	_FP_FRAC_SLL_##wc(X, (_shift+_FP_WORKBITS));			\
+	X##_e -= _FP_EXPBIAS_##fs - 1 + _shift;				\
+	X##_c = FP_CLS_NORMAL;						\
+	FP_SET_EXCEPTION(FP_EX_DENORM);					\
+	if (FP_DENORM_ZERO)						\
+	  {								\
+	    FP_SET_EXCEPTION(FP_EX_INEXACT);				\
+	    X##_c = FP_CLS_ZERO;					\
+	  }								\
+      }									\
+    break;								\
+									\
+  case _FP_EXPMAX_##fs:							\
+    if (_FP_FRAC_ZEROP_##wc(X))						\
+      X##_c = FP_CLS_INF;						\
+    else								\
+      {									\
+	X##_c = FP_CLS_NAN;						\
+	/* Check for signaling NaN */					\
+	if (!(_FP_FRAC_HIGH_RAW_##fs(X) & _FP_QNANBIT_##fs))		\
+	  FP_SET_EXCEPTION(FP_EX_INVALID | FP_EX_INVALID_SNAN);		\
+      }									\
+    break;								\
+  }									\
+} while (0)
+
+/*
+ * Before packing the bits back into the native fp result, take care
+ * of such mundane things as rounding and overflow.  Also, for some
+ * kinds of fp values, the original parts may not have been fully
+ * extracted -- but that is ok, we can regenerate them now.
+ */
+
+#define _FP_PACK_CANONICAL(fs, wc, X)				\
+do {								\
+  switch (X##_c)						\
+  {								\
+  case FP_CLS_NORMAL:						\
+    X##_e += _FP_EXPBIAS_##fs;					\
+    if (X##_e > 0)						\
+      {								\
+	_FP_ROUND(wc, X);					\
+	if (_FP_FRAC_OVERP_##wc(fs, X))				\
+	  {							\
+	    _FP_FRAC_CLEAR_OVERP_##wc(fs, X);			\
+	    X##_e++;						\
+	  }							\
+	_FP_FRAC_SRL_##wc(X, _FP_WORKBITS);			\
+	if (X##_e >= _FP_EXPMAX_##fs)				\
+	  {							\
+	    /* overflow */					\
+	    switch (FP_ROUNDMODE)				\
+	      {							\
+	      case FP_RND_NEAREST:				\
+		X##_c = FP_CLS_INF;				\
+		break;						\
+	      case FP_RND_PINF:					\
+		if (!X##_s) X##_c = FP_CLS_INF;			\
+		break;						\
+	      case FP_RND_MINF:					\
+		if (X##_s) X##_c = FP_CLS_INF;			\
+		break;						\
+	      }							\
+	    if (X##_c == FP_CLS_INF)				\
+	      {							\
+		/* Overflow to infinity */			\
+		X##_e = _FP_EXPMAX_##fs;			\
+		_FP_FRAC_SET_##wc(X, _FP_ZEROFRAC_##wc);	\
+	      }							\
+	    else						\
+	      {							\
+		/* Overflow to maximum normal */		\
+		X##_e = _FP_EXPMAX_##fs - 1;			\
+		_FP_FRAC_SET_##wc(X, _FP_MAXFRAC_##wc);		\
+	      }							\
+	    FP_SET_EXCEPTION(FP_EX_OVERFLOW);			\
+            FP_SET_EXCEPTION(FP_EX_INEXACT);			\
+	  }							\
+      }								\
+    else							\
+      {								\
+	/* we've got a denormalized number */			\
+	X##_e = -X##_e + 1;					\
+	if (X##_e <= _FP_WFRACBITS_##fs)			\
+	  {							\
+	    _FP_FRAC_SRS_##wc(X, X##_e, _FP_WFRACBITS_##fs);	\
+	    if (_FP_FRAC_HIGH_##fs(X)				\
+		& (_FP_OVERFLOW_##fs >> 1))			\
+	      {							\
+	        X##_e = 1;					\
+	        _FP_FRAC_SET_##wc(X, _FP_ZEROFRAC_##wc);	\
+	      }							\
+	    else						\
+	      {							\
+		_FP_ROUND(wc, X);				\
+		if (_FP_FRAC_HIGH_##fs(X)			\
+		   & (_FP_OVERFLOW_##fs >> 1))			\
+		  {						\
+		    X##_e = 1;					\
+		    _FP_FRAC_SET_##wc(X, _FP_ZEROFRAC_##wc);	\
+		    FP_SET_EXCEPTION(FP_EX_INEXACT);		\
+		  }						\
+		else						\
+		  {						\
+		    X##_e = 0;					\
+		    _FP_FRAC_SRL_##wc(X, _FP_WORKBITS);		\
+		  }						\
+	      }							\
+	    if ((FP_CUR_EXCEPTIONS & FP_EX_INEXACT) ||		\
+		(FP_TRAPPING_EXCEPTIONS & FP_EX_UNDERFLOW))	\
+		FP_SET_EXCEPTION(FP_EX_UNDERFLOW);		\
+	  }							\
+	else							\
+	  {							\
+	    /* underflow to zero */				\
+	    X##_e = 0;						\
+	    if (!_FP_FRAC_ZEROP_##wc(X))			\
+	      {							\
