view gcc/config/i386/i386.md @ 112:ab0bcb71f44d

merge gcc 7
author mir3636
date Fri, 10 Nov 2017 19:46:38 +0900
parents 04ced10e8804 1b10fe6932e1
children d34655255c78
line wrap: on
line source

;; GCC machine description for IA-32 and x86-64.
;; Copyright (C) 1988-2017 Free Software Foundation, Inc.
;; Mostly by William Schelter.
;; x86_64 support added by Jan Hubicka
;;
;; 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/>.  */
;;
;; The original PO technology requires these to be ordered by speed,
;; so that assigner will pick the fastest.
;;
;; See file "rtl.def" for documentation on define_insn, match_*, et. al.
;;
;; The special asm out single letter directives following a '%' are:
;; L,W,B,Q,S,T -- print the opcode suffix for specified size of operand.
;; C -- print opcode suffix for set/cmov insn.
;; c -- like C, but print reversed condition
;; F,f -- likewise, but for floating-point.
;; O -- if HAVE_AS_IX86_CMOV_SUN_SYNTAX, expand to "w.", "l." or "q.",
;;      otherwise nothing
;; R -- print the prefix for register names.
;; z -- print the opcode suffix for the size of the current operand.
;; Z -- likewise, with special suffixes for x87 instructions.
;; * -- print a star (in certain assembler syntax)
;; A -- print an absolute memory reference.
;; E -- print address with DImode register names if TARGET_64BIT.
;; w -- print the operand as if it's a "word" (HImode) even if it isn't.
;; s -- print a shift double count, followed by the assemblers argument
;;	delimiter.
;; b -- print the QImode name of the register for the indicated operand.
;;	%b0 would print %al if operands[0] is reg 0.
;; w --  likewise, print the HImode name of the register.
;; k --  likewise, print the SImode name of the register.
;; q --  likewise, print the DImode name of the register.
;; x --  likewise, print the V4SFmode name of the register.
;; t --  likewise, print the V8SFmode name of the register.
;; h -- print the QImode name for a "high" register, either ah, bh, ch or dh.
;; y -- print "st(0)" instead of "st" as a register.
;; d -- print duplicated register operand for AVX instruction.
;; D -- print condition for SSE cmp instruction.
;; P -- if PIC, print an @PLT suffix.
;; p -- print raw symbol name.
;; X -- don't print any sort of PIC '@' suffix for a symbol.
;; & -- print some in-use local-dynamic symbol name.
;; H -- print a memory address offset by 8; used for sse high-parts
;; K -- print HLE lock prefix
;; Y -- print condition for XOP pcom* instruction.
;; + -- print a branch hint as 'cs' or 'ds' prefix
;; ; -- print a semicolon (after prefixes due to bug in older gas).
;; ~ -- print "i" if TARGET_AVX2, "f" otherwise.
;; ^ -- print addr32 prefix if TARGET_64BIT and Pmode != word_mode
;; ! -- print MPX or NOTRACK prefix for jxx/call/ret instructions if required.

(define_c_enum "unspec" [
  ;; Relocation specifiers
  UNSPEC_GOT
  UNSPEC_GOTOFF
  UNSPEC_GOTPCREL
  UNSPEC_GOTTPOFF
  UNSPEC_TPOFF
  UNSPEC_NTPOFF
  UNSPEC_DTPOFF
  UNSPEC_GOTNTPOFF
  UNSPEC_INDNTPOFF
  UNSPEC_PLTOFF
  UNSPEC_MACHOPIC_OFFSET
  UNSPEC_PCREL
  UNSPEC_SIZEOF

  ;; Prologue support
  UNSPEC_STACK_ALLOC
  UNSPEC_SET_GOT
  UNSPEC_SET_RIP
  UNSPEC_SET_GOT_OFFSET
  UNSPEC_MEMORY_BLOCKAGE
  UNSPEC_PROBE_STACK

  ;; TLS support
  UNSPEC_TP
  UNSPEC_TLS_GD
  UNSPEC_TLS_LD_BASE
  UNSPEC_TLSDESC
  UNSPEC_TLS_IE_SUN

  ;; Other random patterns
  UNSPEC_SCAS
  UNSPEC_FNSTSW
  UNSPEC_SAHF
  UNSPEC_PARITY
  UNSPEC_FSTCW
  UNSPEC_FLDCW
  UNSPEC_REP
  UNSPEC_LD_MPIC	; load_macho_picbase
  UNSPEC_TRUNC_NOOP
  UNSPEC_DIV_ALREADY_SPLIT
  UNSPEC_PAUSE
  UNSPEC_LEA_ADDR
  UNSPEC_XBEGIN_ABORT
  UNSPEC_STOS
  UNSPEC_PEEPSIB
  UNSPEC_INSN_FALSE_DEP
  UNSPEC_SBB

  ;; For SSE/MMX support:
  UNSPEC_FIX_NOTRUNC
  UNSPEC_MASKMOV
  UNSPEC_MOVMSK
  UNSPEC_RCP
  UNSPEC_RSQRT
  UNSPEC_PSADBW

  ;; Generic math support
  UNSPEC_COPYSIGN
  UNSPEC_IEEE_MIN	; not commutative
  UNSPEC_IEEE_MAX	; not commutative

  ;; x87 Floating point
  UNSPEC_SIN
  UNSPEC_COS
  UNSPEC_FPATAN
  UNSPEC_FYL2X
  UNSPEC_FYL2XP1
  UNSPEC_FRNDINT
  UNSPEC_FIST
  UNSPEC_F2XM1
  UNSPEC_TAN
  UNSPEC_FXAM

  ;; x87 Rounding
  UNSPEC_FRNDINT_FLOOR
  UNSPEC_FRNDINT_CEIL
  UNSPEC_FRNDINT_TRUNC
  UNSPEC_FRNDINT_MASK_PM
  UNSPEC_FIST_FLOOR
  UNSPEC_FIST_CEIL

  ;; x87 Double output FP
  UNSPEC_SINCOS_COS
  UNSPEC_SINCOS_SIN
  UNSPEC_XTRACT_FRACT
  UNSPEC_XTRACT_EXP
  UNSPEC_FSCALE_FRACT
  UNSPEC_FSCALE_EXP
  UNSPEC_FPREM_F
  UNSPEC_FPREM_U
  UNSPEC_FPREM1_F
  UNSPEC_FPREM1_U

  UNSPEC_C2_FLAG
  UNSPEC_FXAM_MEM

  ;; SSP patterns
  UNSPEC_SP_SET
  UNSPEC_SP_TEST

  ;; For ROUND support
  UNSPEC_ROUND

  ;; For CRC32 support
  UNSPEC_CRC32

  ;; For LZCNT suppoprt
  UNSPEC_LZCNT

  ;; For BMI support
  UNSPEC_TZCNT
  UNSPEC_BEXTR

  ;; For BMI2 support
  UNSPEC_PDEP
  UNSPEC_PEXT

  UNSPEC_BNDMK
  UNSPEC_BNDMK_ADDR
  UNSPEC_BNDSTX
  UNSPEC_BNDLDX
  UNSPEC_BNDLDX_ADDR
  UNSPEC_BNDCL
  UNSPEC_BNDCU
  UNSPEC_BNDCN
  UNSPEC_MPX_FENCE

  ;; IRET support
  UNSPEC_INTERRUPT_RETURN
])

(define_c_enum "unspecv" [
  UNSPECV_UD2
  UNSPECV_BLOCKAGE
  UNSPECV_STACK_PROBE
  UNSPECV_PROBE_STACK_RANGE
  UNSPECV_ALIGN
  UNSPECV_PROLOGUE_USE
  UNSPECV_SPLIT_STACK_RETURN
  UNSPECV_CLD
  UNSPECV_NOPS
  UNSPECV_RDTSC
  UNSPECV_RDTSCP
  UNSPECV_RDPMC
  UNSPECV_LLWP_INTRINSIC
  UNSPECV_SLWP_INTRINSIC
  UNSPECV_LWPVAL_INTRINSIC
  UNSPECV_LWPINS_INTRINSIC
  UNSPECV_RDFSBASE
  UNSPECV_RDGSBASE
  UNSPECV_WRFSBASE
  UNSPECV_WRGSBASE
  UNSPECV_FXSAVE
  UNSPECV_FXRSTOR
  UNSPECV_FXSAVE64
  UNSPECV_FXRSTOR64
  UNSPECV_XSAVE
  UNSPECV_XRSTOR
  UNSPECV_XSAVE64
  UNSPECV_XRSTOR64
  UNSPECV_XSAVEOPT
  UNSPECV_XSAVEOPT64
  UNSPECV_XSAVES
  UNSPECV_XRSTORS
  UNSPECV_XSAVES64
  UNSPECV_XRSTORS64
  UNSPECV_XSAVEC
  UNSPECV_XSAVEC64
  UNSPECV_XGETBV
  UNSPECV_XSETBV

  ;; For atomic compound assignments.
  UNSPECV_FNSTENV
  UNSPECV_FLDENV
  UNSPECV_FNSTSW
  UNSPECV_FNCLEX

  ;; For RDRAND support
  UNSPECV_RDRAND

  ;; For RDSEED support
  UNSPECV_RDSEED

  ;; For RTM support
  UNSPECV_XBEGIN
  UNSPECV_XEND
  UNSPECV_XABORT
  UNSPECV_XTEST

  UNSPECV_NLGR

  ;; For CLWB support
  UNSPECV_CLWB

  ;; For CLFLUSHOPT support
  UNSPECV_CLFLUSHOPT

  ;; For MONITORX and MWAITX support 
  UNSPECV_MONITORX
  UNSPECV_MWAITX

  ;; For CLZERO support
  UNSPECV_CLZERO

  ;; For RDPKRU and WRPKRU support
  UNSPECV_PKU

  ;; For RDPID support
  UNSPECV_RDPID

  ;; For CET support
  UNSPECV_NOP_ENDBR
  UNSPECV_NOP_RDSSP
  UNSPECV_INCSSP
  UNSPECV_SAVEPREVSSP
  UNSPECV_RSTORSSP
  UNSPECV_WRSS
  UNSPECV_WRUSS
  UNSPECV_SETSSBSY
  UNSPECV_CLRSSBSY
])

;; Constants to represent rounding modes in the ROUND instruction
(define_constants
  [(ROUND_FLOOR			0x1)
   (ROUND_CEIL			0x2)
   (ROUND_TRUNC			0x3)
   (ROUND_MXCSR			0x4)
   (ROUND_NO_EXC		0x8)
  ])

;; Constants to represent AVX512F embeded rounding
(define_constants
  [(ROUND_NEAREST_INT			0)
   (ROUND_NEG_INF			1)
   (ROUND_POS_INF			2)
   (ROUND_ZERO				3)
   (NO_ROUND				4)
   (ROUND_SAE				8)
  ])

;; Constants to represent pcomtrue/pcomfalse variants
(define_constants
  [(PCOM_FALSE			0)
   (PCOM_TRUE			1)
   (COM_FALSE_S			2)
   (COM_FALSE_P			3)
   (COM_TRUE_S			4)
   (COM_TRUE_P			5)
  ])

;; Constants used in the XOP pperm instruction
(define_constants
  [(PPERM_SRC			0x00)	/* copy source */
   (PPERM_INVERT		0x20)	/* invert source */
   (PPERM_REVERSE		0x40)	/* bit reverse source */
   (PPERM_REV_INV		0x60)	/* bit reverse & invert src */
   (PPERM_ZERO			0x80)	/* all 0's */
   (PPERM_ONES			0xa0)	/* all 1's */
   (PPERM_SIGN			0xc0)	/* propagate sign bit */
   (PPERM_INV_SIGN		0xe0)	/* invert & propagate sign */
   (PPERM_SRC1			0x00)	/* use first source byte */
   (PPERM_SRC2			0x10)	/* use second source byte */
   ])

;; Registers by name.
(define_constants
  [(AX_REG			 0)
   (DX_REG			 1)
   (CX_REG			 2)
   (BX_REG			 3)
   (SI_REG			 4)
   (DI_REG			 5)
   (BP_REG			 6)
   (SP_REG			 7)
   (ST0_REG			 8)
   (ST1_REG			 9)
   (ST2_REG			10)
   (ST3_REG			11)
   (ST4_REG			12)
   (ST5_REG			13)
   (ST6_REG			14)
   (ST7_REG			15)
   (ARGP_REG			16)
   (FLAGS_REG			17)
   (FPSR_REG			18)
   (FPCR_REG			19)
   (FRAME_REG			20)
   (XMM0_REG			21)
   (XMM1_REG			22)
   (XMM2_REG			23)
   (XMM3_REG			24)
   (XMM4_REG			25)
   (XMM5_REG			26)
   (XMM6_REG			27)
   (XMM7_REG			28)
   (MM0_REG			29)
   (MM1_REG			30)
   (MM2_REG			31)
   (MM3_REG			32)
   (MM4_REG			33)
   (MM5_REG			34)
   (MM6_REG			35)
   (MM7_REG			36)
   (R8_REG			37)
   (R9_REG			38)
   (R10_REG			39)
   (R11_REG			40)
   (R12_REG			41)
   (R13_REG			42)
   (R14_REG			43)
   (R15_REG			44)
   (XMM8_REG			45)
   (XMM9_REG			46)
   (XMM10_REG			47)
   (XMM11_REG			48)
   (XMM12_REG			49)
   (XMM13_REG			50)
   (XMM14_REG			51)
   (XMM15_REG			52)
   (XMM16_REG			53)
   (XMM17_REG			54)
   (XMM18_REG			55)
   (XMM19_REG			56)
   (XMM20_REG			57)
   (XMM21_REG			58)
   (XMM22_REG			59)
   (XMM23_REG			60)
   (XMM24_REG			61)
   (XMM25_REG			62)
   (XMM26_REG			63)
   (XMM27_REG			64)
   (XMM28_REG			65)
   (XMM29_REG			66)
   (XMM30_REG			67)
   (XMM31_REG			68)
   (MASK0_REG			69)
   (MASK1_REG			70)
   (MASK2_REG			71)
   (MASK3_REG			72)
   (MASK4_REG			73)
   (MASK5_REG			74)
   (MASK6_REG			75)
   (MASK7_REG			76)
   (BND0_REG			77)
   (BND1_REG			78)
   (BND2_REG			79)
   (BND3_REG			80)
   (FIRST_PSEUDO_REG		81)
  ])

;; Insns whose names begin with "x86_" are emitted by gen_FOO calls
;; from i386.c.

;; In C guard expressions, put expressions which may be compile-time
;; constants first.  This allows for better optimization.  For
;; example, write "TARGET_64BIT && reload_completed", not
;; "reload_completed && TARGET_64BIT".


;; Processor type.
(define_attr "cpu" "none,pentium,pentiumpro,geode,k6,athlon,k8,core2,nehalem,
		    atom,slm,haswell,generic,amdfam10,bdver1,bdver2,bdver3,
		    bdver4,btver2,znver1"
  (const (symbol_ref "ix86_schedule")))

;; A basic instruction type.  Refinements due to arguments to be
;; provided in other attributes.
(define_attr "type"
  "other,multi,
   alu,alu1,negnot,imov,imovx,lea,
   incdec,ishift,ishiftx,ishift1,rotate,rotatex,rotate1,
   imul,imulx,idiv,icmp,test,ibr,setcc,icmov,
   push,pop,call,callv,leave,
   str,bitmanip,
   fmov,fop,fsgn,fmul,fdiv,fpspc,fcmov,fcmp,
   fxch,fistp,fisttp,frndint,
   sse,ssemov,sseadd,sseadd1,sseiadd,sseiadd1,
   ssemul,sseimul,ssediv,sselog,sselog1,
   sseishft,sseishft1,ssecmp,ssecomi,
   ssecvt,ssecvt1,sseicvt,sseins,
   sseshuf,sseshuf1,ssemuladd,sse4arg,
   lwp,mskmov,msklog,
   mmx,mmxmov,mmxadd,mmxmul,mmxcmp,mmxcvt,mmxshft,
   mpxmov,mpxmk,mpxchk,mpxld,mpxst"
  (const_string "other"))

;; Main data type used by the insn
(define_attr "mode"
  "unknown,none,QI,HI,SI,DI,TI,OI,XI,SF,DF,XF,TF,V16SF,V8SF,V4DF,V4SF,
  V2DF,V2SF,V1DF,V8DF"
  (const_string "unknown"))

;; The CPU unit operations uses.
(define_attr "unit" "integer,i387,sse,mmx,unknown"
  (cond [(eq_attr "type" "fmov,fop,fsgn,fmul,fdiv,fpspc,fcmov,fcmp,
			  fxch,fistp,fisttp,frndint")
	   (const_string "i387")
	 (eq_attr "type" "sse,ssemov,sseadd,sseadd1,sseiadd,sseiadd1,
			  ssemul,sseimul,ssediv,sselog,sselog1,
			  sseishft,sseishft1,ssecmp,ssecomi,
			  ssecvt,ssecvt1,sseicvt,sseins,
			  sseshuf,sseshuf1,ssemuladd,sse4arg,mskmov")
	   (const_string "sse")
	 (eq_attr "type" "mmx,mmxmov,mmxadd,mmxmul,mmxcmp,mmxcvt,mmxshft")
	   (const_string "mmx")
	 (eq_attr "type" "other")
	   (const_string "unknown")]
	 (const_string "integer")))

;; The (bounding maximum) length of an instruction immediate.
(define_attr "length_immediate" ""
  (cond [(eq_attr "type" "incdec,setcc,icmov,str,lea,other,multi,idiv,leave,
			  bitmanip,imulx,msklog,mskmov,mpxmk,mpxmov,mpxchk,
			  mpxld,mpxst")
	   (const_int 0)
	 (eq_attr "unit" "i387,sse,mmx")
	   (const_int 0)
	 (eq_attr "type" "alu,alu1,negnot,imovx,ishift,ishiftx,ishift1,
			  rotate,rotatex,rotate1,imul,icmp,push,pop")
	   (symbol_ref "ix86_attr_length_immediate_default (insn, true)")
	 (eq_attr "type" "imov,test")
	   (symbol_ref "ix86_attr_length_immediate_default (insn, false)")
	 (eq_attr "type" "call")
	   (if_then_else (match_operand 0 "constant_call_address_operand")
	     (const_int 4)
	     (const_int 0))
	 (eq_attr "type" "callv")
	   (if_then_else (match_operand 1 "constant_call_address_operand")
	     (const_int 4)
	     (const_int 0))
	 ;; We don't know the size before shorten_branches.  Expect
	 ;; the instruction to fit for better scheduling.
	 (eq_attr "type" "ibr")
	   (const_int 1)
	 ]
	 (symbol_ref "/* Update immediate_length and other attributes! */
		      gcc_unreachable (),1")))

;; The (bounding maximum) length of an instruction address.
(define_attr "length_address" ""
  (cond [(eq_attr "type" "str,other,multi,fxch")
	   (const_int 0)
	 (and (eq_attr "type" "call")
	      (match_operand 0 "constant_call_address_operand"))
	     (const_int 0)
	 (and (eq_attr "type" "callv")
	      (match_operand 1 "constant_call_address_operand"))
	     (const_int 0)
	 ]
	 (symbol_ref "ix86_attr_length_address_default (insn)")))

;; Set when length prefix is used.
(define_attr "prefix_data16" ""
  (cond [(eq_attr "type" "ssemuladd,sse4arg,sseiadd1,ssecvt1")
	   (const_int 0)
	 (eq_attr "mode" "HI")
	   (const_int 1)
	 (and (eq_attr "unit" "sse") (eq_attr "mode" "V2DF,TI"))
	   (const_int 1)
	]
	(const_int 0)))

;; Set when string REP prefix is used.
(define_attr "prefix_rep" ""
  (cond [(eq_attr "type" "ssemuladd,sse4arg,sseiadd1,ssecvt1")
	   (const_int 0)
	 (and (eq_attr "unit" "sse") (eq_attr "mode" "SF,DF"))
	   (const_int 1)
	 (and (eq_attr "type" "ibr,call,callv")
	      (match_test "ix86_bnd_prefixed_insn_p (insn)"))
	   (const_int 1)
	]
	(const_int 0)))

;; Set when 0f opcode prefix is used.
(define_attr "prefix_0f" ""
  (if_then_else
    (ior (eq_attr "type" "imovx,setcc,icmov,bitmanip,msklog,mskmov,
			  mpxmk,mpxmov,mpxchk,mpxld,mpxst")
	 (eq_attr "unit" "sse,mmx"))
    (const_int 1)
    (const_int 0)))

;; Set when REX opcode prefix is used.
(define_attr "prefix_rex" ""
  (cond [(not (match_test "TARGET_64BIT"))
	   (const_int 0)
	 (and (eq_attr "mode" "DI")
	      (and (eq_attr "type" "!push,pop,call,callv,leave,ibr")
		   (eq_attr "unit" "!mmx")))
	   (const_int 1)
	 (and (eq_attr "mode" "QI")
	      (match_test "x86_extended_QIreg_mentioned_p (insn)"))
	   (const_int 1)
	 (match_test "x86_extended_reg_mentioned_p (insn)")
	   (const_int 1)
	 (and (eq_attr "type" "imovx")
	      (match_operand:QI 1 "ext_QIreg_operand"))
	   (const_int 1)
	]
	(const_int 0)))

;; There are also additional prefixes in 3DNOW, SSSE3.
;; ssemuladd,sse4arg default to 0f24/0f25 and DREX byte,
;; sseiadd1,ssecvt1 to 0f7a with no DREX byte.
;; 3DNOW has 0f0f prefix, SSSE3 and SSE4_{1,2} 0f38/0f3a.
(define_attr "prefix_extra" ""
  (cond [(eq_attr "type" "ssemuladd,sse4arg")
	   (const_int 2)
	 (eq_attr "type" "sseiadd1,ssecvt1")
	   (const_int 1)
	]
	(const_int 0)))

;; Set when BND opcode prefix may be used.
(define_attr "maybe_prefix_bnd" "" (const_int 0))

;; Prefix used: original, VEX or maybe VEX.
(define_attr "prefix" "orig,vex,maybe_vex,evex,maybe_evex"
  (cond [(eq_attr "mode" "OI,V8SF,V4DF")
           (const_string "vex")
         (eq_attr "mode" "XI,V16SF,V8DF")
           (const_string "evex")
        ]
        (const_string "orig")))

;; VEX W bit is used.
(define_attr "prefix_vex_w" "" (const_int 0))

;; The length of VEX prefix
;; Only instructions with 0f prefix can have 2 byte VEX prefix,
;; 0f38/0f3a prefixes can't.  In i386.md 0f3[8a] is
;; still prefix_0f 1, with prefix_extra 1.
(define_attr "length_vex" ""
  (if_then_else (and (eq_attr "prefix_0f" "1")
		     (eq_attr "prefix_extra" "0"))
    (if_then_else (eq_attr "prefix_vex_w" "1")
      (symbol_ref "ix86_attr_length_vex_default (insn, true, true)")
      (symbol_ref "ix86_attr_length_vex_default (insn, true, false)"))
    (if_then_else (eq_attr "prefix_vex_w" "1")
      (symbol_ref "ix86_attr_length_vex_default (insn, false, true)")
      (symbol_ref "ix86_attr_length_vex_default (insn, false, false)"))))

;; 4-bytes evex prefix and 1 byte opcode.
(define_attr "length_evex" "" (const_int 5))

;; Set when modrm byte is used.
(define_attr "modrm" ""
  (cond [(eq_attr "type" "str,leave")
	   (const_int 0)
	 (eq_attr "unit" "i387")
	   (const_int 0)
         (and (eq_attr "type" "incdec")
	      (and (not (match_test "TARGET_64BIT"))
		   (ior (match_operand:SI 1 "register_operand")
			(match_operand:HI 1 "register_operand"))))
	   (const_int 0)
	 (and (eq_attr "type" "push")
	      (not (match_operand 1 "memory_operand")))
	   (const_int 0)
	 (and (eq_attr "type" "pop")
	      (not (match_operand 0 "memory_operand")))
	   (const_int 0)
	 (and (eq_attr "type" "imov")
	      (and (not (eq_attr "mode" "DI"))
		   (ior (and (match_operand 0 "register_operand")
			     (match_operand 1 "immediate_operand"))
		        (ior (and (match_operand 0 "ax_reg_operand")
				  (match_operand 1 "memory_displacement_only_operand"))
			     (and (match_operand 0 "memory_displacement_only_operand")
				  (match_operand 1 "ax_reg_operand"))))))
	   (const_int 0)
	 (and (eq_attr "type" "call")
	      (match_operand 0 "constant_call_address_operand"))
	     (const_int 0)
	 (and (eq_attr "type" "callv")
	      (match_operand 1 "constant_call_address_operand"))
	     (const_int 0)
	 (and (eq_attr "type" "alu,alu1,icmp,test")
	      (match_operand 0 "ax_reg_operand"))
	     (symbol_ref "(get_attr_length_immediate (insn) <= (get_attr_mode (insn) != MODE_QI))")
	 ]
	 (const_int 1)))

(define_attr "modrm_class" "none,incdec,op0,op01,op02,pushpop,unknown"
  (cond [(eq_attr "modrm" "0")
	   (const_string "none")
	 (eq_attr "type" "alu,imul,ishift")
	   (const_string "op02")
	 (eq_attr "type" "imov,imovx,lea,alu1,icmp")
	   (const_string "op01")
	 (eq_attr "type" "incdec")
	   (const_string "incdec")
	 (eq_attr "type" "push,pop")
	   (const_string "pushpop")]
	 (const_string "unknown")))

;; The (bounding maximum) length of an instruction in bytes.
;; ??? fistp and frndint are in fact fldcw/{fistp,frndint}/fldcw sequences.
;; Later we may want to split them and compute proper length as for
;; other insns.
(define_attr "length" ""
  (cond [(eq_attr "type" "other,multi,fistp,frndint")
	   (const_int 16)
	 (eq_attr "type" "fcmp")
	   (const_int 4)
	 (eq_attr "unit" "i387")
	   (plus (const_int 2)
		 (plus (attr "prefix_data16")
		       (attr "length_address")))
	 (ior (eq_attr "prefix" "evex")
	      (and (ior (eq_attr "prefix" "maybe_evex")
			(eq_attr "prefix" "maybe_vex"))
		   (match_test "TARGET_AVX512F")))
	   (plus (attr "length_evex")
		 (plus (attr "length_immediate")
		       (plus (attr "modrm")
			     (attr "length_address"))))
	 (ior (eq_attr "prefix" "vex")
	      (and (ior (eq_attr "prefix" "maybe_vex")
			(eq_attr "prefix" "maybe_evex"))
		   (match_test "TARGET_AVX")))
	   (plus (attr "length_vex")
		 (plus (attr "length_immediate")
		       (plus (attr "modrm")
			     (attr "length_address"))))]
	 (plus (plus (attr "modrm")
		     (plus (attr "prefix_0f")
			   (plus (attr "prefix_rex")
				 (plus (attr "prefix_extra")
				       (const_int 1)))))
	       (plus (attr "prefix_rep")
		     (plus (attr "prefix_data16")
			   (plus (attr "length_immediate")
				 (attr "length_address")))))))

;; The `memory' attribute is `none' if no memory is referenced, `load' or
;; `store' if there is a simple memory reference therein, or `unknown'
;; if the instruction is complex.

(define_attr "memory" "none,load,store,both,unknown"
  (cond [(eq_attr "type" "other,multi,str,lwp")
	   (const_string "unknown")
	 (eq_attr "type" "lea,fcmov,fpspc,mpxmk,mpxchk")
	   (const_string "none")
	 (eq_attr "type" "fistp,leave")
	   (const_string "both")
	 (eq_attr "type" "frndint")
	   (const_string "load")
	 (eq_attr "type" "mpxld")
	   (const_string "load")
	 (eq_attr "type" "mpxst")
	   (const_string "store")
	 (eq_attr "type" "push")
	   (if_then_else (match_operand 1 "memory_operand")
	     (const_string "both")
	     (const_string "store"))
	 (eq_attr "type" "pop")
	   (if_then_else (match_operand 0 "memory_operand")
	     (const_string "both")
	     (const_string "load"))
	 (eq_attr "type" "setcc")
	   (if_then_else (match_operand 0 "memory_operand")
	     (const_string "store")
	     (const_string "none"))
	 (eq_attr "type" "icmp,test,ssecmp,ssecomi,mmxcmp,fcmp")
	   (if_then_else (ior (match_operand 0 "memory_operand")
			      (match_operand 1 "memory_operand"))
	     (const_string "load")
	     (const_string "none"))
	 (eq_attr "type" "ibr")
	   (if_then_else (match_operand 0 "memory_operand")
	     (const_string "load")
	     (const_string "none"))
	 (eq_attr "type" "call")
	   (if_then_else (match_operand 0 "constant_call_address_operand")
	     (const_string "none")
	     (const_string "load"))
	 (eq_attr "type" "callv")
	   (if_then_else (match_operand 1 "constant_call_address_operand")
	     (const_string "none")
	     (const_string "load"))
	 (and (eq_attr "type" "alu1,negnot,ishift1,sselog1,sseshuf1")
	      (match_operand 1 "memory_operand"))
	   (const_string "both")
	 (and (match_operand 0 "memory_operand")
	      (match_operand 1 "memory_operand"))
	   (const_string "both")
	 (match_operand 0 "memory_operand")
	   (const_string "store")
	 (match_operand 1 "memory_operand")
	   (const_string "load")
	 (and (eq_attr "type"
		 "!alu1,negnot,ishift1,
		   imov,imovx,icmp,test,bitmanip,
		   fmov,fcmp,fsgn,
		   sse,ssemov,ssecmp,ssecomi,ssecvt,ssecvt1,sseicvt,
		   sselog1,sseshuf1,sseadd1,sseiadd1,sseishft1,
		   mmx,mmxmov,mmxcmp,mmxcvt,mskmov,msklog,mpxmov")
	      (match_operand 2 "memory_operand"))
	   (const_string "load")
	 (and (eq_attr "type" "icmov,ssemuladd,sse4arg")
	      (match_operand 3 "memory_operand"))
	   (const_string "load")
	]
	(const_string "none")))

;; Indicates if an instruction has both an immediate and a displacement.

(define_attr "imm_disp" "false,true,unknown"
  (cond [(eq_attr "type" "other,multi")
	   (const_string "unknown")
	 (and (eq_attr "type" "icmp,test,imov,alu1,ishift1,rotate1")
	      (and (match_operand 0 "memory_displacement_operand")
		   (match_operand 1 "immediate_operand")))
	   (const_string "true")
	 (and (eq_attr "type" "alu,ishift,ishiftx,rotate,rotatex,imul,idiv")
	      (and (match_operand 0 "memory_displacement_operand")
		   (match_operand 2 "immediate_operand")))
	   (const_string "true")
	]
	(const_string "false")))

;; Indicates if an FP operation has an integer source.

(define_attr "fp_int_src" "false,true"
  (const_string "false"))

;; Defines rounding mode of an FP operation.

(define_attr "i387_cw" "trunc,floor,ceil,mask_pm,uninitialized,any"
  (const_string "any"))

;; Define attribute to classify add/sub insns that consumes carry flag (CF)
(define_attr "use_carry" "0,1" (const_string "0"))

;; Define attribute to indicate unaligned ssemov insns
(define_attr "movu" "0,1" (const_string "0"))

;; Used to control the "enabled" attribute on a per-instruction basis.
(define_attr "isa" "base,x64,x64_sse4,x64_sse4_noavx,x64_avx,nox64,
		    sse2,sse2_noavx,sse3,sse4,sse4_noavx,avx,noavx,
		    avx2,noavx2,bmi,bmi2,fma4,fma,avx512f,noavx512f,
		    avx512bw,noavx512bw,avx512dq,noavx512dq,
		    avx512vl,noavx512vl,x64_avx512dq,x64_avx512bw"
  (const_string "base"))

(define_attr "enabled" ""
  (cond [(eq_attr "isa" "x64") (symbol_ref "TARGET_64BIT")
	 (eq_attr "isa" "x64_sse4")
	   (symbol_ref "TARGET_64BIT && TARGET_SSE4_1")
	 (eq_attr "isa" "x64_sse4_noavx")
	   (symbol_ref "TARGET_64BIT && TARGET_SSE4_1 && !TARGET_AVX")
	 (eq_attr "isa" "x64_avx")
	   (symbol_ref "TARGET_64BIT && TARGET_AVX")
	 (eq_attr "isa" "x64_avx512dq")
	   (symbol_ref "TARGET_64BIT && TARGET_AVX512DQ")
	 (eq_attr "isa" "x64_avx512bw")
	   (symbol_ref "TARGET_64BIT && TARGET_AVX512BW")
	 (eq_attr "isa" "nox64") (symbol_ref "!TARGET_64BIT")
	 (eq_attr "isa" "sse2") (symbol_ref "TARGET_SSE2")
	 (eq_attr "isa" "sse2_noavx")
	   (symbol_ref "TARGET_SSE2 && !TARGET_AVX")
	 (eq_attr "isa" "sse3") (symbol_ref "TARGET_SSE3")
	 (eq_attr "isa" "sse4") (symbol_ref "TARGET_SSE4_1")
	 (eq_attr "isa" "sse4_noavx")
	   (symbol_ref "TARGET_SSE4_1 && !TARGET_AVX")
	 (eq_attr "isa" "avx") (symbol_ref "TARGET_AVX")
	 (eq_attr "isa" "noavx") (symbol_ref "!TARGET_AVX")
	 (eq_attr "isa" "avx2") (symbol_ref "TARGET_AVX2")
	 (eq_attr "isa" "noavx2") (symbol_ref "!TARGET_AVX2")
	 (eq_attr "isa" "bmi") (symbol_ref "TARGET_BMI")
	 (eq_attr "isa" "bmi2") (symbol_ref "TARGET_BMI2")
	 (eq_attr "isa" "fma4") (symbol_ref "TARGET_FMA4")
	 (eq_attr "isa" "fma") (symbol_ref "TARGET_FMA")
	 (eq_attr "isa" "avx512f") (symbol_ref "TARGET_AVX512F")
	 (eq_attr "isa" "noavx512f") (symbol_ref "!TARGET_AVX512F")
	 (eq_attr "isa" "avx512bw") (symbol_ref "TARGET_AVX512BW")
	 (eq_attr "isa" "noavx512bw") (symbol_ref "!TARGET_AVX512BW")
	 (eq_attr "isa" "avx512dq") (symbol_ref "TARGET_AVX512DQ")
	 (eq_attr "isa" "noavx512dq") (symbol_ref "!TARGET_AVX512DQ")
	 (eq_attr "isa" "avx512vl") (symbol_ref "TARGET_AVX512VL")
	 (eq_attr "isa" "noavx512vl") (symbol_ref "!TARGET_AVX512VL")
	]
	(const_int 1)))

(define_attr "preferred_for_size" "" (const_int 1))
(define_attr "preferred_for_speed" "" (const_int 1))

;; Describe a user's asm statement.
(define_asm_attributes
  [(set_attr "length" "128")
   (set_attr "type" "multi")])

(define_code_iterator plusminus [plus minus])

(define_code_iterator sat_plusminus [ss_plus us_plus ss_minus us_minus])

(define_code_iterator multdiv [mult div])

;; Base name for define_insn
(define_code_attr plusminus_insn
  [(plus "add") (ss_plus "ssadd") (us_plus "usadd")
   (minus "sub") (ss_minus "sssub") (us_minus "ussub")])

;; Base name for insn mnemonic.
(define_code_attr plusminus_mnemonic
  [(plus "add") (ss_plus "adds") (us_plus "addus")
   (minus "sub") (ss_minus "subs") (us_minus "subus")])
(define_code_attr multdiv_mnemonic
  [(mult "mul") (div "div")])

;; Mark commutative operators as such in constraints.
(define_code_attr comm [(plus "%") (ss_plus "%") (us_plus "%")
			(minus "") (ss_minus "") (us_minus "")])

;; Mapping of max and min
(define_code_iterator maxmin [smax smin umax umin])

;; Mapping of signed max and min
(define_code_iterator smaxmin [smax smin])

;; Mapping of unsigned max and min
(define_code_iterator umaxmin [umax umin])

;; Base name for integer and FP insn mnemonic
(define_code_attr maxmin_int [(smax "maxs") (smin "mins")
			      (umax "maxu") (umin "minu")])
(define_code_attr maxmin_float [(smax "max") (smin "min")])

(define_int_iterator IEEE_MAXMIN
	[UNSPEC_IEEE_MAX
	 UNSPEC_IEEE_MIN])

(define_int_attr ieee_maxmin
	[(UNSPEC_IEEE_MAX "max")
	 (UNSPEC_IEEE_MIN "min")])

;; Mapping of logic operators
(define_code_iterator any_logic [and ior xor])
(define_code_iterator any_or [ior xor])
(define_code_iterator fpint_logic [and xor])

;; Base name for insn mnemonic.
(define_code_attr logic [(and "and") (ior "or") (xor "xor")])

;; Mapping of logic-shift operators
(define_code_iterator any_lshift [ashift lshiftrt])

;; Mapping of shift-right operators
(define_code_iterator any_shiftrt [lshiftrt ashiftrt])

;; Mapping of all shift operators
(define_code_iterator any_shift [ashift lshiftrt ashiftrt])

;; Base name for define_insn
(define_code_attr shift_insn
  [(ashift "ashl") (lshiftrt "lshr") (ashiftrt "ashr")])

;; Base name for insn mnemonic.
(define_code_attr shift [(ashift "sll") (lshiftrt "shr") (ashiftrt "sar")])
(define_code_attr vshift [(ashift "sll") (lshiftrt "srl") (ashiftrt "sra")])

;; Mapping of rotate operators
(define_code_iterator any_rotate [rotate rotatert])

;; Base name for define_insn
(define_code_attr rotate_insn [(rotate "rotl") (rotatert "rotr")])

;; Base name for insn mnemonic.
(define_code_attr rotate [(rotate "rol") (rotatert "ror")])

;; Mapping of abs neg operators
(define_code_iterator absneg [abs neg])

;; Base name for x87 insn mnemonic.
(define_code_attr absneg_mnemonic [(abs "abs") (neg "chs")])

;; Used in signed and unsigned widening multiplications.
(define_code_iterator any_extend [sign_extend zero_extend])

;; Prefix for insn menmonic.
(define_code_attr sgnprefix [(sign_extend "i") (zero_extend "")])

;; Prefix for define_insn
(define_code_attr u [(sign_extend "") (zero_extend "u")])
(define_code_attr s [(sign_extend "s") (zero_extend "u")])
(define_code_attr u_bool [(sign_extend "false") (zero_extend "true")])

;; Used in signed and unsigned truncations.
(define_code_iterator any_truncate [ss_truncate truncate us_truncate])
;; Instruction suffix for truncations.
(define_code_attr trunsuffix [(ss_truncate "s") (truncate "") (us_truncate "us")])

;; Used in signed and unsigned fix.
(define_code_iterator any_fix [fix unsigned_fix])
(define_code_attr fixsuffix [(fix "") (unsigned_fix "u")])

;; Used in signed and unsigned float.
(define_code_iterator any_float [float unsigned_float])
(define_code_attr floatsuffix [(float "") (unsigned_float "u")])

;; All integer modes.
(define_mode_iterator SWI1248x [QI HI SI DI])

;; All integer modes without QImode.
(define_mode_iterator SWI248x [HI SI DI])

;; All integer modes without QImode and HImode.
(define_mode_iterator SWI48x [SI DI])

;; All integer modes without SImode and DImode.
(define_mode_iterator SWI12 [QI HI])

;; All integer modes without DImode.
(define_mode_iterator SWI124 [QI HI SI])

;; All integer modes without QImode and DImode.
(define_mode_iterator SWI24 [HI SI])

;; Single word integer modes.
(define_mode_iterator SWI [QI HI SI (DI "TARGET_64BIT")])

;; Single word integer modes without QImode.
(define_mode_iterator SWI248 [HI SI (DI "TARGET_64BIT")])

;; Single word integer modes without QImode and HImode.
(define_mode_iterator SWI48 [SI (DI "TARGET_64BIT")])

;; All math-dependant single and double word integer modes.
(define_mode_iterator SDWIM [(QI "TARGET_QIMODE_MATH")
			     (HI "TARGET_HIMODE_MATH")
			     SI DI (TI "TARGET_64BIT")])

;; Math-dependant single word integer modes.
(define_mode_iterator SWIM [(QI "TARGET_QIMODE_MATH")
			    (HI "TARGET_HIMODE_MATH")
			    SI (DI "TARGET_64BIT")])

;; Math-dependant integer modes without DImode.
(define_mode_iterator SWIM124 [(QI "TARGET_QIMODE_MATH")
			       (HI "TARGET_HIMODE_MATH")
			       SI])

;; Math-dependant integer modes with DImode.
(define_mode_iterator SWIM1248x [(QI "TARGET_QIMODE_MATH")
				 (HI "TARGET_HIMODE_MATH")
				 SI (DI "(TARGET_STV && TARGET_SSE2) || TARGET_64BIT")])

;; Math-dependant single word integer modes without QImode.
(define_mode_iterator SWIM248 [(HI "TARGET_HIMODE_MATH")
		      	       SI (DI "TARGET_64BIT")])

;; Double word integer modes.
(define_mode_iterator DWI [(DI "!TARGET_64BIT")
			   (TI "TARGET_64BIT")])

;; GET_MODE_SIZE for selected modes.  As GET_MODE_SIZE is not
;; compile time constant, it is faster to use <MODE_SIZE> than
;; GET_MODE_SIZE (<MODE>mode).  For XFmode which depends on
;; command line options just use GET_MODE_SIZE macro.
(define_mode_attr MODE_SIZE [(QI "1") (HI "2") (SI "4") (DI "8") (TI "16")
			     (SF "4") (DF "8") (XF "GET_MODE_SIZE (XFmode)")
			     (V16QI "16") (V32QI "32") (V64QI "64")
			     (V8HI "16") (V16HI "32") (V32HI "64")
			     (V4SI "16") (V8SI "32") (V16SI "64")
			     (V2DI "16") (V4DI "32") (V8DI "64")
			     (V1TI "16") (V2TI "32") (V4TI "64")
			     (V2DF "16") (V4DF "32") (V8DF "64")
			     (V4SF "16") (V8SF "32") (V16SF "64")])

;; Double word integer modes as mode attribute.
(define_mode_attr DWI [(QI "HI") (HI "SI") (SI "DI") (DI "TI")])
(define_mode_attr dwi [(QI "hi") (HI "si") (SI "di") (DI "ti")])

;; LEA mode corresponding to an integer mode
(define_mode_attr LEAMODE [(QI "SI") (HI "SI") (SI "SI") (DI "DI")])

;; Half mode for double word integer modes.
(define_mode_iterator DWIH [(SI "!TARGET_64BIT")
			    (DI "TARGET_64BIT")])

;; Bound modes.
(define_mode_iterator BND [(BND32 "!TARGET_LP64")
			   (BND64 "TARGET_LP64")])

;; Pointer mode corresponding to bound mode.
(define_mode_attr bnd_ptr [(BND32 "SI") (BND64 "DI")])

;; MPX check types
(define_int_iterator BNDCHECK [UNSPEC_BNDCL UNSPEC_BNDCU UNSPEC_BNDCN])

;; Check name
(define_int_attr bndcheck [(UNSPEC_BNDCL "cl")
			   (UNSPEC_BNDCU "cu")
			   (UNSPEC_BNDCN "cn")])

;; Instruction suffix for integer modes.
(define_mode_attr imodesuffix [(QI "b") (HI "w") (SI "l") (DI "q")])

;; Instruction suffix for masks.
(define_mode_attr mskmodesuffix [(QI "b") (HI "w") (SI "d") (DI "q")])

;; Pointer size prefix for integer modes (Intel asm dialect)
(define_mode_attr iptrsize [(QI "BYTE")
			    (HI "WORD")
			    (SI "DWORD")
			    (DI "QWORD")])

;; Register class for integer modes.
(define_mode_attr r [(QI "q") (HI "r") (SI "r") (DI "r")])

;; Immediate operand constraint for integer modes.
(define_mode_attr i [(QI "n") (HI "n") (SI "e") (DI "e")])

;; General operand constraint for word modes.
(define_mode_attr g [(QI "qmn") (HI "rmn") (SI "rme") (DI "rme")])

;; Immediate operand constraint for double integer modes.
(define_mode_attr di [(SI "nF") (DI "Wd")])

;; Immediate operand constraint for shifts.
(define_mode_attr S [(QI "I") (HI "I") (SI "I") (DI "J") (TI "O")])

;; Print register name in the specified mode.
(define_mode_attr k [(QI "b") (HI "w") (SI "k") (DI "q")])

;; General operand predicate for integer modes.
(define_mode_attr general_operand
	[(QI "general_operand")
	 (HI "general_operand")
	 (SI "x86_64_general_operand")
	 (DI "x86_64_general_operand")
	 (TI "x86_64_general_operand")])

;; General operand predicate for integer modes, where for TImode
;; we need both words of the operand to be general operands.
(define_mode_attr general_hilo_operand
	[(QI "general_operand")
	 (HI "general_operand")
	 (SI "x86_64_general_operand")
	 (DI "x86_64_general_operand")
	 (TI "x86_64_hilo_general_operand")])

;; General sign extend operand predicate for integer modes,
;; which disallows VOIDmode operands and thus it is suitable
;; for use inside sign_extend.
(define_mode_attr general_sext_operand
	[(QI "sext_operand")
	 (HI "sext_operand")
	 (SI "x86_64_sext_operand")
	 (DI "x86_64_sext_operand")])

;; General sign/zero extend operand predicate for integer modes.
(define_mode_attr general_szext_operand
	[(QI "general_operand")
	 (HI "general_operand")
	 (SI "x86_64_szext_general_operand")
	 (DI "x86_64_szext_general_operand")])

;; Immediate operand predicate for integer modes.
(define_mode_attr immediate_operand
	[(QI "immediate_operand")
	 (HI "immediate_operand")
	 (SI "x86_64_immediate_operand")
	 (DI "x86_64_immediate_operand")])

;; Nonmemory operand predicate for integer modes.
(define_mode_attr nonmemory_operand
	[(QI "nonmemory_operand")
	 (HI "nonmemory_operand")
	 (SI "x86_64_nonmemory_operand")
	 (DI "x86_64_nonmemory_operand")])

;; Operand predicate for shifts.
(define_mode_attr shift_operand
	[(QI "nonimmediate_operand")
	 (HI "nonimmediate_operand")
	 (SI "nonimmediate_operand")
	 (DI "shiftdi_operand")
	 (TI "register_operand")])

;; Operand predicate for shift argument.
(define_mode_attr shift_immediate_operand
	[(QI "const_1_to_31_operand")
	 (HI "const_1_to_31_operand")
	 (SI "const_1_to_31_operand")
	 (DI "const_1_to_63_operand")])

;; Input operand predicate for arithmetic left shifts.
(define_mode_attr ashl_input_operand
	[(QI "nonimmediate_operand")
	 (HI "nonimmediate_operand")
	 (SI "nonimmediate_operand")
	 (DI "ashldi_input_operand")
	 (TI "reg_or_pm1_operand")])

;; SSE and x87 SFmode and DFmode floating point modes
(define_mode_iterator MODEF [SF DF])

;; All x87 floating point modes
(define_mode_iterator X87MODEF [SF DF XF])

;; SSE instruction suffix for various modes
(define_mode_attr ssemodesuffix
  [(SF "ss") (DF "sd")
   (V16SF "ps") (V8DF "pd")
   (V8SF "ps") (V4DF "pd")
   (V4SF "ps") (V2DF "pd")
   (V16QI "b") (V8HI "w") (V4SI "d") (V2DI "q")
   (V32QI "b") (V16HI "w") (V8SI "d") (V4DI "q")
   (V64QI "b") (V32HI "w") (V16SI "d") (V8DI "q")])

;; SSE vector suffix for floating point modes
(define_mode_attr ssevecmodesuffix [(SF "ps") (DF "pd")])

;; SSE vector mode corresponding to a scalar mode
(define_mode_attr ssevecmode
  [(QI "V16QI") (HI "V8HI") (SI "V4SI") (DI "V2DI") (SF "V4SF") (DF "V2DF")])
(define_mode_attr ssevecmodelower
  [(QI "v16qi") (HI "v8hi") (SI "v4si") (DI "v2di") (SF "v4sf") (DF "v2df")])

;; AVX512F vector mode corresponding to a scalar mode
(define_mode_attr avx512fvecmode
  [(QI "V64QI") (HI "V32HI") (SI "V16SI") (DI "V8DI") (SF "V16SF") (DF "V8DF")])

;; Instruction suffix for REX 64bit operators.
(define_mode_attr rex64suffix [(SI "") (DI "{q}")])

;; This mode iterator allows :P to be used for patterns that operate on
;; pointer-sized quantities.  Exactly one of the two alternatives will match.
(define_mode_iterator P [(SI "Pmode == SImode") (DI "Pmode == DImode")])

;; This mode iterator allows :W to be used for patterns that operate on
;; word_mode sized quantities.
(define_mode_iterator W
  [(SI "word_mode == SImode") (DI "word_mode == DImode")])

;; This mode iterator allows :PTR to be used for patterns that operate on
;; ptr_mode sized quantities.
(define_mode_iterator PTR
  [(SI "ptr_mode == SImode") (DI "ptr_mode == DImode")])

;; Scheduling descriptions

(include "pentium.md")
(include "ppro.md")
(include "k6.md")
(include "athlon.md")
(include "bdver1.md")
(include "bdver3.md")
(include "btver2.md")
(include "znver1.md")
(include "geode.md")
(include "atom.md")
(include "slm.md")
(include "core2.md")
(include "haswell.md")


;; Operand and operator predicates and constraints

(include "predicates.md")
(include "constraints.md")


;; Compare and branch/compare and store instructions.

