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, 472 insertions, 0 deletions

diff --git a/arch/x86/math-emu/wm_sqrt.S b/arch/x86/math-emu/wm_sqrt.S
new file mode 100644
index 000000000..40526dd85
--- /dev/null
+++ b/arch/x86/math-emu/wm_sqrt.S
@@ -0,0 +1,472 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+	.file	"wm_sqrt.S"
+/*---------------------------------------------------------------------------+
+ |  wm_sqrt.S                                                                |
+ |                                                                           |
+ | Fixed point arithmetic square root evaluation.                            |
+ |                                                                           |
+ | Copyright (C) 1992,1993,1995,1997                                         |
+ |                       W. Metzenthen, 22 Parker St, Ormond, Vic 3163,      |
+ |                       Australia.  E-mail billm@suburbia.net               |
+ |                                                                           |
+ | Call from C as:                                                           |
+ |    int wm_sqrt(FPU_REG *n, unsigned int control_word)                     |
+ |                                                                           |
+ +---------------------------------------------------------------------------*/
+
+/*---------------------------------------------------------------------------+
+ |  wm_sqrt(FPU_REG *n, unsigned int control_word)                           |
+ |    returns the square root of n in n.                                     |
+ |                                                                           |
+ |  Use Newton's method to compute the square root of a number, which must   |
+ |  be in the range  [1.0 .. 4.0),  to 64 bits accuracy.                     |
+ |  Does not check the sign or tag of the argument.                          |
+ |  Sets the exponent, but not the sign or tag of the result.                |
+ |                                                                           |
+ |  The guess is kept in %esi:%edi                                           |
+ +---------------------------------------------------------------------------*/
+
+#include "exception.h"
+#include "fpu_emu.h"
+
+
+#ifndef NON_REENTRANT_FPU
+/*	Local storage on the stack: */
+#define FPU_accum_3	-4(%ebp)	/* ms word */
+#define FPU_accum_2	-8(%ebp)
+#define FPU_accum_1	-12(%ebp)
+#define FPU_accum_0	-16(%ebp)
+
+/*
+ * The de-normalised argument:
+ *                  sq_2                  sq_1              sq_0
+ *        b b b b b b b ... b b b   b b b .... b b b   b 0 0 0 ... 0
+ *           ^ binary point here
+ */
+#define FPU_fsqrt_arg_2	-20(%ebp)	/* ms word */
+#define FPU_fsqrt_arg_1	-24(%ebp)
+#define FPU_fsqrt_arg_0	-28(%ebp)	/* ls word, at most the ms bit is set */
+
+#else
+/*	Local storage in a static area: */
+.data