+	        _FP_FRAC_SET_##wc(X, _FP_MINFRAC_##wc);		\
+	        _FP_ROUND(wc, X);				\
+	        _FP_FRAC_LOW_##wc(X) >>= (_FP_WORKBITS);	\
+	      }							\
+	    FP_SET_EXCEPTION(FP_EX_UNDERFLOW);			\
+	  }							\
+      }								\
+    break;							\
+								\
+  case FP_CLS_ZERO:						\
+    X##_e = 0;							\
+    _FP_FRAC_SET_##wc(X, _FP_ZEROFRAC_##wc);			\
+    break;							\
+								\
+  case FP_CLS_INF:						\
+    X##_e = _FP_EXPMAX_##fs;					\
+    _FP_FRAC_SET_##wc(X, _FP_ZEROFRAC_##wc);			\
+    break;							\
+								\
+  case FP_CLS_NAN:						\
+    X##_e = _FP_EXPMAX_##fs;					\
+    if (!_FP_KEEPNANFRACP)					\
+      {								\
+	_FP_FRAC_SET_##wc(X, _FP_NANFRAC_##fs);			\
+	X##_s = _FP_NANSIGN_##fs;				\
+      }								\
+    else							\
+      _FP_FRAC_HIGH_RAW_##fs(X) |= _FP_QNANBIT_##fs;		\
+    break;							\
+  }								\
+} while (0)
+
+/* This one accepts raw argument and not cooked,  returns
+ * 1 if X is a signaling NaN.
+ */
+#define _FP_ISSIGNAN(fs, wc, X)					\
+({								\
+  int __ret = 0;						\
+  if (X##_e == _FP_EXPMAX_##fs)					\
+    {								\
+      if (!_FP_FRAC_ZEROP_##wc(X)				\
+	  && !(_FP_FRAC_HIGH_RAW_##fs(X) & _FP_QNANBIT_##fs))	\
+	__ret = 1;						\
+    }								\
+  __ret;							\
+})
+
+
+
+
+
+/*
+ * Main addition routine.  The input values should be cooked.
+ */
+
+#define _FP_ADD_INTERNAL(fs, wc, R, X, Y, OP)				     \
+do {									     \
+  switch (_FP_CLS_COMBINE(X##_c, Y##_c))				     \
+  {									     \
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_NORMAL):			     \
+    {									     \
+      /* shift the smaller number so that its exponent matches the larger */ \
+      _FP_I_TYPE diff = X##_e - Y##_e;					     \
+									     \
+      if (diff < 0)							     \
+	{								     \
+	  diff = -diff;							     \
+	  if (diff <= _FP_WFRACBITS_##fs)				     \
+	    _FP_FRAC_SRS_##wc(X, diff, _FP_WFRACBITS_##fs);		     \
+	  else if (!_FP_FRAC_ZEROP_##wc(X))				     \
+	    _FP_FRAC_SET_##wc(X, _FP_MINFRAC_##wc);			     \
+	  R##_e = Y##_e;						     \
+	}								     \
+      else								     \
+	{								     \
+	  if (diff > 0)							     \
+	    {								     \
+	      if (diff <= _FP_WFRACBITS_##fs)				     \
+	        _FP_FRAC_SRS_##wc(Y, diff, _FP_WFRACBITS_##fs);		     \
+	      else if (!_FP_FRAC_ZEROP_##wc(Y))				     \
+	        _FP_FRAC_SET_##wc(Y, _FP_MINFRAC_##wc);			     \
+	    }								     \
+	  R##_e = X##_e;						     \
+	}								     \
+									     \
+      R##_c = FP_CLS_NORMAL;						     \
+									     \
+      if (X##_s == Y##_s)						     \
+	{								     \
+	  R##_s = X##_s;						     \
+	  _FP_FRAC_ADD_##wc(R, X, Y);					     \
+	  if (_FP_FRAC_OVERP_##wc(fs, R))				     \
+	    {								     \
+	      _FP_FRAC_SRS_##wc(R, 1, _FP_WFRACBITS_##fs);		     \
+	      R##_e++;							     \
+	    }								     \
+	}								     \
+      else								     \
+	{								     \
+	  R##_s = X##_s;						     \
+	  _FP_FRAC_SUB_##wc(R, X, Y);					     \
+	  if (_FP_FRAC_ZEROP_##wc(R))					     \
+	    {								     \
+	      /* return an exact zero */				     \
+	      if (FP_ROUNDMODE == FP_RND_MINF)				     \
+		R##_s |= Y##_s;						     \
+	      else							     \
+		R##_s &= Y##_s;						     \
+	      R##_c = FP_CLS_ZERO;					     \
+	    }								     \
+	  else								     \
+	    {								     \
+	      if (_FP_FRAC_NEGP_##wc(R))				     \
+		{							     \
+		  _FP_FRAC_SUB_##wc(R, Y, X);				     \