(define_expand "cbranch<mode>4"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_operand:SDWIM 1 "nonimmediate_operand")
		    (match_operand:SDWIM 2 "<general_operand>")))
   (set (pc) (if_then_else
	       (match_operator 0 "ordered_comparison_operator"
		[(reg:CC FLAGS_REG) (const_int 0)])
	       (label_ref (match_operand 3))
	       (pc)))]
  ""
{
  if (MEM_P (operands[1]) && MEM_P (operands[2]))
    operands[1] = force_reg (<MODE>mode, operands[1]);
  ix86_expand_branch (GET_CODE (operands[0]),
		      operands[1], operands[2], operands[3]);
  DONE;
})

(define_expand "cstore<mode>4"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_operand:SWIM 2 "nonimmediate_operand")
		    (match_operand:SWIM 3 "<general_operand>")))
   (set (match_operand:QI 0 "register_operand")
	(match_operator 1 "ordered_comparison_operator"
	  [(reg:CC FLAGS_REG) (const_int 0)]))]
  ""
{
  if (MEM_P (operands[2]) && MEM_P (operands[3]))
    operands[2] = force_reg (<MODE>mode, operands[2]);
  ix86_expand_setcc (operands[0], GET_CODE (operands[1]),
		     operands[2], operands[3]);
  DONE;
})

(define_expand "cmp<mode>_1"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_operand:SWI48 0 "nonimmediate_operand")
		    (match_operand:SWI48 1 "<general_operand>")))])

(define_insn "*cmp<mode>_ccno_1"
  [(set (reg FLAGS_REG)
	(compare (match_operand:SWI 0 "nonimmediate_operand" "<r>,?m<r>")
		 (match_operand:SWI 1 "const0_operand")))]
  "ix86_match_ccmode (insn, CCNOmode)"
  "@
   test{<imodesuffix>}\t%0, %0
   cmp{<imodesuffix>}\t{%1, %0|%0, %1}"
  [(set_attr "type" "test,icmp")
   (set_attr "length_immediate" "0,1")
   (set_attr "modrm_class" "op0,unknown")
   (set_attr "mode" "<MODE>")])

(define_insn "*cmp<mode>_1"
  [(set (reg FLAGS_REG)
	(compare (match_operand:SWI 0 "nonimmediate_operand" "<r>m,<r>")
		 (match_operand:SWI 1 "<general_operand>" "<r><i>,<r>m")))]
  "ix86_match_ccmode (insn, CCmode)"
  "cmp{<imodesuffix>}\t{%1, %0|%0, %1}"
  [(set_attr "type" "icmp")
   (set_attr "mode" "<MODE>")])

(define_insn "*cmp<mode>_minus_1"
  [(set (reg FLAGS_REG)
	(compare
	  (minus:SWI (match_operand:SWI 0 "nonimmediate_operand" "<r>m,<r>")
		     (match_operand:SWI 1 "<general_operand>" "<r><i>,<r>m"))
	  (const_int 0)))]
  "ix86_match_ccmode (insn, CCGOCmode)"
  "cmp{<imodesuffix>}\t{%1, %0|%0, %1}"
  [(set_attr "type" "icmp")
   (set_attr "mode" "<MODE>")])

(define_insn "*cmpqi_ext_1"
  [(set (reg FLAGS_REG)
	(compare
	  (match_operand:QI 0 "nonimmediate_operand" "QBc,m")
	  (subreg:QI
	    (zero_extract:SI
	      (match_operand 1 "ext_register_operand" "Q,Q")
	      (const_int 8)
	      (const_int 8)) 0)))]
  "ix86_match_ccmode (insn, CCmode)"
  "cmp{b}\t{%h1, %0|%0, %h1}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "icmp")
   (set_attr "mode" "QI")])

(define_insn "*cmpqi_ext_2"
  [(set (reg FLAGS_REG)
	(compare
	  (subreg:QI
	    (zero_extract:SI
	      (match_operand 0 "ext_register_operand" "Q")
	      (const_int 8)
	      (const_int 8)) 0)
	  (match_operand:QI 1 "const0_operand")))]
  "ix86_match_ccmode (insn, CCNOmode)"
  "test{b}\t%h0, %h0"
  [(set_attr "type" "test")
   (set_attr "length_immediate" "0")
   (set_attr "mode" "QI")])

(define_expand "cmpqi_ext_3"
  [(set (reg:CC FLAGS_REG)
	(compare:CC
	  (subreg:QI
	    (zero_extract:SI
	      (match_operand 0 "ext_register_operand")
	      (const_int 8)
	      (const_int 8)) 0)
	  (match_operand:QI 1 "const_int_operand")))])

(define_insn "*cmpqi_ext_3"
  [(set (reg FLAGS_REG)
	(compare
	  (subreg:QI
	    (zero_extract:SI
	      (match_operand 0 "ext_register_operand" "Q,Q")
	      (const_int 8)
	      (const_int 8)) 0)
	  (match_operand:QI 1 "general_operand" "QnBc,m")))]
  "ix86_match_ccmode (insn, CCmode)"
  "cmp{b}\t{%1, %h0|%h0, %1}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "icmp")
   (set_attr "mode" "QI")])

(define_insn "*cmpqi_ext_4"
  [(set (reg FLAGS_REG)
	(compare
	  (subreg:QI
	    (zero_extract:SI
	      (match_operand 0 "ext_register_operand" "Q")
	      (const_int 8)
	      (const_int 8)) 0)
	  (subreg:QI
	    (zero_extract:SI
	      (match_operand 1 "ext_register_operand" "Q")
	      (const_int 8)
	      (const_int 8)) 0)))]
  "ix86_match_ccmode (insn, CCmode)"
  "cmp{b}\t{%h1, %h0|%h0, %h1}"
  [(set_attr "type" "icmp")
   (set_attr "mode" "QI")])

;; These implement float point compares.
;; %%% See if we can get away with VOIDmode operands on the actual insns,
;; which would allow mix and match FP modes on the compares.  Which is what
;; the old patterns did, but with many more of them.

(define_expand "cbranchxf4"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_operand:XF 1 "nonmemory_operand")
		    (match_operand:XF 2 "nonmemory_operand")))
   (set (pc) (if_then_else
              (match_operator 0 "ix86_fp_comparison_operator"
               [(reg:CC FLAGS_REG)
                (const_int 0)])
              (label_ref (match_operand 3))
              (pc)))]
  "TARGET_80387"
{
  ix86_expand_branch (GET_CODE (operands[0]),
		      operands[1], operands[2], operands[3]);
  DONE;
})

(define_expand "cstorexf4"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_operand:XF 2 "nonmemory_operand")
		    (match_operand:XF 3 "nonmemory_operand")))
   (set (match_operand:QI 0 "register_operand")
              (match_operator 1 "ix86_fp_comparison_operator"
               [(reg:CC FLAGS_REG)
                (const_int 0)]))]
  "TARGET_80387"
{
  ix86_expand_setcc (operands[0], GET_CODE (operands[1]),
		     operands[2], operands[3]);
  DONE;
})

(define_expand "cbranch<mode>4"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_operand:MODEF 1 "cmp_fp_expander_operand")
		    (match_operand:MODEF 2 "cmp_fp_expander_operand")))
   (set (pc) (if_then_else
              (match_operator 0 "ix86_fp_comparison_operator"
               [(reg:CC FLAGS_REG)
                (const_int 0)])
              (label_ref (match_operand 3))
              (pc)))]
  "TARGET_80387 || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)"
{
  ix86_expand_branch (GET_CODE (operands[0]),
		      operands[1], operands[2], operands[3]);
  DONE;
})

(define_expand "cstore<mode>4"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_operand:MODEF 2 "cmp_fp_expander_operand")
		    (match_operand:MODEF 3 "cmp_fp_expander_operand")))
   (set (match_operand:QI 0 "register_operand")
              (match_operator 1 "ix86_fp_comparison_operator"
               [(reg:CC FLAGS_REG)
                (const_int 0)]))]
  "TARGET_80387 || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)"
{
  ix86_expand_setcc (operands[0], GET_CODE (operands[1]),
		     operands[2], operands[3]);
  DONE;
})

(define_expand "cbranchcc4"
  [(set (pc) (if_then_else
              (match_operator 0 "comparison_operator"
               [(match_operand 1 "flags_reg_operand")
                (match_operand 2 "const0_operand")])
              (label_ref (match_operand 3))
              (pc)))]
  ""
{
  ix86_expand_branch (GET_CODE (operands[0]),
		      operands[1], operands[2], operands[3]);
  DONE;
})

(define_expand "cstorecc4"
  [(set (match_operand:QI 0 "register_operand")
              (match_operator 1 "comparison_operator"
               [(match_operand 2 "flags_reg_operand")
                (match_operand 3 "const0_operand")]))]
  ""
{
  ix86_expand_setcc (operands[0], GET_CODE (operands[1]),
		     operands[2], operands[3]);
  DONE;
})


;; FP compares, step 1:
;; Set the FP condition codes.
;;
;; CCFPmode	compare with exceptions
;; CCFPUmode	compare with no exceptions

;; We may not use "#" to split and emit these, since the REG_DEAD notes
;; used to manage the reg stack popping would not be preserved.

(define_insn "*cmp<mode>_0_i387"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(unspec:HI
	  [(compare:CCFP
	     (match_operand:X87MODEF 1 "register_operand" "f")
	     (match_operand:X87MODEF 2 "const0_operand"))]
	UNSPEC_FNSTSW))]
  "TARGET_80387"
  "* return output_fp_compare (insn, operands, false, false);"
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "*cmp<mode>_0_cc_i387"
  [(set (reg:CCFP FLAGS_REG)
	(compare:CCFP
	  (match_operand:X87MODEF 1 "register_operand" "f")
	  (match_operand:X87MODEF 2 "const0_operand")))
   (clobber (match_operand:HI 0 "register_operand" "=a"))]
  "TARGET_80387 && TARGET_SAHF && !TARGET_CMOVE"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(unspec:HI
	  [(compare:CCFP (match_dup 1)(match_dup 2))]
	UNSPEC_FNSTSW))
   (set (reg:CC FLAGS_REG)
	(unspec:CC [(match_dup 0)] UNSPEC_SAHF))]
  ""
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "<MODE>")])

(define_insn "*cmpxf_i387"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(unspec:HI
	  [(compare:CCFP
	     (match_operand:XF 1 "register_operand" "f")
	     (match_operand:XF 2 "register_operand" "f"))]
	  UNSPEC_FNSTSW))]
  "TARGET_80387"
  "* return output_fp_compare (insn, operands, false, false);"
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "XF")])

(define_insn_and_split "*cmpxf_cc_i387"
  [(set (reg:CCFP FLAGS_REG)
	(compare:CCFP
	  (match_operand:XF 1 "register_operand" "f")
	  (match_operand:XF 2 "register_operand" "f")))
   (clobber (match_operand:HI 0 "register_operand" "=a"))]
  "TARGET_80387 && TARGET_SAHF && !TARGET_CMOVE"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(unspec:HI
	  [(compare:CCFP (match_dup 1)(match_dup 2))]
	UNSPEC_FNSTSW))
   (set (reg:CC FLAGS_REG)
	(unspec:CC [(match_dup 0)] UNSPEC_SAHF))]
  ""
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "XF")])

(define_insn "*cmp<mode>_i387"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(unspec:HI
	  [(compare:CCFP
	     (match_operand:MODEF 1 "register_operand" "f")
	     (match_operand:MODEF 2 "nonimmediate_operand" "fm"))]
	  UNSPEC_FNSTSW))]
  "TARGET_80387"
  "* return output_fp_compare (insn, operands, false, false);"
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "*cmp<mode>_cc_i387"
  [(set (reg:CCFP FLAGS_REG)
	(compare:CCFP
	  (match_operand:MODEF 1 "register_operand" "f")
	  (match_operand:MODEF 2 "nonimmediate_operand" "fm")))
   (clobber (match_operand:HI 0 "register_operand" "=a"))]
  "TARGET_80387 && TARGET_SAHF && !TARGET_CMOVE"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(unspec:HI
	  [(compare:CCFP (match_dup 1)(match_dup 2))]
	UNSPEC_FNSTSW))
   (set (reg:CC FLAGS_REG)
	(unspec:CC [(match_dup 0)] UNSPEC_SAHF))]
  ""
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "<MODE>")])

(define_insn "*cmpu<mode>_i387"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(unspec:HI
	  [(compare:CCFPU
	     (match_operand:X87MODEF 1 "register_operand" "f")
	     (match_operand:X87MODEF 2 "register_operand" "f"))]
	  UNSPEC_FNSTSW))]
  "TARGET_80387"
  "* return output_fp_compare (insn, operands, false, true);"
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "*cmpu<mode>_cc_i387"
  [(set (reg:CCFPU FLAGS_REG)
	(compare:CCFPU
	  (match_operand:X87MODEF 1 "register_operand" "f")
	  (match_operand:X87MODEF 2 "register_operand" "f")))
   (clobber (match_operand:HI 0 "register_operand" "=a"))]
  "TARGET_80387 && TARGET_SAHF && !TARGET_CMOVE"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(unspec:HI
	  [(compare:CCFPU (match_dup 1)(match_dup 2))]
	UNSPEC_FNSTSW))
   (set (reg:CC FLAGS_REG)
	(unspec:CC [(match_dup 0)] UNSPEC_SAHF))]
  ""
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "mode" "<MODE>")])

(define_insn "*cmp<X87MODEF:mode>_<SWI24:mode>_i387"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(unspec:HI
	  [(compare:CCFP
	     (match_operand:X87MODEF 1 "register_operand" "f")
	     (float:X87MODEF
	       (match_operand:SWI24 2 "memory_operand" "m")))]
	  UNSPEC_FNSTSW))]
  "TARGET_80387
   && (TARGET_USE_<SWI24:MODE>MODE_FIOP
       || optimize_function_for_size_p (cfun))"
  "* return output_fp_compare (insn, operands, false, false);"
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "fp_int_src" "true")
   (set_attr "mode" "<SWI24:MODE>")])

(define_insn_and_split "*cmp<X87MODEF:mode>_<SWI24:mode>_cc_i387"
  [(set (reg:CCFP FLAGS_REG)
	(compare:CCFP
	  (match_operand:X87MODEF 1 "register_operand" "f")
	  (float:X87MODEF
	    (match_operand:SWI24 2 "memory_operand" "m"))))
   (clobber (match_operand:HI 0 "register_operand" "=a"))]
  "TARGET_80387 && TARGET_SAHF && !TARGET_CMOVE
   && (TARGET_USE_<SWI24:MODE>MODE_FIOP
       || optimize_function_for_size_p (cfun))"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(unspec:HI
	  [(compare:CCFP
	     (match_dup 1)
	     (float:X87MODEF (match_dup 2)))]
	UNSPEC_FNSTSW))
   (set (reg:CC FLAGS_REG)
	(unspec:CC [(match_dup 0)] UNSPEC_SAHF))]
  ""
  [(set_attr "type" "multi")
   (set_attr "unit" "i387")
   (set_attr "fp_int_src" "true")
   (set_attr "mode" "<SWI24:MODE>")])

;; FP compares, step 2
;; Move the fpsw to ax.

(define_insn "x86_fnstsw_1"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(unspec:HI [(reg:CCFP FPSR_REG)] UNSPEC_FNSTSW))]
  "TARGET_80387"
  "fnstsw\t%0"
  [(set_attr "length" "2")
   (set_attr "mode" "SI")
   (set_attr "unit" "i387")])

;; FP compares, step 3
;; Get ax into flags, general case.

(define_insn "x86_sahf_1"
  [(set (reg:CC FLAGS_REG)
	(unspec:CC [(match_operand:HI 0 "register_operand" "a")]
		   UNSPEC_SAHF))]
  "TARGET_SAHF"
{
#ifndef HAVE_AS_IX86_SAHF
  if (TARGET_64BIT)
    return ASM_BYTE "0x9e";
  else
#endif
  return "sahf";
}
  [(set_attr "length" "1")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "direct")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "SI")])

;; Pentium Pro can do steps 1 through 3 in one go.
;; (these instructions set flags directly)

(define_mode_iterator FPCMP [CCFP CCFPU])
(define_mode_attr unord [(CCFP "") (CCFPU "u")])

(define_insn "*cmpi<FPCMP:unord><MODEF:mode>"
  [(set (reg:FPCMP FLAGS_REG)
	(compare:FPCMP
	  (match_operand:MODEF 0 "register_operand" "f,v")
	  (match_operand:MODEF 1 "register_ssemem_operand" "f,vm")))]
  "(SSE_FLOAT_MODE_P (<MODEF:MODE>mode) && TARGET_SSE_MATH)
   || (TARGET_80387 && TARGET_CMOVE)"
  "@
   * return output_fp_compare (insn, operands, true, \
			       <FPCMP:MODE>mode == CCFPUmode);
   %v<FPCMP:unord>comi<MODEF:ssemodesuffix>\t{%1, %0|%0, %1}"
  [(set_attr "type" "fcmp,ssecomi")
   (set_attr "prefix" "orig,maybe_vex")
   (set_attr "mode" "<MODEF:MODE>")
   (set_attr "prefix_rep" "*,0")
   (set (attr "prefix_data16")
	(cond [(eq_attr "alternative" "0")
		 (const_string "*")
	       (eq_attr "mode" "DF")
		 (const_string "1")
	      ]
	      (const_string "0")))
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "direct")
   (set_attr "bdver1_decode" "double")
   (set_attr "znver1_decode" "double")
   (set (attr "enabled")
     (if_then_else
       (match_test ("SSE_FLOAT_MODE_P (<MODEF:MODE>mode) && TARGET_SSE_MATH"))
       (if_then_else
	 (eq_attr "alternative" "0")
	 (symbol_ref "TARGET_MIX_SSE_I387")
	 (symbol_ref "true"))
       (if_then_else
	 (eq_attr "alternative" "0")
	 (symbol_ref "true")
	 (symbol_ref "false"))))])

(define_insn "*cmpi<unord>xf_i387"
  [(set (reg:FPCMP FLAGS_REG)
	(compare:FPCMP
	  (match_operand:XF 0 "register_operand" "f")
	  (match_operand:XF 1 "register_operand" "f")))]
  "TARGET_80387 && TARGET_CMOVE"
  "* return output_fp_compare (insn, operands, true,
			       <MODE>mode == CCFPUmode);"
  [(set_attr "type" "fcmp")
   (set_attr "mode" "XF")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "direct")
   (set_attr "bdver1_decode" "double")
   (set_attr "znver1_decode" "double")])

;; Push/pop instructions.

(define_insn "*push<mode>2"
  [(set (match_operand:DWI 0 "push_operand" "=<")
	(match_operand:DWI 1 "general_no_elim_operand" "riF*o"))]
  ""
  "#"
  [(set_attr "type" "multi")
   (set_attr "mode" "<MODE>")])

(define_split
  [(set (match_operand:DWI 0 "push_operand")
        (match_operand:DWI 1 "general_gr_operand"))]
  "reload_completed"
  [(const_int 0)]
  "ix86_split_long_move (operands); DONE;")

(define_insn "*pushdi2_rex64"
  [(set (match_operand:DI 0 "push_operand" "=<,!<")
	(match_operand:DI 1 "general_no_elim_operand" "re*m,n"))]
  "TARGET_64BIT"
  "@
   push{q}\t%1
   #"
  [(set_attr "type" "push,multi")
   (set_attr "mode" "DI")])

;; Convert impossible pushes of immediate to existing instructions.
;; First try to get scratch register and go through it.  In case this
;; fails, push sign extended lower part first and then overwrite
;; upper part by 32bit move.
(define_peephole2
  [(match_scratch:DI 2 "r")
   (set (match_operand:DI 0 "push_operand")
        (match_operand:DI 1 "immediate_operand"))]
  "TARGET_64BIT && !symbolic_operand (operands[1], DImode)
   && !x86_64_immediate_operand (operands[1], DImode)"
  [(set (match_dup 2) (match_dup 1))
   (set (match_dup 0) (match_dup 2))])

;; We need to define this as both peepholer and splitter for case
;; peephole2 pass is not run.
;; "&& 1" is needed to keep it from matching the previous pattern.
(define_peephole2
  [(set (match_operand:DI 0 "push_operand")
        (match_operand:DI 1 "immediate_operand"))]
  "TARGET_64BIT && !symbolic_operand (operands[1], DImode)
   && !x86_64_immediate_operand (operands[1], DImode) && 1"
  [(set (match_dup 0) (match_dup 1))
   (set (match_dup 2) (match_dup 3))]
{
  split_double_mode (DImode, &operands[1], 1, &operands[2], &operands[3]);

  operands[1] = gen_lowpart (DImode, operands[2]);
  operands[2] = gen_rtx_MEM (SImode, gen_rtx_PLUS (Pmode, stack_pointer_rtx,
						   GEN_INT (4)));
})

(define_split
  [(set (match_operand:DI 0 "push_operand")
        (match_operand:DI 1 "immediate_operand"))]
  "TARGET_64BIT && ((optimize > 0 && flag_peephole2)
		    ? epilogue_completed : reload_completed)
   && !symbolic_operand (operands[1], DImode)
   && !x86_64_immediate_operand (operands[1], DImode)"
  [(set (match_dup 0) (match_dup 1))
   (set (match_dup 2) (match_dup 3))]
{
  split_double_mode (DImode, &operands[1], 1, &operands[2], &operands[3]);

  operands[1] = gen_lowpart (DImode, operands[2]);
  operands[2] = gen_rtx_MEM (SImode, gen_rtx_PLUS (Pmode, stack_pointer_rtx,
						   GEN_INT (4)));
})

(define_insn "*pushsi2"
  [(set (match_operand:SI 0 "push_operand" "=<")
	(match_operand:SI 1 "general_no_elim_operand" "ri*m"))]
  "!TARGET_64BIT"
  "push{l}\t%1"
  [(set_attr "type" "push")
   (set_attr "mode" "SI")])

;; emit_push_insn when it calls move_by_pieces requires an insn to
;; "push a byte/word".  But actually we use pushl, which has the effect
;; of rounding the amount pushed up to a word.

;; For TARGET_64BIT we always round up to 8 bytes.
(define_insn "*push<mode>2_rex64"
  [(set (match_operand:SWI124 0 "push_operand" "=X")
	(match_operand:SWI124 1 "nonmemory_no_elim_operand" "r<i>"))]
  "TARGET_64BIT"
  "push{q}\t%q1"
  [(set_attr "type" "push")
   (set_attr "mode" "DI")])

(define_insn "*push<mode>2"
  [(set (match_operand:SWI12 0 "push_operand" "=X")
	(match_operand:SWI12 1 "nonmemory_no_elim_operand" "rn"))]
  "!TARGET_64BIT"
  "push{l}\t%k1"
  [(set_attr "type" "push")
   (set_attr "mode" "SI")])

(define_insn "*push<mode>2_prologue"
  [(set (match_operand:W 0 "push_operand" "=<")
	(match_operand:W 1 "general_no_elim_operand" "r<i>*m"))
   (clobber (mem:BLK (scratch)))]
  ""
  "push{<imodesuffix>}\t%1"
  [(set_attr "type" "push")
   (set_attr "mode" "<MODE>")])

(define_insn "*pop<mode>1"
  [(set (match_operand:W 0 "nonimmediate_operand" "=r*m")
	(match_operand:W 1 "pop_operand" ">"))]
  ""
  "pop{<imodesuffix>}\t%0"
  [(set_attr "type" "pop")
   (set_attr "mode" "<MODE>")])

(define_insn "*pop<mode>1_epilogue"
  [(set (match_operand:W 0 "nonimmediate_operand" "=r*m")
	(match_operand:W 1 "pop_operand" ">"))
   (clobber (mem:BLK (scratch)))]
  ""
  "pop{<imodesuffix>}\t%0"
  [(set_attr "type" "pop")
   (set_attr "mode" "<MODE>")])

(define_insn "*pushfl<mode>2"
  [(set (match_operand:W 0 "push_operand" "=<")
	(match_operand:W 1 "flags_reg_operand"))]
  ""
  "pushf{<imodesuffix>}"
  [(set_attr "type" "push")
   (set_attr "mode" "<MODE>")])

(define_insn "*popfl<mode>1"
  [(set (match_operand:W 0 "flags_reg_operand")
	(match_operand:W 1 "pop_operand" ">"))]
  ""
  "popf{<imodesuffix>}"
  [(set_attr "type" "pop")
   (set_attr "mode" "<MODE>")])


;; Reload patterns to support multi-word load/store
;; with non-offsetable address.
(define_expand "reload_noff_store"
  [(parallel [(match_operand 0 "memory_operand" "=m")
              (match_operand 1 "register_operand" "r")
              (match_operand:DI 2 "register_operand" "=&r")])]
  "TARGET_64BIT"
{
  rtx mem = operands[0];
  rtx addr = XEXP (mem, 0);

  emit_move_insn (operands[2], addr);
  mem = replace_equiv_address_nv (mem, operands[2]);

  emit_insn (gen_rtx_SET (mem, operands[1]));
  DONE;
})

(define_expand "reload_noff_load"
  [(parallel [(match_operand 0 "register_operand" "=r")
              (match_operand 1 "memory_operand" "m")
              (match_operand:DI 2 "register_operand" "=r")])]
  "TARGET_64BIT"
{
  rtx mem = operands[1];
  rtx addr = XEXP (mem, 0);

  emit_move_insn (operands[2], addr);
  mem = replace_equiv_address_nv (mem, operands[2]);

  emit_insn (gen_rtx_SET (operands[0], mem));
  DONE;
})

;; Move instructions.

(define_expand "movxi"
  [(set (match_operand:XI 0 "nonimmediate_operand")
	(match_operand:XI 1 "general_operand"))]
  "TARGET_AVX512F"
  "ix86_expand_vector_move (XImode, operands); DONE;")

(define_expand "movoi"
  [(set (match_operand:OI 0 "nonimmediate_operand")
	(match_operand:OI 1 "general_operand"))]
  "TARGET_AVX"
  "ix86_expand_vector_move (OImode, operands); DONE;")

(define_expand "movti"
  [(set (match_operand:TI 0 "nonimmediate_operand")
	(match_operand:TI 1 "general_operand"))]
  "TARGET_64BIT || TARGET_SSE"
{
  if (TARGET_64BIT)
    ix86_expand_move (TImode, operands);
  else
    ix86_expand_vector_move (TImode, operands);
  DONE;
})

;; This expands to what emit_move_complex would generate if we didn't
;; have a movti pattern.  Having this avoids problems with reload on
;; 32-bit targets when SSE is present, but doesn't seem to be harmful
;; to have around all the time.
(define_expand "movcdi"
  [(set (match_operand:CDI 0 "nonimmediate_operand")
	(match_operand:CDI 1 "general_operand"))]
  ""
{
  if (push_operand (operands[0], CDImode))
    emit_move_complex_push (CDImode, operands[0], operands[1]);
  else
    emit_move_complex_parts (operands[0], operands[1]);
  DONE;
})

(define_expand "mov<mode>"
  [(set (match_operand:SWI1248x 0 "nonimmediate_operand")
	(match_operand:SWI1248x 1 "general_operand"))]
  ""
  "ix86_expand_move (<MODE>mode, operands); DONE;")

(define_insn "*mov<mode>_xor"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(match_operand:SWI48 1 "const0_operand"))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  "xor{l}\t%k0, %k0"
  [(set_attr "type" "alu1")
   (set_attr "modrm_class" "op0")
   (set_attr "mode" "SI")
   (set_attr "length_immediate" "0")])

(define_insn "*mov<mode>_or"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(match_operand:SWI48 1 "constm1_operand"))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  "or{<imodesuffix>}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "mode" "<MODE>")
   (set_attr "length_immediate" "1")])

(define_insn "*movxi_internal_avx512f"
  [(set (match_operand:XI 0 "nonimmediate_operand"		"=v,v ,v ,m")
	(match_operand:XI 1 "nonimmediate_or_sse_const_operand" " C,BC,vm,v"))]
  "TARGET_AVX512F
   && (register_operand (operands[0], XImode)
       || register_operand (operands[1], XImode))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_SSEMOV:
      if (misaligned_operand (operands[0], XImode)
	  || misaligned_operand (operands[1], XImode))
	return "vmovdqu32\t{%1, %0|%0, %1}";
      else
	return "vmovdqa32\t{%1, %0|%0, %1}";

    default:
      gcc_unreachable ();
    }
}
  [(set_attr "type" "sselog1,sselog1,ssemov,ssemov")
   (set_attr "prefix" "evex")
   (set_attr "mode" "XI")])

(define_insn "*movoi_internal_avx"
  [(set (match_operand:OI 0 "nonimmediate_operand"		"=v,v ,v ,m")
	(match_operand:OI 1 "nonimmediate_or_sse_const_operand" " C,BC,vm,v"))]
  "TARGET_AVX
   && (register_operand (operands[0], OImode)
       || register_operand (operands[1], OImode))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_SSEMOV:
      if (misaligned_operand (operands[0], OImode)
	  || misaligned_operand (operands[1], OImode))
	{
	  if (get_attr_mode (insn) == MODE_V8SF)
	    return "vmovups\t{%1, %0|%0, %1}";
	  else if (get_attr_mode (insn) == MODE_XI)
	    return "vmovdqu32\t{%1, %0|%0, %1}";
	  else
	    return "vmovdqu\t{%1, %0|%0, %1}";
	}
      else
	{
	  if (get_attr_mode (insn) == MODE_V8SF)
	    return "vmovaps\t{%1, %0|%0, %1}";
	  else if (get_attr_mode (insn) == MODE_XI)
	    return "vmovdqa32\t{%1, %0|%0, %1}";
	  else
	    return "vmovdqa\t{%1, %0|%0, %1}";
	}

    default:
      gcc_unreachable ();
    }
}
  [(set_attr "isa" "*,avx2,*,*")
   (set_attr "type" "sselog1,sselog1,ssemov,ssemov")
   (set_attr "prefix" "vex")
   (set (attr "mode")
	(cond [(ior (match_operand 0 "ext_sse_reg_operand")
		    (match_operand 1 "ext_sse_reg_operand"))
		 (const_string "XI")
	       (and (eq_attr "alternative" "1")
		    (match_test "TARGET_AVX512VL"))
		 (const_string "XI")
	       (ior (match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL")
		    (and (eq_attr "alternative" "3")
			 (match_test "TARGET_SSE_TYPELESS_STORES")))
		 (const_string "V8SF")
	      ]
	      (const_string "OI")))])

(define_insn "*movti_internal"
  [(set (match_operand:TI 0 "nonimmediate_operand" "=!r ,o ,v,v ,v ,m,?r,?Yd")
	(match_operand:TI 1 "general_operand"	   "riFo,re,C,BC,vm,v,Ye,r"))]
  "(TARGET_64BIT
    && !(MEM_P (operands[0]) && MEM_P (operands[1])))
   || (TARGET_SSE
       && nonimmediate_or_sse_const_operand (operands[1], TImode)
       && (register_operand (operands[0], TImode)
	   || register_operand (operands[1], TImode)))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_MULTI:
      return "#";

    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_SSEMOV:
      /* TDmode values are passed as TImode on the stack.  Moving them
	 to stack may result in unaligned memory access.  */
      if (misaligned_operand (operands[0], TImode)
	  || misaligned_operand (operands[1], TImode))
	{
	  if (get_attr_mode (insn) == MODE_V4SF)
	    return "%vmovups\t{%1, %0|%0, %1}";
	  else if (get_attr_mode (insn) == MODE_XI)
	    return "vmovdqu32\t{%1, %0|%0, %1}";
	  else
	    return "%vmovdqu\t{%1, %0|%0, %1}";
	}
      else
	{
	  if (get_attr_mode (insn) == MODE_V4SF)
	    return "%vmovaps\t{%1, %0|%0, %1}";
	  else if (get_attr_mode (insn) == MODE_XI)
	    return "vmovdqa32\t{%1, %0|%0, %1}";
	  else
	    return "%vmovdqa\t{%1, %0|%0, %1}";
	}

    default:
      gcc_unreachable ();
    }
}
  [(set (attr "isa")
     (cond [(eq_attr "alternative" "0,1,6,7")
	      (const_string "x64")
	    (eq_attr "alternative" "3")
	      (const_string "sse2")
	   ]
	   (const_string "*")))
   (set (attr "type")
     (cond [(eq_attr "alternative" "0,1,6,7")
	      (const_string "multi")
	    (eq_attr "alternative" "2,3")
	      (const_string "sselog1")
	   ]
	   (const_string "ssemov")))
   (set (attr "prefix")
     (if_then_else (eq_attr "type" "sselog1,ssemov")
       (const_string "maybe_vex")
       (const_string "orig")))
   (set (attr "mode")
	(cond [(eq_attr "alternative" "0,1")
		 (const_string "DI")
	       (ior (match_operand 0 "ext_sse_reg_operand")
		    (match_operand 1 "ext_sse_reg_operand"))
		 (const_string "XI")
	       (and (eq_attr "alternative" "3")
		    (match_test "TARGET_AVX512VL"))
		 (const_string "XI")
	       (ior (not (match_test "TARGET_SSE2"))
		    (ior (match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL")
			 (and (eq_attr "alternative" "5")
			      (match_test "TARGET_SSE_TYPELESS_STORES"))))
		 (const_string "V4SF")
	       (match_test "TARGET_AVX")
		 (const_string "TI")
	       (match_test "optimize_function_for_size_p (cfun)")
		 (const_string "V4SF")
	       ]
	       (const_string "TI")))])

(define_split
  [(set (match_operand:TI 0 "sse_reg_operand")
        (match_operand:TI 1 "general_reg_operand"))]
  "TARGET_64BIT && TARGET_SSE4_1 && TARGET_INTER_UNIT_MOVES_TO_VEC
   && reload_completed"
  [(set (match_dup 2)
  	(vec_merge:V2DI
	  (vec_duplicate:V2DI (match_dup 3))
	  (match_dup 2)
	  (const_int 2)))]
{
  operands[2] = lowpart_subreg (V2DImode, operands[0], TImode);
  operands[3] = gen_highpart (DImode, operands[1]);

  emit_move_insn (gen_lowpart (DImode, operands[0]),
  		  gen_lowpart (DImode, operands[1]));
})

(define_insn "*movdi_internal"
  [(set (match_operand:DI 0 "nonimmediate_operand"
    "=r  ,o  ,r,r  ,r,m ,*y,*y,?*y,?m,?r ,?*Ym,*v,*v,*v,m ,m,?r ,?*Yd,?r ,?*Yi,?*Ym,?*Yi,*k,*k ,*r,*m")
	(match_operand:DI 1 "general_operand"
    "riFo,riF,Z,rem,i,re,C ,*y,m  ,*y,*Yn,r   ,C ,*v,m ,*v,v,*Ye,r   ,*Yj,r   ,*Yj ,*Yn ,*r,*km,*k,*k"))]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_MSKMOV:
      return "kmovq\t{%1, %0|%0, %1}";

    case TYPE_MULTI:
      return "#";

    case TYPE_MMX:
      return "pxor\t%0, %0";

    case TYPE_MMXMOV:
      /* Handle broken assemblers that require movd instead of movq.  */
      if (!HAVE_AS_IX86_INTERUNIT_MOVQ
	  && (GENERAL_REG_P (operands[0]) || GENERAL_REG_P (operands[1])))
	return "movd\t{%1, %0|%0, %1}";
      return "movq\t{%1, %0|%0, %1}";

    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_SSEMOV:
      switch (get_attr_mode (insn))
	{
	case MODE_DI:
	  /* Handle broken assemblers that require movd instead of movq.  */
	  if (!HAVE_AS_IX86_INTERUNIT_MOVQ
	      && (GENERAL_REG_P (operands[0]) || GENERAL_REG_P (operands[1])))
	    return "%vmovd\t{%1, %0|%0, %1}";
	  return "%vmovq\t{%1, %0|%0, %1}";
	case MODE_TI:
	  return "%vmovdqa\t{%1, %0|%0, %1}";
	case MODE_XI:
	  return "vmovdqa64\t{%g1, %g0|%g0, %g1}";

	case MODE_V2SF:
	  gcc_assert (!TARGET_AVX);
	  return "movlps\t{%1, %0|%0, %1}";
	case MODE_V4SF:
	  return "%vmovaps\t{%1, %0|%0, %1}";

	default:
	  gcc_unreachable ();
	}

    case TYPE_SSECVT:
      if (SSE_REG_P (operands[0]))
	return "movq2dq\t{%1, %0|%0, %1}";
      else
	return "movdq2q\t{%1, %0|%0, %1}";

    case TYPE_LEA:
      return "lea{q}\t{%E1, %0|%0, %E1}";

    case TYPE_IMOV:
      gcc_assert (!flag_pic || LEGITIMATE_PIC_OPERAND_P (operands[1]));
      if (get_attr_mode (insn) == MODE_SI)
	return "mov{l}\t{%k1, %k0|%k0, %k1}";
      else if (which_alternative == 4)
	return "movabs{q}\t{%1, %0|%0, %1}";
      else if (ix86_use_lea_for_mov (insn, operands))
	return "lea{q}\t{%E1, %0|%0, %E1}";
      else
	return "mov{q}\t{%1, %0|%0, %1}";

    default:
      gcc_unreachable ();
    }
}
  [(set (attr "isa")
     (cond [(eq_attr "alternative" "0,1,17,18")
	      (const_string "nox64")
	    (eq_attr "alternative" "2,3,4,5,10,11,19,20,23,25")
	      (const_string "x64")
	   ]
	   (const_string "*")))
   (set (attr "type")
     (cond [(eq_attr "alternative" "0,1,17,18")
	      (const_string "multi")
	    (eq_attr "alternative" "6")
	      (const_string "mmx")
	    (eq_attr "alternative" "7,8,9,10,11")
	      (const_string "mmxmov")
	    (eq_attr "alternative" "12")
	      (const_string "sselog1")
	    (eq_attr "alternative" "13,14,15,16,19,20")
	      (const_string "ssemov")
	    (eq_attr "alternative" "21,22")
	      (const_string "ssecvt")
	    (eq_attr "alternative" "23,24,25,26")
	      (const_string "mskmov")
	    (and (match_operand 0 "register_operand")
		 (match_operand 1 "pic_32bit_operand"))
	      (const_string "lea")
	   ]
	   (const_string "imov")))
   (set (attr "modrm")
     (if_then_else
       (and (eq_attr "alternative" "4") (eq_attr "type" "imov"))
       (const_string "0")
       (const_string "*")))
   (set (attr "length_immediate")
     (if_then_else
       (and (eq_attr "alternative" "4") (eq_attr "type" "imov"))
       (const_string "8")
       (const_string "*")))
   (set (attr "prefix_rex")
     (if_then_else
       (eq_attr "alternative" "10,11,19,20")
       (const_string "1")
       (const_string "*")))
   (set (attr "prefix")
     (if_then_else (eq_attr "type" "sselog1,ssemov")
       (const_string "maybe_vex")
       (const_string "orig")))
   (set (attr "prefix_data16")
     (if_then_else (and (eq_attr "type" "ssemov") (eq_attr "mode" "DI"))
       (const_string "1")
       (const_string "*")))
   (set (attr "mode")
     (cond [(eq_attr "alternative" "2")
	      (const_string "SI")
	    (eq_attr "alternative" "12,13")
	      (cond [(ior (match_operand 0 "ext_sse_reg_operand")
			  (match_operand 1 "ext_sse_reg_operand"))
		       (const_string "XI")
		     (ior (not (match_test "TARGET_SSE2"))
			  (match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL"))
		       (const_string "V4SF")
		     (match_test "TARGET_AVX")
		       (const_string "TI")
		     (match_test "optimize_function_for_size_p (cfun)")
		       (const_string "V4SF")
		    ]
		    (const_string "TI"))

	    (and (eq_attr "alternative" "14,15,16")
		 (not (match_test "TARGET_SSE2")))
	      (const_string "V2SF")
	   ]
	   (const_string "DI")))
   (set (attr "enabled")
     (cond [(eq_attr "alternative" "15")
              (if_then_else
		(match_test "TARGET_STV && TARGET_SSE2")
		(symbol_ref "false")
		(const_string "*"))
	    (eq_attr "alternative" "16")
              (if_then_else
		(match_test "TARGET_STV && TARGET_SSE2")
		(symbol_ref "true")
		(symbol_ref "false"))
	   ]
	   (const_string "*")))])

(define_split
  [(set (match_operand:<DWI> 0 "general_reg_operand")
        (match_operand:<DWI> 1 "sse_reg_operand"))]
  "TARGET_SSE4_1 && TARGET_INTER_UNIT_MOVES_FROM_VEC
   && reload_completed"
  [(set (match_dup 2)
  	(vec_select:DWIH
	  (match_dup 3)
	  (parallel [(const_int 1)])))]
{
  operands[2] = gen_highpart (<MODE>mode, operands[0]);
  operands[3] = lowpart_subreg (<ssevecmode>mode, operands[1], <DWI>mode);

  emit_move_insn (gen_lowpart (<MODE>mode, operands[0]),
  		  gen_lowpart (<MODE>mode, operands[1]));
})

(define_split
  [(set (match_operand:DWI 0 "nonimmediate_gr_operand")
        (match_operand:DWI 1 "general_gr_operand"))]
  "reload_completed"
  [(const_int 0)]
  "ix86_split_long_move (operands); DONE;")

(define_split
  [(set (match_operand:DI 0 "sse_reg_operand")
        (match_operand:DI 1 "general_reg_operand"))]
  "!TARGET_64BIT && TARGET_SSE4_1 && TARGET_INTER_UNIT_MOVES_TO_VEC
   && reload_completed"
  [(set (match_dup 2)
  	(vec_merge:V4SI
	  (vec_duplicate:V4SI (match_dup 3))
	  (match_dup 2)
	  (const_int 2)))]
{
  operands[2] = lowpart_subreg (V4SImode, operands[0], DImode);
  operands[3] = gen_highpart (SImode, operands[1]);

  emit_move_insn (gen_lowpart (SImode, operands[0]),
  		  gen_lowpart (SImode, operands[1]));
})

;; movabsq $0x0012345678000000, %rax is longer
;; than movl $0x12345678, %eax; shlq $24, %rax.
(define_peephole2
  [(set (match_operand:DI 0 "register_operand")
	(match_operand:DI 1 "const_int_operand"))]
  "TARGET_64BIT
   && optimize_insn_for_size_p ()
   && LEGACY_INT_REG_P (operands[0])
   && !x86_64_immediate_operand (operands[1], DImode)
   && !x86_64_zext_immediate_operand (operands[1], DImode)
   && !((UINTVAL (operands[1]) >> ctz_hwi (UINTVAL (operands[1])))
        & ~(HOST_WIDE_INT) 0xffffffff)
   && peep2_regno_dead_p (0, FLAGS_REG)"
  [(set (match_dup 0) (match_dup 1))
   (parallel [(set (match_dup 0) (ashift:DI (match_dup 0) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  int shift = ctz_hwi (UINTVAL (operands[1]));
  operands[1] = gen_int_mode (UINTVAL (operands[1]) >> shift, DImode);
  operands[2] = gen_int_mode (shift, QImode);
})

(define_insn "*movsi_internal"
  [(set (match_operand:SI 0 "nonimmediate_operand"
    "=r,m ,*y,*y,?*y,?m,?r ,?*Ym,*v,*v,*v,m ,?r ,?*Yi,*k,*k ,*rm")
	(match_operand:SI 1 "general_operand"
    "g ,re,C ,*y,m  ,*y,*Yn,r   ,C ,*v,m ,*v,*Yj,r   ,*r,*km,*k"))]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_MSKMOV:
      return "kmovd\t{%1, %0|%0, %1}";

    case TYPE_SSEMOV:
      switch (get_attr_mode (insn))
	{
	case MODE_SI:
          return "%vmovd\t{%1, %0|%0, %1}";
	case MODE_TI:
	  return "%vmovdqa\t{%1, %0|%0, %1}";
	case MODE_XI:
	  return "vmovdqa32\t{%g1, %g0|%g0, %g1}";

	case MODE_V4SF:
	  return "%vmovaps\t{%1, %0|%0, %1}";

	case MODE_SF:
	  gcc_assert (!TARGET_AVX);
          return "movss\t{%1, %0|%0, %1}";

	default:
	  gcc_unreachable ();
	}

    case TYPE_MMX:
      return "pxor\t%0, %0";

    case TYPE_MMXMOV:
      switch (get_attr_mode (insn))
	{
	case MODE_DI:
	  return "movq\t{%1, %0|%0, %1}";
	case MODE_SI:
	  return "movd\t{%1, %0|%0, %1}";

	default:
	  gcc_unreachable ();
	}

    case TYPE_LEA:
      return "lea{l}\t{%E1, %0|%0, %E1}";

    case TYPE_IMOV:
      gcc_assert (!flag_pic || LEGITIMATE_PIC_OPERAND_P (operands[1]));
      if (ix86_use_lea_for_mov (insn, operands))
	return "lea{l}\t{%E1, %0|%0, %E1}";
      else
	return "mov{l}\t{%1, %0|%0, %1}";

    default:
      gcc_unreachable ();
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "2")
	      (const_string "mmx")
	    (eq_attr "alternative" "3,4,5,6,7")
	      (const_string "mmxmov")
	    (eq_attr "alternative" "8")
	      (const_string "sselog1")
	    (eq_attr "alternative" "9,10,11,12,13")
	      (const_string "ssemov")
	    (eq_attr "alternative" "14,15,16")
	      (const_string "mskmov")
	    (and (match_operand 0 "register_operand")
		 (match_operand 1 "pic_32bit_operand"))
	      (const_string "lea")
	   ]
	   (const_string "imov")))
   (set (attr "prefix")
     (if_then_else (eq_attr "type" "sselog1,ssemov")
       (const_string "maybe_vex")
       (const_string "orig")))
   (set (attr "prefix_data16")
     (if_then_else (and (eq_attr "type" "ssemov") (eq_attr "mode" "SI"))
       (const_string "1")
       (const_string "*")))
   (set (attr "mode")
     (cond [(eq_attr "alternative" "2,3")
	      (const_string "DI")
	    (eq_attr "alternative" "8,9")
	      (cond [(ior (match_operand 0 "ext_sse_reg_operand")
			  (match_operand 1 "ext_sse_reg_operand"))
		       (const_string "XI")
		     (ior (not (match_test "TARGET_SSE2"))
			  (match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL"))
		       (const_string "V4SF")
		     (match_test "TARGET_AVX")
		       (const_string "TI")
		     (match_test "optimize_function_for_size_p (cfun)")
		       (const_string "V4SF")
		    ]
		    (const_string "TI"))

	    (and (eq_attr "alternative" "10,11")
	         (not (match_test "TARGET_SSE2")))
	      (const_string "SF")
	   ]
	   (const_string "SI")))])

(define_insn "*movhi_internal"
  [(set (match_operand:HI 0 "nonimmediate_operand" "=r,r ,r ,m ,k,k ,r,m")
	(match_operand:HI 1 "general_operand"      "r ,rn,rm,rn,r,km,k,k"))]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_IMOVX:
      /* movzwl is faster than movw on p2 due to partial word stalls,
	 though not as fast as an aligned movl.  */
      return "movz{wl|x}\t{%1, %k0|%k0, %1}";

    case TYPE_MSKMOV:
      switch (which_alternative)
	{
	case 4:
	  return "kmovw\t{%k1, %0|%0, %k1}";
	case 6:
	  return "kmovw\t{%1, %k0|%k0, %1}";
	case 5:
	case 7:
	  return "kmovw\t{%1, %0|%0, %1}";
	default:
	  gcc_unreachable ();
	}

    default:
      if (get_attr_mode (insn) == MODE_SI)
	return "mov{l}\t{%k1, %k0|%k0, %k1}";
      else
	return "mov{w}\t{%1, %0|%0, %1}";
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "4,5,6,7")
	      (const_string "mskmov")
	    (match_test "optimize_function_for_size_p (cfun)")
	      (const_string "imov")
	    (and (eq_attr "alternative" "0")
		 (ior (not (match_test "TARGET_PARTIAL_REG_STALL"))
		      (not (match_test "TARGET_HIMODE_MATH"))))
	      (const_string "imov")
	    (and (eq_attr "alternative" "1,2")
		 (match_operand:HI 1 "aligned_operand"))
	      (const_string "imov")
	    (and (match_test "TARGET_MOVX")
		 (eq_attr "alternative" "0,2"))
	      (const_string "imovx")
	   ]
	   (const_string "imov")))
    (set (attr "prefix")
      (if_then_else (eq_attr "alternative" "4,5,6,7")
	(const_string "vex")
	(const_string "orig")))
    (set (attr "mode")
      (cond [(eq_attr "type" "imovx")
	       (const_string "SI")
	     (and (eq_attr "alternative" "1,2")
		  (match_operand:HI 1 "aligned_operand"))
	       (const_string "SI")
	     (and (eq_attr "alternative" "0")
		  (ior (not (match_test "TARGET_PARTIAL_REG_STALL"))
		       (not (match_test "TARGET_HIMODE_MATH"))))
	       (const_string "SI")
	    ]
	    (const_string "HI")))])

;; Situation is quite tricky about when to choose full sized (SImode) move
;; over QImode moves.  For Q_REG -> Q_REG move we use full size only for
;; partial register dependency machines (such as AMD Athlon), where QImode
;; moves issue extra dependency and for partial register stalls machines
;; that don't use QImode patterns (and QImode move cause stall on the next
;; instruction).
;;
;; For loads of Q_REG to NONQ_REG we use full sized moves except for partial
;; register stall machines with, where we use QImode instructions, since
;; partial register stall can be caused there.  Then we use movzx.