+	.align 4,0
+FPU_accum_3:
+	.long	0		/* ms word */
+FPU_accum_2:
+	.long	0
+FPU_accum_1:
+	.long	0
+FPU_accum_0:
+	.long	0
+
+/* The de-normalised argument:
+                    sq_2                  sq_1              sq_0
+          b b b b b b b ... b b b   b b b .... b b b   b 0 0 0 ... 0
+             ^ binary point here
+ */
+FPU_fsqrt_arg_2:
+	.long	0		/* ms word */
+FPU_fsqrt_arg_1:
+	.long	0
+FPU_fsqrt_arg_0:
+	.long	0		/* ls word, at most the ms bit is set */
+#endif /* NON_REENTRANT_FPU */ 
+
+
+.text
+SYM_FUNC_START(wm_sqrt)
+	pushl	%ebp
+	movl	%esp,%ebp
+#ifndef NON_REENTRANT_FPU
+	subl	$28,%esp
+#endif /* NON_REENTRANT_FPU */
+	pushl	%esi
+	pushl	%edi
+	pushl	%ebx
+
+	movl	PARAM1,%esi
+
+	movl	SIGH(%esi),%eax
+	movl	SIGL(%esi),%ecx
+	xorl	%edx,%edx
+
+/* We use a rough linear estimate for the first guess.. */
+
+	cmpw	EXP_BIAS,EXP(%esi)
+	jnz	sqrt_arg_ge_2
+
+	shrl	$1,%eax			/* arg is in the range  [1.0 .. 2.0) */
+	rcrl	$1,%ecx
+	rcrl	$1,%edx
+
+sqrt_arg_ge_2:
+/* From here on, n is never accessed directly again until it is
+   replaced by the answer. */
+
+	movl	%eax,FPU_fsqrt_arg_2		/* ms word of n */
+	movl	%ecx,FPU_fsqrt_arg_1
+	movl	%edx,FPU_fsqrt_arg_0
+
+/* Make a linear first estimate */
+	shrl	$1,%eax
+	addl	$0x40000000,%eax
+	movl	$0xaaaaaaaa,%ecx
+	mull	%ecx
+	shll	%edx			/* max result was 7fff... */
+	testl	$0x80000000,%edx	/* but min was 3fff... */
+	jnz	sqrt_prelim_no_adjust
+
+	movl	$0x80000000,%edx	/* round up */
+
+sqrt_prelim_no_adjust:
+	movl	%edx,%esi	/* Our first guess */
+
+/* We have now computed (approx)   (2 + x) / 3, which forms the basis
+   for a few iterations of Newton's method */
+
+	movl	FPU_fsqrt_arg_2,%ecx	/* ms word */
+
+/*
+ * From our initial estimate, three iterations are enough to get us
+ * to 30 bits or so. This will then allow two iterations at better
+ * precision to complete the process.
+ */
+
+/* Compute  (g + n/g)/2  at each iteration (g is the guess). */
+	shrl	%ecx		/* Doing this first will prevent a divide */
+				/* overflow later. */
+
+	movl	%ecx,%edx	/* msw of the arg / 2 */
+	divl	%esi		/* current estimate */
+	shrl	%esi		/* divide by 2 */
+	addl	%eax,%esi	/* the new estimate */
+
+	movl	%ecx,%edx
+	divl	%esi
+	shrl	%esi
+	addl	%eax,%esi
+
+	movl	%ecx,%edx
+	divl	%esi
+	shrl	%esi
+	addl	%eax,%esi
+
+/*
+ * Now that an estimate accurate to about 30 bits has been obtained (in %esi),
+ * we improve it to 60 bits or so.
+ *
+ * The strategy from now on is to compute new estimates from
+ *      guess := guess + (n - guess^2) / (2 * guess)
+ */
+
+/* First, find the square of the guess */
+	movl	%esi,%eax
+	mull	%esi
+/* guess^2 now in %edx:%eax */
+
+	movl	FPU_fsqrt_arg_1,%ecx
+	subl	%ecx,%eax
+	movl	FPU_fsqrt_arg_2,%ecx	/* ms word of normalized n */
+	sbbl	%ecx,%edx
+	jnc	sqrt_stage_2_positive
+
+/* Subtraction gives a negative result,
+   negate the result before division. */
+	notl	%edx
+	notl	%eax
+	addl	$1,%eax