+		  R##_s = Y##_s;					     \
+		}							     \
+									     \
+	      /* renormalize after subtraction */			     \
+	      _FP_FRAC_CLZ_##wc(diff, R);				     \
+	      diff -= _FP_WFRACXBITS_##fs;				     \
+	      if (diff)							     \
+		{							     \
+		  R##_e -= diff;					     \
+		  _FP_FRAC_SLL_##wc(R, diff);				     \
+		}							     \
+	    }								     \
+	}								     \
+      break;								     \
+    }									     \
+									     \
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_NAN):				     \
+    _FP_CHOOSENAN(fs, wc, R, X, Y, OP);					     \
+    break;								     \
+									     \
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_ZERO):			     \
+    R##_e = X##_e;							     \
+	fallthrough;							     \
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_NORMAL):			     \
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_INF):				     \
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_ZERO):				     \
+    _FP_FRAC_COPY_##wc(R, X);						     \
+    R##_s = X##_s;							     \
+    R##_c = X##_c;							     \
+    break;								     \
+									     \
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_NORMAL):			     \
+    R##_e = Y##_e;							     \
+	fallthrough;							     \
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_NAN):			     \
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_NAN):				     \
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_NAN):				     \
+    _FP_FRAC_COPY_##wc(R, Y);						     \
+    R##_s = Y##_s;							     \
+    R##_c = Y##_c;							     \
+    break;								     \
+									     \
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_INF):				     \
+    if (X##_s != Y##_s)							     \
+      {									     \
+	/* +INF + -INF => NAN */					     \
+	_FP_FRAC_SET_##wc(R, _FP_NANFRAC_##fs);				     \
+	R##_s = _FP_NANSIGN_##fs;					     \
+	R##_c = FP_CLS_NAN;						     \
+	FP_SET_EXCEPTION(FP_EX_INVALID | FP_EX_INVALID_ISI);		     \
+	break;								     \
+      }									     \
+    fallthrough;							     \
+									     \
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_NORMAL):			     \
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_ZERO):				     \
+    R##_s = X##_s;							     \
+    R##_c = FP_CLS_INF;							     \
+    break;								     \
+									     \
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_INF):			     \
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_INF):				     \
+    R##_s = Y##_s;							     \
+    R##_c = FP_CLS_INF;							     \
+    break;								     \
+									     \
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_ZERO):			     \
+    /* make sure the sign is correct */					     \
+    if (FP_ROUNDMODE == FP_RND_MINF)					     \
+      R##_s = X##_s | Y##_s;						     \
+    else								     \
+      R##_s = X##_s & Y##_s;						     \
+    R##_c = FP_CLS_ZERO;						     \
+    break;								     \
+									     \
+  default:								     \
+    abort();								     \
+  }									     \
+} while (0)
+
+#define _FP_ADD(fs, wc, R, X, Y) _FP_ADD_INTERNAL(fs, wc, R, X, Y, '+')
+#define _FP_SUB(fs, wc, R, X, Y)					     \
+  do {									     \
+    if (Y##_c != FP_CLS_NAN) Y##_s ^= 1;				     \
+    _FP_ADD_INTERNAL(fs, wc, R, X, Y, '-');				     \
+  } while (0)
+
+
+/*
+ * Main negation routine.  FIXME -- when we care about setting exception
+ * bits reliably, this will not do.  We should examine all of the fp classes.