(define_insn "*movqi_internal"
  [(set (match_operand:QI 0 "nonimmediate_operand"
			"=Q,R,r,q,q,r,r ,?r,m ,k,k,r,m,k")
	(match_operand:QI 1 "general_operand"
			"Q ,R,r,n,m,q,rn, m,qn,r,k,k,k,m"))]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))"
{
  static char buf[128];
  const char *ops;
  const char *suffix;

  switch (get_attr_type (insn))
    {
    case TYPE_IMOVX:
      gcc_assert (ANY_QI_REG_P (operands[1]) || MEM_P (operands[1]));
      return "movz{bl|x}\t{%1, %k0|%k0, %1}";

    case TYPE_MSKMOV:
      switch (which_alternative)
        {
	case 9:
	  ops = "kmov%s\t{%%k1, %%0|%%0, %%k1}";
	  break;
	case 11:
	  ops = "kmov%s\t{%%1, %%k0|%%k0, %%1}";
	  break;
	case 12:
	case 13:
	  gcc_assert (TARGET_AVX512DQ);
	  /* FALLTHRU */
	case 10:
	  ops = "kmov%s\t{%%1, %%0|%%0, %%1}";
	  break;
	default:
	  gcc_unreachable ();
	}

      suffix = (get_attr_mode (insn) == MODE_HI) ? "w" : "b";

      snprintf (buf, sizeof (buf), ops, suffix);
      return buf;

    default:
      if (get_attr_mode (insn) == MODE_SI)
	return "mov{l}\t{%k1, %k0|%k0, %k1}";
      else
	return "mov{b}\t{%1, %0|%0, %1}";
    }
}
  [(set (attr "isa")
     (cond [(eq_attr "alternative" "1,2")
	      (const_string "x64")
	    (eq_attr "alternative" "12,13")
	      (const_string "avx512dq")
	   ]
	   (const_string "*")))
   (set (attr "type")
     (cond [(eq_attr "alternative" "9,10,11,12,13")
	      (const_string "mskmov")
	    (and (eq_attr "alternative" "7")
		 (not (match_operand:QI 1 "aligned_operand")))
	      (const_string "imovx")
	    (match_test "optimize_function_for_size_p (cfun)")
	      (const_string "imov")
	    (and (eq_attr "alternative" "5")
		 (ior (not (match_test "TARGET_PARTIAL_REG_STALL"))
		      (not (match_test "TARGET_QIMODE_MATH"))))
	      (const_string "imov")
	    (eq_attr "alternative" "5,7")
	      (const_string "imovx")
	    (and (match_test "TARGET_MOVX")
		 (eq_attr "alternative" "4"))
	      (const_string "imovx")
	   ]
	   (const_string "imov")))
   (set (attr "prefix")
     (if_then_else (eq_attr "alternative" "9,10,11")
       (const_string "vex")
       (const_string "orig")))
   (set (attr "mode")
      (cond [(eq_attr "alternative" "5,6,7")
	       (const_string "SI")
	     (eq_attr "alternative" "8")
	       (const_string "QI")
	     (and (eq_attr "alternative" "9,10,11")
		  (not (match_test "TARGET_AVX512DQ")))
	       (const_string "HI")
	     (eq_attr "type" "imovx")
	       (const_string "SI")
	     ;; For -Os, 8-bit immediates are always shorter than 32-bit
	     ;; ones.
	     (and (eq_attr "type" "imov")
		  (and (eq_attr "alternative" "3")
		       (match_test "optimize_function_for_size_p (cfun)")))
	       (const_string "QI")
	     ;; For -Os, movl where one or both operands are NON_Q_REGS
	     ;; and both are LEGACY_REGS is shorter than movb.
	     ;; Otherwise movb and movl sizes are the same, so decide purely
	     ;; based on speed factors.
	     (and (eq_attr "type" "imov")
		  (and (eq_attr "alternative" "1")
		       (match_test "optimize_function_for_size_p (cfun)")))
	       (const_string "SI")
	     (and (eq_attr "type" "imov")
		  (and (eq_attr "alternative" "0,1,2,3")
		       (and (match_test "TARGET_PARTIAL_REG_DEPENDENCY")
			    (not (match_test "TARGET_PARTIAL_REG_STALL")))))
	       (const_string "SI")
	     ;; Avoid partial register stalls when not using QImode arithmetic
	     (and (eq_attr "type" "imov")
		  (and (eq_attr "alternative" "0,1,2,3")
		       (and (match_test "TARGET_PARTIAL_REG_STALL")
			    (not (match_test "TARGET_QIMODE_MATH")))))
	       (const_string "SI")
	   ]
	   (const_string "QI")))])

;; Stores and loads of ax to arbitrary constant address.
;; We fake an second form of instruction to force reload to load address
;; into register when rax is not available
(define_insn "*movabs<mode>_1"
  [(set (mem:SWI1248x (match_operand:DI 0 "x86_64_movabs_operand" "i,r"))
	(match_operand:SWI1248x 1 "nonmemory_operand" "a,r<i>"))]
  "TARGET_LP64 && ix86_check_movabs (insn, 0)"
{
  /* Recover the full memory rtx.  */
  operands[0] = SET_DEST (PATTERN (insn));
  switch (which_alternative)
    {
    case 0:
      return "movabs{<imodesuffix>}\t{%1, %P0|<iptrsize> PTR [%P0], %1}";
    case 1:
      return "mov{<imodesuffix>}\t{%1, %0|%0, %1}";
    default:
      gcc_unreachable ();
    }
}
  [(set_attr "type" "imov")
   (set_attr "modrm" "0,*")
   (set_attr "length_address" "8,0")
   (set_attr "length_immediate" "0,*")
   (set_attr "memory" "store")
   (set_attr "mode" "<MODE>")])

(define_insn "*movabs<mode>_2"
  [(set (match_operand:SWI1248x 0 "register_operand" "=a,r")
        (mem:SWI1248x (match_operand:DI 1 "x86_64_movabs_operand" "i,r")))]
  "TARGET_LP64 && ix86_check_movabs (insn, 1)"
{
  /* Recover the full memory rtx.  */
  operands[1] = SET_SRC (PATTERN (insn));
  switch (which_alternative)
    {
    case 0:
      return "movabs{<imodesuffix>}\t{%P1, %0|%0, <iptrsize> PTR [%P1]}";
    case 1:
      return "mov{<imodesuffix>}\t{%1, %0|%0, %1}";
    default:
      gcc_unreachable ();
    }
}
  [(set_attr "type" "imov")
   (set_attr "modrm" "0,*")
   (set_attr "length_address" "8,0")
   (set_attr "length_immediate" "0")
   (set_attr "memory" "load")
   (set_attr "mode" "<MODE>")])

(define_insn "*swap<mode>"
  [(set (match_operand:SWI48 0 "register_operand" "+r")
	(match_operand:SWI48 1 "register_operand" "+r"))
   (set (match_dup 1)
	(match_dup 0))]
  ""
  "xchg{<imodesuffix>}\t%1, %0"
  [(set_attr "type" "imov")
   (set_attr "mode" "<MODE>")
   (set_attr "pent_pair" "np")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "double")
   (set_attr "bdver1_decode" "double")])

(define_insn "*swap<mode>"
  [(set (match_operand:SWI12 0 "register_operand" "+<r>,r")
	(match_operand:SWI12 1 "register_operand" "+<r>,r"))
   (set (match_dup 1)
	(match_dup 0))]
  ""
  "@
   xchg{<imodesuffix>}\t%1, %0
   xchg{l}\t%k1, %k0"
  [(set_attr "type" "imov")
   (set_attr "mode" "<MODE>,SI")
   (set (attr "preferred_for_size")
     (cond [(eq_attr "alternative" "0")
	      (symbol_ref "false")]
	   (symbol_ref "true")))
   ;; Potential partial reg stall on alternative 1.
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "1")
	      (symbol_ref "!TARGET_PARTIAL_REG_STALL")]
	   (symbol_ref "true")))
   (set_attr "pent_pair" "np")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "double")
   (set_attr "bdver1_decode" "double")])

(define_expand "movstrict<mode>"
  [(set (strict_low_part (match_operand:SWI12 0 "nonimmediate_operand"))
	(match_operand:SWI12 1 "general_operand"))]
  ""
{
  if (TARGET_PARTIAL_REG_STALL && optimize_function_for_speed_p (cfun))
    FAIL;
  if (SUBREG_P (operands[0])
      && GET_MODE_CLASS (GET_MODE (SUBREG_REG (operands[0]))) != MODE_INT)
    FAIL;
  /* Don't generate memory->memory moves, go through a register */
  if (MEM_P (operands[0]) && MEM_P (operands[1]))
    operands[1] = force_reg (<MODE>mode, operands[1]);
})

(define_insn "*movstrict<mode>_1"
  [(set (strict_low_part
	  (match_operand:SWI12 0 "nonimmediate_operand" "+<r>m,<r>"))
	(match_operand:SWI12 1 "general_operand" "<r>n,m"))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "mov{<imodesuffix>}\t{%1, %0|%0, %1}"
  [(set_attr "type" "imov")
   (set_attr "mode" "<MODE>")])

(define_insn "*movstrict<mode>_xor"
  [(set (strict_low_part (match_operand:SWI12 0 "register_operand" "+<r>"))
	(match_operand:SWI12 1 "const0_operand"))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  "xor{<imodesuffix>}\t%0, %0"
  [(set_attr "type" "alu1")
   (set_attr "modrm_class" "op0")
   (set_attr "mode" "<MODE>")
   (set_attr "length_immediate" "0")])

(define_expand "extv<mode>"
  [(set (match_operand:SWI24 0 "register_operand")
	(sign_extract:SWI24 (match_operand:SWI24 1 "register_operand")
			    (match_operand:SI 2 "const_int_operand")
			    (match_operand:SI 3 "const_int_operand")))]
  ""
{
  /* Handle extractions from %ah et al.  */
  if (INTVAL (operands[2]) != 8 || INTVAL (operands[3]) != 8)
    FAIL;

  unsigned int regno = reg_or_subregno (operands[1]);

  /* Be careful to expand only with registers having upper parts.  */
  if (regno <= LAST_VIRTUAL_REGISTER && !QI_REGNO_P (regno))
    operands[1] = copy_to_reg (operands[1]);
})

(define_insn "*extv<mode>"
  [(set (match_operand:SWI24 0 "register_operand" "=R")
	(sign_extract:SWI24 (match_operand 1 "ext_register_operand" "Q")
			    (const_int 8)
			    (const_int 8)))]
  ""
  "movs{bl|x}\t{%h1, %k0|%k0, %h1}"
  [(set_attr "type" "imovx")
   (set_attr "mode" "SI")])

(define_expand "extzv<mode>"
  [(set (match_operand:SWI248 0 "register_operand")
	(zero_extract:SWI248 (match_operand:SWI248 1 "register_operand")
			     (match_operand:SI 2 "const_int_operand")
			     (match_operand:SI 3 "const_int_operand")))]
  ""
{
  if (ix86_expand_pextr (operands))
    DONE;

  /* Handle extractions from %ah et al.  */
  if (INTVAL (operands[2]) != 8 || INTVAL (operands[3]) != 8)
    FAIL;

  unsigned int regno = reg_or_subregno (operands[1]);

  /* Be careful to expand only with registers having upper parts.  */
  if (regno <= LAST_VIRTUAL_REGISTER && !QI_REGNO_P (regno))
    operands[1] = copy_to_reg (operands[1]);
})

(define_insn "*extzvqi_mem_rex64"
  [(set (match_operand:QI 0 "norex_memory_operand" "=Bn")
	(subreg:QI
	  (zero_extract:SI (match_operand 1 "ext_register_operand" "Q")
			   (const_int 8)
			   (const_int 8)) 0))]
  "TARGET_64BIT && reload_completed"
  "mov{b}\t{%h1, %0|%0, %h1}"
  [(set_attr "type" "imov")
   (set_attr "mode" "QI")])

(define_insn "*extzv<mode>"
  [(set (match_operand:SWI248 0 "register_operand" "=R")
	(zero_extract:SWI248 (match_operand 1 "ext_register_operand" "Q")
			     (const_int 8)
			     (const_int 8)))]
  ""
  "movz{bl|x}\t{%h1, %k0|%k0, %h1}"
  [(set_attr "type" "imovx")
   (set_attr "mode" "SI")])

(define_insn "*extzvqi"
  [(set (match_operand:QI 0 "nonimmediate_operand" "=QBc,?R,m")
	(subreg:QI
	  (zero_extract:SI (match_operand 1 "ext_register_operand" "Q,Q,Q")
			   (const_int 8)
			   (const_int 8)) 0))]
  ""
{
  switch (get_attr_type (insn))
    {
    case TYPE_IMOVX:
      return "movz{bl|x}\t{%h1, %k0|%k0, %h1}";
    default:
      return "mov{b}\t{%h1, %0|%0, %h1}";
    }
}
  [(set_attr "isa" "*,*,nox64")
   (set (attr "type")
     (if_then_else (and (match_operand:QI 0 "register_operand")
			(ior (not (match_operand:QI 0 "QIreg_operand"))
			     (match_test "TARGET_MOVX")))
	(const_string "imovx")
	(const_string "imov")))
   (set (attr "mode")
     (if_then_else (eq_attr "type" "imovx")
	(const_string "SI")
	(const_string "QI")))])

(define_peephole2
  [(set (match_operand:QI 0 "register_operand")
	(subreg:QI
	  (zero_extract:SI (match_operand 1 "ext_register_operand")
			   (const_int 8)
			   (const_int 8)) 0))
   (set (match_operand:QI 2 "norex_memory_operand") (match_dup 0))]
  "TARGET_64BIT
   && peep2_reg_dead_p (2, operands[0])"
  [(set (match_dup 2)
	(subreg:QI
	  (zero_extract:SI (match_dup 1)
			   (const_int 8)
			   (const_int 8)) 0))])

(define_expand "insv<mode>"
  [(set (zero_extract:SWI248 (match_operand:SWI248 0 "register_operand")
			     (match_operand:SI 1 "const_int_operand")
			     (match_operand:SI 2 "const_int_operand"))
        (match_operand:SWI248 3 "register_operand"))]
  ""
{
  rtx dst;

  if (ix86_expand_pinsr (operands))
    DONE;

  /* Handle insertions to %ah et al.  */
  if (INTVAL (operands[1]) != 8 || INTVAL (operands[2]) != 8)
    FAIL;

  unsigned int regno = reg_or_subregno (operands[0]);

  /* Be careful to expand only with registers having upper parts.  */
  if (regno <= LAST_VIRTUAL_REGISTER && !QI_REGNO_P (regno))
    dst = copy_to_reg (operands[0]);
  else
    dst = operands[0];

  emit_insn (gen_insv<mode>_1 (dst, operands[3]));

  /* Fix up the destination if needed.  */
  if (dst != operands[0])
    emit_move_insn (operands[0], dst);

  DONE;
})

(define_insn "*insvqi_1_mem_rex64"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (match_operand:QI 1 "norex_memory_operand" "Bn") 0))]
  "TARGET_64BIT && reload_completed"
  "mov{b}\t{%1, %h0|%h0, %1}"
  [(set_attr "type" "imov")
   (set_attr "mode" "QI")])

(define_insn "insv<mode>_1"
  [(set (zero_extract:SWI248 (match_operand 0 "ext_register_operand" "+Q,Q")
			     (const_int 8)
			     (const_int 8))
	(match_operand:SWI248 1 "general_operand" "QnBc,m"))]
  ""
{
  if (CONST_INT_P (operands[1]))
    operands[1] = gen_int_mode (INTVAL (operands[1]), QImode);
  return "mov{b}\t{%b1, %h0|%h0, %b1}";
}
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "imov")
   (set_attr "mode" "QI")])

(define_insn "*insvqi_1"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q,Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (match_operand:QI 1 "general_operand" "QnBc,m") 0))]
  ""
  "mov{b}\t{%1, %h0|%h0, %1}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "imov")
   (set_attr "mode" "QI")])

(define_peephole2
  [(set (match_operand:QI 0 "register_operand")
	(match_operand:QI 1 "norex_memory_operand"))
   (set (zero_extract:SI (match_operand 2 "ext_register_operand")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI (match_dup 0) 0))]
  "TARGET_64BIT
   && peep2_reg_dead_p (2, operands[0])"
  [(set (zero_extract:SI (match_dup 2)
			 (const_int 8)
			 (const_int 8))
	   (subreg:SI (match_dup 1) 0))])

(define_code_iterator any_extract [sign_extract zero_extract])

(define_insn "*insvqi_2"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q")
			 (const_int 8)
			 (const_int 8))
	(any_extract:SI (match_operand 1 "ext_register_operand" "Q")
			(const_int 8)
			(const_int 8)))]
  ""
  "mov{b}\t{%h1, %h0|%h0, %h1}"
  [(set_attr "type" "imov")
   (set_attr "mode" "QI")])

(define_insn "*insvqi_3"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q")
			 (const_int 8)
			 (const_int 8))
	(any_shiftrt:SI (match_operand:SI 1 "register_operand" "Q")
			(const_int 8)))]
  ""
  "mov{b}\t{%h1, %h0|%h0, %h1}"
  [(set_attr "type" "imov")
   (set_attr "mode" "QI")])

;; Floating point push instructions.

(define_insn "*pushtf"
  [(set (match_operand:TF 0 "push_operand" "=<,<")
	(match_operand:TF 1 "general_no_elim_operand" "v,*roC"))]
  "TARGET_64BIT || TARGET_SSE"
{
  /* This insn should be already split before reg-stack.  */
  gcc_unreachable ();
}
  [(set_attr "isa" "*,x64")
   (set_attr "type" "multi")
   (set_attr "unit" "sse,*")
   (set_attr "mode" "TF,DI")])

;; %%% Kill this when call knows how to work this out.
(define_split
  [(set (match_operand:TF 0 "push_operand")
	(match_operand:TF 1 "sse_reg_operand"))]
  "TARGET_SSE && reload_completed"
  [(set (reg:P SP_REG) (plus:P (reg:P SP_REG) (const_int -16)))
   (set (match_dup 0) (match_dup 1))]
{
  /* Preserve memory attributes. */
  operands[0] = replace_equiv_address (operands[0], stack_pointer_rtx);
})

(define_insn_and_split "*pushxf_rounded"
  [(set (mem:XF
	  (pre_modify:P
	    (reg:P SP_REG)
	    (plus:P (reg:P SP_REG) (const_int -16))))
	(match_operand:XF 0 "nonmemory_no_elim_operand" "f,r,*r,C"))]
  "TARGET_64BIT"
  "#"
  "&& 1"
  [(set (reg:P SP_REG) (plus:P (reg:P SP_REG) (const_int -16)))
   (set (match_dup 1) (match_dup 0))]
{
  rtx pat = PATTERN (curr_insn);
  operands[1] = SET_DEST (pat);

  /* Preserve memory attributes. */
  operands[1] = replace_equiv_address (operands[1], stack_pointer_rtx);
}
  [(set_attr "type" "multi")
   (set_attr "unit" "i387,*,*,*")
   (set (attr "mode")
	(cond [(eq_attr "alternative" "1,2,3")
		 (const_string "DI")
	      ]
	      (const_string "XF")))
   (set (attr "preferred_for_size")
     (cond [(eq_attr "alternative" "1")
              (symbol_ref "false")]
           (symbol_ref "true")))])

(define_insn "*pushxf"
  [(set (match_operand:XF 0 "push_operand" "=<,<,<,<,<")
	(match_operand:XF 1 "general_no_elim_operand" "f,r,*r,oF,oC"))]
  ""
{
  /* This insn should be already split before reg-stack.  */
  gcc_unreachable ();
}
  [(set_attr "isa" "*,*,*,nox64,x64")
   (set_attr "type" "multi")
   (set_attr "unit" "i387,*,*,*,*")
   (set (attr "mode")
	(cond [(eq_attr "alternative" "1,2,3,4")
		 (if_then_else (match_test "TARGET_64BIT")
		   (const_string "DI")
		   (const_string "SI"))
	      ]
	      (const_string "XF")))
   (set (attr "preferred_for_size")
     (cond [(eq_attr "alternative" "1")
              (symbol_ref "false")]
           (symbol_ref "true")))])

;; %%% Kill this when call knows how to work this out.
(define_split
  [(set (match_operand:XF 0 "push_operand")
	(match_operand:XF 1 "fp_register_operand"))]
  "reload_completed"
  [(set (reg:P SP_REG) (plus:P (reg:P SP_REG) (match_dup 2)))
   (set (match_dup 0) (match_dup 1))]
{
  operands[2] = GEN_INT (-PUSH_ROUNDING (GET_MODE_SIZE (XFmode)));
  /* Preserve memory attributes. */
  operands[0] = replace_equiv_address (operands[0], stack_pointer_rtx);
})

(define_insn "*pushdf"
  [(set (match_operand:DF 0 "push_operand" "=<,<,<,<,<,<")
	(match_operand:DF 1 "general_no_elim_operand" "f,r,*r,oF,rmC,x"))]
  ""
{
  /* This insn should be already split before reg-stack.  */
  gcc_unreachable ();
}
  [(set_attr "isa" "*,nox64,nox64,nox64,x64,sse2")
   (set_attr "type" "multi")
   (set_attr "unit" "i387,*,*,*,*,sse")
   (set_attr "mode" "DF,SI,SI,SI,DI,DF")
   (set (attr "preferred_for_size")
     (cond [(eq_attr "alternative" "1")
              (symbol_ref "false")]
           (symbol_ref "true")))
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "1")
              (symbol_ref "TARGET_INTEGER_DFMODE_MOVES")]
           (symbol_ref "true")))])
   
;; %%% Kill this when call knows how to work this out.
(define_split
  [(set (match_operand:DF 0 "push_operand")
	(match_operand:DF 1 "any_fp_register_operand"))]
  "reload_completed"
  [(set (reg:P SP_REG) (plus:P (reg:P SP_REG) (const_int -8)))
   (set (match_dup 0) (match_dup 1))]
{
  /* Preserve memory attributes. */
  operands[0] = replace_equiv_address (operands[0], stack_pointer_rtx);
})

(define_insn "*pushsf_rex64"
  [(set (match_operand:SF 0 "push_operand" "=X,X,X")
	(match_operand:SF 1 "nonmemory_no_elim_operand" "f,rF,x"))]
  "TARGET_64BIT"
{
  /* Anything else should be already split before reg-stack.  */
  gcc_assert (which_alternative == 1);
  return "push{q}\t%q1";
}
  [(set_attr "type" "multi,push,multi")
   (set_attr "unit" "i387,*,*")
   (set_attr "mode" "SF,DI,SF")])

(define_insn "*pushsf"
  [(set (match_operand:SF 0 "push_operand" "=<,<,<")
	(match_operand:SF 1 "general_no_elim_operand" "f,rmF,x"))]
  "!TARGET_64BIT"
{
  /* Anything else should be already split before reg-stack.  */
  gcc_assert (which_alternative == 1);
  return "push{l}\t%1";
}
  [(set_attr "type" "multi,push,multi")
   (set_attr "unit" "i387,*,*")
   (set_attr "mode" "SF,SI,SF")])

;; %%% Kill this when call knows how to work this out.
(define_split
  [(set (match_operand:SF 0 "push_operand")
	(match_operand:SF 1 "any_fp_register_operand"))]
  "reload_completed"
  [(set (reg:P SP_REG) (plus:P (reg:P SP_REG) (match_dup 2)))
   (set (match_dup 0) (match_dup 1))]
{
  rtx op = XEXP (operands[0], 0);
  if (GET_CODE (op) == PRE_DEC)
    {
      gcc_assert (!TARGET_64BIT);
      op = GEN_INT (-4);
    }
  else
    {
      op = XEXP (XEXP (op, 1), 1);
      gcc_assert (CONST_INT_P (op));
    }
  operands[2] = op;
  /* Preserve memory attributes. */
  operands[0] = replace_equiv_address (operands[0], stack_pointer_rtx);
})

(define_split
  [(set (match_operand:SF 0 "push_operand")
	(match_operand:SF 1 "memory_operand"))]
  "reload_completed
   && find_constant_src (insn)"
  [(set (match_dup 0) (match_dup 2))]
  "operands[2] = find_constant_src (curr_insn);")

(define_split
  [(set (match_operand 0 "push_operand")
	(match_operand 1 "general_gr_operand"))]
  "reload_completed
   && (GET_MODE (operands[0]) == TFmode
       || GET_MODE (operands[0]) == XFmode
       || GET_MODE (operands[0]) == DFmode)"
  [(const_int 0)]
  "ix86_split_long_move (operands); DONE;")

;; Floating point move instructions.

(define_expand "movtf"
  [(set (match_operand:TF 0 "nonimmediate_operand")
	(match_operand:TF 1 "nonimmediate_operand"))]
  "TARGET_64BIT || TARGET_SSE"
  "ix86_expand_move (TFmode, operands); DONE;")

(define_expand "mov<mode>"
  [(set (match_operand:X87MODEF 0 "nonimmediate_operand")
	(match_operand:X87MODEF 1 "general_operand"))]
  ""
  "ix86_expand_move (<MODE>mode, operands); DONE;")

(define_insn "*movtf_internal"
  [(set (match_operand:TF 0 "nonimmediate_operand" "=v,v ,m,?*r ,!o")
	(match_operand:TF 1 "general_operand"	   "C ,vm,v,*roF,*rC"))]
  "(TARGET_64BIT || TARGET_SSE)
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))
   && (lra_in_progress || reload_completed
       || !CONST_DOUBLE_P (operands[1])
       || ((optimize_function_for_size_p (cfun)
	    || (ix86_cmodel == CM_LARGE || ix86_cmodel == CM_LARGE_PIC))
	   && standard_sse_constant_p (operands[1], TFmode) == 1
	   && !memory_operand (operands[0], TFmode))
       || (!TARGET_MEMORY_MISMATCH_STALL
	   && memory_operand (operands[0], TFmode)))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_SSEMOV:
      /* Handle misaligned load/store since we
         don't have movmisaligntf pattern. */
      if (misaligned_operand (operands[0], TFmode)
	  || misaligned_operand (operands[1], TFmode))
	{
	  if (get_attr_mode (insn) == MODE_V4SF)
	    return "%vmovups\t{%1, %0|%0, %1}";
	  else if (TARGET_AVX512VL
		   && (EXT_REX_SSE_REG_P (operands[0])
		       || EXT_REX_SSE_REG_P (operands[1])))
	    return "vmovdqu64\t{%1, %0|%0, %1}";
	  else
	    return "%vmovdqu\t{%1, %0|%0, %1}";
	}
      else
	{
	  if (get_attr_mode (insn) == MODE_V4SF)
	    return "%vmovaps\t{%1, %0|%0, %1}";
	  else if (TARGET_AVX512VL
		   && (EXT_REX_SSE_REG_P (operands[0])
		       || EXT_REX_SSE_REG_P (operands[1])))
	    return "vmovdqa64\t{%1, %0|%0, %1}";
	  else
	    return "%vmovdqa\t{%1, %0|%0, %1}";
	}

    case TYPE_MULTI:
	return "#";

    default:
      gcc_unreachable ();
    }
}
  [(set_attr "isa" "*,*,*,x64,x64")
   (set_attr "type" "sselog1,ssemov,ssemov,multi,multi")
   (set (attr "prefix")
     (if_then_else (eq_attr "type" "sselog1,ssemov")
       (const_string "maybe_vex")
       (const_string "orig")))
   (set (attr "mode")
        (cond [(eq_attr "alternative" "3,4")
		 (const_string "DI")
	       (match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL")
		 (const_string "V4SF")
	       (and (eq_attr "alternative" "2")
		    (match_test "TARGET_SSE_TYPELESS_STORES"))
		 (const_string "V4SF")
	       (match_test "TARGET_AVX")
		 (const_string "TI")
	       (ior (not (match_test "TARGET_SSE2"))
		    (match_test "optimize_function_for_size_p (cfun)"))
		 (const_string "V4SF")
	       ]
	       (const_string "TI")))])

(define_split
  [(set (match_operand:TF 0 "nonimmediate_gr_operand")
        (match_operand:TF 1 "general_gr_operand"))]
  "reload_completed"
  [(const_int 0)]
  "ix86_split_long_move (operands); DONE;")

;; Possible store forwarding (partial memory) stall
;; in alternatives 4, 6, 7 and 8.
(define_insn "*movxf_internal"
  [(set (match_operand:XF 0 "nonimmediate_operand"
	 "=f,m,f,?r ,!o,?*r ,!o,!o,!o,r  ,o ,o")
	(match_operand:XF 1 "general_operand"
	 "fm,f,G,roF,r ,*roF,*r,F ,C ,roF,rF,rC"))]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))
   && (lra_in_progress || reload_completed
       || !CONST_DOUBLE_P (operands[1])
       || ((optimize_function_for_size_p (cfun)
	    || (ix86_cmodel == CM_LARGE || ix86_cmodel == CM_LARGE_PIC))
	   && standard_80387_constant_p (operands[1]) > 0
	   && !memory_operand (operands[0], XFmode))
       || (!TARGET_MEMORY_MISMATCH_STALL
	   && memory_operand (operands[0], XFmode))
       || !TARGET_HARD_XF_REGS)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_FMOV:
      if (which_alternative == 2)
        return standard_80387_constant_opcode (operands[1]);
      return output_387_reg_move (insn, operands);

    case TYPE_MULTI:
      return "#";

    default:
      gcc_unreachable ();
    }
}
  [(set (attr "isa")
	(cond [(eq_attr "alternative" "7,10")
		 (const_string "nox64")
	       (eq_attr "alternative" "8,11")
		 (const_string "x64")
	      ]
	      (const_string "*")))
   (set (attr "type")
	(cond [(eq_attr "alternative" "3,4,5,6,7,8,9,10,11")
		 (const_string "multi")
	      ]
	      (const_string "fmov")))
   (set (attr "mode")
	(cond [(eq_attr "alternative" "3,4,5,6,7,8,9,10,11")
		 (if_then_else (match_test "TARGET_64BIT")
		   (const_string "DI")
		   (const_string "SI"))
	      ]
	      (const_string "XF")))
   (set (attr "preferred_for_size")
     (cond [(eq_attr "alternative" "3,4")
              (symbol_ref "false")]
           (symbol_ref "true")))
   (set (attr "enabled")
     (cond [(eq_attr "alternative" "9,10,11")
              (if_then_else
		(match_test "TARGET_HARD_XF_REGS")
		(symbol_ref "false")
		(const_string "*"))
            (not (match_test "TARGET_HARD_XF_REGS"))
	      (symbol_ref "false")
	   ]
	   (const_string "*")))])
   
(define_split
  [(set (match_operand:XF 0 "nonimmediate_gr_operand")
        (match_operand:XF 1 "general_gr_operand"))]
  "reload_completed"
  [(const_int 0)]
  "ix86_split_long_move (operands); DONE;")

;; Possible store forwarding (partial memory) stall in alternatives 4, 6 and 7.
(define_insn "*movdf_internal"
  [(set (match_operand:DF 0 "nonimmediate_operand"
    "=Yf*f,m   ,Yf*f,?r ,!o,?*r ,!o,!o,?r,?m,?r,?r,v,v,v,m,*x,*x,*x,m ,r ,Yi,r  ,o ,r  ,m")
	(match_operand:DF 1 "general_operand"
    "Yf*fm,Yf*f,G   ,roF,r ,*roF,*r,F ,rm,rC,C ,F ,C,v,m,v,C ,*x,m ,*x,Yj,r ,roF,rF,rmF,rC"))]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))
   && (lra_in_progress || reload_completed
       || !CONST_DOUBLE_P (operands[1])
       || ((optimize_function_for_size_p (cfun)
	    || (ix86_cmodel == CM_LARGE || ix86_cmodel == CM_LARGE_PIC))
	   && ((IS_STACK_MODE (DFmode)
		&& standard_80387_constant_p (operands[1]) > 0)
	       || (TARGET_SSE2 && TARGET_SSE_MATH
		   && standard_sse_constant_p (operands[1], DFmode) == 1))
	   && !memory_operand (operands[0], DFmode))
       || ((TARGET_64BIT || !TARGET_MEMORY_MISMATCH_STALL)
	   && memory_operand (operands[0], DFmode))
       || !TARGET_HARD_DF_REGS)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_FMOV:
      if (which_alternative == 2)
        return standard_80387_constant_opcode (operands[1]);
      return output_387_reg_move (insn, operands);

    case TYPE_MULTI:
      return "#";

    case TYPE_IMOV:
      if (get_attr_mode (insn) == MODE_SI)
	return "mov{l}\t{%1, %k0|%k0, %1}";
      else if (which_alternative == 11)
	return "movabs{q}\t{%1, %0|%0, %1}";
      else
	return "mov{q}\t{%1, %0|%0, %1}";

    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_SSEMOV:
      switch (get_attr_mode (insn))
	{
	case MODE_DF:
	  if (TARGET_AVX && REG_P (operands[0]) && REG_P (operands[1]))
	    return "vmovsd\t{%1, %0, %0|%0, %0, %1}";
	  return "%vmovsd\t{%1, %0|%0, %1}";

	case MODE_V4SF:
	  return "%vmovaps\t{%1, %0|%0, %1}";
	case MODE_V8DF:
	  return "vmovapd\t{%g1, %g0|%g0, %g1}";
	case MODE_V2DF:
	  return "%vmovapd\t{%1, %0|%0, %1}";

	case MODE_V2SF:
	  gcc_assert (!TARGET_AVX);
	  return "movlps\t{%1, %0|%0, %1}";
	case MODE_V1DF:
	  gcc_assert (!TARGET_AVX);
	  return "movlpd\t{%1, %0|%0, %1}";

	case MODE_DI:
	  /* Handle broken assemblers that require movd instead of movq.  */
	  if (!HAVE_AS_IX86_INTERUNIT_MOVQ
	      && (GENERAL_REG_P (operands[0]) || GENERAL_REG_P (operands[1])))
	    return "%vmovd\t{%1, %0|%0, %1}";
	  return "%vmovq\t{%1, %0|%0, %1}";

	default:
	  gcc_unreachable ();
	}

    default:
      gcc_unreachable ();
    }
}
  [(set (attr "isa")
	(cond [(eq_attr "alternative" "3,4,5,6,7,22,23")
		 (const_string "nox64")
	       (eq_attr "alternative" "8,9,10,11,20,21,24,25")
		 (const_string "x64")
	       (eq_attr "alternative" "12,13,14,15")
		 (const_string "sse2")
	      ]
	      (const_string "*")))
   (set (attr "type")
	(cond [(eq_attr "alternative" "0,1,2")
		 (const_string "fmov")
	       (eq_attr "alternative" "3,4,5,6,7,22,23")
		 (const_string "multi")
	       (eq_attr "alternative" "8,9,10,11,24,25")
		 (const_string "imov")
	       (eq_attr "alternative" "12,16")
		 (const_string "sselog1")
	      ]
	      (const_string "ssemov")))
   (set (attr "modrm")
     (if_then_else (eq_attr "alternative" "11")
       (const_string "0")
       (const_string "*")))
   (set (attr "length_immediate")
     (if_then_else (eq_attr "alternative" "11")
       (const_string "8")
       (const_string "*")))
   (set (attr "prefix")
     (if_then_else (eq_attr "type" "sselog1,ssemov")
       (const_string "maybe_vex")
       (const_string "orig")))
   (set (attr "prefix_data16")
     (if_then_else
       (ior (and (eq_attr "type" "ssemov") (eq_attr "mode" "DI"))
	    (eq_attr "mode" "V1DF"))
       (const_string "1")
       (const_string "*")))
   (set (attr "mode")
	(cond [(eq_attr "alternative" "3,4,5,6,7,10,22,23")
		 (const_string "SI")
	       (eq_attr "alternative" "8,9,11,20,21,24,25")
		 (const_string "DI")

	       /* xorps is one byte shorter for non-AVX targets.  */
	       (eq_attr "alternative" "12,16")
		 (cond [(not (match_test "TARGET_SSE2"))
		 	  (const_string "V4SF")
			(and (match_test "TARGET_AVX512F")
			  (not (match_test "TARGET_PREFER_AVX256")))
			  (const_string "XI")
			(match_test "TARGET_AVX")
			  (const_string "V2DF")
			(match_test "optimize_function_for_size_p (cfun)")
			  (const_string "V4SF")
			(match_test "TARGET_SSE_LOAD0_BY_PXOR")
			  (const_string "TI")
		       ]
		       (const_string "V2DF"))

	       /* For architectures resolving dependencies on
		  whole SSE registers use movapd to break dependency
		  chains, otherwise use short move to avoid extra work.  */

	       /* movaps is one byte shorter for non-AVX targets.  */
	       (eq_attr "alternative" "13,17")
		 (cond [(and (ior (not (match_test "TARGET_PREFER_AVX256"))
				  (not (match_test "TARGET_AVX512VL")))
			     (ior (match_operand 0 "ext_sse_reg_operand")
				  (match_operand 1 "ext_sse_reg_operand")))
			  (const_string "V8DF")
			(ior (not (match_test "TARGET_SSE2"))
			     (match_test "TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL"))
			  (const_string "V4SF")
			(match_test "TARGET_SSE_PARTIAL_REG_DEPENDENCY")
			  (const_string "V2DF")
			(match_test "TARGET_AVX")
			  (const_string "DF")
			(match_test "optimize_function_for_size_p (cfun)")
			  (const_string "V4SF")
		       ]
		       (const_string "DF"))

	       /* For architectures resolving dependencies on register
		  parts we may avoid extra work to zero out upper part
		  of register.  */
	       (eq_attr "alternative" "14,18")
		 (cond [(not (match_test "TARGET_SSE2"))
			  (const_string "V2SF")
			(match_test "TARGET_AVX")
			  (const_string "DF")
			(match_test "TARGET_SSE_SPLIT_REGS")
			  (const_string "V1DF")
		       ]
		       (const_string "DF"))

	       (and (eq_attr "alternative" "15,19")
		    (not (match_test "TARGET_SSE2")))
		 (const_string "V2SF")
	      ]
	      (const_string "DF")))
   (set (attr "preferred_for_size")
     (cond [(eq_attr "alternative" "3,4")
              (symbol_ref "false")]
           (symbol_ref "true")))
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "3,4")
              (symbol_ref "TARGET_INTEGER_DFMODE_MOVES")]
           (symbol_ref "true")))
   (set (attr "enabled")
     (cond [(eq_attr "alternative" "22,23,24,25")
              (if_then_else
		(match_test "TARGET_HARD_DF_REGS")
		(symbol_ref "false")
		(const_string "*"))
            (not (match_test "TARGET_HARD_DF_REGS"))
	      (symbol_ref "false")
	   ]
	   (const_string "*")))])

(define_split
  [(set (match_operand:DF 0 "nonimmediate_gr_operand")
        (match_operand:DF 1 "general_gr_operand"))]
  "!TARGET_64BIT && reload_completed"
  [(const_int 0)]
  "ix86_split_long_move (operands); DONE;")

(define_insn "*movsf_internal"
  [(set (match_operand:SF 0 "nonimmediate_operand"
	  "=Yf*f,m   ,Yf*f,?r ,?m,v,v,v,m,?r,?Yi,!*y,!*y,!m,!r ,!*Ym,r  ,m")
	(match_operand:SF 1 "general_operand"
	  "Yf*fm,Yf*f,G   ,rmF,rF,C,v,m,v,Yj,r  ,*y ,m  ,*y,*Yn,r   ,rmF,rF"))]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))
   && (lra_in_progress || reload_completed
       || !CONST_DOUBLE_P (operands[1])
       || ((optimize_function_for_size_p (cfun)
	    || (ix86_cmodel == CM_LARGE || ix86_cmodel == CM_LARGE_PIC))
	   && ((IS_STACK_MODE (SFmode)
		&& standard_80387_constant_p (operands[1]) > 0)
	       || (TARGET_SSE && TARGET_SSE_MATH
		   && standard_sse_constant_p (operands[1], SFmode) == 1)))
       || memory_operand (operands[0], SFmode)
       || !TARGET_HARD_SF_REGS)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_FMOV:
      if (which_alternative == 2)
        return standard_80387_constant_opcode (operands[1]);
      return output_387_reg_move (insn, operands);

    case TYPE_IMOV:
      return "mov{l}\t{%1, %0|%0, %1}";

    case TYPE_SSELOG1:
      return standard_sse_constant_opcode (insn, operands[1]);

    case TYPE_SSEMOV:
      switch (get_attr_mode (insn))
	{
	case MODE_SF:
	  if (TARGET_AVX && REG_P (operands[0]) && REG_P (operands[1]))
	    return "vmovss\t{%1, %0, %0|%0, %0, %1}";
	  return "%vmovss\t{%1, %0|%0, %1}";

	case MODE_V16SF:
	  return "vmovaps\t{%g1, %g0|%g0, %g1}";
	case MODE_V4SF:
	  return "%vmovaps\t{%1, %0|%0, %1}";

	case MODE_SI:
	  return "%vmovd\t{%1, %0|%0, %1}";

	default:
	  gcc_unreachable ();
	}

    case TYPE_MMXMOV:
      switch (get_attr_mode (insn))
	{
	case MODE_DI:
	  return "movq\t{%1, %0|%0, %1}";
	case MODE_SI:
	  return "movd\t{%1, %0|%0, %1}";

	default:
	  gcc_unreachable ();
	}

    default:
      gcc_unreachable ();
    }
}
  [(set (attr "type")
	(cond [(eq_attr "alternative" "0,1,2")
		 (const_string "fmov")
	       (eq_attr "alternative" "3,4,16,17")
		 (const_string "imov")
	       (eq_attr "alternative" "5")
		 (const_string "sselog1")
	       (eq_attr "alternative" "11,12,13,14,15")
		 (const_string "mmxmov")
	      ]
	      (const_string "ssemov")))
   (set (attr "prefix")
     (if_then_else (eq_attr "type" "sselog1,ssemov")
       (const_string "maybe_vex")
       (const_string "orig")))
   (set (attr "prefix_data16")
     (if_then_else (and (eq_attr "type" "ssemov") (eq_attr "mode" "SI"))
       (const_string "1")
       (const_string "*")))
   (set (attr "mode")
        (cond [(eq_attr "alternative" "3,4,9,10,12,13,14,15,16,17")
		 (const_string "SI")
	       (eq_attr "alternative" "11")
		 (const_string "DI")
	       (eq_attr "alternative" "5")
		 (cond [(not (match_test "TARGET_SSE2"))
 		 	  (const_string "V4SF")
			(and (match_test "TARGET_AVX512F")
			  (not (match_test "TARGET_PREFER_AVX256")))
			  (const_string "V16SF")
			(match_test "TARGET_AVX")
			  (const_string "V4SF")
 			(match_test "optimize_function_for_size_p (cfun)")
			  (const_string "V4SF")
 			(match_test "TARGET_SSE_LOAD0_BY_PXOR")
			  (const_string "TI")
		       ]
		       (const_string "V4SF"))

	       /* For architectures resolving dependencies on
		  whole SSE registers use APS move to break dependency
		  chains, otherwise use short move to avoid extra work.