+	adcl	$0,%edx
+
+	divl	%esi
+	movl	%eax,%ecx
+
+	movl	%edx,%eax
+	divl	%esi
+	jmp	sqrt_stage_2_finish
+
+sqrt_stage_2_positive:
+	divl	%esi
+	movl	%eax,%ecx
+
+	movl	%edx,%eax
+	divl	%esi
+
+	notl	%ecx
+	notl	%eax
+	addl	$1,%eax
+	adcl	$0,%ecx
+
+sqrt_stage_2_finish:
+	sarl	$1,%ecx		/* divide by 2 */
+	rcrl	$1,%eax
+
+	/* Form the new estimate in %esi:%edi */
+	movl	%eax,%edi
+	addl	%ecx,%esi
+
+	jnz	sqrt_stage_2_done	/* result should be [1..2) */
+
+#ifdef PARANOID
+/* It should be possible to get here only if the arg is ffff....ffff */
+	cmpl	$0xffffffff,FPU_fsqrt_arg_1
+	jnz	sqrt_stage_2_error
+#endif /* PARANOID */
+
+/* The best rounded result. */
+	xorl	%eax,%eax
+	decl	%eax
+	movl	%eax,%edi
+	movl	%eax,%esi
+	movl	$0x7fffffff,%eax
+	jmp	sqrt_round_result
+
+#ifdef PARANOID
+sqrt_stage_2_error:
+	pushl	EX_INTERNAL|0x213
+	call	EXCEPTION
+#endif /* PARANOID */ 
+
+sqrt_stage_2_done:
+
+/* Now the square root has been computed to better than 60 bits. */
+
+/* Find the square of the guess. */
+	movl	%edi,%eax		/* ls word of guess */
+	mull	%edi
+	movl	%edx,FPU_accum_1
+
+	movl	%esi,%eax
+	mull	%esi
+	movl	%edx,FPU_accum_3
+	movl	%eax,FPU_accum_2
+
+	movl	%edi,%eax
+	mull	%esi
+	addl	%eax,FPU_accum_1
+	adcl	%edx,FPU_accum_2
+	adcl	$0,FPU_accum_3
+
+/*	movl	%esi,%eax */
+/*	mull	%edi */
+	addl	%eax,FPU_accum_1
+	adcl	%edx,FPU_accum_2
+	adcl	$0,FPU_accum_3
+
+/* guess^2 now in FPU_accum_3:FPU_accum_2:FPU_accum_1 */
+
+	movl	FPU_fsqrt_arg_0,%eax		/* get normalized n */
+	subl	%eax,FPU_accum_1
+	movl	FPU_fsqrt_arg_1,%eax
+	sbbl	%eax,FPU_accum_2
+	movl	FPU_fsqrt_arg_2,%eax		/* ms word of normalized n */
+	sbbl	%eax,FPU_accum_3
+	jnc	sqrt_stage_3_positive
+
+/* Subtraction gives a negative result,
+   negate the result before division */
+	notl	FPU_accum_1
+	notl	FPU_accum_2
+	notl	FPU_accum_3
+	addl	$1,FPU_accum_1
+	adcl	$0,FPU_accum_2
+
+#ifdef PARANOID
+	adcl	$0,FPU_accum_3	/* This must be zero */
+	jz	sqrt_stage_3_no_error
+
+sqrt_stage_3_error:
+	pushl	EX_INTERNAL|0x207
+	call	EXCEPTION
+
+sqrt_stage_3_no_error:
+#endif /* PARANOID */
+
+	movl	FPU_accum_2,%edx
+	movl	FPU_accum_1,%eax
+	divl	%esi
+	movl	%eax,%ecx
+
+	movl	%edx,%eax
+	divl	%esi
+
+	sarl	$1,%ecx		/* divide by 2 */
+	rcrl	$1,%eax
+
+	/* prepare to round the result */
+
+	addl	%ecx,%edi
+	adcl	$0,%esi
+
+	jmp	sqrt_stage_3_finished
+
+sqrt_stage_3_positive:
+	movl	FPU_accum_2,%edx
+	movl	FPU_accum_1,%eax
+	divl	%esi
+	movl	%eax,%ecx
+
+	movl	%edx,%eax
+	divl	%esi
+
+	sarl	$1,%ecx		/* divide by 2 */
+	rcrl	$1,%eax
+
+	/* prepare to round the result */
+
+	notl	%eax		/* Negate the correction term */
+	notl	%ecx
+	addl	$1,%eax
+	adcl	$0,%ecx		/* carry here ==> correction == 0 */
+	adcl	$0xffffffff,%esi
+
+	addl	%ecx,%edi
+	adcl	$0,%esi
+
+sqrt_stage_3_finished:
+
+/*
+ * The result in %esi:%edi:%esi should be good to about 90 bits here,
+ * and the rounding information here does not have sufficient accuracy
+ * in a few rare cases.