+ */
+
+#define _FP_NEG(fs, wc, R, X)		\
+  do {					\
+    _FP_FRAC_COPY_##wc(R, X);		\
+    R##_c = X##_c;			\
+    R##_e = X##_e;			\
+    R##_s = 1 ^ X##_s;			\
+  } while (0)
+
+
+/*
+ * Main multiplication routine.  The input values should be cooked.
+ */
+
+#define _FP_MUL(fs, wc, R, X, Y)			\
+do {							\
+  R##_s = X##_s ^ Y##_s;				\
+  switch (_FP_CLS_COMBINE(X##_c, Y##_c))		\
+  {							\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_NORMAL):	\
+    R##_c = FP_CLS_NORMAL;				\
+    R##_e = X##_e + Y##_e + 1;				\
+							\
+    _FP_MUL_MEAT_##fs(R,X,Y);				\
+							\
+    if (_FP_FRAC_OVERP_##wc(fs, R))			\
+      _FP_FRAC_SRS_##wc(R, 1, _FP_WFRACBITS_##fs);	\
+    else						\
+      R##_e--;						\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_NAN):		\
+    _FP_CHOOSENAN(fs, wc, R, X, Y, '*');		\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_NORMAL):	\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_INF):		\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_ZERO):		\
+    R##_s = X##_s;					\
+	  fallthrough;					\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_INF):		\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_NORMAL):	\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_NORMAL):	\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_ZERO):	\
+    _FP_FRAC_COPY_##wc(R, X);				\
+    R##_c = X##_c;					\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_NAN):	\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_NAN):		\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_NAN):		\
+    R##_s = Y##_s;					\
+	  fallthrough;					\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_INF):	\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_ZERO):	\
+    _FP_FRAC_COPY_##wc(R, Y);				\
+    R##_c = Y##_c;					\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_ZERO):		\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_INF):		\
+    R##_s = _FP_NANSIGN_##fs;				\
+    R##_c = FP_CLS_NAN;					\
+    _FP_FRAC_SET_##wc(R, _FP_NANFRAC_##fs);		\
+    FP_SET_EXCEPTION(FP_EX_INVALID | FP_EX_INVALID_IMZ);\
+    break;						\
+							\
+  default:						\
+    abort();						\
+  }							\
+} while (0)
+
+
+/*
+ * Main division routine.  The input values should be cooked.
+ */
+
+#define _FP_DIV(fs, wc, R, X, Y)			\
+do {							\
+  R##_s = X##_s ^ Y##_s;				\
+  switch (_FP_CLS_COMBINE(X##_c, Y##_c))		\
+  {							\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_NORMAL):	\
+    R##_c = FP_CLS_NORMAL;				\
+    R##_e = X##_e - Y##_e;				\
+							\
+    _FP_DIV_MEAT_##fs(R,X,Y);				\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_NAN):		\
+    _FP_CHOOSENAN(fs, wc, R, X, Y, '/');		\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_NORMAL):	\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_INF):		\
+  case _FP_CLS_COMBINE(FP_CLS_NAN,FP_CLS_ZERO):		\
+    R##_s = X##_s;					\
+    _FP_FRAC_COPY_##wc(R, X);				\
+    R##_c = X##_c;					\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_NAN):	\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_NAN):		\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_NAN):		\
+    R##_s = Y##_s;					\
+    _FP_FRAC_COPY_##wc(R, Y);				\
+    R##_c = Y##_c;					\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_INF):	\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_INF):		\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_NORMAL):	\
+    R##_c = FP_CLS_ZERO;				\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_NORMAL,FP_CLS_ZERO):	\
+    FP_SET_EXCEPTION(FP_EX_DIVZERO);			\
+	fallthrough;					\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_ZERO):		\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_NORMAL):	\
+    R##_c = FP_CLS_INF;					\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_INF):		\
+    R##_s = _FP_NANSIGN_##fs;				\
+    R##_c = FP_CLS_NAN;					\
+    _FP_FRAC_SET_##wc(R, _FP_NANFRAC_##fs);		\
+    FP_SET_EXCEPTION(FP_EX_INVALID | FP_EX_INVALID_IDI);\
+    break;						\
+							\
+  case _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_ZERO):	\
+    R##_s = _FP_NANSIGN_##fs;				\
+    R##_c = FP_CLS_NAN;					\
+    _FP_FRAC_SET_##wc(R, _FP_NANFRAC_##fs);		\
+    FP_SET_EXCEPTION(FP_EX_INVALID | FP_EX_INVALID_ZDZ);\
+    break;						\
+							\
+  default:						\
+    abort();						\
+  }							\
+} while (0)
+
+
+/*
+ * Main differential comparison routine.  The inputs should be raw not
+ * cooked.  The return is -1,0,1 for normal values, 2 otherwise.