		  Do the same for architectures resolving dependencies on
		  the parts.  While in DF mode it is better to always handle
		  just register parts, the SF mode is different due to lack
		  of instructions to load just part of the register.  It is
		  better to maintain the whole registers in single format
		  to avoid problems on using packed logical operations.  */
	       (eq_attr "alternative" "6")
		 (cond [(and (ior (not (match_test "TARGET_PREFER_AVX256"))
				  (not (match_test "TARGET_AVX512VL")))
			     (ior (match_operand 0 "ext_sse_reg_operand")
				  (match_operand 1 "ext_sse_reg_operand")))
			  (const_string "V16SF")
			(ior (match_test "TARGET_SSE_PARTIAL_REG_DEPENDENCY")
			     (match_test "TARGET_SSE_SPLIT_REGS"))
			  (const_string "V4SF")
		       ]
		       (const_string "SF"))
	      ]
	      (const_string "SF")))
   (set (attr "enabled")
     (cond [(eq_attr "alternative" "16,17")
              (if_then_else
		(match_test "TARGET_HARD_SF_REGS")
		(symbol_ref "false")
		(const_string "*"))
            (not (match_test "TARGET_HARD_SF_REGS"))
	      (symbol_ref "false")
	   ]
	   (const_string "*")))])

(define_split
  [(set (match_operand 0 "any_fp_register_operand")
	(match_operand 1 "nonimmediate_operand"))]
  "reload_completed
   && (GET_MODE (operands[0]) == TFmode
       || GET_MODE (operands[0]) == XFmode
       || GET_MODE (operands[0]) == DFmode
       || GET_MODE (operands[0]) == SFmode)
   && ix86_standard_x87sse_constant_load_p (insn, operands[0])"
  [(set (match_dup 0) (match_dup 2))]
  "operands[2] = find_constant_src (curr_insn);")

(define_split
  [(set (match_operand 0 "any_fp_register_operand")
	(float_extend (match_operand 1 "nonimmediate_operand")))]
  "reload_completed
   && (GET_MODE (operands[0]) == TFmode
       || GET_MODE (operands[0]) == XFmode
       || GET_MODE (operands[0]) == DFmode)
   && ix86_standard_x87sse_constant_load_p (insn, operands[0])"
  [(set (match_dup 0) (match_dup 2))]
  "operands[2] = find_constant_src (curr_insn);")

;; Split the load of -0.0 or -1.0 into fldz;fchs or fld1;fchs sequence
(define_split
  [(set (match_operand:X87MODEF 0 "fp_register_operand")
	(match_operand:X87MODEF 1 "immediate_operand"))]
  "reload_completed
   && (standard_80387_constant_p (operands[1]) == 8
       || standard_80387_constant_p (operands[1]) == 9)"
  [(set (match_dup 0)(match_dup 1))
   (set (match_dup 0)
	(neg:X87MODEF (match_dup 0)))]
{
  if (real_isnegzero (CONST_DOUBLE_REAL_VALUE (operands[1])))
    operands[1] = CONST0_RTX (<MODE>mode);
  else
    operands[1] = CONST1_RTX (<MODE>mode);
})

(define_insn "swapxf"
  [(set (match_operand:XF 0 "register_operand" "+f")
	(match_operand:XF 1 "register_operand" "+f"))
   (set (match_dup 1)
	(match_dup 0))]
  "TARGET_80387"
{
  if (STACK_TOP_P (operands[0]))
    return "fxch\t%1";
  else
    return "fxch\t%0";
}
  [(set_attr "type" "fxch")
   (set_attr "mode" "XF")])

(define_insn "*swap<mode>"
  [(set (match_operand:MODEF 0 "fp_register_operand" "+f")
	(match_operand:MODEF 1 "fp_register_operand" "+f"))
   (set (match_dup 1)
	(match_dup 0))]
  "TARGET_80387 || reload_completed"
{
  if (STACK_TOP_P (operands[0]))
    return "fxch\t%1";
  else
    return "fxch\t%0";
}
  [(set_attr "type" "fxch")
   (set_attr "mode" "<MODE>")])

;; Zero extension instructions

(define_expand "zero_extendsidi2"
  [(set (match_operand:DI 0 "nonimmediate_operand")
	(zero_extend:DI (match_operand:SI 1 "nonimmediate_operand")))])

(define_insn "*zero_extendsidi2"
  [(set (match_operand:DI 0 "nonimmediate_operand"
		"=r,?r,?o,r   ,o,?*Ym,?!*y,?r ,?*Yi,*x,*x,*v,*r")
	(zero_extend:DI
	 (match_operand:SI 1 "x86_64_zext_operand"
	        "0 ,rm,r ,rmWz,0,r   ,m   ,*Yj,r   ,m ,*x,*v,*k")))]
  ""
{
  switch (get_attr_type (insn))
    {
    case TYPE_IMOVX:
      if (ix86_use_lea_for_mov (insn, operands))
	return "lea{l}\t{%E1, %k0|%k0, %E1}";
      else
	return "mov{l}\t{%1, %k0|%k0, %1}";

    case TYPE_MULTI:
      return "#";

    case TYPE_MMXMOV:
      return "movd\t{%1, %0|%0, %1}";

    case TYPE_SSEMOV:
      if (SSE_REG_P (operands[0]) && SSE_REG_P (operands[1]))
	{
	  if (EXT_REX_SSE_REG_P (operands[0])
	      || EXT_REX_SSE_REG_P (operands[1]))
	    return "vpmovzxdq\t{%t1, %g0|%g0, %t1}";
	  else
	    return "%vpmovzxdq\t{%1, %0|%0, %1}";
	}

      if (GENERAL_REG_P (operands[0]))
	return "%vmovd\t{%1, %k0|%k0, %1}";

      return "%vmovd\t{%1, %0|%0, %1}";

    case TYPE_MSKMOV:
      return "kmovd\t{%1, %k0|%k0, %1}";

    default:
      gcc_unreachable ();
    }
}
  [(set (attr "isa")
     (cond [(eq_attr "alternative" "0,1,2")
	      (const_string "nox64")
	    (eq_attr "alternative" "3")
	      (const_string "x64")
	    (eq_attr "alternative" "9")
	      (const_string "sse2")
	    (eq_attr "alternative" "10")
	      (const_string "sse4")
	    (eq_attr "alternative" "11")
	      (const_string "avx512f")
	    (eq_attr "alternative" "12")
	      (const_string "x64_avx512bw")
	   ]
	   (const_string "*")))
   (set (attr "type")
     (cond [(eq_attr "alternative" "0,1,2,4")
	      (const_string "multi")
	    (eq_attr "alternative" "5,6")
	      (const_string "mmxmov")
	    (eq_attr "alternative" "7")
	      (if_then_else (match_test "TARGET_64BIT")
		(const_string "ssemov")
		(const_string "multi"))
	    (eq_attr "alternative" "8,9,10,11")
	      (const_string "ssemov")
	    (eq_attr "alternative" "12")
	      (const_string "mskmov")
	   ]
	   (const_string "imovx")))
   (set (attr "prefix_extra")
     (if_then_else (eq_attr "alternative" "10,11")
       (const_string "1")
       (const_string "*")))
   (set (attr "prefix")
     (if_then_else (eq_attr "type" "ssemov")
       (const_string "maybe_vex")
       (const_string "orig")))
   (set (attr "prefix_0f")
     (if_then_else (eq_attr "type" "imovx")
       (const_string "0")
       (const_string "*")))
   (set (attr "mode")
     (cond [(eq_attr "alternative" "5,6")
	      (const_string "DI")
	    (and (eq_attr "alternative" "7")
		 (match_test "TARGET_64BIT"))
	      (const_string "TI")
	    (eq_attr "alternative" "8,10,11")
	      (const_string "TI")
	   ]
	   (const_string "SI")))])

(define_split
  [(set (match_operand:DI 0 "memory_operand")
     	(zero_extend:DI (match_operand:SI 1 "memory_operand")))]
  "reload_completed"
  [(set (match_dup 4) (const_int 0))]
  "split_double_mode (DImode, &operands[0], 1, &operands[3], &operands[4]);")

(define_split
  [(set (match_operand:DI 0 "general_reg_operand")
	(zero_extend:DI (match_operand:SI 1 "general_reg_operand")))]
  "!TARGET_64BIT && reload_completed
   && REGNO (operands[0]) == REGNO (operands[1])"
  [(set (match_dup 4) (const_int 0))]
  "split_double_mode (DImode, &operands[0], 1, &operands[3], &operands[4]);")

(define_split
  [(set (match_operand:DI 0 "nonimmediate_gr_operand")
	(zero_extend:DI (match_operand:SI 1 "nonimmediate_operand")))]
  "!TARGET_64BIT && reload_completed
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  [(set (match_dup 3) (match_dup 1))
   (set (match_dup 4) (const_int 0))]
  "split_double_mode (DImode, &operands[0], 1, &operands[3], &operands[4]);")

(define_peephole2
  [(set (match_operand:DI 0 "general_reg_operand")
	(zero_extend:DI (match_operand:SI 1 "nonimmediate_gr_operand")))
   (set (match_operand:DI 2 "sse_reg_operand") (match_dup 0))]
  "TARGET_64BIT && TARGET_SSE2 && TARGET_INTER_UNIT_MOVES_TO_VEC
   && peep2_reg_dead_p (2, operands[0])"
  [(set (match_dup 2)
	(zero_extend:DI (match_dup 1)))])

(define_mode_attr kmov_isa
  [(QI "avx512dq") (HI "avx512f") (SI "avx512bw") (DI "avx512bw")])

(define_insn "zero_extend<mode>di2"
  [(set (match_operand:DI 0 "register_operand" "=r,*r")
	(zero_extend:DI
	 (match_operand:SWI12 1 "nonimmediate_operand" "<r>m,*k")))]
  "TARGET_64BIT"
  "@
   movz{<imodesuffix>l|x}\t{%1, %k0|%k0, %1}
   kmov<mskmodesuffix>\t{%1, %k0|%k0, %1}"
  [(set_attr "isa" "*,<kmov_isa>")
   (set_attr "type" "imovx,mskmov")
   (set_attr "mode" "SI,<MODE>")])

(define_expand "zero_extend<mode>si2"
  [(set (match_operand:SI 0 "register_operand")
	(zero_extend:SI (match_operand:SWI12 1 "nonimmediate_operand")))]
  ""
{
  if (TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun))
    {
      operands[1] = force_reg (<MODE>mode, operands[1]);
      emit_insn (gen_zero_extend<mode>si2_and (operands[0], operands[1]));
      DONE;
    }
})

(define_insn_and_split "zero_extend<mode>si2_and"
  [(set (match_operand:SI 0 "register_operand" "=r,?&<r>")
	(zero_extend:SI
	  (match_operand:SWI12 1 "nonimmediate_operand" "0,<r>m")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun)"
  "#"
  "&& reload_completed"
  [(parallel [(set (match_dup 0) (and:SI (match_dup 0) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  if (!REG_P (operands[1])
      || REGNO (operands[0]) != REGNO (operands[1]))
    {
      ix86_expand_clear (operands[0]);

      gcc_assert (!TARGET_PARTIAL_REG_STALL);
      emit_insn (gen_movstrict<mode>
		  (gen_lowpart (<MODE>mode, operands[0]), operands[1]));
      DONE;
    }

  operands[2] = GEN_INT (GET_MODE_MASK (<MODE>mode));
}
  [(set_attr "type" "alu1")
   (set_attr "mode" "SI")])

(define_insn "*zero_extend<mode>si2"
  [(set (match_operand:SI 0 "register_operand" "=r,*r")
	(zero_extend:SI
	  (match_operand:SWI12 1 "nonimmediate_operand" "<r>m,*k")))]
  "!(TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun))"
  "@
   movz{<imodesuffix>l|x}\t{%1, %0|%0, %1}
   kmov<mskmodesuffix>\t{%1, %0|%0, %1}"
  [(set_attr "isa" "*,<kmov_isa>")
   (set_attr "type" "imovx,mskmov")
   (set_attr "mode" "SI,<MODE>")])

(define_expand "zero_extendqihi2"
  [(set (match_operand:HI 0 "register_operand")
	(zero_extend:HI (match_operand:QI 1 "nonimmediate_operand")))]
  ""
{
  if (TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun))
    {
      operands[1] = force_reg (QImode, operands[1]);
      emit_insn (gen_zero_extendqihi2_and (operands[0], operands[1]));
      DONE;
    }
})

(define_insn_and_split "zero_extendqihi2_and"
  [(set (match_operand:HI 0 "register_operand" "=r,?&q")
	(zero_extend:HI (match_operand:QI 1 "nonimmediate_operand" "0,qm")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun)"
  "#"
  "&& reload_completed"
  [(parallel [(set (match_dup 0) (and:SI (match_dup 0) (const_int 255)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  if (!REG_P (operands[1])
      || REGNO (operands[0]) != REGNO (operands[1]))
    {
      ix86_expand_clear (operands[0]);

      gcc_assert (!TARGET_PARTIAL_REG_STALL);
      emit_insn (gen_movstrictqi
		  (gen_lowpart (QImode, operands[0]), operands[1]));
      DONE;
    }

  operands[0] = gen_lowpart (SImode, operands[0]);
}
  [(set_attr "type" "alu1")
   (set_attr "mode" "SI")])

; zero extend to SImode to avoid partial register stalls
(define_insn "*zero_extendqihi2"
  [(set (match_operand:HI 0 "register_operand" "=r,*r")
	(zero_extend:HI (match_operand:QI 1 "nonimmediate_operand" "qm,*k")))]
  "!(TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun))"
  "@
   movz{bl|x}\t{%1, %k0|%k0, %1}
   kmovb\t{%1, %k0|%k0, %1}"
  [(set_attr "isa" "*,avx512dq")
   (set_attr "type" "imovx,mskmov")
   (set_attr "mode" "SI,QI")])

(define_insn_and_split "*zext<mode>_doubleword_and"
  [(set (match_operand:DI 0 "register_operand" "=&<r>")
	(zero_extend:DI (match_operand:SWI12 1 "nonimmediate_operand" "<r>m")))]
  "!TARGET_64BIT && TARGET_STV && TARGET_SSE2
   && TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun)"
  "#"
  "&& reload_completed && GENERAL_REG_P (operands[0])"
  [(set (match_dup 2) (const_int 0))]
{
  split_double_mode (DImode, &operands[0], 1, &operands[0], &operands[2]);

  emit_move_insn (operands[0], const0_rtx);

  gcc_assert (!TARGET_PARTIAL_REG_STALL);
  emit_insn (gen_movstrict<mode>
	     (gen_lowpart (<MODE>mode, operands[0]), operands[1]));
})

(define_insn_and_split "*zext<mode>_doubleword"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (match_operand:SWI12 1 "nonimmediate_operand" "<r>m")))]
  "!TARGET_64BIT && TARGET_STV && TARGET_SSE2
   && !(TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun))"
  "#"
  "&& reload_completed && GENERAL_REG_P (operands[0])"
  [(set (match_dup 0) (zero_extend:SI (match_dup 1)))
   (set (match_dup 2) (const_int 0))]
  "split_double_mode (DImode, &operands[0], 1, &operands[0], &operands[2]);")

(define_insn_and_split "*zextsi_doubleword"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (match_operand:SI 1 "nonimmediate_operand" "rm")))]
  "!TARGET_64BIT && TARGET_STV && TARGET_SSE2"
  "#"
  "&& reload_completed && GENERAL_REG_P (operands[0])"
  [(set (match_dup 0) (match_dup 1))
   (set (match_dup 2) (const_int 0))]
  "split_double_mode (DImode, &operands[0], 1, &operands[0], &operands[2]);")

;; Sign extension instructions

(define_expand "extendsidi2"
  [(set (match_operand:DI 0 "register_operand")
	(sign_extend:DI (match_operand:SI 1 "register_operand")))]
  ""
{
  if (!TARGET_64BIT)
    {
      emit_insn (gen_extendsidi2_1 (operands[0], operands[1]));
      DONE;
    }
})

(define_insn "*extendsidi2_rex64"
  [(set (match_operand:DI 0 "register_operand" "=*a,r")
	(sign_extend:DI (match_operand:SI 1 "nonimmediate_operand" "*0,rm")))]
  "TARGET_64BIT"
  "@
   {cltq|cdqe}
   movs{lq|x}\t{%1, %0|%0, %1}"
  [(set_attr "type" "imovx")
   (set_attr "mode" "DI")
   (set_attr "prefix_0f" "0")
   (set_attr "modrm" "0,1")])

(define_insn "extendsidi2_1"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=*A,r,?r,?*o")
	(sign_extend:DI (match_operand:SI 1 "register_operand" "0,0,r,r")))
   (clobber (reg:CC FLAGS_REG))
   (clobber (match_scratch:SI 2 "=X,X,X,&r"))]
  "!TARGET_64BIT"
  "#")

;; Split the memory case.  If the source register doesn't die, it will stay
;; this way, if it does die, following peephole2s take care of it.
(define_split
  [(set (match_operand:DI 0 "memory_operand")
	(sign_extend:DI (match_operand:SI 1 "register_operand")))
   (clobber (reg:CC FLAGS_REG))
   (clobber (match_operand:SI 2 "register_operand"))]
  "reload_completed"
  [(const_int 0)]
{
  split_double_mode (DImode, &operands[0], 1, &operands[3], &operands[4]);

  emit_move_insn (operands[3], operands[1]);

  /* Generate a cltd if possible and doing so it profitable.  */
  if ((optimize_function_for_size_p (cfun) || TARGET_USE_CLTD)
      && REGNO (operands[1]) == AX_REG
      && REGNO (operands[2]) == DX_REG)
    {
      emit_insn (gen_ashrsi3_cvt (operands[2], operands[1], GEN_INT (31)));
    }
  else
    {
      emit_move_insn (operands[2], operands[1]);
      emit_insn (gen_ashrsi3_cvt (operands[2], operands[2], GEN_INT (31)));
    }
  emit_move_insn (operands[4], operands[2]);
  DONE;
})

;; Peepholes for the case where the source register does die, after
;; being split with the above splitter.
(define_peephole2
  [(set (match_operand:SI 0 "memory_operand")
	(match_operand:SI 1 "general_reg_operand"))
   (set (match_operand:SI 2 "general_reg_operand") (match_dup 1))
   (parallel [(set (match_dup 2)
		   (ashiftrt:SI (match_dup 2) (const_int 31)))
	       (clobber (reg:CC FLAGS_REG))])
   (set (match_operand:SI 3 "memory_operand") (match_dup 2))]
  "REGNO (operands[1]) != REGNO (operands[2])
   && peep2_reg_dead_p (2, operands[1])
   && peep2_reg_dead_p (4, operands[2])
   && !reg_mentioned_p (operands[2], operands[3])"
  [(set (match_dup 0) (match_dup 1))
   (parallel [(set (match_dup 1) (ashiftrt:SI (match_dup 1) (const_int 31)))
	      (clobber (reg:CC FLAGS_REG))])
   (set (match_dup 3) (match_dup 1))])

(define_peephole2
  [(set (match_operand:SI 0 "memory_operand")
	(match_operand:SI 1 "general_reg_operand"))
   (parallel [(set (match_operand:SI 2 "general_reg_operand")
		   (ashiftrt:SI (match_dup 1) (const_int 31)))
	       (clobber (reg:CC FLAGS_REG))])
   (set (match_operand:SI 3 "memory_operand") (match_dup 2))]
  "/* cltd is shorter than sarl $31, %eax */
   !optimize_function_for_size_p (cfun)
   && REGNO (operands[1]) == AX_REG
   && REGNO (operands[2]) == DX_REG
   && peep2_reg_dead_p (2, operands[1])
   && peep2_reg_dead_p (3, operands[2])
   && !reg_mentioned_p (operands[2], operands[3])"
  [(set (match_dup 0) (match_dup 1))
   (parallel [(set (match_dup 1) (ashiftrt:SI (match_dup 1) (const_int 31)))
	      (clobber (reg:CC FLAGS_REG))])
   (set (match_dup 3) (match_dup 1))])

;; Extend to register case.  Optimize case where source and destination
;; registers match and cases where we can use cltd.
(define_split
  [(set (match_operand:DI 0 "register_operand")
	(sign_extend:DI (match_operand:SI 1 "register_operand")))
   (clobber (reg:CC FLAGS_REG))
   (clobber (match_scratch:SI 2))]
  "reload_completed"
  [(const_int 0)]
{
  split_double_mode (DImode, &operands[0], 1, &operands[3], &operands[4]);

  if (REGNO (operands[3]) != REGNO (operands[1]))
    emit_move_insn (operands[3], operands[1]);

  /* Generate a cltd if possible and doing so it profitable.  */
  if ((optimize_function_for_size_p (cfun) || TARGET_USE_CLTD)
      && REGNO (operands[3]) == AX_REG
      && REGNO (operands[4]) == DX_REG)
    {
      emit_insn (gen_ashrsi3_cvt (operands[4], operands[3], GEN_INT (31)));
      DONE;
    }

  if (REGNO (operands[4]) != REGNO (operands[1]))
    emit_move_insn (operands[4], operands[1]);

  emit_insn (gen_ashrsi3_cvt (operands[4], operands[4], GEN_INT (31)));
  DONE;
})

(define_insn "extend<mode>di2"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(sign_extend:DI
	 (match_operand:SWI12 1 "nonimmediate_operand" "<r>m")))]
  "TARGET_64BIT"
  "movs{<imodesuffix>q|x}\t{%1, %0|%0, %1}"
  [(set_attr "type" "imovx")
   (set_attr "mode" "DI")])

(define_insn "extendhisi2"
  [(set (match_operand:SI 0 "register_operand" "=*a,r")
	(sign_extend:SI (match_operand:HI 1 "nonimmediate_operand" "*0,rm")))]
  ""
{
  switch (get_attr_prefix_0f (insn))
    {
    case 0:
      return "{cwtl|cwde}";
    default:
      return "movs{wl|x}\t{%1, %0|%0, %1}";
    }
}
  [(set_attr "type" "imovx")
   (set_attr "mode" "SI")
   (set (attr "prefix_0f")
     ;; movsx is short decodable while cwtl is vector decoded.
     (if_then_else (and (eq_attr "cpu" "!k6")
			(eq_attr "alternative" "0"))
	(const_string "0")
	(const_string "1")))
   (set (attr "znver1_decode")
     (if_then_else (eq_attr "prefix_0f" "0")
	(const_string "double")
	(const_string "direct")))
   (set (attr "modrm")
     (if_then_else (eq_attr "prefix_0f" "0")
	(const_string "0")
	(const_string "1")))])

(define_insn "*extendhisi2_zext"
  [(set (match_operand:DI 0 "register_operand" "=*a,r")
	(zero_extend:DI
	 (sign_extend:SI
	  (match_operand:HI 1 "nonimmediate_operand" "*0,rm"))))]
  "TARGET_64BIT"
{
  switch (get_attr_prefix_0f (insn))
    {
    case 0:
      return "{cwtl|cwde}";
    default:
      return "movs{wl|x}\t{%1, %k0|%k0, %1}";
    }
}
  [(set_attr "type" "imovx")
   (set_attr "mode" "SI")
   (set (attr "prefix_0f")
     ;; movsx is short decodable while cwtl is vector decoded.
     (if_then_else (and (eq_attr "cpu" "!k6")
			(eq_attr "alternative" "0"))
	(const_string "0")
	(const_string "1")))
   (set (attr "modrm")
     (if_then_else (eq_attr "prefix_0f" "0")
	(const_string "0")
	(const_string "1")))])

(define_insn "extendqisi2"
  [(set (match_operand:SI 0 "register_operand" "=r")
	(sign_extend:SI (match_operand:QI 1 "nonimmediate_operand" "qm")))]
  ""
  "movs{bl|x}\t{%1, %0|%0, %1}"
   [(set_attr "type" "imovx")
    (set_attr "mode" "SI")])

(define_insn "*extendqisi2_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (sign_extend:SI (match_operand:QI 1 "nonimmediate_operand" "qm"))))]
  "TARGET_64BIT"
  "movs{bl|x}\t{%1, %k0|%k0, %1}"
   [(set_attr "type" "imovx")
    (set_attr "mode" "SI")])

(define_insn "extendqihi2"
  [(set (match_operand:HI 0 "register_operand" "=*a,r")
	(sign_extend:HI (match_operand:QI 1 "nonimmediate_operand" "*0,qm")))]
  ""
{
  switch (get_attr_prefix_0f (insn))
    {
    case 0:
      return "{cbtw|cbw}";
    default:
      return "movs{bw|x}\t{%1, %0|%0, %1}";
    }
}
  [(set_attr "type" "imovx")
   (set_attr "mode" "HI")
   (set (attr "prefix_0f")
     ;; movsx is short decodable while cwtl is vector decoded.
     (if_then_else (and (eq_attr "cpu" "!k6")
			(eq_attr "alternative" "0"))
	(const_string "0")
	(const_string "1")))
   (set (attr "modrm")
     (if_then_else (eq_attr "prefix_0f" "0")
	(const_string "0")
	(const_string "1")))])

;; Conversions between float and double.

;; These are all no-ops in the model used for the 80387.
;; So just emit moves.

;; %%% Kill these when call knows how to work out a DFmode push earlier.
(define_split
  [(set (match_operand:DF 0 "push_operand")
	(float_extend:DF (match_operand:SF 1 "fp_register_operand")))]
  "reload_completed"
  [(set (reg:P SP_REG) (plus:P (reg:P SP_REG) (const_int -8)))
   (set (mem:DF (reg:P SP_REG)) (float_extend:DF (match_dup 1)))])

(define_split
  [(set (match_operand:XF 0 "push_operand")
	(float_extend:XF (match_operand:MODEF 1 "fp_register_operand")))]
  "reload_completed"
  [(set (reg:P SP_REG) (plus:P (reg:P SP_REG) (match_dup 2)))
   (set (mem:XF (reg:P SP_REG)) (float_extend:XF (match_dup 1)))]
  "operands[2] = GEN_INT (-GET_MODE_SIZE (XFmode));")

(define_expand "extendsfdf2"
  [(set (match_operand:DF 0 "nonimm_ssenomem_operand")
        (float_extend:DF (match_operand:SF 1 "general_operand")))]
  "TARGET_80387 || (TARGET_SSE2 && TARGET_SSE_MATH)"
{
  /* ??? Needed for compress_float_constant since all fp constants
     are TARGET_LEGITIMATE_CONSTANT_P.  */
  if (CONST_DOUBLE_P (operands[1]))
    {
      if ((!TARGET_SSE2 || TARGET_MIX_SSE_I387)
	  && standard_80387_constant_p (operands[1]) > 0)
	{
	  operands[1] = simplify_const_unary_operation
	    (FLOAT_EXTEND, DFmode, operands[1], SFmode);
	  emit_move_insn_1 (operands[0], operands[1]);
	  DONE;
	}
      operands[1] = validize_mem (force_const_mem (SFmode, operands[1]));
    }
})

/* For converting SF(xmm2) to DF(xmm1), use the following code instead of
   cvtss2sd:
      unpcklps xmm2,xmm2   ; packed conversion might crash on signaling NaNs
      cvtps2pd xmm2,xmm1
   We do the conversion post reload to avoid producing of 128bit spills
   that might lead to ICE on 32bit target.  The sequence unlikely combine
   anyway.  */
(define_split
  [(set (match_operand:DF 0 "sse_reg_operand")
        (float_extend:DF
	  (match_operand:SF 1 "nonimmediate_operand")))]
  "TARGET_USE_VECTOR_FP_CONVERTS
   && optimize_insn_for_speed_p ()
   && reload_completed
   && (!EXT_REX_SSE_REG_P (operands[0])
       || TARGET_AVX512VL)"
   [(set (match_dup 2)
	 (float_extend:V2DF
	   (vec_select:V2SF
	     (match_dup 3)
	     (parallel [(const_int 0) (const_int 1)]))))]
{
  operands[2] = lowpart_subreg (V2DFmode, operands[0], DFmode);
  operands[3] = lowpart_subreg (V4SFmode, operands[0], DFmode);
  /* Use movss for loading from memory, unpcklps reg, reg for registers.
     Try to avoid move when unpacking can be done in source.  */
  if (REG_P (operands[1]))
    {
      /* If it is unsafe to overwrite upper half of source, we need
	 to move to destination and unpack there.  */
      if (REGNO (operands[0]) != REGNO (operands[1])
	  || (EXT_REX_SSE_REG_P (operands[1])
	      && !TARGET_AVX512VL))
	{
	  rtx tmp = lowpart_subreg (SFmode, operands[0], DFmode);
	  emit_move_insn (tmp, operands[1]);
	}
      else
	operands[3] = lowpart_subreg (V4SFmode, operands[1], SFmode);
      /* FIXME: vec_interleave_lowv4sf for AVX512VL should allow
	 =v, v, then vbroadcastss will be only needed for AVX512F without
	 AVX512VL.  */
      if (!EXT_REX_SSE_REGNO_P (REGNO (operands[3])))
	emit_insn (gen_vec_interleave_lowv4sf (operands[3], operands[3],
					       operands[3]));
      else
	{
	  rtx tmp = lowpart_subreg (V16SFmode, operands[3], V4SFmode);
	  emit_insn (gen_avx512f_vec_dupv16sf_1 (tmp, tmp));
	}
    }
  else
    emit_insn (gen_vec_setv4sf_0 (operands[3],
				  CONST0_RTX (V4SFmode), operands[1]));
})

;; It's more profitable to split and then extend in the same register.
(define_peephole2
  [(set (match_operand:DF 0 "sse_reg_operand")
	(float_extend:DF
	  (match_operand:SF 1 "memory_operand")))]
  "TARGET_SPLIT_MEM_OPND_FOR_FP_CONVERTS
   && optimize_insn_for_speed_p ()"
  [(set (match_dup 2) (match_dup 1))
   (set (match_dup 0) (float_extend:DF (match_dup 2)))]
  "operands[2] = lowpart_subreg (SFmode, operands[0], DFmode);")

(define_insn "*extendsfdf2"
  [(set (match_operand:DF 0 "nonimm_ssenomem_operand" "=f,m,v")
        (float_extend:DF
	  (match_operand:SF 1 "nonimmediate_operand" "fm,f,vm")))]
  "TARGET_80387 || (TARGET_SSE2 && TARGET_SSE_MATH)"
{
  switch (which_alternative)
    {
    case 0:
    case 1:
      return output_387_reg_move (insn, operands);

    case 2:
      return "%vcvtss2sd\t{%1, %d0|%d0, %1}";

    default:
      gcc_unreachable ();
    }
}
  [(set_attr "type" "fmov,fmov,ssecvt")
   (set_attr "prefix" "orig,orig,maybe_vex")
   (set_attr "mode" "SF,XF,DF")
   (set (attr "enabled")
     (if_then_else
       (match_test ("TARGET_SSE2 && TARGET_SSE_MATH"))
       (if_then_else
	 (eq_attr "alternative" "0,1")
	 (symbol_ref "TARGET_MIX_SSE_I387")
	 (symbol_ref "true"))
       (if_then_else
	 (eq_attr "alternative" "0,1")
	 (symbol_ref "true")
	 (symbol_ref "false"))))])

(define_expand "extend<mode>xf2"
  [(set (match_operand:XF 0 "nonimmediate_operand")
        (float_extend:XF (match_operand:MODEF 1 "general_operand")))]
  "TARGET_80387"
{
  /* ??? Needed for compress_float_constant since all fp constants
     are TARGET_LEGITIMATE_CONSTANT_P.  */
  if (CONST_DOUBLE_P (operands[1]))
    {
      if (standard_80387_constant_p (operands[1]) > 0)
	{
	  operands[1] = simplify_const_unary_operation
	    (FLOAT_EXTEND, XFmode, operands[1], <MODE>mode);
	  emit_move_insn_1 (operands[0], operands[1]);
	  DONE;
	}
      operands[1] = validize_mem (force_const_mem (<MODE>mode, operands[1]));
    }
})

(define_insn "*extend<mode>xf2_i387"
  [(set (match_operand:XF 0 "nonimmediate_operand" "=f,m")
        (float_extend:XF
	  (match_operand:MODEF 1 "nonimmediate_operand" "fm,f")))]
  "TARGET_80387"
  "* return output_387_reg_move (insn, operands);"
  [(set_attr "type" "fmov")
   (set_attr "mode" "<MODE>,XF")])

;; %%% This seems like bad news.
;; This cannot output into an f-reg because there is no way to be sure
;; of truncating in that case.  Otherwise this is just like a simple move
;; insn.  So we pretend we can output to a reg in order to get better
;; register preferencing, but we really use a stack slot.

;; Conversion from DFmode to SFmode.

(define_expand "truncdfsf2"
  [(set (match_operand:SF 0 "nonimmediate_operand")
	(float_truncate:SF
	  (match_operand:DF 1 "nonimmediate_operand")))]
  "TARGET_80387 || (TARGET_SSE2 && TARGET_SSE_MATH)"
{
  if (TARGET_SSE2 && TARGET_SSE_MATH && !TARGET_MIX_SSE_I387)
    ;
  else if (flag_unsafe_math_optimizations)
    ;
  else
    {
      rtx temp = assign_386_stack_local (SFmode, SLOT_TEMP);
      emit_insn (gen_truncdfsf2_with_temp (operands[0], operands[1], temp));
      DONE;
    }
})

/* For converting DF(xmm2) to SF(xmm1), use the following code instead of
   cvtsd2ss:
      unpcklpd xmm2,xmm2   ; packed conversion might crash on signaling NaNs
      cvtpd2ps xmm2,xmm1
   We do the conversion post reload to avoid producing of 128bit spills
   that might lead to ICE on 32bit target.  The sequence unlikely combine
   anyway.  */
(define_split
  [(set (match_operand:SF 0 "sse_reg_operand")
        (float_truncate:SF
	  (match_operand:DF 1 "nonimmediate_operand")))]
  "TARGET_USE_VECTOR_FP_CONVERTS
   && optimize_insn_for_speed_p ()
   && reload_completed
   && (!EXT_REX_SSE_REG_P (operands[0])
       || TARGET_AVX512VL)"
   [(set (match_dup 2)
	 (vec_concat:V4SF
	   (float_truncate:V2SF
	     (match_dup 4))
	   (match_dup 3)))]
{
  operands[2] = lowpart_subreg (V4SFmode, operands[0], SFmode);
  operands[3] = CONST0_RTX (V2SFmode);
  operands[4] = lowpart_subreg (V2DFmode, operands[0], SFmode);
  /* Use movsd for loading from memory, unpcklpd for registers.
     Try to avoid move when unpacking can be done in source, or SSE3
     movddup is available.  */
  if (REG_P (operands[1]))
    {
      if (!TARGET_SSE3
	  && REGNO (operands[0]) != REGNO (operands[1]))
	{
	  rtx tmp = lowpart_subreg (DFmode, operands[0], SFmode);
	  emit_move_insn (tmp, operands[1]);
	  operands[1] = tmp;
	}
      else if (!TARGET_SSE3)
	operands[4] = lowpart_subreg (V2DFmode, operands[1], DFmode);
      emit_insn (gen_vec_dupv2df (operands[4], operands[1]));
    }
  else
    emit_insn (gen_vec_concatv2df (operands[4], operands[1],
				   CONST0_RTX (DFmode)));
})

;; It's more profitable to split and then extend in the same register.
(define_peephole2
  [(set (match_operand:SF 0 "sse_reg_operand")
	(float_truncate:SF
	  (match_operand:DF 1 "memory_operand")))]
  "TARGET_SPLIT_MEM_OPND_FOR_FP_CONVERTS
   && optimize_insn_for_speed_p ()"
  [(set (match_dup 2) (match_dup 1))
   (set (match_dup 0) (float_truncate:SF (match_dup 2)))]
  "operands[2] = lowpart_subreg (DFmode, operands[0], SFmode);")

(define_expand "truncdfsf2_with_temp"
  [(parallel [(set (match_operand:SF 0)
		   (float_truncate:SF (match_operand:DF 1)))
	      (clobber (match_operand:SF 2))])])

;; SSE alternative doesn't depend on flag_unsafe_math_optimizations,
;; because nothing we do there is unsafe.
(define_insn "*truncdfsf_fast_mixed"
  [(set (match_operand:SF 0 "nonimmediate_operand"   "=fm,v")
        (float_truncate:SF
          (match_operand:DF 1 "nonimmediate_operand" "f  ,vm")))]
  "TARGET_SSE2 && TARGET_SSE_MATH"
{
  switch (which_alternative)
    {
    case 0:
      return output_387_reg_move (insn, operands);
    case 1:
      return "%vcvtsd2ss\t{%1, %d0|%d0, %1}";
    default:
      gcc_unreachable ();
    }
}
  [(set_attr "type" "fmov,ssecvt")
   (set_attr "prefix" "orig,maybe_vex")
   (set_attr "mode" "SF")
   (set (attr "enabled")
     (cond [(eq_attr "alternative" "0")
              (symbol_ref "TARGET_MIX_SSE_I387
			   && flag_unsafe_math_optimizations")
	   ]
           (symbol_ref "true")))])

(define_insn "*truncdfsf_fast_i387"
  [(set (match_operand:SF 0 "nonimmediate_operand"   "=fm")
        (float_truncate:SF
          (match_operand:DF 1 "nonimmediate_operand" "f")))]
  "TARGET_80387 && flag_unsafe_math_optimizations"
  "* return output_387_reg_move (insn, operands);"
  [(set_attr "type" "fmov")
   (set_attr "mode" "SF")])

(define_insn "*truncdfsf_mixed"
  [(set (match_operand:SF 0 "nonimmediate_operand"   "=m,v ,?f,?v,?*r")
	(float_truncate:SF
	  (match_operand:DF 1 "nonimmediate_operand" "f ,vm,f ,f ,f")))
   (clobber (match_operand:SF 2 "memory_operand"     "=X,X ,m ,m ,m"))]
  "TARGET_MIX_SSE_I387"
{
  switch (which_alternative)
    {
    case 0:
      return output_387_reg_move (insn, operands);
    case 1:
      return "%vcvtsd2ss\t{%1, %d0|%d0, %1}";

    default:
      return "#";
    }
}
  [(set_attr "isa" "*,sse2,*,*,*")
   (set_attr "type" "fmov,ssecvt,multi,multi,multi")
   (set_attr "unit" "*,*,i387,i387,i387")
   (set_attr "prefix" "orig,maybe_vex,orig,orig,orig")
   (set_attr "mode" "SF")])

(define_insn "*truncdfsf_i387"
  [(set (match_operand:SF 0 "nonimmediate_operand"   "=m,?f,?v,?*r")
	(float_truncate:SF
	  (match_operand:DF 1 "nonimmediate_operand" "f ,f ,f ,f")))
   (clobber (match_operand:SF 2 "memory_operand"     "=X,m ,m ,m"))]
  "TARGET_80387"
{
  switch (which_alternative)
    {
    case 0:
      return output_387_reg_move (insn, operands);

    default:
      return "#";
    }
}
  [(set_attr "type" "fmov,multi,multi,multi")
   (set_attr "unit" "*,i387,i387,i387")
   (set_attr "mode" "SF")])

(define_insn "*truncdfsf2_i387_1"
  [(set (match_operand:SF 0 "memory_operand" "=m")
	(float_truncate:SF
	  (match_operand:DF 1 "register_operand" "f")))]
  "TARGET_80387
   && !(TARGET_SSE2 && TARGET_SSE_MATH)
   && !TARGET_MIX_SSE_I387"
  "* return output_387_reg_move (insn, operands);"
  [(set_attr "type" "fmov")
   (set_attr "mode" "SF")])

(define_split
  [(set (match_operand:SF 0 "register_operand")
	(float_truncate:SF
	 (match_operand:DF 1 "fp_register_operand")))
   (clobber (match_operand 2))]
  "reload_completed"
  [(set (match_dup 2) (match_dup 1))
   (set (match_dup 0) (match_dup 2))]
  "operands[1] = gen_rtx_REG (SFmode, REGNO (operands[1]));")

;; Conversion from XFmode to {SF,DF}mode

(define_expand "truncxf<mode>2"
  [(parallel [(set (match_operand:MODEF 0 "nonimmediate_operand")
		   (float_truncate:MODEF
		     (match_operand:XF 1 "register_operand")))
	      (clobber (match_dup 2))])]
  "TARGET_80387"
{
  if (flag_unsafe_math_optimizations)
    {
      rtx reg = REG_P (operands[0]) ? operands[0] : gen_reg_rtx (<MODE>mode);
      emit_insn (gen_truncxf<mode>2_i387_noop (reg, operands[1]));
      if (reg != operands[0])
	emit_move_insn (operands[0], reg);
      DONE;
    }
  else
    operands[2] = assign_386_stack_local (<MODE>mode, SLOT_TEMP);
})

(define_insn "*truncxfsf2_mixed"
  [(set (match_operand:SF 0 "nonimmediate_operand" "=m,?f,?v,?*r")
	(float_truncate:SF
	  (match_operand:XF 1 "register_operand"   "f ,f ,f ,f")))
   (clobber (match_operand:SF 2 "memory_operand"   "=X,m ,m ,m"))]
  "TARGET_80387"
{
  gcc_assert (!which_alternative);
  return output_387_reg_move (insn, operands);
}
  [(set_attr "type" "fmov,multi,multi,multi")
   (set_attr "unit" "*,i387,i387,i387")
   (set_attr "mode" "SF")])

(define_insn "*truncxfdf2_mixed"
  [(set (match_operand:DF 0 "nonimmediate_operand" "=m,?f,?v,?*r")
	(float_truncate:DF
	  (match_operand:XF 1 "register_operand"   "f ,f ,f  ,f")))
   (clobber (match_operand:DF 2 "memory_operand"   "=X,m ,m  ,m"))]
  "TARGET_80387"
{
  gcc_assert (!which_alternative);
  return output_387_reg_move (insn, operands);
}
  [(set_attr "isa" "*,*,sse2,*")
   (set_attr "type" "fmov,multi,multi,multi")
   (set_attr "unit" "*,i387,i387,i387")
   (set_attr "mode" "DF")])

(define_insn "truncxf<mode>2_i387_noop"
  [(set (match_operand:MODEF 0 "register_operand" "=f")
	(float_truncate:MODEF
	  (match_operand:XF 1 "register_operand" "f")))]
  "TARGET_80387 && flag_unsafe_math_optimizations"
  "* return output_387_reg_move (insn, operands);"
  [(set_attr "type" "fmov")
   (set_attr "mode" "<MODE>")])

(define_insn "*truncxf<mode>2_i387"
  [(set (match_operand:MODEF 0 "memory_operand" "=m")
	(float_truncate:MODEF
	  (match_operand:XF 1 "register_operand" "f")))]
  "TARGET_80387"
  "* return output_387_reg_move (insn, operands);"
  [(set_attr "type" "fmov")
   (set_attr "mode" "<MODE>")])

(define_split
  [(set (match_operand:MODEF 0 "register_operand")
	(float_truncate:MODEF
	  (match_operand:XF 1 "register_operand")))
   (clobber (match_operand:MODEF 2 "memory_operand"))]
  "TARGET_80387 && reload_completed"
  [(set (match_dup 2) (float_truncate:MODEF (match_dup 1)))
   (set (match_dup 0) (match_dup 2))])

(define_split
  [(set (match_operand:MODEF 0 "memory_operand")
	(float_truncate:MODEF
	  (match_operand:XF 1 "register_operand")))
   (clobber (match_operand:MODEF 2 "memory_operand"))]
  "TARGET_80387"
  [(set (match_dup 0) (float_truncate:MODEF (match_dup 1)))])

;; Signed conversion to DImode.

(define_expand "fix_truncxfdi2"
  [(parallel [(set (match_operand:DI 0 "nonimmediate_operand")
                   (fix:DI (match_operand:XF 1 "register_operand")))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_80387"
{
  if (TARGET_FISTTP)
   {
     emit_insn (gen_fix_truncdi_fisttp_i387_1 (operands[0], operands[1]));
     DONE;
   }
})

(define_expand "fix_trunc<mode>di2"
  [(parallel [(set (match_operand:DI 0 "nonimmediate_operand")
                   (fix:DI (match_operand:MODEF 1 "register_operand")))
              (clobber (reg:CC FLAGS_REG))])]
  "TARGET_80387 || (TARGET_64BIT && SSE_FLOAT_MODE_P (<MODE>mode))"
{
  if (TARGET_FISTTP
      && !(TARGET_64BIT && SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH))
   {
     emit_insn (gen_fix_truncdi_fisttp_i387_1 (operands[0], operands[1]));
     DONE;
   }
  if (TARGET_64BIT && SSE_FLOAT_MODE_P (<MODE>mode))
   {
     rtx out = REG_P (operands[0]) ? operands[0] : gen_reg_rtx (DImode);
     emit_insn (gen_fix_trunc<mode>di_sse (out, operands[1]));
     if (out != operands[0])
	emit_move_insn (operands[0], out);
     DONE;
   }
})

;; Signed conversion to SImode.

(define_expand "fix_truncxfsi2"
  [(parallel [(set (match_operand:SI 0 "nonimmediate_operand")
                   (fix:SI (match_operand:XF 1 "register_operand")))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_80387"
{
  if (TARGET_FISTTP)
   {
     emit_insn (gen_fix_truncsi_fisttp_i387_1 (operands[0], operands[1]));
     DONE;
   }
})

(define_expand "fix_trunc<mode>si2"
  [(parallel [(set (match_operand:SI 0 "nonimmediate_operand")
	           (fix:SI (match_operand:MODEF 1 "register_operand")))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_80387 || SSE_FLOAT_MODE_P (<MODE>mode)"
{
  if (TARGET_FISTTP
      && !(SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH))
   {
     emit_insn (gen_fix_truncsi_fisttp_i387_1 (operands[0], operands[1]));
     DONE;
   }
  if (SSE_FLOAT_MODE_P (<MODE>mode))
   {
     rtx out = REG_P (operands[0]) ? operands[0] : gen_reg_rtx (SImode);
     emit_insn (gen_fix_trunc<mode>si_sse (out, operands[1]));
     if (out != operands[0])
	emit_move_insn (operands[0], out);
     DONE;
   }
})

;; Signed conversion to HImode.

(define_expand "fix_trunc<mode>hi2"
  [(parallel [(set (match_operand:HI 0 "nonimmediate_operand")
	           (fix:HI (match_operand:X87MODEF 1 "register_operand")))
              (clobber (reg:CC FLAGS_REG))])]
  "TARGET_80387
   && !(SSE_FLOAT_MODE_P (<MODE>mode) && (!TARGET_FISTTP || TARGET_SSE_MATH))"
{
  if (TARGET_FISTTP)
   {
     emit_insn (gen_fix_trunchi_fisttp_i387_1 (operands[0], operands[1]));
     DONE;
   }
})

;; Unsigned conversion to SImode.