+ */
+	cmpl	$0xffffffe0,%eax
+	ja	sqrt_near_exact_x
+
+	cmpl	$0x00000020,%eax
+	jb	sqrt_near_exact
+
+	cmpl	$0x7fffffe0,%eax
+	jb	sqrt_round_result
+
+	cmpl	$0x80000020,%eax
+	jb	sqrt_get_more_precision
+
+sqrt_round_result:
+/* Set up for rounding operations */
+	movl	%eax,%edx
+	movl	%esi,%eax
+	movl	%edi,%ebx
+	movl	PARAM1,%edi
+	movw	EXP_BIAS,EXP(%edi)	/* Result is in  [1.0 .. 2.0) */
+	jmp	fpu_reg_round
+
+
+sqrt_near_exact_x:
+/* First, the estimate must be rounded up. */
+	addl	$1,%edi
+	adcl	$0,%esi
+
+sqrt_near_exact:
+/*
+ * This is an easy case because x^1/2 is monotonic.
+ * We need just find the square of our estimate, compare it
+ * with the argument, and deduce whether our estimate is
+ * above, below, or exact. We use the fact that the estimate
+ * is known to be accurate to about 90 bits.
+ */
+	movl	%edi,%eax		/* ls word of guess */
+	mull	%edi
+	movl	%edx,%ebx		/* 2nd ls word of square */
+	movl	%eax,%ecx		/* ls word of square */
+
+	movl	%edi,%eax
+	mull	%esi
+	addl	%eax,%ebx
+	addl	%eax,%ebx
+
+#ifdef PARANOID
+	cmp	$0xffffffb0,%ebx
+	jb	sqrt_near_exact_ok
+
+	cmp	$0x00000050,%ebx
+	ja	sqrt_near_exact_ok
+
+	pushl	EX_INTERNAL|0x214
+	call	EXCEPTION
+
+sqrt_near_exact_ok:
+#endif /* PARANOID */ 
+
+	or	%ebx,%ebx
+	js	sqrt_near_exact_small
+
+	jnz	sqrt_near_exact_large
+
+	or	%ebx,%edx
+	jnz	sqrt_near_exact_large
+
+/* Our estimate is exactly the right answer */
+	xorl	%eax,%eax
+	jmp	sqrt_round_result
+
+sqrt_near_exact_small:
+/* Our estimate is too small */
+	movl	$0x000000ff,%eax
+	jmp	sqrt_round_result
+	
+sqrt_near_exact_large:
+/* Our estimate is too large, we need to decrement it */
+	subl	$1,%edi
+	sbbl	$0,%esi
+	movl	$0xffffff00,%eax
+	jmp	sqrt_round_result
+
+
+sqrt_get_more_precision:
+/* This case is almost the same as the above, except we start
+   with an extra bit of precision in the estimate. */
+	stc			/* The extra bit. */
+	rcll	$1,%edi		/* Shift the estimate left one bit */
+	rcll	$1,%esi
+
+	movl	%edi,%eax		/* ls word of guess */
+	mull	%edi
+	movl	%edx,%ebx		/* 2nd ls word of square */
+	movl	%eax,%ecx		/* ls word of square */
+
+	movl	%edi,%eax
+	mull	%esi
+	addl	%eax,%ebx
+	addl	%eax,%ebx
+
+/* Put our estimate back to its original value */
+	stc			/* The ms bit. */
+	rcrl	$1,%esi		/* Shift the estimate left one bit */
+	rcrl	$1,%edi
+
+#ifdef PARANOID
+	cmp	$0xffffff60,%ebx
+	jb	sqrt_more_prec_ok
+
+	cmp	$0x000000a0,%ebx
+	ja	sqrt_more_prec_ok
+
+	pushl	EX_INTERNAL|0x215
+	call	EXCEPTION
+
+sqrt_more_prec_ok:
+#endif /* PARANOID */ 
+
+	or	%ebx,%ebx
+	js	sqrt_more_prec_small
+
+	jnz	sqrt_more_prec_large
+
+	or	%ebx,%ecx
+	jnz	sqrt_more_prec_large
+
+/* Our estimate is exactly the right answer */
+	movl	$0x80000000,%eax
+	jmp	sqrt_round_result
+
+sqrt_more_prec_small:
+/* Our estimate is too small */
+	movl	$0x800000ff,%eax
+	jmp	sqrt_round_result
+	
+sqrt_more_prec_large:
+/* Our estimate is too large */
+	movl	$0x7fffff00,%eax
+	jmp	sqrt_round_result
+SYM_FUNC_END(wm_sqrt)