+ */
+
+#define _FP_CMP(fs, wc, ret, X, Y, un)					\
+  do {									\
+    /* NANs are unordered */						\
+    if ((X##_e == _FP_EXPMAX_##fs && !_FP_FRAC_ZEROP_##wc(X))		\
+	|| (Y##_e == _FP_EXPMAX_##fs && !_FP_FRAC_ZEROP_##wc(Y)))	\
+      {									\
+	ret = un;							\
+      }									\
+    else								\
+      {									\
+	int __is_zero_x;						\
+	int __is_zero_y;						\
+									\
+	__is_zero_x = (!X##_e && _FP_FRAC_ZEROP_##wc(X)) ? 1 : 0;	\
+	__is_zero_y = (!Y##_e && _FP_FRAC_ZEROP_##wc(Y)) ? 1 : 0;	\
+									\
+	if (__is_zero_x && __is_zero_y)					\
+		ret = 0;						\
+	else if (__is_zero_x)						\
+		ret = Y##_s ? 1 : -1;					\
+	else if (__is_zero_y)						\
+		ret = X##_s ? -1 : 1;					\
+	else if (X##_s != Y##_s)					\
+	  ret = X##_s ? -1 : 1;						\
+	else if (X##_e > Y##_e)						\
+	  ret = X##_s ? -1 : 1;						\
+	else if (X##_e < Y##_e)						\
+	  ret = X##_s ? 1 : -1;						\
+	else if (_FP_FRAC_GT_##wc(X, Y))				\
+	  ret = X##_s ? -1 : 1;						\
+	else if (_FP_FRAC_GT_##wc(Y, X))				\
+	  ret = X##_s ? 1 : -1;						\
+	else								\
+	  ret = 0;							\
+      }									\
+  } while (0)
+
+
+/* Simplification for strict equality.  */
+
+#define _FP_CMP_EQ(fs, wc, ret, X, Y)					  \
+  do {									  \
+    /* NANs are unordered */						  \
+    if ((X##_e == _FP_EXPMAX_##fs && !_FP_FRAC_ZEROP_##wc(X))		  \
+	|| (Y##_e == _FP_EXPMAX_##fs && !_FP_FRAC_ZEROP_##wc(Y)))	  \
+      {									  \
+	ret = 1;							  \
+      }									  \
+    else								  \
+      {									  \
+	ret = !(X##_e == Y##_e						  \
+		&& _FP_FRAC_EQ_##wc(X, Y)				  \
+		&& (X##_s == Y##_s || !X##_e && _FP_FRAC_ZEROP_##wc(X))); \
+      }									  \
+  } while (0)
+
+/*
+ * Main square root routine.  The input value should be cooked.