(define_expand "fixuns_trunc<mode>si2"
  [(parallel
    [(set (match_operand:SI 0 "register_operand")
	  (unsigned_fix:SI
	    (match_operand:MODEF 1 "nonimmediate_operand")))
     (use (match_dup 2))
     (clobber (match_scratch:<ssevecmode> 3))
     (clobber (match_scratch:<ssevecmode> 4))])]
  "!TARGET_64BIT && TARGET_SSE2 && TARGET_SSE_MATH"
{
  machine_mode mode = <MODE>mode;
  machine_mode vecmode = <ssevecmode>mode;
  REAL_VALUE_TYPE TWO31r;
  rtx two31;

  if (optimize_insn_for_size_p ())
    FAIL;

  real_ldexp (&TWO31r, &dconst1, 31);
  two31 = const_double_from_real_value (TWO31r, mode);
  two31 = ix86_build_const_vector (vecmode, true, two31);
  operands[2] = force_reg (vecmode, two31);
})

(define_insn_and_split "*fixuns_trunc<mode>_1"
  [(set (match_operand:SI 0 "register_operand" "=&x,&x")
	(unsigned_fix:SI
	  (match_operand:MODEF 3 "nonimmediate_operand" "xm,xm")))
   (use (match_operand:<ssevecmode> 4  "nonimmediate_operand" "m,x"))
   (clobber (match_scratch:<ssevecmode> 1 "=x,&x"))
   (clobber (match_scratch:<ssevecmode> 2 "=x,x"))]
  "!TARGET_64BIT && TARGET_SSE2 && TARGET_SSE_MATH
   && optimize_function_for_speed_p (cfun)"
  "#"
  "&& reload_completed"
  [(const_int 0)]
{
  ix86_split_convert_uns_si_sse (operands);
  DONE;
})

;; Unsigned conversion to HImode.
;; Without these patterns, we'll try the unsigned SI conversion which
;; is complex for SSE, rather than the signed SI conversion, which isn't.

(define_expand "fixuns_trunc<mode>hi2"
  [(set (match_dup 2)
	(fix:SI (match_operand:MODEF 1 "nonimmediate_operand")))
   (set (match_operand:HI 0 "nonimmediate_operand")
	(subreg:HI (match_dup 2) 0))]
  "SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH"
  "operands[2] = gen_reg_rtx (SImode);")

;; When SSE is available, it is always faster to use it!
(define_insn "fix_trunc<MODEF:mode><SWI48:mode>_sse"
  [(set (match_operand:SWI48 0 "register_operand" "=r,r")
	(fix:SWI48 (match_operand:MODEF 1 "nonimmediate_operand" "v,m")))]
  "SSE_FLOAT_MODE_P (<MODEF:MODE>mode)
   && (!TARGET_FISTTP || TARGET_SSE_MATH)"
  "%vcvtt<MODEF:ssemodesuffix>2si<SWI48:rex64suffix>\t{%1, %0|%0, %1}"
  [(set_attr "type" "sseicvt")
   (set_attr "prefix" "maybe_vex")
   (set (attr "prefix_rex")
	(if_then_else
	  (match_test "<SWI48:MODE>mode == DImode")
	  (const_string "1")
	  (const_string "*")))
   (set_attr "mode" "<MODEF:MODE>")
   (set_attr "athlon_decode" "double,vector")
   (set_attr "amdfam10_decode" "double,double")
   (set_attr "bdver1_decode" "double,double")])

;; Avoid vector decoded forms of the instruction.
(define_peephole2
  [(match_scratch:MODEF 2 "x")
   (set (match_operand:SWI48 0 "register_operand")
	(fix:SWI48 (match_operand:MODEF 1 "memory_operand")))]
  "TARGET_AVOID_VECTOR_DECODE
   && SSE_FLOAT_MODE_P (<MODEF:MODE>mode)
   && optimize_insn_for_speed_p ()"
  [(set (match_dup 2) (match_dup 1))
   (set (match_dup 0) (fix:SWI48 (match_dup 2)))])

(define_insn_and_split "fix_trunc<mode>_fisttp_i387_1"
  [(set (match_operand:SWI248x 0 "nonimmediate_operand")
	(fix:SWI248x (match_operand 1 "register_operand")))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && TARGET_FISTTP
   && !((SSE_FLOAT_MODE_P (GET_MODE (operands[1]))
	 && (TARGET_64BIT || <MODE>mode != DImode))
	&& TARGET_SSE_MATH)
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(const_int 0)]
{
  if (memory_operand (operands[0], VOIDmode))
    emit_insn (gen_fix_trunc<mode>_i387_fisttp (operands[0], operands[1]));
  else
    {
      operands[2] = assign_386_stack_local (<MODE>mode, SLOT_TEMP);
      emit_insn (gen_fix_trunc<mode>_i387_fisttp_with_temp (operands[0],
							    operands[1],
							    operands[2]));
    }
  DONE;
}
  [(set_attr "type" "fisttp")
   (set_attr "mode" "<MODE>")])

(define_insn "fix_trunc<mode>_i387_fisttp"
  [(set (match_operand:SWI248x 0 "memory_operand" "=m")
	(fix:SWI248x (match_operand 1 "register_operand" "f")))
   (clobber (match_scratch:XF 2 "=&1f"))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && TARGET_FISTTP
   && !((SSE_FLOAT_MODE_P (GET_MODE (operands[1]))
	 && (TARGET_64BIT || <MODE>mode != DImode))
	&& TARGET_SSE_MATH)"
  "* return output_fix_trunc (insn, operands, true);"
  [(set_attr "type" "fisttp")
   (set_attr "mode" "<MODE>")])

(define_insn "fix_trunc<mode>_i387_fisttp_with_temp"
  [(set (match_operand:SWI248x 0 "nonimmediate_operand" "=m,?r")
	(fix:SWI248x (match_operand 1 "register_operand" "f,f")))
   (clobber (match_operand:SWI248x 2 "memory_operand" "=X,m"))
   (clobber (match_scratch:XF 3 "=&1f,&1f"))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && TARGET_FISTTP
   && !((SSE_FLOAT_MODE_P (GET_MODE (operands[1]))
	&& (TARGET_64BIT || <MODE>mode != DImode))
	&& TARGET_SSE_MATH)"
  "#"
  [(set_attr "type" "fisttp")
   (set_attr "mode" "<MODE>")])

(define_split
  [(set (match_operand:SWI248x 0 "register_operand")
	(fix:SWI248x (match_operand 1 "register_operand")))
   (clobber (match_operand:SWI248x 2 "memory_operand"))
   (clobber (match_scratch 3))]
  "reload_completed"
  [(parallel [(set (match_dup 2) (fix:SWI248x (match_dup 1)))
	      (clobber (match_dup 3))])
   (set (match_dup 0) (match_dup 2))])

(define_split
  [(set (match_operand:SWI248x 0 "memory_operand")
	(fix:SWI248x (match_operand 1 "register_operand")))
   (clobber (match_operand:SWI248x 2 "memory_operand"))
   (clobber (match_scratch 3))]
  "reload_completed"
  [(parallel [(set (match_dup 0) (fix:SWI248x (match_dup 1)))
	      (clobber (match_dup 3))])])

;; See the comments in i386.h near OPTIMIZE_MODE_SWITCHING for the description
;; of the machinery. Please note the clobber of FLAGS_REG. In i387 control
;; word calculation (inserted by LCM in mode switching pass) a FLAGS_REG
;; clobbering insns can be used. Look at emit_i387_cw_initialization ()
;; function in i386.c.
(define_insn_and_split "*fix_trunc<mode>_i387_1"
  [(set (match_operand:SWI248x 0 "nonimmediate_operand")
	(fix:SWI248x (match_operand 1 "register_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && !TARGET_FISTTP
   && !(SSE_FLOAT_MODE_P (GET_MODE (operands[1]))
	 && (TARGET_64BIT || <MODE>mode != DImode))
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(const_int 0)]
{
  ix86_optimize_mode_switching[I387_TRUNC] = 1;

  operands[2] = assign_386_stack_local (HImode, SLOT_CW_STORED);
  operands[3] = assign_386_stack_local (HImode, SLOT_CW_TRUNC);
  if (memory_operand (operands[0], VOIDmode))
    emit_insn (gen_fix_trunc<mode>_i387 (operands[0], operands[1],
					 operands[2], operands[3]));
  else
    {
      operands[4] = assign_386_stack_local (<MODE>mode, SLOT_TEMP);
      emit_insn (gen_fix_trunc<mode>_i387_with_temp (operands[0], operands[1],
						     operands[2], operands[3],
						     operands[4]));
    }
  DONE;
}
  [(set_attr "type" "fistp")
   (set_attr "i387_cw" "trunc")
   (set_attr "mode" "<MODE>")])

(define_insn "fix_truncdi_i387"
  [(set (match_operand:DI 0 "memory_operand" "=m")
	(fix:DI (match_operand 1 "register_operand" "f")))
   (use (match_operand:HI 2 "memory_operand" "m"))
   (use (match_operand:HI 3 "memory_operand" "m"))
   (clobber (match_scratch:XF 4 "=&1f"))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && !TARGET_FISTTP
   && !(TARGET_64BIT && SSE_FLOAT_MODE_P (GET_MODE (operands[1])))"
  "* return output_fix_trunc (insn, operands, false);"
  [(set_attr "type" "fistp")
   (set_attr "i387_cw" "trunc")
   (set_attr "mode" "DI")])

(define_insn "fix_truncdi_i387_with_temp"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=m,?r")
	(fix:DI (match_operand 1 "register_operand" "f,f")))
   (use (match_operand:HI 2 "memory_operand" "m,m"))
   (use (match_operand:HI 3 "memory_operand" "m,m"))
   (clobber (match_operand:DI 4 "memory_operand" "=X,m"))
   (clobber (match_scratch:XF 5 "=&1f,&1f"))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && !TARGET_FISTTP
   && !(TARGET_64BIT && SSE_FLOAT_MODE_P (GET_MODE (operands[1])))"
  "#"
  [(set_attr "type" "fistp")
   (set_attr "i387_cw" "trunc")
   (set_attr "mode" "DI")])

(define_split
  [(set (match_operand:DI 0 "register_operand")
	(fix:DI (match_operand 1 "register_operand")))
   (use (match_operand:HI 2 "memory_operand"))
   (use (match_operand:HI 3 "memory_operand"))
   (clobber (match_operand:DI 4 "memory_operand"))
   (clobber (match_scratch 5))]
  "reload_completed"
  [(parallel [(set (match_dup 4) (fix:DI (match_dup 1)))
	      (use (match_dup 2))
	      (use (match_dup 3))
	      (clobber (match_dup 5))])
   (set (match_dup 0) (match_dup 4))])

(define_split
  [(set (match_operand:DI 0 "memory_operand")
	(fix:DI (match_operand 1 "register_operand")))
   (use (match_operand:HI 2 "memory_operand"))
   (use (match_operand:HI 3 "memory_operand"))
   (clobber (match_operand:DI 4 "memory_operand"))
   (clobber (match_scratch 5))]
  "reload_completed"
  [(parallel [(set (match_dup 0) (fix:DI (match_dup 1)))
	      (use (match_dup 2))
	      (use (match_dup 3))
	      (clobber (match_dup 5))])])

(define_insn "fix_trunc<mode>_i387"
  [(set (match_operand:SWI24 0 "memory_operand" "=m")
	(fix:SWI24 (match_operand 1 "register_operand" "f")))
   (use (match_operand:HI 2 "memory_operand" "m"))
   (use (match_operand:HI 3 "memory_operand" "m"))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && !TARGET_FISTTP
   && !SSE_FLOAT_MODE_P (GET_MODE (operands[1]))"
  "* return output_fix_trunc (insn, operands, false);"
  [(set_attr "type" "fistp")
   (set_attr "i387_cw" "trunc")
   (set_attr "mode" "<MODE>")])

(define_insn "fix_trunc<mode>_i387_with_temp"
  [(set (match_operand:SWI24 0 "nonimmediate_operand" "=m,?r")
	(fix:SWI24 (match_operand 1 "register_operand" "f,f")))
   (use (match_operand:HI 2 "memory_operand" "m,m"))
   (use (match_operand:HI 3 "memory_operand" "m,m"))
   (clobber (match_operand:SWI24 4 "memory_operand" "=X,m"))]
  "X87_FLOAT_MODE_P (GET_MODE (operands[1]))
   && !TARGET_FISTTP
   && !SSE_FLOAT_MODE_P (GET_MODE (operands[1]))"
  "#"
  [(set_attr "type" "fistp")
   (set_attr "i387_cw" "trunc")
   (set_attr "mode" "<MODE>")])

(define_split
  [(set (match_operand:SWI24 0 "register_operand")
	(fix:SWI24 (match_operand 1 "register_operand")))
   (use (match_operand:HI 2 "memory_operand"))
   (use (match_operand:HI 3 "memory_operand"))
   (clobber (match_operand:SWI24 4 "memory_operand"))]
  "reload_completed"
  [(parallel [(set (match_dup 4) (fix:SWI24 (match_dup 1)))
	      (use (match_dup 2))
	      (use (match_dup 3))])
   (set (match_dup 0) (match_dup 4))])

(define_split
  [(set (match_operand:SWI24 0 "memory_operand")
	(fix:SWI24 (match_operand 1 "register_operand")))
   (use (match_operand:HI 2 "memory_operand"))
   (use (match_operand:HI 3 "memory_operand"))
   (clobber (match_operand:SWI24 4 "memory_operand"))]
  "reload_completed"
  [(parallel [(set (match_dup 0) (fix:SWI24 (match_dup 1)))
	      (use (match_dup 2))
	      (use (match_dup 3))])])

(define_insn "x86_fnstcw_1"
  [(set (match_operand:HI 0 "memory_operand" "=m")
	(unspec:HI [(reg:HI FPCR_REG)] UNSPEC_FSTCW))]
  "TARGET_80387"
  "fnstcw\t%0"
  [(set (attr "length")
	(symbol_ref "ix86_attr_length_address_default (insn) + 2"))
   (set_attr "mode" "HI")
   (set_attr "unit" "i387")
   (set_attr "bdver1_decode" "vector")])

(define_insn "x86_fldcw_1"
  [(set (reg:HI FPCR_REG)
	(unspec:HI [(match_operand:HI 0 "memory_operand" "m")] UNSPEC_FLDCW))]
  "TARGET_80387"
  "fldcw\t%0"
  [(set (attr "length")
	(symbol_ref "ix86_attr_length_address_default (insn) + 2"))
   (set_attr "mode" "HI")
   (set_attr "unit" "i387")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "vector")
   (set_attr "bdver1_decode" "vector")])

;; Conversion between fixed point and floating point.

;; Even though we only accept memory inputs, the backend _really_
;; wants to be able to do this between registers.  Thankfully, LRA
;; will fix this up for us during register allocation.

(define_insn "floathi<mode>2"
  [(set (match_operand:X87MODEF 0 "register_operand" "=f")
	(float:X87MODEF (match_operand:HI 1 "nonimmediate_operand" "m")))]
  "TARGET_80387
   && (!(SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)
       || TARGET_MIX_SSE_I387)"
  "fild%Z1\t%1"
  [(set_attr "type" "fmov")
   (set_attr "mode" "<MODE>")
   (set_attr "znver1_decode" "double")
   (set_attr "fp_int_src" "true")])

(define_insn "float<SWI48x:mode>xf2"
  [(set (match_operand:XF 0 "register_operand" "=f")
	(float:XF (match_operand:SWI48x 1 "nonimmediate_operand" "m")))]
  "TARGET_80387"
  "fild%Z1\t%1"
  [(set_attr "type" "fmov")
   (set_attr "mode" "XF")
   (set_attr "znver1_decode" "double")
   (set_attr "fp_int_src" "true")])

(define_expand "float<SWI48:mode><MODEF:mode>2"
  [(set (match_operand:MODEF 0 "register_operand")
	(float:MODEF (match_operand:SWI48 1 "nonimmediate_operand")))]
  "TARGET_80387 || (SSE_FLOAT_MODE_P (<MODEF:MODE>mode) && TARGET_SSE_MATH)"
{
  if (!(SSE_FLOAT_MODE_P (<MODEF:MODE>mode) && TARGET_SSE_MATH)
      && !X87_ENABLE_FLOAT (<MODEF:MODE>mode, <SWI48:MODE>mode))
    {
      rtx reg = gen_reg_rtx (XFmode);
      rtx (*insn)(rtx, rtx);

      emit_insn (gen_float<SWI48:mode>xf2 (reg, operands[1]));

      if (<MODEF:MODE>mode == SFmode)
	insn = gen_truncxfsf2;
      else if (<MODEF:MODE>mode == DFmode)
	insn = gen_truncxfdf2;
      else
	gcc_unreachable ();

      emit_insn (insn (operands[0], reg));
      DONE;
    }
})

(define_insn "*float<SWI48:mode><MODEF:mode>2_mixed"
  [(set (match_operand:MODEF 0 "register_operand" "=f,Yc,v")
	(float:MODEF
	  (match_operand:SWI48 1 "nonimmediate_operand" "m,r,m")))]
  "SSE_FLOAT_MODE_P (<MODEF:MODE>mode) && TARGET_SSE_MATH"
  "@
   fild%Z1\t%1
   %vcvtsi2<MODEF:ssemodesuffix><SWI48:rex64suffix>\t{%1, %d0|%d0, %1}
   %vcvtsi2<MODEF:ssemodesuffix><SWI48:rex64suffix>\t{%1, %d0|%d0, %1}"
  [(set_attr "type" "fmov,sseicvt,sseicvt")
   (set_attr "prefix" "orig,maybe_vex,maybe_vex")
   (set_attr "mode" "<MODEF:MODE>")
   (set (attr "prefix_rex")
     (if_then_else
       (and (eq_attr "prefix" "maybe_vex")
	    (match_test "<SWI48:MODE>mode == DImode"))
       (const_string "1")
       (const_string "*")))
   (set_attr "unit" "i387,*,*")
   (set_attr "athlon_decode" "*,double,direct")
   (set_attr "amdfam10_decode" "*,vector,double")
   (set_attr "bdver1_decode" "*,double,direct")
   (set_attr "znver1_decode" "double,*,*")
   (set_attr "fp_int_src" "true")
   (set (attr "enabled")
     (cond [(eq_attr "alternative" "0")
              (symbol_ref "TARGET_MIX_SSE_I387
                           && X87_ENABLE_FLOAT (<MODEF:MODE>mode,
                                                <SWI48:MODE>mode)")
           ]
           (symbol_ref "true")))])

(define_insn "*float<SWI48x:mode><MODEF:mode>2_i387"
  [(set (match_operand:MODEF 0 "register_operand" "=f")
	(float:MODEF (match_operand:SWI48x 1 "nonimmediate_operand" "m")))]
  "TARGET_80387 && X87_ENABLE_FLOAT (<MODEF:MODE>mode, <SWI48x:MODE>mode)"
  "fild%Z1\t%1"
  [(set_attr "type" "fmov")
   (set_attr "mode" "<MODEF:MODE>")
   (set_attr "znver1_decode" "double")
   (set_attr "fp_int_src" "true")])

;; Try TARGET_USE_VECTOR_CONVERTS, but not so hard as to require extra memory
;; slots when !TARGET_INTER_UNIT_MOVES_TO_VEC disables the general_regs
;; alternative in sse2_loadld.
(define_split
  [(set (match_operand:MODEF 0 "sse_reg_operand")
	(float:MODEF (match_operand:SI 1 "nonimmediate_operand")))]
  "TARGET_SSE2
   && TARGET_USE_VECTOR_CONVERTS
   && optimize_function_for_speed_p (cfun)
   && reload_completed
   && (MEM_P (operands[1]) || TARGET_INTER_UNIT_MOVES_TO_VEC)
   && (!EXT_REX_SSE_REG_P (operands[0])
       || TARGET_AVX512VL)"
  [(const_int 0)]
{
  operands[3] = lowpart_subreg (<ssevecmode>mode, operands[0], <MODE>mode);
  operands[4] = lowpart_subreg (V4SImode, operands[0], <MODE>mode);

  emit_insn (gen_sse2_loadld (operands[4],
			      CONST0_RTX (V4SImode), operands[1]));

  if (<ssevecmode>mode == V4SFmode)
    emit_insn (gen_floatv4siv4sf2 (operands[3], operands[4]));
  else
    emit_insn (gen_sse2_cvtdq2pd (operands[3], operands[4]));
  DONE;
})

;; Avoid partial SSE register dependency stalls.  This splitter should split
;; late in the pass sequence (after register rename pass), so allocated
;; registers won't change anymore

(define_split
  [(set (match_operand:MODEF 0 "sse_reg_operand")
	(float:MODEF (match_operand:SWI48 1 "nonimmediate_operand")))]
  "TARGET_SSE_PARTIAL_REG_DEPENDENCY && epilogue_completed
   && optimize_function_for_speed_p (cfun)
   && (!EXT_REX_SSE_REG_P (operands[0])
       || TARGET_AVX512VL)"
  [(set (match_dup 0)
	(vec_merge:<MODEF:ssevecmode>
	  (vec_duplicate:<MODEF:ssevecmode>
	    (float:MODEF
	      (match_dup 1)))
	  (match_dup 0)
	  (const_int 1)))]
{
  const machine_mode vmode = <MODEF:ssevecmode>mode;

  operands[0] = lowpart_subreg (vmode, operands[0], <MODEF:MODE>mode);
  emit_move_insn (operands[0], CONST0_RTX (vmode));
})

;; Break partial reg stall for cvtsd2ss.  This splitter should split
;; late in the pass sequence (after register rename pass),
;; so allocated registers won't change anymore.

(define_split
  [(set (match_operand:SF 0 "sse_reg_operand")
        (float_truncate:SF
	  (match_operand:DF 1 "nonimmediate_operand")))]
  "TARGET_SSE_PARTIAL_REG_DEPENDENCY && epilogue_completed
   && optimize_function_for_speed_p (cfun)
   && (!REG_P (operands[1])
       || REGNO (operands[0]) != REGNO (operands[1]))
   && (!EXT_REX_SSE_REG_P (operands[0])
       || TARGET_AVX512VL)"
  [(set (match_dup 0)
	(vec_merge:V4SF
	  (vec_duplicate:V4SF
	    (float_truncate:SF
	      (match_dup 1)))
	  (match_dup 0)
	  (const_int 1)))]
{
  operands[0] = lowpart_subreg (V4SFmode, operands[0], SFmode);
  emit_move_insn (operands[0], CONST0_RTX (V4SFmode));
})

;; Break partial reg stall for cvtss2sd.  This splitter should split
;; late in the pass sequence (after register rename pass),
;; so allocated registers won't change anymore.

(define_split
  [(set (match_operand:DF 0 "sse_reg_operand")
        (float_extend:DF
          (match_operand:SF 1 "nonimmediate_operand")))]
  "TARGET_SSE_PARTIAL_REG_DEPENDENCY && epilogue_completed
   && optimize_function_for_speed_p (cfun)
   && (!REG_P (operands[1])
       || REGNO (operands[0]) != REGNO (operands[1]))
   && (!EXT_REX_SSE_REG_P (operands[0])
       || TARGET_AVX512VL)"
  [(set (match_dup 0)
        (vec_merge:V2DF
	  (vec_duplicate:V2DF
	    (float_extend:DF
	      (match_dup 1)))
	  (match_dup 0)
          (const_int 1)))]
{
  operands[0] = lowpart_subreg (V2DFmode, operands[0], DFmode);
  emit_move_insn (operands[0], CONST0_RTX (V2DFmode));
})

;; Avoid store forwarding (partial memory) stall penalty
;; by passing DImode value through XMM registers.  */

(define_insn "floatdi<X87MODEF:mode>2_i387_with_xmm"
  [(set (match_operand:X87MODEF 0 "register_operand" "=f,f")
	(float:X87MODEF
	  (match_operand:DI 1 "nonimmediate_operand" "m,?r")))
   (clobber (match_scratch:V4SI 3 "=X,x"))
   (clobber (match_scratch:V4SI 4 "=X,x"))
   (clobber (match_operand:DI 2 "memory_operand" "=X,m"))]
  "TARGET_80387 && X87_ENABLE_FLOAT (<X87MODEF:MODE>mode, DImode)
   && TARGET_SSE2 && TARGET_INTER_UNIT_MOVES_TO_VEC
   && !TARGET_64BIT && optimize_function_for_speed_p (cfun)"
  "#"
  [(set_attr "type" "multi")
   (set_attr "mode" "<X87MODEF:MODE>")
   (set_attr "unit" "i387")
   (set_attr "fp_int_src" "true")])

(define_split
  [(set (match_operand:X87MODEF 0 "fp_register_operand")
	(float:X87MODEF (match_operand:DI 1 "register_operand")))
   (clobber (match_scratch:V4SI 3))
   (clobber (match_scratch:V4SI 4))
   (clobber (match_operand:DI 2 "memory_operand"))]
  "TARGET_80387 && X87_ENABLE_FLOAT (<X87MODEF:MODE>mode, DImode)
   && TARGET_SSE2 && TARGET_INTER_UNIT_MOVES_TO_VEC
   && !TARGET_64BIT && optimize_function_for_speed_p (cfun)
   && reload_completed"
  [(set (match_dup 2) (match_dup 3))
   (set (match_dup 0) (float:X87MODEF (match_dup 2)))]
{
  /* The DImode arrived in a pair of integral registers (e.g. %edx:%eax).
     Assemble the 64-bit DImode value in an xmm register.  */
  emit_insn (gen_sse2_loadld (operands[3], CONST0_RTX (V4SImode),
			      gen_lowpart (SImode, operands[1])));
  emit_insn (gen_sse2_loadld (operands[4], CONST0_RTX (V4SImode),
			      gen_highpart (SImode, operands[1])));
  emit_insn (gen_vec_interleave_lowv4si (operands[3], operands[3],
					 operands[4]));

  operands[3] = gen_lowpart (DImode, operands[3]);
})

(define_split
  [(set (match_operand:X87MODEF 0 "fp_register_operand")
	(float:X87MODEF (match_operand:DI 1 "memory_operand")))
   (clobber (match_scratch:V4SI 3))
   (clobber (match_scratch:V4SI 4))
   (clobber (match_operand:DI 2 "memory_operand"))]
  "TARGET_80387 && X87_ENABLE_FLOAT (<X87MODEF:MODE>mode, DImode)
   && TARGET_SSE2 && TARGET_INTER_UNIT_MOVES_TO_VEC
   && !TARGET_64BIT && optimize_function_for_speed_p (cfun)
   && reload_completed"
  [(set (match_dup 0) (float:X87MODEF (match_dup 1)))])

(define_expand "floatuns<SWI12:mode><MODEF:mode>2"
  [(set (match_operand:MODEF 0 "register_operand")
	(unsigned_float:MODEF
	  (match_operand:SWI12 1 "nonimmediate_operand")))]
  "!TARGET_64BIT
   && SSE_FLOAT_MODE_P (<MODEF:MODE>mode) && TARGET_SSE_MATH"
{
  operands[1] = convert_to_mode (SImode, operands[1], 1);
  emit_insn (gen_floatsi<MODEF:mode>2 (operands[0], operands[1]));
  DONE;
})

;; Avoid store forwarding (partial memory) stall penalty by extending
;; SImode value to DImode through XMM register instead of pushing two
;; SImode values to stack. Also note that fild loads from memory only.

(define_insn_and_split "*floatunssi<mode>2_i387_with_xmm"
  [(set (match_operand:X87MODEF 0 "register_operand" "=f")
	(unsigned_float:X87MODEF
	  (match_operand:SI 1 "nonimmediate_operand" "rm")))
   (clobber (match_scratch:DI 3 "=x"))
   (clobber (match_operand:DI 2 "memory_operand" "=m"))]
  "!TARGET_64BIT
   && TARGET_80387 && X87_ENABLE_FLOAT (<X87MODEF:MODE>mode, DImode)
   && TARGET_SSE2 && TARGET_INTER_UNIT_MOVES_TO_VEC"
  "#"
  "&& reload_completed"
  [(set (match_dup 3) (zero_extend:DI (match_dup 1)))
   (set (match_dup 2) (match_dup 3))
   (set (match_dup 0)
	(float:X87MODEF (match_dup 2)))]
  ""
  [(set_attr "type" "multi")
   (set_attr "mode" "<MODE>")])

(define_expand "floatunssi<mode>2"
  [(parallel
     [(set (match_operand:X87MODEF 0 "register_operand")
	   (unsigned_float:X87MODEF
	     (match_operand:SI 1 "nonimmediate_operand")))
      (clobber (match_scratch:DI 3))
      (clobber (match_dup 2))])]
  "!TARGET_64BIT
   && ((TARGET_80387 && X87_ENABLE_FLOAT (<X87MODEF:MODE>mode, DImode)
	&& TARGET_SSE2 && TARGET_INTER_UNIT_MOVES_TO_VEC)
       || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH))"
{
  if (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)
    {
      ix86_expand_convert_uns_si<mode>_sse (operands[0], operands[1]);
      DONE;
    }
  else
    operands[2] = assign_386_stack_local (DImode, SLOT_TEMP);
})

(define_expand "floatunsdisf2"
  [(use (match_operand:SF 0 "register_operand"))
   (use (match_operand:DI 1 "nonimmediate_operand"))]
  "TARGET_64BIT && TARGET_SSE && TARGET_SSE_MATH"
  "x86_emit_floatuns (operands); DONE;")

(define_expand "floatunsdidf2"
  [(use (match_operand:DF 0 "register_operand"))
   (use (match_operand:DI 1 "nonimmediate_operand"))]
  "(TARGET_64BIT || TARGET_KEEPS_VECTOR_ALIGNED_STACK)
   && TARGET_SSE2 && TARGET_SSE_MATH"
{
  if (TARGET_64BIT)
    x86_emit_floatuns (operands);
  else
    ix86_expand_convert_uns_didf_sse (operands[0], operands[1]);
  DONE;
})

;; Load effective address instructions

(define_insn_and_split "*lea<mode>"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(match_operand:SWI48 1 "address_no_seg_operand" "Ts"))]
  ""
{
  if (SImode_address_operand (operands[1], VOIDmode))
    {
      gcc_assert (TARGET_64BIT);
      return "lea{l}\t{%E1, %k0|%k0, %E1}";
    }
  else 
    return "lea{<imodesuffix>}\t{%E1, %0|%0, %E1}";
}
  "reload_completed && ix86_avoid_lea_for_addr (insn, operands)"
  [(const_int 0)]
{
  machine_mode mode = <MODE>mode;
  rtx pat;

  /* ix86_avoid_lea_for_addr re-recognizes insn and may
     change operands[] array behind our back.  */
  pat = PATTERN (curr_insn);

  operands[0] = SET_DEST (pat);
  operands[1] = SET_SRC (pat);

  /* Emit all operations in SImode for zero-extended addresses.  */
  if (SImode_address_operand (operands[1], VOIDmode))
    mode = SImode;

  ix86_split_lea_for_addr (curr_insn, operands, mode);

  /* Zero-extend return register to DImode for zero-extended addresses.  */
  if (mode != <MODE>mode)
    emit_insn (gen_zero_extendsidi2
    	       (operands[0], gen_lowpart (mode, operands[0])));

  DONE;
}
  [(set_attr "type" "lea")
   (set (attr "mode")
     (if_then_else
       (match_operand 1 "SImode_address_operand")
       (const_string "SI")
       (const_string "<MODE>")))])

;; Add instructions

(define_expand "add<mode>3"
  [(set (match_operand:SDWIM 0 "nonimmediate_operand")
	(plus:SDWIM (match_operand:SDWIM 1 "nonimmediate_operand")
		    (match_operand:SDWIM 2 "<general_hilo_operand>")))]
  ""
  "ix86_expand_binary_operator (PLUS, <MODE>mode, operands); DONE;")

(define_insn_and_split "*add<dwi>3_doubleword"
  [(set (match_operand:<DWI> 0 "nonimmediate_operand" "=r,o")
	(plus:<DWI>
	  (match_operand:<DWI> 1 "nonimmediate_operand" "%0,0")
	  (match_operand:<DWI> 2 "x86_64_hilo_general_operand"
							"ro<di>,r<di>")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (PLUS, <DWI>mode, operands)"
  "#"
  "reload_completed"
  [(parallel [(set (reg:CCC FLAGS_REG)
		   (compare:CCC
		     (plus:DWIH (match_dup 1) (match_dup 2))
		     (match_dup 1)))
	      (set (match_dup 0)
		   (plus:DWIH (match_dup 1) (match_dup 2)))])
   (parallel [(set (match_dup 3)
		   (plus:DWIH
		     (plus:DWIH
		       (ltu:DWIH (reg:CC FLAGS_REG) (const_int 0))
		       (match_dup 4))
		     (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  split_double_mode (<DWI>mode, &operands[0], 3, &operands[0], &operands[3]);
  if (operands[2] == const0_rtx)
    {
      ix86_expand_binary_operator (PLUS, <MODE>mode, &operands[3]);
      DONE;
    }
})

(define_insn "*add<mode>_1"
  [(set (match_operand:SWI48 0 "nonimmediate_operand" "=r,rm,r,r")
	(plus:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand" "%0,0,r,r")
	  (match_operand:SWI48 2 "x86_64_general_operand" "rme,re,0,le")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
      return "#";

    case TYPE_INCDEC:
      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (operands[2] == const1_rtx)
        return "inc{<imodesuffix>}\t%0";
      else
        {
	  gcc_assert (operands[2] == constm1_rtx);
          return "dec{<imodesuffix>}\t%0";
	}

    default:
      /* For most processors, ADD is faster than LEA.  This alternative
	 was added to use ADD as much as possible.  */
      if (which_alternative == 2)
        std::swap (operands[1], operands[2]);
        
      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (x86_maybe_negate_const_int (&operands[2], <MODE>mode))
        return "sub{<imodesuffix>}\t{%2, %0|%0, %2}";

      return "add{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "3")
              (const_string "lea")
	    (match_operand:SWI48 2 "incdec_operand")
	      (const_string "incdec")
	   ]
	   (const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "<MODE>")])

;; It may seem that nonimmediate operand is proper one for operand 1.
;; The addsi_1 pattern allows nonimmediate operand at that place and
;; we take care in ix86_binary_operator_ok to not allow two memory
;; operands so proper swapping will be done in reload.  This allow
;; patterns constructed from addsi_1 to match.

(define_insn "addsi_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r,r,r")
	(zero_extend:DI
	  (plus:SI (match_operand:SI 1 "nonimmediate_operand" "%0,r,r")
		   (match_operand:SI 2 "x86_64_general_operand" "rme,0,le"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (PLUS, SImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
      return "#";

    case TYPE_INCDEC:
      if (operands[2] == const1_rtx)
        return "inc{l}\t%k0";
      else
        {
	  gcc_assert (operands[2] == constm1_rtx);
          return "dec{l}\t%k0";
	}

    default:
      /* For most processors, ADD is faster than LEA.  This alternative
	 was added to use ADD as much as possible.  */
      if (which_alternative == 1)
        std::swap (operands[1], operands[2]);

      if (x86_maybe_negate_const_int (&operands[2], SImode))
        return "sub{l}\t{%2, %k0|%k0, %2}";

      return "add{l}\t{%2, %k0|%k0, %2}";
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "2")
	      (const_string "lea")
	    (match_operand:SI 2 "incdec_operand")
	      (const_string "incdec")
	   ]
	   (const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "SI")])

(define_insn "*addhi_1"
  [(set (match_operand:HI 0 "nonimmediate_operand" "=rm,r,r,Yp")
	(plus:HI (match_operand:HI 1 "nonimmediate_operand" "%0,0,r,Yp")
		 (match_operand:HI 2 "general_operand" "rn,rm,0,ln")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (PLUS, HImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
      return "#";

    case TYPE_INCDEC:
      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (operands[2] == const1_rtx)
	return "inc{w}\t%0";
      else
	{
	  gcc_assert (operands[2] == constm1_rtx);
	  return "dec{w}\t%0";
	}

    default:
      /* For most processors, ADD is faster than LEA.  This alternative
	 was added to use ADD as much as possible.  */
      if (which_alternative == 2)
        std::swap (operands[1], operands[2]);

      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (x86_maybe_negate_const_int (&operands[2], HImode))
	return "sub{w}\t{%2, %0|%0, %2}";

      return "add{w}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "3")
              (const_string "lea")
	    (match_operand:HI 2 "incdec_operand")
	      (const_string "incdec")
	   ]
	   (const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "HI,HI,HI,SI")])

(define_insn "*addqi_1"
  [(set (match_operand:QI 0 "nonimmediate_operand" "=qm,q,q,r,r,Yp")
	(plus:QI (match_operand:QI 1 "nonimmediate_operand" "%0,0,q,0,r,Yp")
		 (match_operand:QI 2 "general_operand" "qn,qm,0,rn,0,ln")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (PLUS, QImode, operands)"
{
  bool widen = (get_attr_mode (insn) != MODE_QI);

  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
      return "#";

    case TYPE_INCDEC:
      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (operands[2] == const1_rtx)
	return widen ? "inc{l}\t%k0" : "inc{b}\t%0";
      else
	{
	  gcc_assert (operands[2] == constm1_rtx);
	  return widen ? "dec{l}\t%k0" : "dec{b}\t%0";
	}

    default:
      /* For most processors, ADD is faster than LEA.  These alternatives
	 were added to use ADD as much as possible.  */
      if (which_alternative == 2 || which_alternative == 4)
        std::swap (operands[1], operands[2]);

      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (x86_maybe_negate_const_int (&operands[2], QImode))
	{
	  if (widen)
	    return "sub{l}\t{%2, %k0|%k0, %2}";
	  else
	    return "sub{b}\t{%2, %0|%0, %2}";
	}
      if (widen)
        return "add{l}\t{%k2, %k0|%k0, %k2}";
      else
        return "add{b}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "5")
              (const_string "lea")
	    (match_operand:QI 2 "incdec_operand")
	      (const_string "incdec")
	   ]
	   (const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "QI,QI,QI,SI,SI,SI")
   ;; Potential partial reg stall on alternatives 3 and 4.
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "3,4")
	      (symbol_ref "!TARGET_PARTIAL_REG_STALL")]
	   (symbol_ref "true")))])

(define_insn "*addqi_1_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+qm,q"))
	(plus:QI (match_dup 0)
		 (match_operand:QI 1 "general_operand" "qn,qm")))
   (clobber (reg:CC FLAGS_REG))]
  "(! TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[1] == const1_rtx)
	return "inc{b}\t%0";
      else
	{
	  gcc_assert (operands[1] == constm1_rtx);
	  return "dec{b}\t%0";
	}

    default:
      if (x86_maybe_negate_const_int (&operands[1], QImode))
	return "sub{b}\t{%1, %0|%0, %1}";

      return "add{b}\t{%1, %0|%0, %1}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:QI 1 "incdec_operand")
	(const_string "incdec")
	(const_string "alu1")))
   (set (attr "memory")
     (if_then_else (match_operand 1 "memory_operand")
        (const_string "load")
        (const_string "none")))
   (set_attr "mode" "QI")])

;; Split non destructive adds if we cannot use lea.
(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(plus:SWI48 (match_operand:SWI48 1 "register_operand")
		    (match_operand:SWI48 2 "x86_64_nonmemory_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed && ix86_avoid_lea_for_add (insn, operands)"
  [(set (match_dup 0) (match_dup 1))
   (parallel [(set (match_dup 0) (plus:SWI48 (match_dup 0) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])])

;; Split non destructive adds if we cannot use lea.
(define_split
  [(set (match_operand:DI 0 "register_operand")
  	(zero_extend:DI
	  (plus:SI (match_operand:SI 1 "register_operand")
		   (match_operand:SI 2 "x86_64_nonmemory_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && reload_completed && ix86_avoid_lea_for_add (insn, operands)"
  [(set (match_dup 3) (match_dup 1))
   (parallel [(set (match_dup 0)
   	     	   (zero_extend:DI (plus:SI (match_dup 3) (match_dup 2))))
	      (clobber (reg:CC FLAGS_REG))])]
  "operands[3] = gen_lowpart (SImode, operands[0]);")

;; Convert add to the lea pattern to avoid flags dependency.
(define_split
  [(set (match_operand:SWI 0 "register_operand")
	(plus:SWI (match_operand:SWI 1 "register_operand")
		  (match_operand:SWI 2 "<nonmemory_operand>")))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed && ix86_lea_for_add_ok (insn, operands)" 
  [(set (match_dup 0)
	(plus:<LEAMODE> (match_dup 1) (match_dup 2)))]
{
  if (<MODE>mode != <LEAMODE>mode)
    {
      operands[0] = gen_lowpart (<LEAMODE>mode, operands[0]);
      operands[1] = gen_lowpart (<LEAMODE>mode, operands[1]);
      operands[2] = gen_lowpart (<LEAMODE>mode, operands[2]);
    }
})

;; Convert add to the lea pattern to avoid flags dependency.
(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (plus:SI (match_operand:SI 1 "register_operand")
		   (match_operand:SI 2 "x86_64_nonmemory_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && reload_completed && ix86_lea_for_add_ok (insn, operands)"
  [(set (match_dup 0)
	(zero_extend:DI (plus:SI (match_dup 1) (match_dup 2))))])

(define_insn "*add<mode>_2"
  [(set (reg FLAGS_REG)
	(compare
	  (plus:SWI
	    (match_operand:SWI 1 "nonimmediate_operand" "%0,0,<r>")
	    (match_operand:SWI 2 "<general_operand>" "<g>,<r><i>,0"))
	  (const_int 0)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>,<r>m,<r>")
	(plus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (PLUS, <MODE>mode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == const1_rtx)
        return "inc{<imodesuffix>}\t%0";
      else
        {
	  gcc_assert (operands[2] == constm1_rtx);
          return "dec{<imodesuffix>}\t%0";
	}

    default:
      if (which_alternative == 2)
        std::swap (operands[1], operands[2]);
        
      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (x86_maybe_negate_const_int (&operands[2], <MODE>mode))
        return "sub{<imodesuffix>}\t{%2, %0|%0, %2}";

      return "add{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:SWI 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "<MODE>")])

;; See comment for addsi_1_zext why we do use nonimmediate_operand
(define_insn "*addsi_2_zext"
  [(set (reg FLAGS_REG)
	(compare
	  (plus:SI (match_operand:SI 1 "nonimmediate_operand" "%0,r")
		   (match_operand:SI 2 "x86_64_general_operand" "rme,0"))
	  (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r,r")
	(zero_extend:DI (plus:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (PLUS, SImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == const1_rtx)
        return "inc{l}\t%k0";
      else
	{
	  gcc_assert (operands[2] == constm1_rtx);
          return "dec{l}\t%k0";
	}

    default:
      if (which_alternative == 1)
        std::swap (operands[1], operands[2]);

      if (x86_maybe_negate_const_int (&operands[2], SImode))
        return "sub{l}\t{%2, %k0|%k0, %2}";

      return "add{l}\t{%2, %k0|%k0, %2}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:SI 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "SI")])

(define_insn "*add<mode>_3"
  [(set (reg FLAGS_REG)
	(compare
	  (neg:SWI (match_operand:SWI 2 "<general_operand>" "<g>,0"))
	  (match_operand:SWI 1 "nonimmediate_operand" "%0,<r>")))
   (clobber (match_scratch:SWI 0 "=<r>,<r>"))]
  "ix86_match_ccmode (insn, CCZmode)
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == const1_rtx)
        return "inc{<imodesuffix>}\t%0";
      else
        {
	  gcc_assert (operands[2] == constm1_rtx);
          return "dec{<imodesuffix>}\t%0";
	}

    default:
      if (which_alternative == 1)
        std::swap (operands[1], operands[2]);

      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (x86_maybe_negate_const_int (&operands[2], <MODE>mode))
        return "sub{<imodesuffix>}\t{%2, %0|%0, %2}";

      return "add{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:SWI 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "<MODE>")])

;; See comment for addsi_1_zext why we do use nonimmediate_operand
(define_insn "*addsi_3_zext"
  [(set (reg FLAGS_REG)
	(compare
	  (neg:SI (match_operand:SI 2 "x86_64_general_operand" "rme,0"))
	  (match_operand:SI 1 "nonimmediate_operand" "%0,r")))
   (set (match_operand:DI 0 "register_operand" "=r,r")
	(zero_extend:DI (plus:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCZmode)
   && ix86_binary_operator_ok (PLUS, SImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == const1_rtx)
        return "inc{l}\t%k0";
      else
        {
	  gcc_assert (operands[2] == constm1_rtx);
          return "dec{l}\t%k0";
	}

    default:
      if (which_alternative == 1)
        std::swap (operands[1], operands[2]);

      if (x86_maybe_negate_const_int (&operands[2], SImode))
        return "sub{l}\t{%2, %k0|%k0, %2}";

      return "add{l}\t{%2, %k0|%k0, %2}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:SI 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "SI")])

; For comparisons against 1, -1 and 128, we may generate better code
; by converting cmp to add, inc or dec as done by peephole2.  This pattern
; is matched then.  We can't accept general immediate, because for
; case of overflows,  the result is messed up.
; Also carry flag is reversed compared to cmp, so this conversion is valid
; only for comparisons not depending on it.

(define_insn "*adddi_4"
  [(set (reg FLAGS_REG)
	(compare
	  (match_operand:DI 1 "nonimmediate_operand" "0")
	  (match_operand:DI 2 "x86_64_immediate_operand" "e")))
   (clobber (match_scratch:DI 0 "=rm"))]
  "TARGET_64BIT
   && ix86_match_ccmode (insn, CCGCmode)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == constm1_rtx)
        return "inc{q}\t%0";
      else
        {
	  gcc_assert (operands[2] == const1_rtx);
          return "dec{q}\t%0";
	}

    default:
      if (x86_maybe_negate_const_int (&operands[2], DImode))
	return "add{q}\t{%2, %0|%0, %2}";

      return "sub{q}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:DI 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "DI")])

; For comparisons against 1, -1 and 128, we may generate better code
; by converting cmp to add, inc or dec as done by peephole2.  This pattern
; is matched then.  We can't accept general immediate, because for
; case of overflows,  the result is messed up.
; Also carry flag is reversed compared to cmp, so this conversion is valid
; only for comparisons not depending on it.