+ */
+
+#define _FP_SQRT(fs, wc, R, X)						\
+do {									\
+    _FP_FRAC_DECL_##wc(T); _FP_FRAC_DECL_##wc(S);			\
+    _FP_W_TYPE q;							\
+    switch (X##_c)							\
+    {									\
+    case FP_CLS_NAN:							\
+	_FP_FRAC_COPY_##wc(R, X);					\
+	R##_s = X##_s;							\
+    	R##_c = FP_CLS_NAN;						\
+    	break;								\
+    case FP_CLS_INF:							\
+    	if (X##_s)							\
+    	  {								\
+    	    R##_s = _FP_NANSIGN_##fs;					\
+	    R##_c = FP_CLS_NAN; /* NAN */				\
+	    _FP_FRAC_SET_##wc(R, _FP_NANFRAC_##fs);			\
+	    FP_SET_EXCEPTION(FP_EX_INVALID);				\
+    	  }								\
+    	else								\
+    	  {								\
+    	    R##_s = 0;							\
+    	    R##_c = FP_CLS_INF; /* sqrt(+inf) = +inf */			\
+    	  }								\
+    	break;								\
+    case FP_CLS_ZERO:							\
+	R##_s = X##_s;							\
+	R##_c = FP_CLS_ZERO; /* sqrt(+-0) = +-0 */			\
+	break;								\
+    case FP_CLS_NORMAL:							\
+    	R##_s = 0;							\
+        if (X##_s)							\
+          {								\
+	    R##_c = FP_CLS_NAN; /* sNAN */				\
+	    R##_s = _FP_NANSIGN_##fs;					\
+	    _FP_FRAC_SET_##wc(R, _FP_NANFRAC_##fs);			\
+	    FP_SET_EXCEPTION(FP_EX_INVALID);				\
+	    break;							\
+          }								\
+    	R##_c = FP_CLS_NORMAL;						\
+        if (X##_e & 1)							\
+          _FP_FRAC_SLL_##wc(X, 1);					\
+        R##_e = X##_e >> 1;						\
+        _FP_FRAC_SET_##wc(S, _FP_ZEROFRAC_##wc);			\
+        _FP_FRAC_SET_##wc(R, _FP_ZEROFRAC_##wc);			\
+        q = _FP_OVERFLOW_##fs >> 1;					\
+        _FP_SQRT_MEAT_##wc(R, S, T, X, q);				\
+    }									\
+  } while (0)
+
+/*
+ * Convert from FP to integer
+ */
+
+/* RSIGNED can have following values:
+ * 0:  the number is required to be 0..(2^rsize)-1, if not, NV is set plus
+ *     the result is either 0 or (2^rsize)-1 depending on the sign in such case.
+ * 1:  the number is required to be -(2^(rsize-1))..(2^(rsize-1))-1, if not, NV is
+ *     set plus the result is either -(2^(rsize-1)) or (2^(rsize-1))-1 depending
+ *     on the sign in such case.
+ * 2:  the number is required to be -(2^(rsize-1))..(2^(rsize-1))-1, if not, NV is
+ *     set plus the result is truncated to fit into destination.
+ * -1: the number is required to be -(2^(rsize-1))..(2^rsize)-1, if not, NV is
+ *     set plus the result is either -(2^(rsize-1)) or (2^(rsize-1))-1 depending
+ *     on the sign in such case.
+ */
+#define _FP_TO_INT(fs, wc, r, X, rsize, rsigned)				\
+  do {										\
+    switch (X##_c)								\
+      {										\
+      case FP_CLS_NORMAL:							\
+	if (X##_e < 0)								\
+	  {									\
+	    FP_SET_EXCEPTION(FP_EX_INEXACT);					\
+	    fallthrough;							\
+	  case FP_CLS_ZERO:							\
+	    r = 0;								\
+	  }									\
+	else if (X##_e >= rsize - (rsigned > 0 || X##_s)			\
+		 || (!rsigned && X##_s))					\
+	  {	/* overflow */							\
+	    fallthrough;							\
+	  case FP_CLS_NAN:                                                      \
+	  case FP_CLS_INF:							\
+	    if (rsigned == 2)							\
+	      {									\
+		if (X##_c != FP_CLS_NORMAL					\