(define_insn "*add<mode>_4"
  [(set (reg FLAGS_REG)
	(compare
	  (match_operand:SWI124 1 "nonimmediate_operand" "0")
	  (match_operand:SWI124 2 "const_int_operand" "n")))
   (clobber (match_scratch:SWI124 0 "=<r>m"))]
  "ix86_match_ccmode (insn, CCGCmode)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == constm1_rtx)
        return "inc{<imodesuffix>}\t%0";
      else
        {
	  gcc_assert (operands[2] == const1_rtx);
          return "dec{<imodesuffix>}\t%0";
	}

    default:
      if (x86_maybe_negate_const_int (&operands[2], <MODE>mode))
	return "add{<imodesuffix>}\t{%2, %0|%0, %2}";

      return "sub{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:<MODE> 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "<MODE>")])

(define_insn "*add<mode>_5"
  [(set (reg FLAGS_REG)
	(compare
	  (plus:SWI
	    (match_operand:SWI 1 "nonimmediate_operand" "%0,<r>")
	    (match_operand:SWI 2 "<general_operand>" "<g>,0"))
	  (const_int 0)))
   (clobber (match_scratch:SWI 0 "=<r>,<r>"))]
  "ix86_match_ccmode (insn, CCGOCmode)
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == const1_rtx)
        return "inc{<imodesuffix>}\t%0";
      else
        {
          gcc_assert (operands[2] == constm1_rtx);
          return "dec{<imodesuffix>}\t%0";
	}

    default:
      if (which_alternative == 1)
        std::swap (operands[1], operands[2]);

      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      if (x86_maybe_negate_const_int (&operands[2], <MODE>mode))
        return "sub{<imodesuffix>}\t{%2, %0|%0, %2}";

      return "add{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (if_then_else (match_operand:SWI 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set (attr "length_immediate")
      (if_then_else
	(and (eq_attr "type" "alu") (match_operand 2 "const128_operand"))
	(const_string "1")
	(const_string "*")))
   (set_attr "mode" "<MODE>")])

(define_insn "addqi_ext_1"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q,Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (plus:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "0,0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (match_operand:QI 2 "general_operand" "QnBc,m")) 0))
   (clobber (reg:CC FLAGS_REG))]
  "/* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   rtx_equal_p (operands[0], operands[1])"
{
  switch (get_attr_type (insn))
    {
    case TYPE_INCDEC:
      if (operands[2] == const1_rtx)
	return "inc{b}\t%h0";
      else
        {
	  gcc_assert (operands[2] == constm1_rtx);
          return "dec{b}\t%h0";
        }

    default:
      return "add{b}\t{%2, %h0|%h0, %2}";
    }
}
  [(set_attr "isa" "*,nox64")
   (set (attr "type")
     (if_then_else (match_operand:QI 2 "incdec_operand")
	(const_string "incdec")
	(const_string "alu")))
   (set_attr "mode" "QI")])

(define_insn "*addqi_ext_2"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (plus:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "%0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (subreg:QI
	      (zero_extract:SI (match_operand 2 "ext_register_operand" "Q")
			       (const_int 8)
			       (const_int 8)) 0)) 0))
  (clobber (reg:CC FLAGS_REG))]
  "/* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   rtx_equal_p (operands[0], operands[1])
   || rtx_equal_p (operands[0], operands[2])"
  "add{b}\t{%h2, %h0|%h0, %h2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "QI")])

;; Add with jump on overflow.
(define_expand "addv<mode>4"
  [(parallel [(set (reg:CCO FLAGS_REG)
		   (eq:CCO (plus:<DWI>
			      (sign_extend:<DWI>
				 (match_operand:SWI 1 "nonimmediate_operand"))
			      (match_dup 4))
			   (sign_extend:<DWI>
			      (plus:SWI (match_dup 1)
					(match_operand:SWI 2
					   "<general_operand>")))))
	      (set (match_operand:SWI 0 "register_operand")
		   (plus:SWI (match_dup 1) (match_dup 2)))])
   (set (pc) (if_then_else
	       (eq (reg:CCO FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 3))
	       (pc)))]
  ""
{
  ix86_fixup_binary_operands_no_copy (PLUS, <MODE>mode, operands);
  if (CONST_INT_P (operands[2]))
    operands[4] = operands[2];
  else
    operands[4] = gen_rtx_SIGN_EXTEND (<DWI>mode, operands[2]);
})

(define_insn "*addv<mode>4"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (plus:<DWI>
		   (sign_extend:<DWI>
		      (match_operand:SWI 1 "nonimmediate_operand" "%0,0"))
		   (sign_extend:<DWI>
		      (match_operand:SWI 2 "<general_sext_operand>"
					   "<r>mWe,<r>We")))
		(sign_extend:<DWI>
		   (plus:SWI (match_dup 1) (match_dup 2)))))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>,<r>m")
	(plus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)"
  "add{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*addv<mode>4_1"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (plus:<DWI>
		   (sign_extend:<DWI>
		      (match_operand:SWI 1 "nonimmediate_operand" "0"))
		   (match_operand:<DWI> 3 "const_int_operand" "i"))
		(sign_extend:<DWI>
		   (plus:SWI (match_dup 1)
			     (match_operand:SWI 2 "x86_64_immediate_operand"
						  "<i>")))))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(plus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)
   && CONST_INT_P (operands[2])
   && INTVAL (operands[2]) == INTVAL (operands[3])"
  "add{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")
   (set (attr "length_immediate")
	(cond [(match_test "IN_RANGE (INTVAL (operands[2]), -128, 127)")
		  (const_string "1")
	       (match_test "<MODE_SIZE> == 8")
		  (const_string "4")]
	      (const_string "<MODE_SIZE>")))])

(define_expand "uaddv<mode>4"
  [(parallel [(set (reg:CCC FLAGS_REG)
		   (compare:CCC
		     (plus:SWI
		       (match_operand:SWI 1 "nonimmediate_operand")
		       (match_operand:SWI 2 "<general_operand>"))
		     (match_dup 1)))
	      (set (match_operand:SWI 0 "register_operand")
		   (plus:SWI (match_dup 1) (match_dup 2)))])
   (set (pc) (if_then_else
	       (ltu (reg:CCC FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 3))
	       (pc)))]
  ""
  "ix86_fixup_binary_operands_no_copy (PLUS, <MODE>mode, operands);")

;; The lea patterns for modes less than 32 bits need to be matched by
;; several insns converted to real lea by splitters.

(define_insn_and_split "*lea<mode>_general_1"
  [(set (match_operand:SWI12 0 "register_operand" "=r")
	(plus:SWI12
	  (plus:SWI12 (match_operand:SWI12 1 "index_register_operand" "l")
		      (match_operand:SWI12 2 "register_operand" "r"))
	  (match_operand:SWI12 3 "immediate_operand" "i")))]
  "!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun)"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(plus:SI
	  (plus:SI (match_dup 1) (match_dup 2))
	  (match_dup 3)))]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[2] = gen_lowpart (SImode, operands[2]);
  operands[3] = gen_lowpart (SImode, operands[3]);
}
  [(set_attr "type" "lea")
   (set_attr "mode" "SI")])

(define_insn_and_split "*lea<mode>_general_2"
  [(set (match_operand:SWI12 0 "register_operand" "=r")
	(plus:SWI12
	  (mult:SWI12 (match_operand:SWI12 1 "index_register_operand" "l")
		      (match_operand 2 "const248_operand" "n"))
	  (match_operand:SWI12 3 "nonmemory_operand" "ri")))]
  "!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun)"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(plus:SI
	  (mult:SI (match_dup 1) (match_dup 2))
	  (match_dup 3)))]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[3] = gen_lowpart (SImode, operands[3]);
}
  [(set_attr "type" "lea")
   (set_attr "mode" "SI")])

(define_insn_and_split "*lea<mode>_general_2b"
  [(set (match_operand:SWI12 0 "register_operand" "=r")
	(plus:SWI12
	  (ashift:SWI12 (match_operand:SWI12 1 "index_register_operand" "l")
			(match_operand 2 "const123_operand" "n"))
	  (match_operand:SWI12 3 "nonmemory_operand" "ri")))]
  "!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun)"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(plus:SI
	  (ashift:SI (match_dup 1) (match_dup 2))
	  (match_dup 3)))]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[3] = gen_lowpart (SImode, operands[3]);
}
  [(set_attr "type" "lea")
   (set_attr "mode" "SI")])

(define_insn_and_split "*lea<mode>_general_3"
  [(set (match_operand:SWI12 0 "register_operand" "=r")
	(plus:SWI12
	  (plus:SWI12
	    (mult:SWI12 (match_operand:SWI12 1 "index_register_operand" "l")
			(match_operand 2 "const248_operand" "n"))
	    (match_operand:SWI12 3 "register_operand" "r"))
	  (match_operand:SWI12 4 "immediate_operand" "i")))]
  "!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun)"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(plus:SI
	  (plus:SI
	    (mult:SI (match_dup 1) (match_dup 2))
	    (match_dup 3))
	  (match_dup 4)))]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[3] = gen_lowpart (SImode, operands[3]);
  operands[4] = gen_lowpart (SImode, operands[4]);
}
  [(set_attr "type" "lea")
   (set_attr "mode" "SI")])

(define_insn_and_split "*lea<mode>_general_3b"
  [(set (match_operand:SWI12 0 "register_operand" "=r")
	(plus:SWI12
	  (plus:SWI12
	    (ashift:SWI12 (match_operand:SWI12 1 "index_register_operand" "l")
			  (match_operand 2 "const123_operand" "n"))
	    (match_operand:SWI12 3 "register_operand" "r"))
	  (match_operand:SWI12 4 "immediate_operand" "i")))]
  "!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun)"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(plus:SI
	  (plus:SI
	    (ashift:SI (match_dup 1) (match_dup 2))
	    (match_dup 3))
	  (match_dup 4)))]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[3] = gen_lowpart (SImode, operands[3]);
  operands[4] = gen_lowpart (SImode, operands[4]);
}
  [(set_attr "type" "lea")
   (set_attr "mode" "SI")])

(define_insn_and_split "*lea<mode>_general_4"
  [(set (match_operand:SWI12 0 "register_operand" "=r")
	(any_or:SWI12
	  (ashift:SWI12
	    (match_operand:SWI12 1 "index_register_operand" "l")
	    (match_operand 2 "const_0_to_3_operand" "n"))
	  (match_operand 3 "const_int_operand" "n")))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && ((unsigned HOST_WIDE_INT) INTVAL (operands[3])
       < (HOST_WIDE_INT_1U << INTVAL (operands[2])))"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(plus:SI
	  (mult:SI (match_dup 1) (match_dup 2))
	  (match_dup 3)))]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[2] = GEN_INT (1 << INTVAL (operands[2]));
}
  [(set_attr "type" "lea")
   (set_attr "mode" "SI")])

(define_insn_and_split "*lea<mode>_general_4"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(any_or:SWI48
	  (ashift:SWI48
	    (match_operand:SWI48 1 "index_register_operand" "l")
	    (match_operand 2 "const_0_to_3_operand" "n"))
	  (match_operand 3 "const_int_operand" "n")))]
  "(unsigned HOST_WIDE_INT) INTVAL (operands[3])
   < (HOST_WIDE_INT_1U << INTVAL (operands[2]))"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(plus:SWI48
	  (mult:SWI48 (match_dup 1) (match_dup 2))
	  (match_dup 3)))]
  "operands[2] = GEN_INT (1 << INTVAL (operands[2]));"
  [(set_attr "type" "lea")
   (set_attr "mode" "<MODE>")])

;; Subtract instructions

(define_expand "sub<mode>3"
  [(set (match_operand:SDWIM 0 "nonimmediate_operand")
	(minus:SDWIM (match_operand:SDWIM 1 "nonimmediate_operand")
		     (match_operand:SDWIM 2 "<general_hilo_operand>")))]
  ""
  "ix86_expand_binary_operator (MINUS, <MODE>mode, operands); DONE;")

(define_insn_and_split "*sub<dwi>3_doubleword"
  [(set (match_operand:<DWI> 0 "nonimmediate_operand" "=r,o")
	(minus:<DWI>
	  (match_operand:<DWI> 1 "nonimmediate_operand" "0,0")
	  (match_operand:<DWI> 2 "x86_64_hilo_general_operand"
							"ro<di>,r<di>")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (MINUS, <MODE>mode, operands)"
  "#"
  "reload_completed"
  [(parallel [(set (reg:CC FLAGS_REG)
		   (compare:CC (match_dup 1) (match_dup 2)))
	      (set (match_dup 0)
		   (minus:DWIH (match_dup 1) (match_dup 2)))])
   (parallel [(set (match_dup 3)
		   (minus:DWIH
		     (minus:DWIH
		       (match_dup 4)
		       (ltu:DWIH (reg:CC FLAGS_REG) (const_int 0)))
		     (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  split_double_mode (<DWI>mode, &operands[0], 3, &operands[0], &operands[3]);
  if (operands[2] == const0_rtx)
    {
      ix86_expand_binary_operator (MINUS, <MODE>mode, &operands[3]);
      DONE;
    }
})

(define_insn "*sub<mode>_1"
  [(set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(minus:SWI
	  (match_operand:SWI 1 "nonimmediate_operand" "0,0")
	  (match_operand:SWI 2 "<general_operand>" "<r><i>,<r>m")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (MINUS, <MODE>mode, operands)"
  "sub{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*subsi_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (minus:SI (match_operand:SI 1 "register_operand" "0")
		    (match_operand:SI 2 "x86_64_general_operand" "rme"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (MINUS, SImode, operands)"
  "sub{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*subqi_1_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+qm,q"))
	(minus:QI (match_dup 0)
		  (match_operand:QI 1 "general_operand" "qn,qm")))
   (clobber (reg:CC FLAGS_REG))]
  "(! TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "sub{b}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "mode" "QI")])

(define_insn "*sub<mode>_2"
  [(set (reg FLAGS_REG)
	(compare
	  (minus:SWI
	    (match_operand:SWI 1 "nonimmediate_operand" "0,0")
	    (match_operand:SWI 2 "<general_operand>" "<r><i>,<r>m"))
	  (const_int 0)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(minus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (MINUS, <MODE>mode, operands)"
  "sub{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*subsi_2_zext"
  [(set (reg FLAGS_REG)
	(compare
	  (minus:SI (match_operand:SI 1 "register_operand" "0")
		    (match_operand:SI 2 "x86_64_general_operand" "rme"))
	  (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (minus:SI (match_dup 1)
		    (match_dup 2))))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (MINUS, SImode, operands)"
  "sub{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

;; Subtract with jump on overflow.
(define_expand "subv<mode>4"
  [(parallel [(set (reg:CCO FLAGS_REG)
		   (eq:CCO (minus:<DWI>
			      (sign_extend:<DWI>
				 (match_operand:SWI 1 "nonimmediate_operand"))
			      (match_dup 4))
			   (sign_extend:<DWI>
			      (minus:SWI (match_dup 1)
					 (match_operand:SWI 2
					    "<general_operand>")))))
	      (set (match_operand:SWI 0 "register_operand")
		   (minus:SWI (match_dup 1) (match_dup 2)))])
   (set (pc) (if_then_else
	       (eq (reg:CCO FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 3))
	       (pc)))]
  ""
{
  ix86_fixup_binary_operands_no_copy (MINUS, <MODE>mode, operands);
  if (CONST_INT_P (operands[2]))
    operands[4] = operands[2];
  else
    operands[4] = gen_rtx_SIGN_EXTEND (<DWI>mode, operands[2]);
})

(define_insn "*subv<mode>4"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (minus:<DWI>
		   (sign_extend:<DWI>
		      (match_operand:SWI 1 "nonimmediate_operand" "0,0"))
		   (sign_extend:<DWI>
		      (match_operand:SWI 2 "<general_sext_operand>"
					   "<r>We,<r>m")))
		(sign_extend:<DWI>
		   (minus:SWI (match_dup 1) (match_dup 2)))))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(minus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_binary_operator_ok (MINUS, <MODE>mode, operands)"
  "sub{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*subv<mode>4_1"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (minus:<DWI>
		   (sign_extend:<DWI>
		      (match_operand:SWI 1 "nonimmediate_operand" "0"))
		   (match_operand:<DWI> 3 "const_int_operand" "i"))
		(sign_extend:<DWI>
		   (minus:SWI (match_dup 1)
			      (match_operand:SWI 2 "x86_64_immediate_operand"
						   "<i>")))))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(minus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_binary_operator_ok (MINUS, <MODE>mode, operands)
   && CONST_INT_P (operands[2])
   && INTVAL (operands[2]) == INTVAL (operands[3])"
  "sub{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")
   (set (attr "length_immediate")
	(cond [(match_test "IN_RANGE (INTVAL (operands[2]), -128, 127)")
		  (const_string "1")
	       (match_test "<MODE_SIZE> == 8")
		  (const_string "4")]
	      (const_string "<MODE_SIZE>")))])

(define_expand "usubv<mode>4"
  [(parallel [(set (reg:CC FLAGS_REG)
		   (compare:CC
		     (match_operand:SWI 1 "nonimmediate_operand")
		     (match_operand:SWI 2 "<general_operand>")))
	      (set (match_operand:SWI 0 "register_operand")
		   (minus:SWI (match_dup 1) (match_dup 2)))])
   (set (pc) (if_then_else
	       (ltu (reg:CC FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 3))
	       (pc)))]
  ""
  "ix86_fixup_binary_operands_no_copy (MINUS, <MODE>mode, operands);")

(define_insn "*sub<mode>_3"
  [(set (reg FLAGS_REG)
	(compare (match_operand:SWI 1 "nonimmediate_operand" "0,0")
		 (match_operand:SWI 2 "<general_operand>" "<r><i>,<r>m")))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(minus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_match_ccmode (insn, CCmode)
   && ix86_binary_operator_ok (MINUS, <MODE>mode, operands)"
  "sub{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_peephole2
  [(parallel
     [(set (reg:CC FLAGS_REG)
	   (compare:CC (match_operand:SWI 0 "general_reg_operand")
		       (match_operand:SWI 1 "general_gr_operand")))
      (set (match_dup 0)
	   (minus:SWI (match_dup 0) (match_dup 1)))])]
  "find_regno_note (peep2_next_insn (0), REG_UNUSED, REGNO (operands[0])) != 0"
  [(set (reg:CC FLAGS_REG)
	(compare:CC (match_dup 0) (match_dup 1)))])

(define_insn "*subsi_3_zext"
  [(set (reg FLAGS_REG)
	(compare (match_operand:SI 1 "register_operand" "0")
		 (match_operand:SI 2 "x86_64_general_operand" "rme")))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (minus:SI (match_dup 1)
		    (match_dup 2))))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCmode)
   && ix86_binary_operator_ok (MINUS, SImode, operands)"
  "sub{l}\t{%2, %1|%1, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

;; Add with carry and subtract with borrow

(define_insn "add<mode>3_carry"
  [(set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(plus:SWI
	  (plus:SWI
	    (match_operator:SWI 4 "ix86_carry_flag_operator"
	     [(match_operand 3 "flags_reg_operand") (const_int 0)])
	    (match_operand:SWI 1 "nonimmediate_operand" "%0,0"))
	  (match_operand:SWI 2 "<general_operand>" "<r><i>,<r>m")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)"
  "adc{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "use_carry" "1")
   (set_attr "pent_pair" "pu")
   (set_attr "mode" "<MODE>")])

(define_insn "*addsi3_carry_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (plus:SI
	    (plus:SI (match_operator:SI 3 "ix86_carry_flag_operator"
		      [(reg FLAGS_REG) (const_int 0)])
		     (match_operand:SI 1 "register_operand" "%0"))
	    (match_operand:SI 2 "x86_64_general_operand" "rme"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (PLUS, SImode, operands)"
  "adc{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "use_carry" "1")
   (set_attr "pent_pair" "pu")
   (set_attr "mode" "SI")])

;; There is no point to generate ADCX instruction. ADC is shorter and faster.

(define_insn "addcarry<mode>"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extend:<DWI>
	    (plus:SWI48
	      (plus:SWI48
		(match_operator:SWI48 5 "ix86_carry_flag_operator"
		  [(match_operand 3 "flags_reg_operand") (const_int 0)])
		(match_operand:SWI48 1 "nonimmediate_operand" "%0"))
	      (match_operand:SWI48 2 "nonimmediate_operand" "rm")))
	  (plus:<DWI>
	    (zero_extend:<DWI> (match_dup 2))
	    (match_operator:<DWI> 4 "ix86_carry_flag_operator"
	      [(match_dup 3) (const_int 0)]))))
   (set (match_operand:SWI48 0 "register_operand" "=r")
	(plus:SWI48 (plus:SWI48 (match_op_dup 5
				 [(match_dup 3) (const_int 0)])
				(match_dup 1))
		    (match_dup 2)))]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)"
  "adc{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "use_carry" "1")
   (set_attr "pent_pair" "pu")
   (set_attr "mode" "<MODE>")])

(define_expand "addcarry<mode>_0"
  [(parallel
     [(set (reg:CCC FLAGS_REG)
	   (compare:CCC
	     (plus:SWI48
	       (match_operand:SWI48 1 "nonimmediate_operand")
	       (match_operand:SWI48 2 "x86_64_general_operand"))
	     (match_dup 1)))
      (set (match_operand:SWI48 0 "register_operand")
	   (plus:SWI48 (match_dup 1) (match_dup 2)))])]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)")

(define_insn "sub<mode>3_carry"
  [(set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(minus:SWI
	  (minus:SWI
	    (match_operand:SWI 1 "nonimmediate_operand" "0,0")
	    (match_operator:SWI 4 "ix86_carry_flag_operator"
	     [(match_operand 3 "flags_reg_operand") (const_int 0)]))
	  (match_operand:SWI 2 "<general_operand>" "<r><i>,<r>m")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (MINUS, <MODE>mode, operands)"
  "sbb{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "use_carry" "1")
   (set_attr "pent_pair" "pu")
   (set_attr "mode" "<MODE>")])

(define_insn "*subsi3_carry_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (minus:SI
	    (minus:SI
	      (match_operand:SI 1 "register_operand" "0")
	      (match_operator:SI 3 "ix86_carry_flag_operator"
	       [(reg FLAGS_REG) (const_int 0)]))
	    (match_operand:SI 2 "x86_64_general_operand" "rme"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (MINUS, SImode, operands)"
  "sbb{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "use_carry" "1")
   (set_attr "pent_pair" "pu")
   (set_attr "mode" "SI")])

(define_insn "sub<mode>3_carry_ccc"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extend:<DWI> (match_operand:DWIH 1 "register_operand" "0"))
	  (plus:<DWI>
	    (ltu:<DWI> (reg:CC FLAGS_REG) (const_int 0))
	    (zero_extend:<DWI>
	      (match_operand:DWIH 2 "x86_64_sext_operand" "rmWe")))))
   (clobber (match_scratch:DWIH 0 "=r"))]
  ""
  "sbb{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*sub<mode>3_carry_ccc_1"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extend:<DWI> (match_operand:DWIH 1 "register_operand" "0"))
	  (plus:<DWI>
	    (ltu:<DWI> (reg:CC FLAGS_REG) (const_int 0))
	    (match_operand:<DWI> 2 "x86_64_dwzext_immediate_operand" "Wf"))))
   (clobber (match_scratch:DWIH 0 "=r"))]
  ""
{
  operands[3] = simplify_subreg (<MODE>mode, operands[2], <DWI>mode, 0);
  return "sbb{<imodesuffix>}\t{%3, %0|%0, %3}";
}
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

;; The sign flag is set from the
;; (compare (match_dup 1) (plus:DWIH (ltu:DWIH ...) (match_dup 2)))
;; result, the overflow flag likewise, but the overflow flag is also
;; set if the (plus:DWIH (ltu:DWIH ...) (match_dup 2)) overflows.
(define_insn "sub<mode>3_carry_ccgz"
  [(set (reg:CCGZ FLAGS_REG)
	(unspec:CCGZ [(match_operand:DWIH 1 "register_operand" "0")
		      (match_operand:DWIH 2 "x86_64_general_operand" "rme")
		      (ltu:DWIH (reg:CC FLAGS_REG) (const_int 0))]
		     UNSPEC_SBB))
   (clobber (match_scratch:DWIH 0 "=r"))]
  ""
  "sbb{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "subborrow<mode>"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extend:<DWI>
	    (match_operand:SWI48 1 "nonimmediate_operand" "0"))
	  (plus:<DWI>
	    (match_operator:<DWI> 4 "ix86_carry_flag_operator"
	      [(match_operand 3 "flags_reg_operand") (const_int 0)])
	    (zero_extend:<DWI>
	      (match_operand:SWI48 2 "nonimmediate_operand" "rm")))))
   (set (match_operand:SWI48 0 "register_operand" "=r")
	(minus:SWI48 (minus:SWI48
		       (match_dup 1)
		       (match_operator:SWI48 5 "ix86_carry_flag_operator"
			 [(match_dup 3) (const_int 0)]))
		     (match_dup 2)))]
  "ix86_binary_operator_ok (MINUS, <MODE>mode, operands)"
  "sbb{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "use_carry" "1")
   (set_attr "pent_pair" "pu")
   (set_attr "mode" "<MODE>")])

(define_expand "subborrow<mode>_0"
  [(parallel
     [(set (reg:CC FLAGS_REG)
	   (compare:CC
	     (match_operand:SWI48 1 "nonimmediate_operand")
	     (match_operand:SWI48 2 "<general_operand>")))
      (set (match_operand:SWI48 0 "register_operand")
	   (minus:SWI48 (match_dup 1) (match_dup 2)))])]
  "ix86_binary_operator_ok (MINUS, <MODE>mode, operands)")

;; Overflow setting add instructions

(define_expand "addqi3_cconly_overflow"
  [(parallel
     [(set (reg:CCC FLAGS_REG)
	   (compare:CCC
	     (plus:QI
	       (match_operand:QI 0 "nonimmediate_operand")
	       (match_operand:QI 1 "general_operand"))
	     (match_dup 0)))
      (clobber (match_scratch:QI 2))])]
  "!(MEM_P (operands[0]) && MEM_P (operands[1]))")

(define_insn "*add<mode>3_cconly_overflow_1"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (plus:SWI
	    (match_operand:SWI 1 "nonimmediate_operand" "%0")
	    (match_operand:SWI 2 "<general_operand>" "<g>"))
	  (match_dup 1)))
   (clobber (match_scratch:SWI 0 "=<r>"))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "add{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*add<mode>3_cc_overflow_1"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	    (plus:SWI
		(match_operand:SWI 1 "nonimmediate_operand" "%0,0")
		(match_operand:SWI 2 "<general_operand>" "<r><i>,<r>m"))
	    (match_dup 1)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(plus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)"
  "add{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*addsi3_zext_cc_overflow_1"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (plus:SI
	    (match_operand:SI 1 "nonimmediate_operand" "%0")
	    (match_operand:SI 2 "x86_64_general_operand" "rme"))
	  (match_dup 1)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (plus:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT && ix86_binary_operator_ok (PLUS, SImode, operands)"
  "add{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*add<mode>3_cconly_overflow_2"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (plus:SWI
	    (match_operand:SWI 1 "nonimmediate_operand" "%0")
	    (match_operand:SWI 2 "<general_operand>" "<g>"))
	  (match_dup 2)))
   (clobber (match_scratch:SWI 0 "=<r>"))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "add{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*add<mode>3_cc_overflow_2"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	    (plus:SWI
		(match_operand:SWI 1 "nonimmediate_operand" "%0,0")
		(match_operand:SWI 2 "<general_operand>" "<r><i>,<r>m"))
	    (match_dup 2)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m,<r>")
	(plus:SWI (match_dup 1) (match_dup 2)))]
  "ix86_binary_operator_ok (PLUS, <MODE>mode, operands)"
  "add{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*addsi3_zext_cc_overflow_2"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (plus:SI
	    (match_operand:SI 1 "nonimmediate_operand" "%0")
	    (match_operand:SI 2 "x86_64_general_operand" "rme"))
	  (match_dup 2)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (plus:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT && ix86_binary_operator_ok (PLUS, SImode, operands)"
  "add{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

;; The patterns that match these are at the end of this file.

(define_expand "<plusminus_insn>xf3"
  [(set (match_operand:XF 0 "register_operand")
	(plusminus:XF
	  (match_operand:XF 1 "register_operand")
	  (match_operand:XF 2 "register_operand")))]
  "TARGET_80387")

(define_expand "<plusminus_insn><mode>3"
  [(set (match_operand:MODEF 0 "register_operand")
	(plusminus:MODEF
	  (match_operand:MODEF 1 "register_operand")
	  (match_operand:MODEF 2 "nonimmediate_operand")))]
  "(TARGET_80387 && X87_ENABLE_ARITH (<MODE>mode))
    || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)")

;; Multiply instructions

(define_expand "mul<mode>3"
  [(parallel [(set (match_operand:SWIM248 0 "register_operand")
		   (mult:SWIM248
		     (match_operand:SWIM248 1 "register_operand")
		     (match_operand:SWIM248 2 "<general_operand>")))
	      (clobber (reg:CC FLAGS_REG))])])

(define_expand "mulqi3"
  [(parallel [(set (match_operand:QI 0 "register_operand")
		   (mult:QI
		     (match_operand:QI 1 "register_operand")
		     (match_operand:QI 2 "nonimmediate_operand")))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_QIMODE_MATH")

;; On AMDFAM10
;; IMUL reg32/64, reg32/64, imm8 	Direct
;; IMUL reg32/64, mem32/64, imm8 	VectorPath
;; IMUL reg32/64, reg32/64, imm32 	Direct
;; IMUL reg32/64, mem32/64, imm32 	VectorPath
;; IMUL reg32/64, reg32/64 		Direct
;; IMUL reg32/64, mem32/64 		Direct
;;
;; On BDVER1, all above IMULs use DirectPath
;;
;; On AMDFAM10
;; IMUL reg16, reg16, imm8 	VectorPath
;; IMUL reg16, mem16, imm8 	VectorPath
;; IMUL reg16, reg16, imm16 	VectorPath
;; IMUL reg16, mem16, imm16 	VectorPath
;; IMUL reg16, reg16 		Direct
;; IMUL reg16, mem16 		Direct
;;
;; On BDVER1, all HI MULs use DoublePath

(define_insn "*mul<mode>3_1"
  [(set (match_operand:SWIM248 0 "register_operand" "=r,r,r")
	(mult:SWIM248
	  (match_operand:SWIM248 1 "nonimmediate_operand" "%rm,rm,0")
	  (match_operand:SWIM248 2 "<general_operand>" "K,<i>,mr")))
   (clobber (reg:CC FLAGS_REG))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "@
   imul{<imodesuffix>}\t{%2, %1, %0|%0, %1, %2}
   imul{<imodesuffix>}\t{%2, %1, %0|%0, %1, %2}
   imul{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "imul")
   (set_attr "prefix_0f" "0,0,1")
   (set (attr "athlon_decode")
	(cond [(eq_attr "cpu" "athlon")
		  (const_string "vector")
	       (eq_attr "alternative" "1")
		  (const_string "vector")
	       (and (eq_attr "alternative" "2")
	       	    (ior (match_test "<MODE>mode == HImode")
		         (match_operand 1 "memory_operand")))
		  (const_string "vector")]
	      (const_string "direct")))
   (set (attr "amdfam10_decode")
	(cond [(and (eq_attr "alternative" "0,1")
	      	    (ior (match_test "<MODE>mode == HImode")
		         (match_operand 1 "memory_operand")))
		  (const_string "vector")]
	      (const_string "direct")))
   (set (attr "bdver1_decode")
   	(if_then_else
	  (match_test "<MODE>mode == HImode")
	    (const_string "double")
	    (const_string "direct")))
   (set_attr "mode" "<MODE>")])

(define_insn "*mulsi3_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r,r,r")
	(zero_extend:DI
	  (mult:SI (match_operand:SI 1 "nonimmediate_operand" "%rm,rm,0")
		   (match_operand:SI 2 "x86_64_general_operand" "K,e,mr"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "@
   imul{l}\t{%2, %1, %k0|%k0, %1, %2}
   imul{l}\t{%2, %1, %k0|%k0, %1, %2}
   imul{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "imul")
   (set_attr "prefix_0f" "0,0,1")
   (set (attr "athlon_decode")
	(cond [(eq_attr "cpu" "athlon")
		  (const_string "vector")
	       (eq_attr "alternative" "1")
		  (const_string "vector")
	       (and (eq_attr "alternative" "2")
		    (match_operand 1 "memory_operand"))
		  (const_string "vector")]
	      (const_string "direct")))
   (set (attr "amdfam10_decode")
	(cond [(and (eq_attr "alternative" "0,1")
		    (match_operand 1 "memory_operand"))
		  (const_string "vector")]
	      (const_string "direct")))
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "SI")])

;;On AMDFAM10 and BDVER1
;; MUL reg8 	Direct
;; MUL mem8 	Direct

(define_insn "*mulqi3_1"
  [(set (match_operand:QI 0 "register_operand" "=a")
	(mult:QI (match_operand:QI 1 "nonimmediate_operand" "%0")
		 (match_operand:QI 2 "nonimmediate_operand" "qm")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_QIMODE_MATH
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "mul{b}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
        (const_string "vector")
        (const_string "direct")))
   (set_attr "amdfam10_decode" "direct")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "QI")])

;; Multiply with jump on overflow.
(define_expand "mulv<mode>4"
  [(parallel [(set (reg:CCO FLAGS_REG)
		   (eq:CCO (mult:<DWI>
			      (sign_extend:<DWI>
				 (match_operand:SWI248 1 "register_operand"))
			      (match_dup 4))
			   (sign_extend:<DWI>
			      (mult:SWI248 (match_dup 1)
					   (match_operand:SWI248 2
					      "<general_operand>")))))
	      (set (match_operand:SWI248 0 "register_operand")
		   (mult:SWI248 (match_dup 1) (match_dup 2)))])
   (set (pc) (if_then_else
	       (eq (reg:CCO FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 3))
	       (pc)))]
  ""
{
  if (CONST_INT_P (operands[2]))
    operands[4] = operands[2];
  else
    operands[4] = gen_rtx_SIGN_EXTEND (<DWI>mode, operands[2]);
})

(define_insn "*mulv<mode>4"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (mult:<DWI>
		   (sign_extend:<DWI>
		      (match_operand:SWI48 1 "nonimmediate_operand" "%rm,0"))
		   (sign_extend:<DWI>
		      (match_operand:SWI48 2 "x86_64_sext_operand" "We,mr")))
		(sign_extend:<DWI>
		   (mult:SWI48 (match_dup 1) (match_dup 2)))))
   (set (match_operand:SWI48 0 "register_operand" "=r,r")
	(mult:SWI48 (match_dup 1) (match_dup 2)))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "@
   imul{<imodesuffix>}\t{%2, %1, %0|%0, %1, %2}
   imul{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "imul")
   (set_attr "prefix_0f" "0,1")
   (set (attr "athlon_decode")
	(cond [(eq_attr "cpu" "athlon")
		  (const_string "vector")
	       (eq_attr "alternative" "0")
		  (const_string "vector")
	       (and (eq_attr "alternative" "1")
		    (match_operand 1 "memory_operand"))
		  (const_string "vector")]
	      (const_string "direct")))
   (set (attr "amdfam10_decode")
	(cond [(and (eq_attr "alternative" "1")
		    (match_operand 1 "memory_operand"))
		  (const_string "vector")]
	      (const_string "direct")))
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "<MODE>")])

(define_insn "*mulvhi4"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (mult:SI
		   (sign_extend:SI
		      (match_operand:HI 1 "nonimmediate_operand" "%0"))
		   (sign_extend:SI
		      (match_operand:HI 2 "nonimmediate_operand" "mr")))
		(sign_extend:SI
		   (mult:HI (match_dup 1) (match_dup 2)))))
   (set (match_operand:HI 0 "register_operand" "=r")
	(mult:HI (match_dup 1) (match_dup 2)))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "imul{w}\t{%2, %0|%0, %2}"
  [(set_attr "type" "imul")
   (set_attr "prefix_0f" "1")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "direct")
   (set_attr "bdver1_decode" "double")
   (set_attr "mode" "HI")])

(define_insn "*mulv<mode>4_1"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (mult:<DWI>
		   (sign_extend:<DWI>
		      (match_operand:SWI248 1 "nonimmediate_operand" "rm,rm"))
		   (match_operand:<DWI> 3 "const_int_operand" "K,i"))
		(sign_extend:<DWI>
		   (mult:SWI248 (match_dup 1)
				(match_operand:SWI248 2
				   "<immediate_operand>" "K,<i>")))))
   (set (match_operand:SWI248 0 "register_operand" "=r,r")
	(mult:SWI248 (match_dup 1) (match_dup 2)))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))
   && CONST_INT_P (operands[2])
   && INTVAL (operands[2]) == INTVAL (operands[3])"
  "imul{<imodesuffix>}\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "type" "imul")
   (set (attr "prefix_0f")
   	(if_then_else
	  (match_test "<MODE>mode == HImode")
	    (const_string "0")
	    (const_string "*")))
   (set (attr "athlon_decode")
	(cond [(eq_attr "cpu" "athlon")
		  (const_string "vector")
	       (eq_attr "alternative" "1")
		  (const_string "vector")]
	      (const_string "direct")))
   (set (attr "amdfam10_decode")
	(cond [(ior (match_test "<MODE>mode == HImode")
		    (match_operand 1 "memory_operand"))
		  (const_string "vector")]
	      (const_string "direct")))
   (set (attr "bdver1_decode")
   	(if_then_else
	  (match_test "<MODE>mode == HImode")
	    (const_string "double")
	    (const_string "direct")))
   (set_attr "mode" "<MODE>")
   (set (attr "length_immediate")
	(cond [(eq_attr "alternative" "0")
		  (const_string "1")
	       (match_test "<MODE_SIZE> == 8")
		  (const_string "4")]
	      (const_string "<MODE_SIZE>")))])

(define_expand "umulv<mode>4"
  [(parallel [(set (reg:CCO FLAGS_REG)
		   (eq:CCO (mult:<DWI>
			      (zero_extend:<DWI>
				 (match_operand:SWI248 1
						      "nonimmediate_operand"))
			      (zero_extend:<DWI>
				 (match_operand:SWI248 2
						      "nonimmediate_operand")))
			   (zero_extend:<DWI>
			      (mult:SWI248 (match_dup 1) (match_dup 2)))))
	      (set (match_operand:SWI248 0 "register_operand")
		   (mult:SWI248 (match_dup 1) (match_dup 2)))
	      (clobber (match_scratch:SWI248 4))])
   (set (pc) (if_then_else
	       (eq (reg:CCO FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 3))
	       (pc)))]
  ""
{
  if (MEM_P (operands[1]) && MEM_P (operands[2]))
    operands[1] = force_reg (<MODE>mode, operands[1]);
})

(define_insn "*umulv<mode>4"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (mult:<DWI>
		   (zero_extend:<DWI>
		      (match_operand:SWI248 1 "nonimmediate_operand" "%0"))
		   (zero_extend:<DWI>
		      (match_operand:SWI248 2 "nonimmediate_operand" "rm")))
		(zero_extend:<DWI>
		   (mult:SWI248 (match_dup 1) (match_dup 2)))))
   (set (match_operand:SWI248 0 "register_operand" "=a")
	(mult:SWI248 (match_dup 1) (match_dup 2)))
   (clobber (match_scratch:SWI248 3 "=d"))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "mul{<imodesuffix>}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
       (const_string "vector")
       (const_string "double")))
   (set_attr "amdfam10_decode" "double")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "<MODE>")])

(define_expand "<u>mulvqi4"
  [(parallel [(set (reg:CCO FLAGS_REG)
		   (eq:CCO (mult:HI
			      (any_extend:HI
				 (match_operand:QI 1 "nonimmediate_operand"))
			      (any_extend:HI
				 (match_operand:QI 2 "nonimmediate_operand")))
			   (any_extend:HI
			      (mult:QI (match_dup 1) (match_dup 2)))))
	      (set (match_operand:QI 0 "register_operand")
		   (mult:QI (match_dup 1) (match_dup 2)))])
   (set (pc) (if_then_else
	       (eq (reg:CCO FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 3))
	       (pc)))]
  "TARGET_QIMODE_MATH"
{
  if (MEM_P (operands[1]) && MEM_P (operands[2]))
    operands[1] = force_reg (QImode, operands[1]);
})

(define_insn "*<u>mulvqi4"
  [(set (reg:CCO FLAGS_REG)
	(eq:CCO (mult:HI
		   (any_extend:HI
		      (match_operand:QI 1 "nonimmediate_operand" "%0"))
		   (any_extend:HI
		      (match_operand:QI 2 "nonimmediate_operand" "qm")))
		(any_extend:HI
		   (mult:QI (match_dup 1) (match_dup 2)))))
   (set (match_operand:QI 0 "register_operand" "=a")
	(mult:QI (match_dup 1) (match_dup 2)))]
  "TARGET_QIMODE_MATH
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "<sgnprefix>mul{b}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
	(const_string "vector")
	(const_string "direct")))
   (set_attr "amdfam10_decode" "direct")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "QI")])

(define_expand "<u>mul<mode><dwi>3"
  [(parallel [(set (match_operand:<DWI> 0 "register_operand")
		   (mult:<DWI>
		     (any_extend:<DWI>
		       (match_operand:DWIH 1 "nonimmediate_operand"))
		     (any_extend:<DWI>
		       (match_operand:DWIH 2 "register_operand"))))
	      (clobber (reg:CC FLAGS_REG))])])

(define_expand "<u>mulqihi3"
  [(parallel [(set (match_operand:HI 0 "register_operand")
		   (mult:HI
		     (any_extend:HI
		       (match_operand:QI 1 "nonimmediate_operand"))
		     (any_extend:HI
		       (match_operand:QI 2 "register_operand"))))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_QIMODE_MATH")

(define_insn "*bmi2_umul<mode><dwi>3_1"
  [(set (match_operand:DWIH 0 "register_operand" "=r")
	(mult:DWIH
	  (match_operand:DWIH 2 "nonimmediate_operand" "%d")
	  (match_operand:DWIH 3 "nonimmediate_operand" "rm")))
   (set (match_operand:DWIH 1 "register_operand" "=r")
	(truncate:DWIH
	  (lshiftrt:<DWI>
	    (mult:<DWI> (zero_extend:<DWI> (match_dup 2))
			(zero_extend:<DWI> (match_dup 3)))
	    (match_operand:QI 4 "const_int_operand" "n"))))]
  "TARGET_BMI2 && INTVAL (operands[4]) == <MODE_SIZE> * BITS_PER_UNIT
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "mulx\t{%3, %0, %1|%1, %0, %3}"
  [(set_attr "type" "imulx")
   (set_attr "prefix" "vex")
   (set_attr "mode" "<MODE>")])

(define_insn "*umul<mode><dwi>3_1"
  [(set (match_operand:<DWI> 0 "register_operand" "=r,A")
	(mult:<DWI>
	  (zero_extend:<DWI>
	    (match_operand:DWIH 1 "nonimmediate_operand" "%d,0"))
	  (zero_extend:<DWI>
	    (match_operand:DWIH 2 "nonimmediate_operand" "rm,rm"))))
   (clobber (reg:CC FLAGS_REG))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "@
   #
   mul{<imodesuffix>}\t%2"
  [(set_attr "isa" "bmi2,*")
   (set_attr "type" "imulx,imul")
   (set_attr "length_immediate" "*,0")
   (set (attr "athlon_decode")
	(cond [(eq_attr "alternative" "1")
		 (if_then_else (eq_attr "cpu" "athlon")
		   (const_string "vector")
		   (const_string "double"))]
	      (const_string "*")))
   (set_attr "amdfam10_decode" "*,double")
   (set_attr "bdver1_decode" "*,direct")
   (set_attr "prefix" "vex,orig")
   (set_attr "mode" "<MODE>")])

;; Convert mul to the mulx pattern to avoid flags dependency.
(define_split
 [(set (match_operand:<DWI> 0 "register_operand")
       (mult:<DWI>
	 (zero_extend:<DWI>
	   (match_operand:DWIH 1 "register_operand"))
	 (zero_extend:<DWI>
	   (match_operand:DWIH 2 "nonimmediate_operand"))))
  (clobber (reg:CC FLAGS_REG))]
 "TARGET_BMI2 && reload_completed
  && REGNO (operands[1]) == DX_REG"
  [(parallel [(set (match_dup 3)
		   (mult:DWIH (match_dup 1) (match_dup 2)))
	      (set (match_dup 4)
		   (truncate:DWIH
		     (lshiftrt:<DWI>
		       (mult:<DWI> (zero_extend:<DWI> (match_dup 1))
				   (zero_extend:<DWI> (match_dup 2)))
		       (match_dup 5))))])]
{
  split_double_mode (<DWI>mode, &operands[0], 1, &operands[3], &operands[4]);

  operands[5] = GEN_INT (<MODE_SIZE> * BITS_PER_UNIT);
})

(define_insn "*mul<mode><dwi>3_1"
  [(set (match_operand:<DWI> 0 "register_operand" "=A")
	(mult:<DWI>
	  (sign_extend:<DWI>
	    (match_operand:DWIH 1 "nonimmediate_operand" "%0"))
	  (sign_extend:<DWI>
	    (match_operand:DWIH 2 "nonimmediate_operand" "rm"))))
   (clobber (reg:CC FLAGS_REG))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "imul{<imodesuffix>}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
        (const_string "vector")
        (const_string "double")))
   (set_attr "amdfam10_decode" "double")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "<MODE>")])

(define_insn "*<u>mulqihi3_1"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(mult:HI
	  (any_extend:HI
	    (match_operand:QI 1 "nonimmediate_operand" "%0"))
	  (any_extend:HI
	    (match_operand:QI 2 "nonimmediate_operand" "qm"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_QIMODE_MATH
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "<sgnprefix>mul{b}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
        (const_string "vector")
        (const_string "direct")))
   (set_attr "amdfam10_decode" "direct")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "QI")])

(define_expand "<s>mul<mode>3_highpart"
  [(parallel [(set (match_operand:SWI48 0 "register_operand")
		   (truncate:SWI48
		     (lshiftrt:<DWI>
		       (mult:<DWI>
			 (any_extend:<DWI>
			   (match_operand:SWI48 1 "nonimmediate_operand"))
			 (any_extend:<DWI>
			   (match_operand:SWI48 2 "register_operand")))
		       (match_dup 3))))
	      (clobber (match_scratch:SWI48 4))
	      (clobber (reg:CC FLAGS_REG))])]
  ""
  "operands[3] = GEN_INT (GET_MODE_BITSIZE (<MODE>mode));")

(define_insn "*<s>muldi3_highpart_1"
  [(set (match_operand:DI 0 "register_operand" "=d")
	(truncate:DI
	  (lshiftrt:TI
	    (mult:TI
	      (any_extend:TI
		(match_operand:DI 1 "nonimmediate_operand" "%a"))
	      (any_extend:TI
		(match_operand:DI 2 "nonimmediate_operand" "rm")))
	    (const_int 64))))
   (clobber (match_scratch:DI 3 "=1"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "<sgnprefix>mul{q}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
        (const_string "vector")
        (const_string "double")))
   (set_attr "amdfam10_decode" "double")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "DI")])

(define_insn "*<s>mulsi3_highpart_zext"
  [(set (match_operand:DI 0 "register_operand" "=d")
	(zero_extend:DI (truncate:SI
	  (lshiftrt:DI
	    (mult:DI (any_extend:DI
		       (match_operand:SI 1 "nonimmediate_operand" "%a"))
		     (any_extend:DI
		       (match_operand:SI 2 "nonimmediate_operand" "rm")))
	    (const_int 32)))))
   (clobber (match_scratch:SI 3 "=1"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "<sgnprefix>mul{l}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
        (const_string "vector")
        (const_string "double")))
   (set_attr "amdfam10_decode" "double")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "SI")])

(define_insn "*<s>mulsi3_highpart_1"
  [(set (match_operand:SI 0 "register_operand" "=d")
	(truncate:SI
	  (lshiftrt:DI
	    (mult:DI
	      (any_extend:DI
		(match_operand:SI 1 "nonimmediate_operand" "%a"))
	      (any_extend:DI
		(match_operand:SI 2 "nonimmediate_operand" "rm")))
	    (const_int 32))))
   (clobber (match_scratch:SI 3 "=1"))
   (clobber (reg:CC FLAGS_REG))]
  "!(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "<sgnprefix>mul{l}\t%2"
  [(set_attr "type" "imul")
   (set_attr "length_immediate" "0")
   (set (attr "athlon_decode")
     (if_then_else (eq_attr "cpu" "athlon")
        (const_string "vector")
        (const_string "double")))
   (set_attr "amdfam10_decode" "double")
   (set_attr "bdver1_decode" "direct")
   (set_attr "mode" "SI")])

;; The patterns that match these are at the end of this file.

(define_expand "mulxf3"
  [(set (match_operand:XF 0 "register_operand")
	(mult:XF (match_operand:XF 1 "register_operand")
		 (match_operand:XF 2 "register_operand")))]
  "TARGET_80387")

(define_expand "mul<mode>3"
  [(set (match_operand:MODEF 0 "register_operand")
	(mult:MODEF (match_operand:MODEF 1 "register_operand")
		    (match_operand:MODEF 2 "nonimmediate_operand")))]
  "(TARGET_80387 && X87_ENABLE_ARITH (<MODE>mode))
    || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)")

;; Divide instructions

;; The patterns that match these are at the end of this file.

(define_expand "divxf3"
  [(set (match_operand:XF 0 "register_operand")
	(div:XF (match_operand:XF 1 "register_operand")
		(match_operand:XF 2 "register_operand")))]
  "TARGET_80387")

(define_expand "div<mode>3"
  [(set (match_operand:MODEF 0 "register_operand")
	(div:MODEF (match_operand:MODEF 1 "register_operand")
		   (match_operand:MODEF 2 "nonimmediate_operand")))]
  "(TARGET_80387 && X87_ENABLE_ARITH (<MODE>mode))
    || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)"
{
  if (<MODE>mode == SFmode
      && TARGET_SSE && TARGET_SSE_MATH
      && TARGET_RECIP_DIV
      && optimize_insn_for_speed_p ()
      && flag_finite_math_only && !flag_trapping_math
      && flag_unsafe_math_optimizations)
    {
      ix86_emit_swdivsf (operands[0], operands[1],
			 operands[2], SFmode);
      DONE;
    }
})

;; Divmod instructions.

(define_expand "divmod<mode>4"
  [(parallel [(set (match_operand:SWIM248 0 "register_operand")
		   (div:SWIM248
		     (match_operand:SWIM248 1 "register_operand")
		     (match_operand:SWIM248 2 "nonimmediate_operand")))
	      (set (match_operand:SWIM248 3 "register_operand")
		   (mod:SWIM248 (match_dup 1) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])])

;; Split with 8bit unsigned divide:
;; 	if (dividend an divisor are in [0-255])
;;	   use 8bit unsigned integer divide
;;	 else
;;	   use original integer divide
(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(div:SWI48 (match_operand:SWI48 2 "register_operand")
		    (match_operand:SWI48 3 "nonimmediate_operand")))
   (set (match_operand:SWI48 1 "register_operand")
	(mod:SWI48 (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_8BIT_IDIV
   && TARGET_QIMODE_MATH
   && can_create_pseudo_p ()
   && !optimize_insn_for_size_p ()"
  [(const_int 0)]
  "ix86_split_idivmod (<MODE>mode, operands, true); DONE;")

(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (div:SI (match_operand:SI 2 "register_operand")
		  (match_operand:SI 3 "nonimmediate_operand"))))
   (set (match_operand:SI 1 "register_operand")
	(mod:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_8BIT_IDIV
   && TARGET_QIMODE_MATH
   && can_create_pseudo_p ()
   && !optimize_insn_for_size_p ()"
  [(const_int 0)]
  "ix86_split_idivmod (SImode, operands, true); DONE;")

(define_split
  [(set (match_operand:DI 1 "register_operand")
	(zero_extend:DI
	  (mod:SI (match_operand:SI 2 "register_operand")
		  (match_operand:SI 3 "nonimmediate_operand"))))
   (set (match_operand:SI 0 "register_operand")
	(div:SI  (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_8BIT_IDIV
   && TARGET_QIMODE_MATH
   && can_create_pseudo_p ()
   && !optimize_insn_for_size_p ()"
  [(const_int 0)]
  "ix86_split_idivmod (SImode, operands, true); DONE;")

(define_insn_and_split "divmod<mode>4_1"
  [(set (match_operand:SWI48 0 "register_operand" "=a")
	(div:SWI48 (match_operand:SWI48 2 "register_operand" "0")
		   (match_operand:SWI48 3 "nonimmediate_operand" "rm")))
   (set (match_operand:SWI48 1 "register_operand" "=&d")
	(mod:SWI48 (match_dup 2) (match_dup 3)))
   (unspec [(const_int 0)] UNSPEC_DIV_ALREADY_SPLIT)
   (clobber (reg:CC FLAGS_REG))]
  ""
  "#"
  "reload_completed"
  [(parallel [(set (match_dup 1)
		   (ashiftrt:SWI48 (match_dup 4) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 0)
	           (div:SWI48 (match_dup 2) (match_dup 3)))
	      (set (match_dup 1)
		   (mod:SWI48 (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[5] = GEN_INT (GET_MODE_BITSIZE (<MODE>mode)-1);

  if (optimize_function_for_size_p (cfun) || TARGET_USE_CLTD)
    operands[4] = operands[2];
  else
    {
      /* Avoid use of cltd in favor of a mov+shift.  */
      emit_move_insn (operands[1], operands[2]);
      operands[4] = operands[1];
    }
}
  [(set_attr "type" "multi")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "divmodsi4_zext_1"
  [(set (match_operand:DI 0 "register_operand" "=a")
	(zero_extend:DI
	  (div:SI (match_operand:SI 2 "register_operand" "0")
		  (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 1 "register_operand" "=&d")
	(mod:SI (match_dup 2) (match_dup 3)))
   (unspec [(const_int 0)] UNSPEC_DIV_ALREADY_SPLIT)
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(parallel [(set (match_dup 1)
		   (ashiftrt:SI (match_dup 4) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 0)
		   (zero_extend:DI (div:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 1)
		   (mod:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[5] = GEN_INT (GET_MODE_BITSIZE (SImode)-1);

  if (optimize_function_for_size_p (cfun) || TARGET_USE_CLTD)
    operands[4] = operands[2];
  else
    {
      /* Avoid use of cltd in favor of a mov+shift.  */
      emit_move_insn (operands[1], operands[2]);
      operands[4] = operands[1];
    }
}
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn_and_split "divmodsi4_zext_2"
  [(set (match_operand:DI 1 "register_operand" "=&d")
	(zero_extend:DI
	  (mod:SI (match_operand:SI 2 "register_operand" "0")
		  (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 0 "register_operand" "=a")
	(div:SI (match_dup 2) (match_dup 3)))
   (unspec [(const_int 0)] UNSPEC_DIV_ALREADY_SPLIT)
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(parallel [(set (match_dup 6)
		   (ashiftrt:SI (match_dup 4) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 1)
		   (zero_extend:DI (mod:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 0)
		   (div:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 6))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[5] = GEN_INT (GET_MODE_BITSIZE (SImode)-1);
  operands[6] = gen_lowpart (SImode, operands[1]);

  if (optimize_function_for_size_p (cfun) || TARGET_USE_CLTD)
    operands[4] = operands[2];
  else
    {
      /* Avoid use of cltd in favor of a mov+shift.  */
      emit_move_insn (operands[6], operands[2]);
      operands[4] = operands[6];
    }
}
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn_and_split "*divmod<mode>4"
  [(set (match_operand:SWIM248 0 "register_operand" "=a")
	(div:SWIM248 (match_operand:SWIM248 2 "register_operand" "0")
		    (match_operand:SWIM248 3 "nonimmediate_operand" "rm")))
   (set (match_operand:SWIM248 1 "register_operand" "=&d")
	(mod:SWIM248 (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "#"
  "reload_completed"
  [(parallel [(set (match_dup 1)
		   (ashiftrt:SWIM248 (match_dup 4) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 0)
	           (div:SWIM248 (match_dup 2) (match_dup 3)))
	      (set (match_dup 1)
		   (mod:SWIM248 (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[5] = GEN_INT (GET_MODE_BITSIZE (<MODE>mode)-1);

  if (<MODE>mode != HImode
      && (optimize_function_for_size_p (cfun) || TARGET_USE_CLTD))
    operands[4] = operands[2];
  else
    {
      /* Avoid use of cltd in favor of a mov+shift.  */
      emit_move_insn (operands[1], operands[2]);
      operands[4] = operands[1];
    }
}
  [(set_attr "type" "multi")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "*divmodsi4_zext_1"
  [(set (match_operand:DI 0 "register_operand" "=a")
	(zero_extend:DI
	  (div:SI (match_operand:SI 2 "register_operand" "0")
		  (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 1 "register_operand" "=&d")
	(mod:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(parallel [(set (match_dup 1)
		   (ashiftrt:SI (match_dup 4) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 0)
		   (zero_extend:DI (div:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 1)
		   (mod:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[5] = GEN_INT (GET_MODE_BITSIZE (SImode)-1);

  if (optimize_function_for_size_p (cfun) || TARGET_USE_CLTD)
    operands[4] = operands[2];
  else
    {
      /* Avoid use of cltd in favor of a mov+shift.  */
      emit_move_insn (operands[1], operands[2]);
      operands[4] = operands[1];
    }
}
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn_and_split "*divmodsi4_zext_2"
  [(set (match_operand:DI 1 "register_operand" "=&d")
	(zero_extend:DI
	  (mod:SI (match_operand:SI 2 "register_operand" "0")
		  (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 0 "register_operand" "=a")
	(div:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(parallel [(set (match_dup 6)
		   (ashiftrt:SI (match_dup 4) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 1)
		   (zero_extend:DI (mod:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 0)
		   (div:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 6))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[5] = GEN_INT (GET_MODE_BITSIZE (SImode)-1);
  operands[6] = gen_lowpart (SImode, operands[1]);

  if (optimize_function_for_size_p (cfun) || TARGET_USE_CLTD)
    operands[4] = operands[2];
  else
    {
      /* Avoid use of cltd in favor of a mov+shift.  */
      emit_move_insn (operands[6], operands[2]);
      operands[4] = operands[6];
    }
}
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn "*divmod<mode>4_noext"
  [(set (match_operand:SWIM248 0 "register_operand" "=a")
	(div:SWIM248 (match_operand:SWIM248 2 "register_operand" "0")
		    (match_operand:SWIM248 3 "nonimmediate_operand" "rm")))
   (set (match_operand:SWIM248 1 "register_operand" "=d")
	(mod:SWIM248 (match_dup 2) (match_dup 3)))
   (use (match_operand:SWIM248 4 "register_operand" "1"))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "idiv{<imodesuffix>}\t%3"
  [(set_attr "type" "idiv")
   (set_attr "mode" "<MODE>")])

(define_insn "*divmodsi4_noext_zext_1"
  [(set (match_operand:DI 0 "register_operand" "=a")
	(zero_extend:DI
	  (div:SI (match_operand:SI 2 "register_operand" "0")
		  (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 1 "register_operand" "=d")
	(mod:SI (match_dup 2) (match_dup 3)))
   (use (match_operand:SI 4 "register_operand" "1"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "idiv{l}\t%3"
  [(set_attr "type" "idiv")
   (set_attr "mode" "SI")])

(define_insn "*divmodsi4_noext_zext_2"
  [(set (match_operand:DI 1 "register_operand" "=d")
	(zero_extend:DI
	  (mod:SI (match_operand:SI 2 "register_operand" "0")
		  (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 0 "register_operand" "=a")
	(div:SI (match_dup 2) (match_dup 3)))
   (use (match_operand:SI 4 "register_operand" "1"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "idiv{l}\t%3"
  [(set_attr "type" "idiv")
   (set_attr "mode" "SI")])

(define_expand "divmodqi4"
  [(parallel [(set (match_operand:QI 0 "register_operand")
		   (div:QI
		     (match_operand:QI 1 "register_operand")
		     (match_operand:QI 2 "nonimmediate_operand")))
	      (set (match_operand:QI 3 "register_operand")
		   (mod:QI (match_dup 1) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_QIMODE_MATH"
{
  rtx div, mod;
  rtx tmp0, tmp1;
  
  tmp0 = gen_reg_rtx (HImode);
  tmp1 = gen_reg_rtx (HImode);

  /* Extend operands[1] to HImode.  Generate 8bit divide.  Result is in AX.  */
  emit_insn (gen_extendqihi2 (tmp1, operands[1]));
  emit_insn (gen_divmodhiqi3 (tmp0, tmp1, operands[2]));

  /* Extract remainder from AH.  */
  tmp1 = gen_rtx_ZERO_EXTRACT (SImode, tmp0, GEN_INT (8), GEN_INT (8));
  tmp1 = lowpart_subreg (QImode, tmp1, SImode);
  rtx_insn *insn = emit_move_insn (operands[3], tmp1);

  mod = gen_rtx_MOD (QImode, operands[1], operands[2]);
  set_unique_reg_note (insn, REG_EQUAL, mod);

  /* Extract quotient from AL.  */
  insn = emit_move_insn (operands[0], gen_lowpart (QImode, tmp0));

  div = gen_rtx_DIV (QImode, operands[1], operands[2]);
  set_unique_reg_note (insn, REG_EQUAL, div);

  DONE;
})

;; Divide AX by r/m8, with result stored in
;; AL <- Quotient
;; AH <- Remainder
;; Change div/mod to HImode and extend the second argument to HImode
;; so that mode of div/mod matches with mode of arguments.  Otherwise
;; combine may fail.
(define_insn "divmodhiqi3"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(ior:HI
	  (ashift:HI
	    (zero_extend:HI
	      (truncate:QI
		(mod:HI (match_operand:HI 1 "register_operand" "0")
			(sign_extend:HI
			  (match_operand:QI 2 "nonimmediate_operand" "qm")))))
	    (const_int 8))
	  (zero_extend:HI
	    (truncate:QI
	      (div:HI (match_dup 1) (sign_extend:HI (match_dup 2)))))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_QIMODE_MATH"
  "idiv{b}\t%2"
  [(set_attr "type" "idiv")
   (set_attr "mode" "QI")])

(define_expand "udivmod<mode>4"
  [(parallel [(set (match_operand:SWIM248 0 "register_operand")
		   (udiv:SWIM248
		     (match_operand:SWIM248 1 "register_operand")
		     (match_operand:SWIM248 2 "nonimmediate_operand")))
	      (set (match_operand:SWIM248 3 "register_operand")
		   (umod:SWIM248 (match_dup 1) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])])

;; Split with 8bit unsigned divide:
;; 	if (dividend an divisor are in [0-255])
;;	   use 8bit unsigned integer divide
;;	 else
;;	   use original integer divide
(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(udiv:SWI48 (match_operand:SWI48 2 "register_operand")
		    (match_operand:SWI48 3 "nonimmediate_operand")))
   (set (match_operand:SWI48 1 "register_operand")
	(umod:SWI48 (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_8BIT_IDIV
   && TARGET_QIMODE_MATH
   && can_create_pseudo_p ()
   && !optimize_insn_for_size_p ()"
  [(const_int 0)]
  "ix86_split_idivmod (<MODE>mode, operands, false); DONE;")

(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (udiv:SI (match_operand:SI 2 "register_operand")
		   (match_operand:SI 3 "nonimmediate_operand"))))
   (set (match_operand:SI 1 "register_operand")
	(umod:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && TARGET_USE_8BIT_IDIV
   && TARGET_QIMODE_MATH
   && can_create_pseudo_p ()
   && !optimize_insn_for_size_p ()"
  [(const_int 0)]
  "ix86_split_idivmod (SImode, operands, false); DONE;")

(define_split
  [(set (match_operand:DI 1 "register_operand")
	(zero_extend:DI
	  (umod:SI (match_operand:SI 2 "register_operand")
		   (match_operand:SI 3 "nonimmediate_operand"))))
   (set (match_operand:SI 0 "register_operand")
	(udiv:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && TARGET_USE_8BIT_IDIV
   && TARGET_QIMODE_MATH
   && can_create_pseudo_p ()
   && !optimize_insn_for_size_p ()"
  [(const_int 0)]
  "ix86_split_idivmod (SImode, operands, false); DONE;")

(define_insn_and_split "udivmod<mode>4_1"
  [(set (match_operand:SWI48 0 "register_operand" "=a")
	(udiv:SWI48 (match_operand:SWI48 2 "register_operand" "0")
		    (match_operand:SWI48 3 "nonimmediate_operand" "rm")))
   (set (match_operand:SWI48 1 "register_operand" "=&d")
	(umod:SWI48 (match_dup 2) (match_dup 3)))
   (unspec [(const_int 0)] UNSPEC_DIV_ALREADY_SPLIT)
   (clobber (reg:CC FLAGS_REG))]
  ""
  "#"
  "reload_completed"
  [(set (match_dup 1) (const_int 0))
   (parallel [(set (match_dup 0)
		   (udiv:SWI48 (match_dup 2) (match_dup 3)))
	      (set (match_dup 1)
		   (umod:SWI48 (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
  ""
  [(set_attr "type" "multi")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "udivmodsi4_zext_1"
  [(set (match_operand:DI 0 "register_operand" "=a")
	(zero_extend:DI
	  (udiv:SI (match_operand:SI 2 "register_operand" "0")
		   (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 1 "register_operand" "=&d")
	(umod:SI (match_dup 2) (match_dup 3)))
   (unspec [(const_int 0)] UNSPEC_DIV_ALREADY_SPLIT)
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(set (match_dup 1) (const_int 0))
   (parallel [(set (match_dup 0)
		   (zero_extend:DI (udiv:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 1)
		   (umod:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
  ""
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn_and_split "udivmodsi4_zext_2"
  [(set (match_operand:DI 1 "register_operand" "=&d")
	(zero_extend:DI
	  (umod:SI (match_operand:SI 2 "register_operand" "0")
		 (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 0 "register_operand" "=a")
	(udiv:SI (match_dup 2) (match_dup 3)))
   (unspec [(const_int 0)] UNSPEC_DIV_ALREADY_SPLIT)
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(set (match_dup 4) (const_int 0))
   (parallel [(set (match_dup 1)
		   (zero_extend:DI (umod:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 0)
		   (udiv:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 4))
	      (clobber (reg:CC FLAGS_REG))])]
  "operands[4] = gen_lowpart (SImode, operands[1]);"
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn_and_split "*udivmod<mode>4"
  [(set (match_operand:SWIM248 0 "register_operand" "=a")
	(udiv:SWIM248 (match_operand:SWIM248 2 "register_operand" "0")
		      (match_operand:SWIM248 3 "nonimmediate_operand" "rm")))
   (set (match_operand:SWIM248 1 "register_operand" "=&d")
	(umod:SWIM248 (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "#"
  "reload_completed"
  [(set (match_dup 1) (const_int 0))
   (parallel [(set (match_dup 0)
		   (udiv:SWIM248 (match_dup 2) (match_dup 3)))
	      (set (match_dup 1)
		   (umod:SWIM248 (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
  ""
  [(set_attr "type" "multi")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "*udivmodsi4_zext_1"
  [(set (match_operand:DI 0 "register_operand" "=a")
	(zero_extend:DI
	  (udiv:SI (match_operand:SI 2 "register_operand" "0")
		   (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 1 "register_operand" "=&d")
	(umod:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(set (match_dup 1) (const_int 0))
   (parallel [(set (match_dup 0)
		   (zero_extend:DI (udiv:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 1)
		   (umod:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
  ""
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn_and_split "*udivmodsi4_zext_2"
  [(set (match_operand:DI 1 "register_operand" "=&d")
	(zero_extend:DI
	  (umod:SI (match_operand:SI 2 "register_operand" "0")
		   (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 0 "register_operand" "=a")
	(udiv:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "#"
  "reload_completed"
  [(set (match_dup 4) (const_int 0))
   (parallel [(set (match_dup 1)
		   (zero_extend:DI (umod:SI (match_dup 2) (match_dup 3))))
	      (set (match_dup 0)
		   (udiv:SI (match_dup 2) (match_dup 3)))
	      (use (match_dup 4))
	      (clobber (reg:CC FLAGS_REG))])]
  "operands[4] = gen_lowpart (SImode, operands[1]);"
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

;; Optimize division or modulo by constant power of 2, if the constant
;; materializes only after expansion.
(define_insn_and_split "*udivmod<mode>4_pow2"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(udiv:SWI48 (match_operand:SWI48 2 "register_operand" "0")
		    (match_operand:SWI48 3 "const_int_operand" "n")))
   (set (match_operand:SWI48 1 "register_operand" "=r")
	(umod:SWI48 (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "IN_RANGE (INTVAL (operands[3]), 2, HOST_WIDE_INT_UC (0x80000000))
   && (UINTVAL (operands[3]) & (UINTVAL (operands[3]) - 1)) == 0"
  "#"
  "&& 1"
  [(set (match_dup 1) (match_dup 2))
   (parallel [(set (match_dup 0) (lshiftrt:<MODE> (match_dup 2) (match_dup 4)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 1) (and:<MODE> (match_dup 1) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  int v = exact_log2 (UINTVAL (operands[3]));
  operands[4] = GEN_INT (v);
  operands[5] = GEN_INT ((HOST_WIDE_INT_1U << v) - 1);
}
  [(set_attr "type" "multi")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "*udivmodsi4_pow2_zext_1"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (udiv:SI (match_operand:SI 2 "register_operand" "0")
		   (match_operand:SI 3 "const_int_operand" "n"))))
   (set (match_operand:SI 1 "register_operand" "=r")
	(umod:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && IN_RANGE (INTVAL (operands[3]), 2, HOST_WIDE_INT_UC (0x80000000))
   && (UINTVAL (operands[3]) & (UINTVAL (operands[3]) - 1)) == 0"
  "#"
  "&& 1"
  [(set (match_dup 1) (match_dup 2))
   (parallel [(set (match_dup 0)
		   (zero_extend:DI (lshiftrt:SI (match_dup 2) (match_dup 4))))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 1) (and:SI (match_dup 1) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  int v = exact_log2 (UINTVAL (operands[3]));
  operands[4] = GEN_INT (v);
  operands[5] = GEN_INT ((HOST_WIDE_INT_1U << v) - 1);
}
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn_and_split "*udivmodsi4_pow2_zext_2"
  [(set (match_operand:DI 1 "register_operand" "=r")
	(zero_extend:DI
	  (umod:SI (match_operand:SI 2 "register_operand" "0")
		   (match_operand:SI 3 "const_int_operand" "n"))))
   (set (match_operand:SI 0 "register_operand" "=r")
	(umod:SI (match_dup 2) (match_dup 3)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT
   && IN_RANGE (INTVAL (operands[3]), 2, HOST_WIDE_INT_UC (0x80000000))
   && (UINTVAL (operands[3]) & (UINTVAL (operands[3]) - 1)) == 0"
  "#"
  "&& 1"
  [(set (match_dup 1) (match_dup 2))
   (parallel [(set (match_dup 0) (lshiftrt:SI (match_dup 2) (match_dup 4)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 1)
		   (zero_extend:DI (and:SI (match_dup 1) (match_dup 5))))
	      (clobber (reg:CC FLAGS_REG))])]
{
  int v = exact_log2 (UINTVAL (operands[3]));
  operands[4] = GEN_INT (v);
  operands[5] = GEN_INT ((HOST_WIDE_INT_1U << v) - 1);
}
  [(set_attr "type" "multi")
   (set_attr "mode" "SI")])

(define_insn "*udivmod<mode>4_noext"
  [(set (match_operand:SWIM248 0 "register_operand" "=a")
	(udiv:SWIM248 (match_operand:SWIM248 2 "register_operand" "0")
		      (match_operand:SWIM248 3 "nonimmediate_operand" "rm")))
   (set (match_operand:SWIM248 1 "register_operand" "=d")
	(umod:SWIM248 (match_dup 2) (match_dup 3)))
   (use (match_operand:SWIM248 4 "register_operand" "1"))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "div{<imodesuffix>}\t%3"
  [(set_attr "type" "idiv")
   (set_attr "mode" "<MODE>")])

(define_insn "*udivmodsi4_noext_zext_1"
  [(set (match_operand:DI 0 "register_operand" "=a")
	(zero_extend:DI
	  (udiv:SI (match_operand:SI 2 "register_operand" "0")
		   (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 1 "register_operand" "=d")
	(umod:SI (match_dup 2) (match_dup 3)))
   (use (match_operand:SI 4 "register_operand" "1"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "div{l}\t%3"
  [(set_attr "type" "idiv")
   (set_attr "mode" "SI")])

(define_insn "*udivmodsi4_noext_zext_2"
  [(set (match_operand:DI 1 "register_operand" "=d")
	(zero_extend:DI
	  (umod:SI (match_operand:SI 2 "register_operand" "0")
		   (match_operand:SI 3 "nonimmediate_operand" "rm"))))
   (set (match_operand:SI 0 "register_operand" "=a")
	(udiv:SI (match_dup 2) (match_dup 3)))
   (use (match_operand:SI 4 "register_operand" "1"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "div{l}\t%3"
  [(set_attr "type" "idiv")
   (set_attr "mode" "SI")])

(define_expand "udivmodqi4"
  [(parallel [(set (match_operand:QI 0 "register_operand")
		   (udiv:QI
		     (match_operand:QI 1 "register_operand")
		     (match_operand:QI 2 "nonimmediate_operand")))
	      (set (match_operand:QI 3 "register_operand")
		   (umod:QI (match_dup 1) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_QIMODE_MATH"
{
  rtx div, mod;
  rtx tmp0, tmp1;
  
  tmp0 = gen_reg_rtx (HImode);
  tmp1 = gen_reg_rtx (HImode);

  /* Extend operands[1] to HImode.  Generate 8bit divide.  Result is in AX.  */
  emit_insn (gen_zero_extendqihi2 (tmp1, operands[1]));
  emit_insn (gen_udivmodhiqi3 (tmp0, tmp1, operands[2]));

  /* Extract remainder from AH.  */
  tmp1 = gen_rtx_ZERO_EXTRACT (SImode, tmp0, GEN_INT (8), GEN_INT (8));
  tmp1 = lowpart_subreg (QImode, tmp1, SImode);
  rtx_insn *insn = emit_move_insn (operands[3], tmp1);

  mod = gen_rtx_UMOD (QImode, operands[1], operands[2]);
  set_unique_reg_note (insn, REG_EQUAL, mod);

  /* Extract quotient from AL.  */
  insn = emit_move_insn (operands[0], gen_lowpart (QImode, tmp0));

  div = gen_rtx_UDIV (QImode, operands[1], operands[2]);
  set_unique_reg_note (insn, REG_EQUAL, div);

  DONE;
})

(define_insn "udivmodhiqi3"
  [(set (match_operand:HI 0 "register_operand" "=a")
	(ior:HI
	  (ashift:HI
	    (zero_extend:HI
	      (truncate:QI
		(mod:HI (match_operand:HI 1 "register_operand" "0")
			(zero_extend:HI
			  (match_operand:QI 2 "nonimmediate_operand" "qm")))))
	    (const_int 8))
	  (zero_extend:HI
	    (truncate:QI
	      (div:HI (match_dup 1) (zero_extend:HI (match_dup 2)))))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_QIMODE_MATH"
  "div{b}\t%2"
  [(set_attr "type" "idiv")
   (set_attr "mode" "QI")])

;; We cannot use div/idiv for double division, because it causes
;; "division by zero" on the overflow and that's not what we expect
;; from truncate.  Because true (non truncating) double division is
;; never generated, we can't create this insn anyway.
;
;(define_insn ""
;  [(set (match_operand:SI 0 "register_operand" "=a")
;	(truncate:SI
;	  (udiv:DI (match_operand:DI 1 "register_operand" "A")
;		   (zero_extend:DI
;		     (match_operand:SI 2 "nonimmediate_operand" "rm")))))
;   (set (match_operand:SI 3 "register_operand" "=d")
;	(truncate:SI
;	  (umod:DI (match_dup 1) (zero_extend:DI (match_dup 2)))))
;   (clobber (reg:CC FLAGS_REG))]
;  ""
;  "div{l}\t{%2, %0|%0, %2}"
;  [(set_attr "type" "idiv")])

;;- Logical AND instructions

;; On Pentium, "test imm, reg" is pairable only with eax, ax, and al.
;; Note that this excludes ah.

(define_expand "testsi_ccno_1"
  [(set (reg:CCNO FLAGS_REG)
	(compare:CCNO
	  (and:SI (match_operand:SI 0 "nonimmediate_operand")
		  (match_operand:SI 1 "x86_64_nonmemory_operand"))
	  (const_int 0)))])

(define_expand "testqi_ccz_1"
  [(set (reg:CCZ FLAGS_REG)
        (compare:CCZ (and:QI (match_operand:QI 0 "nonimmediate_operand")
			     (match_operand:QI 1 "nonmemory_operand"))
		 (const_int 0)))])

(define_expand "testdi_ccno_1"
  [(set (reg:CCNO FLAGS_REG)
	(compare:CCNO
	  (and:DI (match_operand:DI 0 "nonimmediate_operand")
		  (match_operand:DI 1 "x86_64_szext_general_operand"))
	  (const_int 0)))]
  "TARGET_64BIT && !(MEM_P (operands[0]) && MEM_P (operands[1]))")

(define_insn "*testdi_1"
  [(set (reg FLAGS_REG)
	(compare
	 (and:DI
	  (match_operand:DI 0 "nonimmediate_operand" "%!*a,r,!*a,r,rm")
	  (match_operand:DI 1 "x86_64_szext_general_operand" "Z,Z,e,e,re"))
	 (const_int 0)))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCNOmode)
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "@
   test{l}\t{%k1, %k0|%k0, %k1}
   test{l}\t{%k1, %k0|%k0, %k1}
   test{q}\t{%1, %0|%0, %1}
   test{q}\t{%1, %0|%0, %1}
   test{q}\t{%1, %0|%0, %1}"
  [(set_attr "type" "test")
   (set_attr "modrm" "0,1,0,1,1")
   (set_attr "mode" "SI,SI,DI,DI,DI")])

(define_insn "*testqi_1_maybe_si"
  [(set (reg FLAGS_REG)
        (compare
	  (and:QI
	    (match_operand:QI 0 "nonimmediate_operand" "%!*a,q,qm,r")
	    (match_operand:QI 1 "general_operand" "n,n,qn,n"))
	  (const_int 0)))]
   "!(MEM_P (operands[0]) && MEM_P (operands[1]))
    && ix86_match_ccmode (insn,
 			 CONST_INT_P (operands[1])
 			 && INTVAL (operands[1]) >= 0 ? CCNOmode : CCZmode)"
{
  if (which_alternative == 3)
    {
      if (CONST_INT_P (operands[1]) && INTVAL (operands[1]) < 0)
	operands[1] = GEN_INT (INTVAL (operands[1]) & 0xff);
      return "test{l}\t{%1, %k0|%k0, %1}";
    }
  return "test{b}\t{%1, %0|%0, %1}";
}
  [(set_attr "type" "test")
   (set_attr "modrm" "0,1,1,1")
   (set_attr "mode" "QI,QI,QI,SI")
   (set_attr "pent_pair" "uv,np,uv,np")])

(define_insn "*test<mode>_1"
  [(set (reg FLAGS_REG)
	(compare
	 (and:SWI124
	  (match_operand:SWI124 0 "nonimmediate_operand" "%!*a,<r>,<r>m")
	  (match_operand:SWI124 1 "<general_operand>" "<i>,<i>,<r><i>"))
	 (const_int 0)))]
  "ix86_match_ccmode (insn, CCNOmode)
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "test{<imodesuffix>}\t{%1, %0|%0, %1}"
  [(set_attr "type" "test")
   (set_attr "modrm" "0,1,1")
   (set_attr "mode" "<MODE>")
   (set_attr "pent_pair" "uv,np,uv")])

(define_expand "testqi_ext_1_ccno"
  [(set (reg:CCNO FLAGS_REG)
	(compare:CCNO
	  (and:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 0 "ext_register_operand")
			       (const_int 8)
			       (const_int 8)) 0)
	      (match_operand 1 "const_int_operand"))
	  (const_int 0)))])

(define_insn "*testqi_ext_1"
  [(set (reg FLAGS_REG)
	(compare
	  (and:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 0 "ext_register_operand" "Q,Q")
			       (const_int 8)
			       (const_int 8)) 0)
	    (match_operand:QI 1 "general_operand" "QnBc,m"))
	  (const_int 0)))]
  "ix86_match_ccmode (insn, CCNOmode)"
  "test{b}\t{%1, %h0|%h0, %1}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "test")
   (set_attr "mode" "QI")])

(define_insn "*testqi_ext_2"
  [(set (reg FLAGS_REG)
	(compare
	  (and:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 0 "ext_register_operand" "Q")
			       (const_int 8)
			       (const_int 8)) 0)
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "Q")
			       (const_int 8)
			       (const_int 8)) 0))
	  (const_int 0)))]
  "ix86_match_ccmode (insn, CCNOmode)"
  "test{b}\t{%h1, %h0|%h0, %h1}"
  [(set_attr "type" "test")
   (set_attr "mode" "QI")])

;; Combine likes to form bit extractions for some tests.  Humor it.
(define_insn_and_split "*testqi_ext_3"
  [(set (match_operand 0 "flags_reg_operand")
        (match_operator 1 "compare_operator"
	  [(zero_extract:SWI248
	     (match_operand 2 "nonimmediate_operand" "rm")
	     (match_operand 3 "const_int_operand" "n")
	     (match_operand 4 "const_int_operand" "n"))
	   (const_int 0)]))]
  "ix86_match_ccmode (insn, CCNOmode)
   && ((TARGET_64BIT && GET_MODE (operands[2]) == DImode)
       || GET_MODE (operands[2]) == SImode
       || GET_MODE (operands[2]) == HImode
       || GET_MODE (operands[2]) == QImode)
   /* Ensure that resulting mask is zero or sign extended operand.  */
   && INTVAL (operands[4]) >= 0
   && ((INTVAL (operands[3]) > 0
	&& INTVAL (operands[3]) + INTVAL (operands[4]) <= 32)
       || (<MODE>mode == DImode
	   && INTVAL (operands[3]) > 32
	   && INTVAL (operands[3]) + INTVAL (operands[4]) == 64))"
  "#"
  "&& 1"
  [(set (match_dup 0) (match_op_dup 1 [(match_dup 2) (const_int 0)]))]
{
  rtx val = operands[2];
  HOST_WIDE_INT len = INTVAL (operands[3]);
  HOST_WIDE_INT pos = INTVAL (operands[4]);
  machine_mode mode = GET_MODE (val);

  if (SUBREG_P (val))
    {
      machine_mode submode = GET_MODE (SUBREG_REG (val));

      /* Narrow paradoxical subregs to prevent partial register stalls.  */
      if (GET_MODE_BITSIZE (mode) > GET_MODE_BITSIZE (submode)
	  && GET_MODE_CLASS (submode) == MODE_INT)
	{
	  val = SUBREG_REG (val);
	  mode = submode;
	}
    }

  /* Small HImode tests can be converted to QImode.  */
  if (register_operand (val, HImode) && pos + len <= 8)
    {
      val = gen_lowpart (QImode, val);
      mode = QImode;
    }

  gcc_assert (pos + len <= GET_MODE_PRECISION (mode));

  wide_int mask
    = wi::shifted_mask (pos, len, false, GET_MODE_PRECISION (mode));

  operands[2] = gen_rtx_AND (mode, val, immed_wide_int_const (mask, mode));
})

;; Convert HImode/SImode test instructions with immediate to QImode ones.
;; i386 does not allow to encode test with 8bit sign extended immediate, so
;; this is relatively important trick.
;; Do the conversion only post-reload to avoid limiting of the register class
;; to QI regs.
(define_split
  [(set (match_operand 0 "flags_reg_operand")
	(match_operator 1 "compare_operator"
	  [(and (match_operand 2 "QIreg_operand")
	        (match_operand 3 "const_int_operand"))
	   (const_int 0)]))]
   "reload_completed
    && GET_MODE (operands[2]) != QImode
    && ((ix86_match_ccmode (insn, CCZmode)
    	 && !(INTVAL (operands[3]) & ~(255 << 8)))
	|| (ix86_match_ccmode (insn, CCNOmode)
	    && !(INTVAL (operands[3]) & ~(127 << 8))))"
  [(set (match_dup 0)
	(match_op_dup 1
	  [(and:QI
	     (subreg:QI
	       (zero_extract:SI (match_dup 2)
				(const_int 8)
				(const_int 8)) 0)
	     (match_dup 3))
	   (const_int 0)]))]
{
  operands[2] = gen_lowpart (SImode, operands[2]);
  operands[3] = gen_int_mode (INTVAL (operands[3]) >> 8, QImode);
})

(define_split
  [(set (match_operand 0 "flags_reg_operand")
	(match_operator 1 "compare_operator"
	  [(and (match_operand 2 "nonimmediate_operand")
	        (match_operand 3 "const_int_operand"))
	   (const_int 0)]))]
   "reload_completed
    && GET_MODE (operands[2]) != QImode
    && (!REG_P (operands[2]) || ANY_QI_REG_P (operands[2]))
    && ((ix86_match_ccmode (insn, CCZmode)
	 && !(INTVAL (operands[3]) & ~255))
	|| (ix86_match_ccmode (insn, CCNOmode)
	    && !(INTVAL (operands[3]) & ~127)))"
  [(set (match_dup 0)
	(match_op_dup 1 [(and:QI (match_dup 2) (match_dup 3))
			 (const_int 0)]))]
{
  operands[2] = gen_lowpart (QImode, operands[2]);
  operands[3] = gen_int_mode (INTVAL (operands[3]), QImode);
})

;; %%% This used to optimize known byte-wide and operations to memory,
;; and sometimes to QImode registers.  If this is considered useful,
;; it should be done with splitters.

(define_expand "and<mode>3"
  [(set (match_operand:SWIM1248x 0 "nonimmediate_operand")
	(and:SWIM1248x (match_operand:SWIM1248x 1 "nonimmediate_operand")
		      (match_operand:SWIM1248x 2 "<general_szext_operand>")))]
  ""
{
  machine_mode mode = <MODE>mode;
  rtx (*insn) (rtx, rtx);

  if (CONST_INT_P (operands[2]) && REG_P (operands[0]))
    {
      HOST_WIDE_INT ival = INTVAL (operands[2]);

      if (ival == (HOST_WIDE_INT) 0xffffffff)
	mode = SImode;
      else if (ival == 0xffff)
	mode = HImode;
      else if (ival == 0xff)
	mode = QImode;
      }

  if (mode == <MODE>mode)
    {
      ix86_expand_binary_operator (AND, <MODE>mode, operands);
      DONE;
    }

  if (<MODE>mode == DImode)
    insn = (mode == SImode)
	   ? gen_zero_extendsidi2
	   : (mode == HImode)
	   ? gen_zero_extendhidi2
	   : gen_zero_extendqidi2;
  else if (<MODE>mode == SImode)
    insn = (mode == HImode)
	   ? gen_zero_extendhisi2
	   : gen_zero_extendqisi2;
  else if (<MODE>mode == HImode)
    insn = gen_zero_extendqihi2;
  else
    gcc_unreachable ();

  emit_insn (insn (operands[0], gen_lowpart (mode, operands[1])));
  DONE;
})

(define_insn_and_split "*anddi3_doubleword"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=r,rm,r")
	(and:DI
	 (match_operand:DI 1 "nonimmediate_operand" "%0,0,0")
	 (match_operand:DI 2 "x86_64_szext_general_operand" "Z,re,rm")))
   (clobber (reg:CC FLAGS_REG))]
  "!TARGET_64BIT && TARGET_STV && TARGET_SSE2
   && ix86_binary_operator_ok (AND, DImode, operands)"
  "#"
  "&& reload_completed"
  [(const_int 0)]
{
  split_double_mode (DImode, &operands[0], 3, &operands[0], &operands[3]);
  if (operands[2] == const0_rtx)
    {
      operands[1] = const0_rtx;
      ix86_expand_move (SImode, &operands[0]);
    }
  else if (operands[2] != constm1_rtx)
    ix86_expand_binary_operator (AND, SImode, &operands[0]);
  else if (operands[5] == constm1_rtx)
    emit_note (NOTE_INSN_DELETED);
  if (operands[5] == const0_rtx)
    {
      operands[4] = const0_rtx;
      ix86_expand_move (SImode, &operands[3]);
    }
  else if (operands[5] != constm1_rtx)
    ix86_expand_binary_operator (AND, SImode, &operands[3]);
  DONE;
})

(define_insn "*anddi_1"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=r,rm,r,r")
	(and:DI
	 (match_operand:DI 1 "nonimmediate_operand" "%0,0,0,qm")
	 (match_operand:DI 2 "x86_64_szext_general_operand" "Z,re,rm,L")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (AND, DImode, operands)"
  "@
   and{l}\t{%k2, %k0|%k0, %k2}
   and{q}\t{%2, %0|%0, %2}
   and{q}\t{%2, %0|%0, %2}
   #"
  [(set_attr "type" "alu,alu,alu,imovx")
   (set_attr "length_immediate" "*,*,*,0")
   (set (attr "prefix_rex")
     (if_then_else
       (and (eq_attr "type" "imovx")
	    (and (match_test "INTVAL (operands[2]) == 0xff")
		 (match_operand 1 "ext_QIreg_operand")))
       (const_string "1")
       (const_string "*")))
   (set_attr "mode" "SI,DI,DI,SI")])

(define_insn_and_split "*anddi_1_btr"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
	(and:DI
	 (match_operand:DI 1 "nonimmediate_operand" "%0")
	 (match_operand:DI 2 "const_int_operand" "n")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && TARGET_USE_BT
   && ix86_binary_operator_ok (AND, DImode, operands)
   && IN_RANGE (exact_log2 (~INTVAL (operands[2])), 31, 63)"
  "#"
  "&& reload_completed"
  [(parallel [(set (zero_extract:DI (match_dup 0)
				    (const_int 1)
				    (match_dup 3))
		   (const_int 0))
	      (clobber (reg:CC FLAGS_REG))])]
  "operands[3] = GEN_INT (exact_log2 (~INTVAL (operands[2])));"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "DI")])

;; Turn *anddi_1 into *andsi_1_zext if possible.
(define_split
  [(set (match_operand:DI 0 "register_operand")
	(and:DI (subreg:DI (match_operand:SI 1 "register_operand") 0)
		(match_operand:DI 2 "x86_64_zext_immediate_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  [(parallel [(set (match_dup 0)
		   (zero_extend:DI (and:SI (match_dup 1) (match_dup 2))))
	      (clobber (reg:CC FLAGS_REG))])]
  "operands[2] = gen_lowpart (SImode, operands[2]);")

;; See comment for addsi_1_zext why we do use nonimmediate_operand
(define_insn "*andsi_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (and:SI (match_operand:SI 1 "nonimmediate_operand" "%0")
		  (match_operand:SI 2 "x86_64_general_operand" "rme"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (AND, SImode, operands)"
  "and{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*and<mode>_1"
  [(set (match_operand:SWI24 0 "nonimmediate_operand" "=rm,r,Ya")
	(and:SWI24 (match_operand:SWI24 1 "nonimmediate_operand" "%0,0,qm")
		   (match_operand:SWI24 2 "<general_operand>" "r<i>,rm,L")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (AND, <MODE>mode, operands)"
  "@
   and{<imodesuffix>}\t{%2, %0|%0, %2}
   and{<imodesuffix>}\t{%2, %0|%0, %2}
   #"
  [(set_attr "type" "alu,alu,imovx")
   (set_attr "length_immediate" "*,*,0")
   (set (attr "prefix_rex")
     (if_then_else
       (and (eq_attr "type" "imovx")
	    (and (match_test "INTVAL (operands[2]) == 0xff")
		 (match_operand 1 "ext_QIreg_operand")))
       (const_string "1")
       (const_string "*")))
   (set_attr "mode" "<MODE>,<MODE>,SI")])

(define_insn "*andqi_1"
  [(set (match_operand:QI 0 "nonimmediate_operand" "=qm,q,r")
	(and:QI (match_operand:QI 1 "nonimmediate_operand" "%0,0,0")
		(match_operand:QI 2 "general_operand" "qn,qmn,rn")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (AND, QImode, operands)"
  "@
   and{b}\t{%2, %0|%0, %2}
   and{b}\t{%2, %0|%0, %2}
   and{l}\t{%k2, %k0|%k0, %k2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "QI,QI,SI")
   ;; Potential partial reg stall on alternative 2.
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "2")
	      (symbol_ref "!TARGET_PARTIAL_REG_STALL")]
	   (symbol_ref "true")))])

(define_insn "*andqi_1_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+qm,q"))
	(and:QI (match_dup 0)
		(match_operand:QI 1 "general_operand" "qn,qmn")))
   (clobber (reg:CC FLAGS_REG))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "and{b}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "mode" "QI")])

(define_split
  [(set (match_operand:SWI248 0 "register_operand")
	(and:SWI248 (match_operand:SWI248 1 "nonimmediate_operand")
		    (match_operand:SWI248 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed
   && (!REG_P (operands[1])
       || REGNO (operands[0]) != REGNO (operands[1]))"
  [(const_int 0)]
{
  HOST_WIDE_INT ival = INTVAL (operands[2]);
  machine_mode mode;
  rtx (*insn) (rtx, rtx);

  if (ival == (HOST_WIDE_INT) 0xffffffff)
    mode = SImode;
  else if (ival == 0xffff)
    mode = HImode;
  else
    {
      gcc_assert (ival == 0xff);
      mode = QImode;
    }

  if (<MODE>mode == DImode)
    insn = (mode == SImode)
	   ? gen_zero_extendsidi2
	   : (mode == HImode)
	   ? gen_zero_extendhidi2
	   : gen_zero_extendqidi2;
  else
    {
      if (<MODE>mode != SImode)
	/* Zero extend to SImode to avoid partial register stalls.  */
	operands[0] = gen_lowpart (SImode, operands[0]);

      insn = (mode == HImode)
	     ? gen_zero_extendhisi2
	     : gen_zero_extendqisi2;
    }
  emit_insn (insn (operands[0], gen_lowpart (mode, operands[1])));
  DONE;
})

(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(and:SWI48 (match_dup 0)
		   (const_int -65536)))
   (clobber (reg:CC FLAGS_REG))]
  "(TARGET_FAST_PREFIX && !TARGET_PARTIAL_REG_STALL)
    || optimize_function_for_size_p (cfun)"
  [(set (strict_low_part (match_dup 1)) (const_int 0))]
  "operands[1] = gen_lowpart (HImode, operands[0]);")

(define_split
  [(set (match_operand:SWI248 0 "any_QIreg_operand")
	(and:SWI248 (match_dup 0)
		    (const_int -256)))
   (clobber (reg:CC FLAGS_REG))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && reload_completed"
  [(set (strict_low_part (match_dup 1)) (const_int 0))]
  "operands[1] = gen_lowpart (QImode, operands[0]);")

(define_split
  [(set (match_operand:SWI248 0 "QIreg_operand")
	(and:SWI248 (match_dup 0)
		    (const_int -65281)))
   (clobber (reg:CC FLAGS_REG))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && reload_completed"
  [(parallel
     [(set (zero_extract:SI (match_dup 0)
			    (const_int 8)
			    (const_int 8))
	   (subreg:SI
	     (xor:QI
	       (subreg:QI
		 (zero_extract:SI (match_dup 0)
				  (const_int 8)
				  (const_int 8)) 0)
	       (subreg:QI
		 (zero_extract:SI (match_dup 0)
				  (const_int 8)
				  (const_int 8)) 0)) 0))
      (clobber (reg:CC FLAGS_REG))])]
  "operands[0] = gen_lowpart (SImode, operands[0]);")