+		    || X##_e >= rsize - 1 + _FP_WFRACBITS_##fs)			\
+		  r = 0;							\
+		else								\
+		  {								\
+		    _FP_FRAC_SLL_##wc(X, (X##_e - _FP_WFRACBITS_##fs + 1));	\
+		    _FP_FRAC_ASSEMBLE_##wc(r, X, rsize);			\
+		  }								\
+	      }									\
+	    else if (rsigned)							\
+	      {									\
+		r = 1;								\
+		r <<= rsize - 1;						\
+		r -= 1 - X##_s;							\
+	      }									\
+	    else								\
+	      {									\
+		r = 0;								\
+		if (!X##_s)							\
+		  r = ~r;							\
+	      }									\
+	    FP_SET_EXCEPTION(FP_EX_INVALID);					\
+	  }									\
+	else									\
+	  {									\
+	    if (_FP_W_TYPE_SIZE*wc < rsize)					\
+	      {									\
+		_FP_FRAC_ASSEMBLE_##wc(r, X, rsize);				\
+		r <<= X##_e - _FP_WFRACBITS_##fs;				\
+	      }									\
+	    else								\
+	      {									\
+		if (X##_e >= _FP_WFRACBITS_##fs)				\
+		  _FP_FRAC_SLL_##wc(X, (X##_e - _FP_WFRACBITS_##fs + 1));	\
+		else if (X##_e < _FP_WFRACBITS_##fs - 1)			\
+		  {								\
+		    _FP_FRAC_SRS_##wc(X, (_FP_WFRACBITS_##fs - X##_e - 2),	\
+				      _FP_WFRACBITS_##fs);			\
+		    if (_FP_FRAC_LOW_##wc(X) & 1)				\
+		      FP_SET_EXCEPTION(FP_EX_INEXACT);				\
+		    _FP_FRAC_SRL_##wc(X, 1);					\
+		  }								\
+		_FP_FRAC_ASSEMBLE_##wc(r, X, rsize);				\
+	      }									\
+	    if (rsigned && X##_s)						\
+	      r = -r;								\
+	  }									\
+	break;									\
+      }										\
+  } while (0)
+
+#define _FP_TO_INT_ROUND(fs, wc, r, X, rsize, rsigned)				\
+  do {										\
+    r = 0;									\
+    switch (X##_c)								\
+      {										\
+      case FP_CLS_NORMAL:							\
+	if (X##_e >= _FP_FRACBITS_##fs - 1)					\
+	  {									\
+	    if (X##_e < rsize - 1 + _FP_WFRACBITS_##fs)				\
+	      {									\
+		if (X##_e >= _FP_WFRACBITS_##fs - 1)				\
+		  {								\
+		    _FP_FRAC_ASSEMBLE_##wc(r, X, rsize);			\
+		    r <<= X##_e - _FP_WFRACBITS_##fs + 1;			\
+		  }								\
+		else								\
+		  {								\
+		    _FP_FRAC_SRL_##wc(X, _FP_WORKBITS - X##_e			\
+				      + _FP_FRACBITS_##fs - 1);			\
+		    _FP_FRAC_ASSEMBLE_##wc(r, X, rsize);			\
+		  }								\
+	      }									\
+	  }									\
+	else									\
+	  {									\
+	    int _lz0, _lz1;							\
+	    if (X##_e <= -_FP_WORKBITS - 1)					\
+	      _FP_FRAC_SET_##wc(X, _FP_MINFRAC_##wc);				\
+	    else								\
+	      _FP_FRAC_SRS_##wc(X, _FP_FRACBITS_##fs - 1 - X##_e,		\
+				_FP_WFRACBITS_##fs);				\
+	    _FP_FRAC_CLZ_##wc(_lz0, X);						\
+	    _FP_ROUND(wc, X);							\
+	    _FP_FRAC_CLZ_##wc(_lz1, X);						\
+	    if (_lz1 < _lz0)							\
+	      X##_e++; /* For overflow detection.  */				\
+	    _FP_FRAC_SRL_##wc(X, _FP_WORKBITS);					\
+	    _FP_FRAC_ASSEMBLE_##wc(r, X, rsize);				\
+	  }									\
+	if (rsigned && X##_s)							\
+	  r = -r;								\
+	if (X##_e >= rsize - (rsigned > 0 || X##_s)				\
+	    || (!rsigned && X##_s))						\
+	  {	/* overflow */							\
+	    fallthrough;							\
+	  case FP_CLS_NAN:                                                      \
+	  case FP_CLS_INF:							\
+	    if (!rsigned)							\
+	      {									\
+		r = 0;								\
+		if (!X##_s)							\
+		  r = ~r;							\
+	      }									\
+	    else if (rsigned != 2)						\
+	      {									\
+		r = 1;								\
+		r <<= rsize - 1;						\
+		r -= 1 - X##_s;							\
+	      }									\
+	    FP_SET_EXCEPTION(FP_EX_INVALID);					\
+	  }									\
+	break;									\
+      case FP_CLS_ZERO:								\
+        break;									\
+      }										\
+  } while (0)
+
+#define _FP_FROM_INT(fs, wc, X, r, rsize, rtype)			\
+  do {									\
+    if (r)								\
+      {									\
+        unsigned rtype ur_;						\
+	X##_c = FP_CLS_NORMAL;						\
+									\
+	if ((X##_s = (r < 0)))						\
+	  ur_ = (unsigned rtype) -r;					\
+	else								\
+	  ur_ = (unsigned rtype) r;					\
+	(void) (((rsize) <= _FP_W_TYPE_SIZE)				\
+		? ({ __FP_CLZ(X##_e, ur_); })				\
+		: ({							\
+		     __FP_CLZ_2(X##_e, (_FP_W_TYPE)(ur_ >> _FP_W_TYPE_SIZE),  \
+							    (_FP_W_TYPE)ur_); \
+		  }));							\
+	if (rsize < _FP_W_TYPE_SIZE)					\
+		X##_e -= (_FP_W_TYPE_SIZE - rsize);			\
+	X##_e = rsize - X##_e - 1;					\
+									\
+	if (_FP_FRACBITS_##fs < rsize && _FP_WFRACBITS_##fs <= X##_e)	\
+	  __FP_FRAC_SRS_1(ur_, (X##_e - _FP_WFRACBITS_##fs + 1), rsize);\
+	_FP_FRAC_DISASSEMBLE_##wc(X, ur_, rsize);			\
+	if ((_FP_WFRACBITS_##fs - X##_e - 1) > 0)			\
+	  _FP_FRAC_SLL_##wc(X, (_FP_WFRACBITS_##fs - X##_e - 1));	\
+      }									\
+    else								\
+      {									\
+	X##_c = FP_CLS_ZERO, X##_s = 0;					\
+      }									\
+  } while (0)
+
+
+#define FP_CONV(dfs,sfs,dwc,swc,D,S)			\
+  do {							\
+    _FP_FRAC_CONV_##dwc##_##swc(dfs, sfs, D, S);	\
+    D##_e = S##_e;					\
+    D##_c = S##_c;					\
+    D##_s = S##_s;					\
+  } while (0)
+
+/*
+ * Helper primitives.
+ */
+
+/* Count leading zeros in a word.  */
+
+#ifndef __FP_CLZ
+#if _FP_W_TYPE_SIZE < 64
+/* this is just to shut the compiler up about shifts > word length -- PMM 02/1998 */
+#define __FP_CLZ(r, x)				\
+  do {						\
+    _FP_W_TYPE _t = (x);			\
+    r = _FP_W_TYPE_SIZE - 1;			\
+    if (_t > 0xffff) r -= 16;			\
+    if (_t > 0xffff) _t >>= 16;			\
+    if (_t > 0xff) r -= 8;			\
+    if (_t > 0xff) _t >>= 8;			\
+    if (_t & 0xf0) r -= 4;			\
+    if (_t & 0xf0) _t >>= 4;			\
+    if (_t & 0xc) r -= 2;			\
+    if (_t & 0xc) _t >>= 2;			\
+    if (_t & 0x2) r -= 1;			\
+  } while (0)
+#else /* not _FP_W_TYPE_SIZE < 64 */
+#define __FP_CLZ(r, x)				\
+  do {						\
+    _FP_W_TYPE _t = (x);			\
+    r = _FP_W_TYPE_SIZE - 1;			\
+    if (_t > 0xffffffff) r -= 32;		\
+    if (_t > 0xffffffff) _t >>= 32;		\
+    if (_t > 0xffff) r -= 16;			\
+    if (_t > 0xffff) _t >>= 16;			\
+    if (_t > 0xff) r -= 8;			\
+    if (_t > 0xff) _t >>= 8;			\
+    if (_t & 0xf0) r -= 4;			\
+    if (_t & 0xf0) _t >>= 4;			\
+    if (_t & 0xc) r -= 2;			\
+    if (_t & 0xc) _t >>= 2;			\
+    if (_t & 0x2) r -= 1;			\
+  } while (0)
+#endif /* not _FP_W_TYPE_SIZE < 64 */
+#endif /* ndef __FP_CLZ */
+
+#define _FP_DIV_HELP_imm(q, r, n, d)		\
+  do {						\
+    q = n / d, r = n % d;			\
+  } while (0)
+
+#endif /* __MATH_EMU_OP_COMMON_H__ */