(define_insn "*anddi_2"
  [(set (reg FLAGS_REG)
	(compare
	 (and:DI
	  (match_operand:DI 1 "nonimmediate_operand" "%0,0,0")
	  (match_operand:DI 2 "x86_64_szext_general_operand" "Z,rem,re"))
	 (const_int 0)))
   (set (match_operand:DI 0 "nonimmediate_operand" "=r,r,rm")
	(and:DI (match_dup 1) (match_dup 2)))]
  "TARGET_64BIT
   && ix86_match_ccmode
	(insn,
	 /* If we are going to emit andl instead of andq, and the operands[2]
	    constant might have the SImode sign bit set, make sure the sign
	    flag isn't tested, because the instruction will set the sign flag
	    based on bit 31 rather than bit 63.  If it isn't CONST_INT,
	    conservatively assume it might have bit 31 set.  */
	 (satisfies_constraint_Z (operands[2])
	  && (!CONST_INT_P (operands[2])
	      || val_signbit_known_set_p (SImode, INTVAL (operands[2]))))
	 ? CCZmode : CCNOmode)
   && ix86_binary_operator_ok (AND, DImode, operands)"
  "@
   and{l}\t{%k2, %k0|%k0, %k2}
   and{q}\t{%2, %0|%0, %2}
   and{q}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI,DI,DI")])

;; See comment for addsi_1_zext why we do use nonimmediate_operand
(define_insn "*andsi_2_zext"
  [(set (reg FLAGS_REG)
	(compare (and:SI
		  (match_operand:SI 1 "nonimmediate_operand" "%0")
		  (match_operand:SI 2 "x86_64_general_operand" "rme"))
		 (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (and:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCNOmode)
   && ix86_binary_operator_ok (AND, SImode, operands)"
  "and{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*andqi_2_maybe_si"
  [(set (reg FLAGS_REG)
	(compare (and:QI
		  (match_operand:QI 1 "nonimmediate_operand" "%0,0,0")
		  (match_operand:QI 2 "general_operand" "qmn,qn,n"))
		 (const_int 0)))
   (set (match_operand:QI 0 "nonimmediate_operand" "=q,qm,*r")
	(and:QI (match_dup 1) (match_dup 2)))]
  "ix86_binary_operator_ok (AND, QImode, operands)
   && ix86_match_ccmode (insn,
			 CONST_INT_P (operands[2])
			 && INTVAL (operands[2]) >= 0 ? CCNOmode : CCZmode)"
{
  if (which_alternative == 2)
    {
      if (CONST_INT_P (operands[2]) && INTVAL (operands[2]) < 0)
        operands[2] = GEN_INT (INTVAL (operands[2]) & 0xff);
      return "and{l}\t{%2, %k0|%k0, %2}";
    }
  return "and{b}\t{%2, %0|%0, %2}";
}
  [(set_attr "type" "alu")
   (set_attr "mode" "QI,QI,SI")])

(define_insn "*and<mode>_2"
  [(set (reg FLAGS_REG)
	(compare (and:SWI124
		  (match_operand:SWI124 1 "nonimmediate_operand" "%0,0")
		  (match_operand:SWI124 2 "<general_operand>" "<g>,<r><i>"))
		 (const_int 0)))
   (set (match_operand:SWI124 0 "nonimmediate_operand" "=<r>,<r>m")
	(and:SWI124 (match_dup 1) (match_dup 2)))]
  "ix86_match_ccmode (insn, CCNOmode)
   && ix86_binary_operator_ok (AND, <MODE>mode, operands)"
  "and{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*andqi_2_slp"
  [(set (reg FLAGS_REG)
	(compare (and:QI
		   (match_operand:QI 0 "nonimmediate_operand" "+q,qm")
		   (match_operand:QI 1 "nonimmediate_operand" "qmn,qn"))
		 (const_int 0)))
   (set (strict_low_part (match_dup 0))
	(and:QI (match_dup 0) (match_dup 1)))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && ix86_match_ccmode (insn, CCNOmode)
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "and{b}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "mode" "QI")])

(define_insn "andqi_ext_1"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q,Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (and:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "0,0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (match_operand:QI 2 "general_operand" "QnBc,m")) 0))
   (clobber (reg:CC FLAGS_REG))]
  "/* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   rtx_equal_p (operands[0], operands[1])"
  "and{b}\t{%2, %h0|%h0, %2}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "alu")
   (set_attr "mode" "QI")])

;; Generated by peephole translating test to and.  This shows up
;; often in fp comparisons.
(define_insn "*andqi_ext_1_cc"
  [(set (reg FLAGS_REG)
	(compare
	  (and:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "0,0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (match_operand:QI 2 "general_operand" "QnBc,m"))
	  (const_int 0)))
   (set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q,Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (and:QI
	    (subreg:QI
	      (zero_extract:SI (match_dup 1)
			       (const_int 8)
			       (const_int 8)) 0)
	    (match_dup 2)) 0))]
  "ix86_match_ccmode (insn, CCNOmode)
   /* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   && rtx_equal_p (operands[0], operands[1])"
  "and{b}\t{%2, %h0|%h0, %2}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "alu")
   (set_attr "mode" "QI")])

(define_insn "*andqi_ext_2"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (and:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "%0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (subreg:QI
	      (zero_extract:SI (match_operand 2 "ext_register_operand" "Q")
			       (const_int 8)
			       (const_int 8)) 0)) 0))
   (clobber (reg:CC FLAGS_REG))]
  "/* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   rtx_equal_p (operands[0], operands[1])
   || rtx_equal_p (operands[0], operands[2])"
  "and{b}\t{%h2, %h0|%h0, %h2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "QI")])

;; Convert wide AND instructions with immediate operand to shorter QImode
;; equivalents when possible.
;; Don't do the splitting with memory operands, since it introduces risk
;; of memory mismatch stalls.  We may want to do the splitting for optimizing
;; for size, but that can (should?) be handled by generic code instead.
(define_split
  [(set (match_operand:SWI248 0 "QIreg_operand")
	(and:SWI248 (match_operand:SWI248 1 "register_operand")
		    (match_operand:SWI248 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
   "reload_completed
    && (!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
    && !(~INTVAL (operands[2]) & ~(255 << 8))"
  [(parallel
     [(set (zero_extract:SI (match_dup 0)
			    (const_int 8)
			    (const_int 8))
	   (subreg:SI
	     (and:QI
	       (subreg:QI
		 (zero_extract:SI (match_dup 1)
				  (const_int 8)
				  (const_int 8)) 0)
	       (match_dup 2)) 0))
      (clobber (reg:CC FLAGS_REG))])]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[2] = gen_int_mode (INTVAL (operands[2]) >> 8, QImode);
})

;; Since AND can be encoded with sign extended immediate, this is only
;; profitable when 7th bit is not set.
(define_split
  [(set (match_operand:SWI248 0 "any_QIreg_operand")
	(and:SWI248 (match_operand:SWI248 1 "general_operand")
		    (match_operand:SWI248 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
   "reload_completed
    && (!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
    && !(~INTVAL (operands[2]) & ~255)
    && !(INTVAL (operands[2]) & 128)"
  [(parallel [(set (strict_low_part (match_dup 0))
		   (and:QI (match_dup 1)
			   (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[0] = gen_lowpart (QImode, operands[0]);
  operands[1] = gen_lowpart (QImode, operands[1]);
  operands[2] = gen_int_mode (INTVAL (operands[2]), QImode);
})

(define_insn "*andndi3_doubleword"
  [(set (match_operand:DI 0 "register_operand" "=r,&r")
	(and:DI
	  (not:DI (match_operand:DI 1 "register_operand" "r,0"))
	  (match_operand:DI 2 "nonimmediate_operand" "rm,rm")))
   (clobber (reg:CC FLAGS_REG))]
  "!TARGET_64BIT && TARGET_STV && TARGET_SSE2"
  "#"
  [(set_attr "isa" "bmi,*")])

(define_split
  [(set (match_operand:DI 0 "register_operand")
	(and:DI
	  (not:DI (match_operand:DI 1 "register_operand"))
	  (match_operand:DI 2 "nonimmediate_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "!TARGET_64BIT && TARGET_BMI && TARGET_STV && TARGET_SSE2
   && reload_completed"
  [(parallel [(set (match_dup 0)
		   (and:SI (not:SI (match_dup 1)) (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])
   (parallel [(set (match_dup 3)
		   (and:SI (not:SI (match_dup 4)) (match_dup 5)))
	      (clobber (reg:CC FLAGS_REG))])]
  "split_double_mode (DImode, &operands[0], 3, &operands[0], &operands[3]);")

(define_split
  [(set (match_operand:DI 0 "register_operand")
	(and:DI
	  (not:DI (match_dup 0))
	  (match_operand:DI 1 "nonimmediate_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "!TARGET_64BIT && !TARGET_BMI && TARGET_STV && TARGET_SSE2
   && reload_completed"
  [(set (match_dup 0) (not:SI (match_dup 0)))
   (parallel [(set (match_dup 0)
		   (and:SI (match_dup 0) (match_dup 1)))
	      (clobber (reg:CC FLAGS_REG))])
   (set (match_dup 2) (not:SI (match_dup 2)))
   (parallel [(set (match_dup 2)
		   (and:SI (match_dup 2) (match_dup 3)))
	      (clobber (reg:CC FLAGS_REG))])]
  "split_double_mode (DImode, &operands[0], 2, &operands[0], &operands[2]);")

(define_insn "*andn<mode>_1"
  [(set (match_operand:SWI48 0 "register_operand" "=r,r")
	(and:SWI48
	  (not:SWI48 (match_operand:SWI48 1 "register_operand" "r,r"))
	  (match_operand:SWI48 2 "nonimmediate_operand" "r,m")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_BMI"
  "andn\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "type" "bitmanip")
   (set_attr "btver2_decode" "direct, double")
   (set_attr "mode" "<MODE>")])

(define_insn "*andn<mode>_1"
  [(set (match_operand:SWI12 0 "register_operand" "=r")
	(and:SWI12
	  (not:SWI12 (match_operand:SWI12 1 "register_operand" "r"))
	  (match_operand:SWI12 2 "register_operand" "r")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_BMI"
  "andn\t{%k2, %k1, %k0|%k0, %k1, %k2}"
  [(set_attr "type" "bitmanip")
   (set_attr "btver2_decode" "direct")
   (set_attr "mode" "SI")])

(define_insn "*andn_<mode>_ccno"
  [(set (reg FLAGS_REG)
	(compare
	  (and:SWI48
	    (not:SWI48 (match_operand:SWI48 1 "register_operand" "r,r"))
	    (match_operand:SWI48 2 "nonimmediate_operand" "r,m"))
	  (const_int 0)))
   (clobber (match_scratch:SWI48 0 "=r,r"))]
  "TARGET_BMI && ix86_match_ccmode (insn, CCNOmode)"
  "andn\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "type" "bitmanip")
   (set_attr "btver2_decode" "direct, double")
   (set_attr "mode" "<MODE>")])

;; Logical inclusive and exclusive OR instructions

;; %%% This used to optimize known byte-wide and operations to memory.
;; If this is considered useful, it should be done with splitters.

(define_expand "<code><mode>3"
  [(set (match_operand:SWIM1248x 0 "nonimmediate_operand")
	(any_or:SWIM1248x (match_operand:SWIM1248x 1 "nonimmediate_operand")
			     (match_operand:SWIM1248x 2 "<general_operand>")))]
  ""
  "ix86_expand_binary_operator (<CODE>, <MODE>mode, operands); DONE;")

(define_insn_and_split "*<code>di3_doubleword"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=r,rm,r")
	(any_or:DI
	 (match_operand:DI 1 "nonimmediate_operand" "%0,0,0")
	 (match_operand:DI 2 "x86_64_szext_general_operand" "Z,re,rm")))
   (clobber (reg:CC FLAGS_REG))]
  "!TARGET_64BIT && TARGET_STV && TARGET_SSE2
   && ix86_binary_operator_ok (<CODE>, DImode, operands)"
  "#"
  "&& reload_completed"
  [(const_int 0)]
{
  split_double_mode (DImode, &operands[0], 3, &operands[0], &operands[3]);
  if (operands[2] == constm1_rtx)
    {
      if (<CODE> == IOR)
	{
	  operands[1] = constm1_rtx;
	  ix86_expand_move (SImode, &operands[0]);
	}
      else
	ix86_expand_unary_operator (NOT, SImode, &operands[0]);
    }
  else if (operands[2] != const0_rtx)
    ix86_expand_binary_operator (<CODE>, SImode, &operands[0]);
  else if (operands[5] == const0_rtx)
    emit_note (NOTE_INSN_DELETED);
  if (operands[5] == constm1_rtx)
    {
      if (<CODE> == IOR)
	{
	  operands[4] = constm1_rtx;
	  ix86_expand_move (SImode, &operands[3]);
	}
      else
	ix86_expand_unary_operator (NOT, SImode, &operands[3]);
    }
  else if (operands[5] != const0_rtx)
    ix86_expand_binary_operator (<CODE>, SImode, &operands[3]);
  DONE;
})

(define_insn "*<code><mode>_1"
  [(set (match_operand:SWI248 0 "nonimmediate_operand" "=r,rm")
	(any_or:SWI248
	 (match_operand:SWI248 1 "nonimmediate_operand" "%0,0")
	 (match_operand:SWI248 2 "<general_operand>" "<g>,r<i>")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
  "<logic>{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn_and_split "*iordi_1_bts"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
	(ior:DI
	 (match_operand:DI 1 "nonimmediate_operand" "%0")
	 (match_operand:DI 2 "const_int_operand" "n")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && TARGET_USE_BT
   && ix86_binary_operator_ok (IOR, DImode, operands)
   && IN_RANGE (exact_log2 (INTVAL (operands[2])), 31, 63)"
  "#"
  "&& reload_completed"
  [(parallel [(set (zero_extract:DI (match_dup 0)
				    (const_int 1)
				    (match_dup 3))
		   (const_int 1))
	      (clobber (reg:CC FLAGS_REG))])]
  "operands[3] = GEN_INT (exact_log2 (INTVAL (operands[2])));"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "DI")])

(define_insn_and_split "*xordi_1_btc"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
	(xor:DI
	 (match_operand:DI 1 "nonimmediate_operand" "%0")
	 (match_operand:DI 2 "const_int_operand" "n")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && TARGET_USE_BT
   && ix86_binary_operator_ok (XOR, DImode, operands)
   && IN_RANGE (exact_log2 (INTVAL (operands[2])), 31, 63)"
  "#"
  "&& reload_completed"
  [(parallel [(set (zero_extract:DI (match_dup 0)
				    (const_int 1)
				    (match_dup 3))
		   (not:DI (zero_extract:DI (match_dup 0)
					    (const_int 1)
					    (match_dup 3))))
	      (clobber (reg:CC FLAGS_REG))])]
  "operands[3] = GEN_INT (exact_log2 (INTVAL (operands[2])));"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "DI")])

;; See comment for addsi_1_zext why we do use nonimmediate_operand
(define_insn "*<code>si_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	 (any_or:SI (match_operand:SI 1 "nonimmediate_operand" "%0")
		    (match_operand:SI 2 "x86_64_general_operand" "rme"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (<CODE>, SImode, operands)"
  "<logic>{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*<code>si_1_zext_imm"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(any_or:DI
	 (zero_extend:DI (match_operand:SI 1 "register_operand" "%0"))
	 (match_operand:DI 2 "x86_64_zext_immediate_operand" "Z")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (<CODE>, SImode, operands)"
  "<logic>{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*<code>qi_1"
  [(set (match_operand:QI 0 "nonimmediate_operand" "=q,m,r")
	(any_or:QI (match_operand:QI 1 "nonimmediate_operand" "%0,0,0")
		   (match_operand:QI 2 "general_operand" "qmn,qn,rn")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, QImode, operands)"
  "@
   <logic>{b}\t{%2, %0|%0, %2}
   <logic>{b}\t{%2, %0|%0, %2}
   <logic>{l}\t{%k2, %k0|%k0, %k2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "QI,QI,SI")
   ;; Potential partial reg stall on alternative 2.
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "2")
	      (symbol_ref "!TARGET_PARTIAL_REG_STALL")]
	   (symbol_ref "true")))])

(define_insn "*<code>qi_1_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+q,m"))
	(any_or:QI (match_dup 0)
		   (match_operand:QI 1 "general_operand" "qmn,qn")))
   (clobber (reg:CC FLAGS_REG))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "<logic>{b}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "mode" "QI")])

(define_insn "*<code><mode>_2"
  [(set (reg FLAGS_REG)
	(compare (any_or:SWI
		  (match_operand:SWI 1 "nonimmediate_operand" "%0,0")
		  (match_operand:SWI 2 "<general_operand>" "<g>,<r><i>"))
		 (const_int 0)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>,<r>m")
	(any_or:SWI (match_dup 1) (match_dup 2)))]
  "ix86_match_ccmode (insn, CCNOmode)
   && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
  "<logic>{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

;; See comment for addsi_1_zext why we do use nonimmediate_operand
;; ??? Special case for immediate operand is missing - it is tricky.
(define_insn "*<code>si_2_zext"
  [(set (reg FLAGS_REG)
	(compare (any_or:SI (match_operand:SI 1 "nonimmediate_operand" "%0")
			    (match_operand:SI 2 "x86_64_general_operand" "rme"))
		 (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (any_or:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCNOmode)
   && ix86_binary_operator_ok (<CODE>, SImode, operands)"
  "<logic>{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*<code>si_2_zext_imm"
  [(set (reg FLAGS_REG)
	(compare (any_or:SI
		  (match_operand:SI 1 "nonimmediate_operand" "%0")
		  (match_operand:SI 2 "x86_64_zext_immediate_operand" "Z"))
		 (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(any_or:DI (zero_extend:DI (match_dup 1)) (match_dup 2)))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCNOmode)
   && ix86_binary_operator_ok (<CODE>, SImode, operands)"
  "<logic>{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "SI")])

(define_insn "*<code>qi_2_slp"
  [(set (reg FLAGS_REG)
	(compare (any_or:QI (match_operand:QI 0 "nonimmediate_operand" "+q,qm")
			    (match_operand:QI 1 "general_operand" "qmn,qn"))
		 (const_int 0)))
   (set (strict_low_part (match_dup 0))
	(any_or:QI (match_dup 0) (match_dup 1)))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   && ix86_match_ccmode (insn, CCNOmode)
   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
  "<logic>{b}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "mode" "QI")])

(define_insn "*<code><mode>_3"
  [(set (reg FLAGS_REG)
	(compare (any_or:SWI
		  (match_operand:SWI 1 "nonimmediate_operand" "%0")
		  (match_operand:SWI 2 "<general_operand>" "<g>"))
		 (const_int 0)))
   (clobber (match_scratch:SWI 0 "=<r>"))]
  "ix86_match_ccmode (insn, CCNOmode)
   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
  "<logic>{<imodesuffix>}\t{%2, %0|%0, %2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "<MODE>")])

(define_insn "*<code>qi_ext_1"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q,Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (any_or:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "0,0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (match_operand:QI 2 "general_operand" "QnBc,m")) 0))
   (clobber (reg:CC FLAGS_REG))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   /* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   && rtx_equal_p (operands[0], operands[1])"
  "<logic>{b}\t{%2, %h0|%h0, %2}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "alu")
   (set_attr "mode" "QI")])

(define_insn "*<code>qi_ext_2"
  [(set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (any_or:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "%0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (subreg:QI
	      (zero_extract:SI (match_operand 2 "ext_register_operand" "Q")
			       (const_int 8)
			       (const_int 8)) 0)) 0))
   (clobber (reg:CC FLAGS_REG))]
  "(!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
   /* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   && (rtx_equal_p (operands[0], operands[1])
       || rtx_equal_p (operands[0], operands[2]))"
  "<logic>{b}\t{%h2, %h0|%h0, %h2}"
  [(set_attr "type" "alu")
   (set_attr "mode" "QI")])

;; Convert wide OR instructions with immediate operand to shorter QImode
;; equivalents when possible.
;; Don't do the splitting with memory operands, since it introduces risk
;; of memory mismatch stalls.  We may want to do the splitting for optimizing
;; for size, but that can (should?) be handled by generic code instead.
(define_split
  [(set (match_operand:SWI248 0 "QIreg_operand")
	(any_or:SWI248 (match_operand:SWI248 1 "register_operand")
		       (match_operand:SWI248 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
   "reload_completed
    && (!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
    && !(INTVAL (operands[2]) & ~(255 << 8))"
  [(parallel
     [(set (zero_extract:SI (match_dup 0)
			    (const_int 8)
			    (const_int 8))
	   (subreg:SI
	     (any_or:QI
	       (subreg:QI
		 (zero_extract:SI (match_dup 1)
				  (const_int 8)
				  (const_int 8)) 0)
	       (match_dup 2)) 0))
      (clobber (reg:CC FLAGS_REG))])]
{
  operands[0] = gen_lowpart (SImode, operands[0]);
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[2] = gen_int_mode (INTVAL (operands[2]) >> 8, QImode);
})

;; Since OR can be encoded with sign extended immediate, this is only
;; profitable when 7th bit is set.
(define_split
  [(set (match_operand:SWI248 0 "any_QIreg_operand")
	(any_or:SWI248 (match_operand:SWI248 1 "general_operand")
		       (match_operand:SWI248 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
   "reload_completed
    && (!TARGET_PARTIAL_REG_STALL || optimize_function_for_size_p (cfun))
    && !(INTVAL (operands[2]) & ~255)
    && (INTVAL (operands[2]) & 128)"
  [(parallel [(set (strict_low_part (match_dup 0))
		   (any_or:QI (match_dup 1)
			      (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[0] = gen_lowpart (QImode, operands[0]);
  operands[1] = gen_lowpart (QImode, operands[1]);
  operands[2] = gen_int_mode (INTVAL (operands[2]), QImode);
})

(define_expand "xorqi_ext_1_cc"
  [(parallel [
     (set (reg:CCNO FLAGS_REG)
	  (compare:CCNO
	    (xor:QI
	      (subreg:QI
		(zero_extract:SI (match_operand 1 "ext_register_operand")
				 (const_int 8)
				 (const_int 8)) 0)
	      (match_operand 2 "const_int_operand"))
	    (const_int 0)))
     (set (zero_extract:SI (match_operand 0 "ext_register_operand")
			   (const_int 8)
			   (const_int 8))
	  (subreg:SI
	    (xor:QI
	      (subreg:QI
		(zero_extract:SI (match_dup 1)
				 (const_int 8)
				 (const_int 8)) 0)
	    (match_dup 2)) 0))])])

(define_insn "*xorqi_ext_1_cc"
  [(set (reg FLAGS_REG)
	(compare
	  (xor:QI
	    (subreg:QI
	      (zero_extract:SI (match_operand 1 "ext_register_operand" "0,0")
			       (const_int 8)
			       (const_int 8)) 0)
	    (match_operand:QI 2 "general_operand" "QnBc,m"))
	  (const_int 0)))
   (set (zero_extract:SI (match_operand 0 "ext_register_operand" "+Q,Q")
			 (const_int 8)
			 (const_int 8))
	(subreg:SI
	  (xor:QI
	    (subreg:QI
	      (zero_extract:SI (match_dup 1)
			       (const_int 8)
			       (const_int 8)) 0)
	  (match_dup 2)) 0))]
  "ix86_match_ccmode (insn, CCNOmode)
   /* FIXME: without this LRA can't reload this pattern, see PR82524.  */
   && rtx_equal_p (operands[0], operands[1])"
  "xor{b}\t{%2, %h0|%h0, %2}"
  [(set_attr "isa" "*,nox64")
   (set_attr "type" "alu")
   (set_attr "mode" "QI")])

;; Negation instructions

(define_expand "neg<mode>2"
  [(set (match_operand:SDWIM 0 "nonimmediate_operand")
	(neg:SDWIM (match_operand:SDWIM 1 "nonimmediate_operand")))]
  ""
  "ix86_expand_unary_operator (NEG, <MODE>mode, operands); DONE;")

(define_insn_and_split "*neg<dwi>2_doubleword"
  [(set (match_operand:<DWI> 0 "nonimmediate_operand" "=ro")
	(neg:<DWI> (match_operand:<DWI> 1 "nonimmediate_operand" "0")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_unary_operator_ok (NEG, <DWI>mode, operands)"
  "#"
  "reload_completed"
  [(parallel
    [(set (reg:CCZ FLAGS_REG)
	  (compare:CCZ (neg:DWIH (match_dup 1)) (const_int 0)))
     (set (match_dup 0) (neg:DWIH (match_dup 1)))])
   (parallel
    [(set (match_dup 2)
	  (plus:DWIH (plus:DWIH (ltu:DWIH (reg:CC FLAGS_REG) (const_int 0))
				(match_dup 3))
		     (const_int 0)))
     (clobber (reg:CC FLAGS_REG))])
   (parallel
    [(set (match_dup 2)
	  (neg:DWIH (match_dup 2)))
     (clobber (reg:CC FLAGS_REG))])]
  "split_double_mode (<DWI>mode, &operands[0], 2, &operands[0], &operands[2]);")

(define_insn "*neg<mode>2_1"
  [(set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(neg:SWI (match_operand:SWI 1 "nonimmediate_operand" "0")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_unary_operator_ok (NEG, <MODE>mode, operands)"
  "neg{<imodesuffix>}\t%0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "<MODE>")])

;; Combine is quite creative about this pattern.
(define_insn "*negsi2_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(lshiftrt:DI
	  (neg:DI (ashift:DI (match_operand:DI 1 "register_operand" "0")
			     (const_int 32)))
	(const_int 32)))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_unary_operator_ok (NEG, SImode, operands)"
  "neg{l}\t%k0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "SI")])

;; The problem with neg is that it does not perform (compare x 0),
;; it really performs (compare 0 x), which leaves us with the zero
;; flag being the only useful item.

(define_insn "*neg<mode>2_cmpz"
  [(set (reg:CCZ FLAGS_REG)
	(compare:CCZ
	  (neg:SWI (match_operand:SWI 1 "nonimmediate_operand" "0"))
		   (const_int 0)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(neg:SWI (match_dup 1)))]
  "ix86_unary_operator_ok (NEG, <MODE>mode, operands)"
  "neg{<imodesuffix>}\t%0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "<MODE>")])

(define_insn "*negsi2_cmpz_zext"
  [(set (reg:CCZ FLAGS_REG)
	(compare:CCZ
	  (lshiftrt:DI
	    (neg:DI (ashift:DI
		      (match_operand:DI 1 "register_operand" "0")
		      (const_int 32)))
	    (const_int 32))
	  (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(lshiftrt:DI (neg:DI (ashift:DI (match_dup 1)
					(const_int 32)))
		     (const_int 32)))]
  "TARGET_64BIT && ix86_unary_operator_ok (NEG, SImode, operands)"
  "neg{l}\t%k0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "SI")])

;; Negate with jump on overflow.
(define_expand "negv<mode>3"
  [(parallel [(set (reg:CCO FLAGS_REG)
		   (ne:CCO (match_operand:SWI 1 "register_operand")
			   (match_dup 3)))
	      (set (match_operand:SWI 0 "register_operand")
		   (neg:SWI (match_dup 1)))])
   (set (pc) (if_then_else
	       (eq (reg:CCO FLAGS_REG) (const_int 0))
	       (label_ref (match_operand 2))
	       (pc)))]
  ""
{
  operands[3]
    = gen_int_mode (HOST_WIDE_INT_1U << (GET_MODE_BITSIZE (<MODE>mode) - 1),
		    <MODE>mode);
})

(define_insn "*negv<mode>3"
  [(set (reg:CCO FLAGS_REG)
	(ne:CCO (match_operand:SWI 1 "nonimmediate_operand" "0")
		(match_operand:SWI 2 "const_int_operand")))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(neg:SWI (match_dup 1)))]
  "ix86_unary_operator_ok (NEG, <MODE>mode, operands)
   && mode_signbit_p (<MODE>mode, operands[2])"
  "neg{<imodesuffix>}\t%0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "<MODE>")])

;; Changing of sign for FP values is doable using integer unit too.

(define_expand "<code><mode>2"
  [(set (match_operand:X87MODEF 0 "register_operand")
	(absneg:X87MODEF (match_operand:X87MODEF 1 "register_operand")))]
  "TARGET_80387 || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)"
  "ix86_expand_fp_absneg_operator (<CODE>, <MODE>mode, operands); DONE;")

(define_insn "*absneg<mode>2"
  [(set (match_operand:MODEF 0 "register_operand" "=Yv,Yv,f,!r")
	(match_operator:MODEF 3 "absneg_operator"
	  [(match_operand:MODEF 1 "register_operand" "0,Yv,0,0")]))
   (use (match_operand:<ssevecmode> 2 "nonimmediate_operand" "Yvm,0,X,X"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_80387 || (SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)"
  "#"
  [(set (attr "enabled")
     (if_then_else
       (match_test ("SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH"))
       (if_then_else
	 (eq_attr "alternative" "2")
	 (symbol_ref "TARGET_MIX_SSE_I387")
	 (symbol_ref "true"))
       (if_then_else
	 (eq_attr "alternative" "2,3")
	 (symbol_ref "true")
	 (symbol_ref "false"))))])

(define_insn "*absnegxf2_i387"
  [(set (match_operand:XF 0 "register_operand" "=f,!r")
	(match_operator:XF 3 "absneg_operator"
	  [(match_operand:XF 1 "register_operand" "0,0")]))
   (use (match_operand 2))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_80387"
  "#")

(define_expand "<code>tf2"
  [(set (match_operand:TF 0 "register_operand")
	(absneg:TF (match_operand:TF 1 "register_operand")))]
  "TARGET_SSE"
  "ix86_expand_fp_absneg_operator (<CODE>, TFmode, operands); DONE;")

(define_insn "*absnegtf2_sse"
  [(set (match_operand:TF 0 "register_operand" "=Yv,Yv")
	(match_operator:TF 3 "absneg_operator"
	  [(match_operand:TF 1 "register_operand" "0,Yv")]))
   (use (match_operand:TF 2 "nonimmediate_operand" "Yvm,0"))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_SSE"
  "#")

;; Splitters for fp abs and neg.

(define_split
  [(set (match_operand 0 "fp_register_operand")
	(match_operator 1 "absneg_operator" [(match_dup 0)]))
   (use (match_operand 2))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  [(set (match_dup 0) (match_op_dup 1 [(match_dup 0)]))])

(define_split
  [(set (match_operand 0 "sse_reg_operand")
	(match_operator 3 "absneg_operator"
	  [(match_operand 1 "register_operand")]))
   (use (match_operand 2 "nonimmediate_operand"))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  [(set (match_dup 0) (match_dup 3))]
{
  machine_mode mode = GET_MODE (operands[0]);
  machine_mode vmode = GET_MODE (operands[2]);
  rtx tmp;

  operands[0] = lowpart_subreg (vmode, operands[0], mode);
  operands[1] = lowpart_subreg (vmode, operands[1], mode);
  if (operands_match_p (operands[0], operands[2]))
    std::swap (operands[1], operands[2]);
  if (GET_CODE (operands[3]) == ABS)
    tmp = gen_rtx_AND (vmode, operands[1], operands[2]);
  else
    tmp = gen_rtx_XOR (vmode, operands[1], operands[2]);
  operands[3] = tmp;
})

(define_split
  [(set (match_operand:SF 0 "general_reg_operand")
	(match_operator:SF 1 "absneg_operator" [(match_dup 0)]))
   (use (match_operand:V4SF 2))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  [(parallel [(set (match_dup 0) (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
{
  rtx tmp;
  operands[0] = gen_lowpart (SImode, operands[0]);
  if (GET_CODE (operands[1]) == ABS)
    {
      tmp = gen_int_mode (0x7fffffff, SImode);
      tmp = gen_rtx_AND (SImode, operands[0], tmp);
    }
  else
    {
      tmp = gen_int_mode (0x80000000, SImode);
      tmp = gen_rtx_XOR (SImode, operands[0], tmp);
    }
  operands[1] = tmp;
})

(define_split
  [(set (match_operand:DF 0 "general_reg_operand")
	(match_operator:DF 1 "absneg_operator" [(match_dup 0)]))
   (use (match_operand 2))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  [(parallel [(set (match_dup 0) (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
{
  rtx tmp;
  if (TARGET_64BIT)
    {
      tmp = gen_lowpart (DImode, operands[0]);
      tmp = gen_rtx_ZERO_EXTRACT (DImode, tmp, const1_rtx, GEN_INT (63));
      operands[0] = tmp;

      if (GET_CODE (operands[1]) == ABS)
	tmp = const0_rtx;
      else
	tmp = gen_rtx_NOT (DImode, tmp);
    }
  else
    {
      operands[0] = gen_highpart (SImode, operands[0]);
      if (GET_CODE (operands[1]) == ABS)
	{
	  tmp = gen_int_mode (0x7fffffff, SImode);
	  tmp = gen_rtx_AND (SImode, operands[0], tmp);
	}
      else
	{
	  tmp = gen_int_mode (0x80000000, SImode);
	  tmp = gen_rtx_XOR (SImode, operands[0], tmp);
	}
    }
  operands[1] = tmp;
})

(define_split
  [(set (match_operand:XF 0 "general_reg_operand")
	(match_operator:XF 1 "absneg_operator" [(match_dup 0)]))
   (use (match_operand 2))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed"
  [(parallel [(set (match_dup 0) (match_dup 1))
	      (clobber (reg:CC FLAGS_REG))])]
{
  rtx tmp;
  operands[0] = gen_rtx_REG (SImode,
			     REGNO (operands[0]) + (TARGET_64BIT ? 1 : 2));
  if (GET_CODE (operands[1]) == ABS)
    {
      tmp = GEN_INT (0x7fff);
      tmp = gen_rtx_AND (SImode, operands[0], tmp);
    }
  else
    {
      tmp = GEN_INT (0x8000);
      tmp = gen_rtx_XOR (SImode, operands[0], tmp);
    }
  operands[1] = tmp;
})

;; Conditionalize these after reload. If they match before reload, we
;; lose the clobber and ability to use integer instructions.

(define_insn "*<code><mode>2_1"
  [(set (match_operand:X87MODEF 0 "register_operand" "=f")
	(absneg:X87MODEF (match_operand:X87MODEF 1 "register_operand" "0")))]
  "TARGET_80387
   && (reload_completed
       || !(SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH))"
  "f<absneg_mnemonic>"
  [(set_attr "type" "fsgn")
   (set_attr "mode" "<MODE>")])

(define_insn "*<code>extendsfdf2"
  [(set (match_operand:DF 0 "register_operand" "=f")
	(absneg:DF (float_extend:DF
		     (match_operand:SF 1 "register_operand" "0"))))]
  "TARGET_80387 && (!TARGET_SSE_MATH || TARGET_MIX_SSE_I387)"
  "f<absneg_mnemonic>"
  [(set_attr "type" "fsgn")
   (set_attr "mode" "DF")])

(define_insn "*<code>extendsfxf2"
  [(set (match_operand:XF 0 "register_operand" "=f")
	(absneg:XF (float_extend:XF
		     (match_operand:SF 1 "register_operand" "0"))))]
  "TARGET_80387"
  "f<absneg_mnemonic>"
  [(set_attr "type" "fsgn")
   (set_attr "mode" "XF")])

(define_insn "*<code>extenddfxf2"
  [(set (match_operand:XF 0 "register_operand" "=f")
	(absneg:XF (float_extend:XF
		     (match_operand:DF 1 "register_operand" "0"))))]
  "TARGET_80387"
  "f<absneg_mnemonic>"
  [(set_attr "type" "fsgn")
   (set_attr "mode" "XF")])

;; Copysign instructions

(define_mode_iterator CSGNMODE [SF DF TF])
(define_mode_attr CSGNVMODE [(SF "V4SF") (DF "V2DF") (TF "TF")])

(define_expand "copysign<mode>3"
  [(match_operand:CSGNMODE 0 "register_operand")
   (match_operand:CSGNMODE 1 "nonmemory_operand")
   (match_operand:CSGNMODE 2 "register_operand")]
  "(SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)
   || (TARGET_SSE && (<MODE>mode == TFmode))"
  "ix86_expand_copysign (operands); DONE;")

(define_insn_and_split "copysign<mode>3_const"
  [(set (match_operand:CSGNMODE 0 "register_operand" "=Yv")
	(unspec:CSGNMODE
	  [(match_operand:<CSGNVMODE> 1 "vector_move_operand" "YvmC")
	   (match_operand:CSGNMODE 2 "register_operand" "0")
	   (match_operand:<CSGNVMODE> 3 "nonimmediate_operand" "Yvm")]
	  UNSPEC_COPYSIGN))]
  "(SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)
   || (TARGET_SSE && (<MODE>mode == TFmode))"
  "#"
  "&& reload_completed"
  [(const_int 0)]
  "ix86_split_copysign_const (operands); DONE;")

(define_insn "copysign<mode>3_var"
  [(set (match_operand:CSGNMODE 0 "register_operand" "=Yv,Yv,Yv,Yv,Yv")
	(unspec:CSGNMODE
	  [(match_operand:CSGNMODE 2 "register_operand"	"Yv,0,0,Yv,Yv")
	   (match_operand:CSGNMODE 3 "register_operand"	"1,1,Yv,1,Yv")
	   (match_operand:<CSGNVMODE> 4
	     "nonimmediate_operand" "X,Yvm,Yvm,0,0")
	   (match_operand:<CSGNVMODE> 5
	     "nonimmediate_operand" "0,Yvm,1,Yvm,1")]
	  UNSPEC_COPYSIGN))
   (clobber (match_scratch:<CSGNVMODE> 1 "=Yv,Yv,Yv,Yv,Yv"))]
  "(SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)
   || (TARGET_SSE && (<MODE>mode == TFmode))"
  "#")

(define_split
  [(set (match_operand:CSGNMODE 0 "register_operand")
	(unspec:CSGNMODE
	  [(match_operand:CSGNMODE 2 "register_operand")
	   (match_operand:CSGNMODE 3 "register_operand")
	   (match_operand:<CSGNVMODE> 4)
	   (match_operand:<CSGNVMODE> 5)]
	  UNSPEC_COPYSIGN))
   (clobber (match_scratch:<CSGNVMODE> 1))]
  "((SSE_FLOAT_MODE_P (<MODE>mode) && TARGET_SSE_MATH)
    || (TARGET_SSE && (<MODE>mode == TFmode)))
   && reload_completed"
  [(const_int 0)]
  "ix86_split_copysign_var (operands); DONE;")

;; One complement instructions

(define_expand "one_cmpl<mode>2"
  [(set (match_operand:SWIM1248x 0 "nonimmediate_operand")
	(not:SWIM1248x (match_operand:SWIM1248x 1 "nonimmediate_operand")))]
  ""
  "ix86_expand_unary_operator (NOT, <MODE>mode, operands); DONE;")

(define_insn_and_split "*one_cmpldi2_doubleword"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
	(not:DI (match_operand:DI 1 "nonimmediate_operand" "0")))]
  "!TARGET_64BIT && TARGET_STV && TARGET_SSE2
   && ix86_unary_operator_ok (NOT, DImode, operands)"
  "#"
  "&& reload_completed"
  [(set (match_dup 0)
	(not:SI (match_dup 1)))
   (set (match_dup 2)
	(not:SI (match_dup 3)))]
  "split_double_mode (DImode, &operands[0], 2, &operands[0], &operands[2]);")

(define_insn "*one_cmpl<mode>2_1"
  [(set (match_operand:SWI248 0 "nonimmediate_operand" "=rm")
	(not:SWI248 (match_operand:SWI248 1 "nonimmediate_operand" "0")))]
  "ix86_unary_operator_ok (NOT, <MODE>mode, operands)"
  "not{<imodesuffix>}\t%0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "<MODE>")])

;; ??? Currently never generated - xor is used instead.
(define_insn "*one_cmplsi2_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (not:SI (match_operand:SI 1 "register_operand" "0"))))]
  "TARGET_64BIT && ix86_unary_operator_ok (NOT, SImode, operands)"
  "not{l}\t%k0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "SI")])

(define_insn "*one_cmplqi2_1"
  [(set (match_operand:QI 0 "nonimmediate_operand" "=qm,r")
	(not:QI (match_operand:QI 1 "nonimmediate_operand" "0,0")))]
  "ix86_unary_operator_ok (NOT, QImode, operands)"
  "@
   not{b}\t%0
   not{l}\t%k0"
  [(set_attr "type" "negnot")
   (set_attr "mode" "QI,SI")
   ;; Potential partial reg stall on alternative 1.
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "1")
	      (symbol_ref "!TARGET_PARTIAL_REG_STALL")]
	   (symbol_ref "true")))])

(define_insn "*one_cmpl<mode>2_2"
  [(set (reg FLAGS_REG)
	(compare (not:SWI (match_operand:SWI 1 "nonimmediate_operand" "0"))
		 (const_int 0)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(not:SWI (match_dup 1)))]
  "ix86_match_ccmode (insn, CCNOmode)
   && ix86_unary_operator_ok (NOT, <MODE>mode, operands)"
  "#"
  [(set_attr "type" "alu1")
   (set_attr "mode" "<MODE>")])

(define_split
  [(set (match_operand 0 "flags_reg_operand")
	(match_operator 2 "compare_operator"
	  [(not:SWI (match_operand:SWI 3 "nonimmediate_operand"))
	   (const_int 0)]))
   (set (match_operand:SWI 1 "nonimmediate_operand")
	(not:SWI (match_dup 3)))]
  "ix86_match_ccmode (insn, CCNOmode)"
  [(parallel [(set (match_dup 0)
		   (match_op_dup 2 [(xor:SWI (match_dup 3) (const_int -1))
				    (const_int 0)]))
	      (set (match_dup 1)
		   (xor:SWI (match_dup 3) (const_int -1)))])])

;; ??? Currently never generated - xor is used instead.
(define_insn "*one_cmplsi2_2_zext"
  [(set (reg FLAGS_REG)
	(compare (not:SI (match_operand:SI 1 "register_operand" "0"))
		 (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (not:SI (match_dup 1))))]
  "TARGET_64BIT && ix86_match_ccmode (insn, CCNOmode)
   && ix86_unary_operator_ok (NOT, SImode, operands)"
  "#"
  [(set_attr "type" "alu1")
   (set_attr "mode" "SI")])

(define_split
  [(set (match_operand 0 "flags_reg_operand")
	(match_operator 2 "compare_operator"
	  [(not:SI (match_operand:SI 3 "register_operand"))
	   (const_int 0)]))
   (set (match_operand:DI 1 "register_operand")
	(zero_extend:DI (not:SI (match_dup 3))))]
  "ix86_match_ccmode (insn, CCNOmode)"
  [(parallel [(set (match_dup 0)
		   (match_op_dup 2 [(xor:SI (match_dup 3) (const_int -1))
				    (const_int 0)]))
	      (set (match_dup 1)
		   (zero_extend:DI (xor:SI (match_dup 3) (const_int -1))))])])

;; Shift instructions

;; DImode shifts are implemented using the i386 "shift double" opcode,
;; which is written as "sh[lr]d[lw] imm,reg,reg/mem".  If the shift count
;; is variable, then the count is in %cl and the "imm" operand is dropped
;; from the assembler input.
;;
;; This instruction shifts the target reg/mem as usual, but instead of
;; shifting in zeros, bits are shifted in from reg operand.  If the insn
;; is a left shift double, bits are taken from the high order bits of
;; reg, else if the insn is a shift right double, bits are taken from the
;; low order bits of reg.  So if %eax is "1234" and %edx is "5678",
;; "shldl $8,%edx,%eax" leaves %edx unchanged and sets %eax to "2345".
;;
;; Since sh[lr]d does not change the `reg' operand, that is done
;; separately, making all shifts emit pairs of shift double and normal
;; shift.  Since sh[lr]d does not shift more than 31 bits, and we wish to
;; support a 63 bit shift, each shift where the count is in a reg expands
;; to a pair of shifts, a branch, a shift by 32 and a label.
;;
;; If the shift count is a constant, we need never emit more than one
;; shift pair, instead using moves and sign extension for counts greater
;; than 31.

(define_expand "ashl<mode>3"
  [(set (match_operand:SDWIM 0 "<shift_operand>")
	(ashift:SDWIM (match_operand:SDWIM 1 "<ashl_input_operand>")
		      (match_operand:QI 2 "nonmemory_operand")))]
  ""
  "ix86_expand_binary_operator (ASHIFT, <MODE>mode, operands); DONE;")

(define_insn "*ashl<mode>3_doubleword"
  [(set (match_operand:DWI 0 "register_operand" "=&r")
	(ashift:DWI (match_operand:DWI 1 "reg_or_pm1_operand" "0n")
		    (match_operand:QI 2 "nonmemory_operand" "<S>c")))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "#"
  [(set_attr "type" "multi")])

(define_split
  [(set (match_operand:DWI 0 "register_operand")
	(ashift:DWI (match_operand:DWI 1 "nonmemory_operand")
		    (match_operand:QI 2 "nonmemory_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "(optimize && flag_peephole2) ? epilogue_completed : reload_completed"
  [(const_int 0)]
  "ix86_split_ashl (operands, NULL_RTX, <MODE>mode); DONE;")

;; By default we don't ask for a scratch register, because when DWImode
;; values are manipulated, registers are already at a premium.  But if
;; we have one handy, we won't turn it away.

(define_peephole2
  [(match_scratch:DWIH 3 "r")
   (parallel [(set (match_operand:<DWI> 0 "register_operand")
		   (ashift:<DWI>
		     (match_operand:<DWI> 1 "nonmemory_operand")
		     (match_operand:QI 2 "nonmemory_operand")))
	      (clobber (reg:CC FLAGS_REG))])
   (match_dup 3)]
  "TARGET_CMOVE"
  [(const_int 0)]
  "ix86_split_ashl (operands, operands[3], <DWI>mode); DONE;")

(define_insn "x86_64_shld"
  [(set (match_operand:DI 0 "nonimmediate_operand" "+r*m")
        (ior:DI (ashift:DI (match_dup 0)
		  (match_operand:QI 2 "nonmemory_operand" "Jc"))
		(lshiftrt:DI (match_operand:DI 1 "register_operand" "r")
		  (minus:QI (const_int 64) (match_dup 2)))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "shld{q}\t{%s2%1, %0|%0, %1, %2}"
  [(set_attr "type" "ishift")
   (set_attr "prefix_0f" "1")
   (set_attr "mode" "DI")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "vector")
   (set_attr "bdver1_decode" "vector")])

(define_insn "x86_shld"
  [(set (match_operand:SI 0 "nonimmediate_operand" "+r*m")
        (ior:SI (ashift:SI (match_dup 0)
		  (match_operand:QI 2 "nonmemory_operand" "Ic"))
		(lshiftrt:SI (match_operand:SI 1 "register_operand" "r")
		  (minus:QI (const_int 32) (match_dup 2)))))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "shld{l}\t{%s2%1, %0|%0, %1, %2}"
  [(set_attr "type" "ishift")
   (set_attr "prefix_0f" "1")
   (set_attr "mode" "SI")
   (set_attr "pent_pair" "np")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "vector")
   (set_attr "bdver1_decode" "vector")])

(define_expand "x86_shift<mode>_adj_1"
  [(set (reg:CCZ FLAGS_REG)
	(compare:CCZ (and:QI (match_operand:QI 2 "register_operand")
			     (match_dup 4))
		     (const_int 0)))
   (set (match_operand:SWI48 0 "register_operand")
        (if_then_else:SWI48 (ne (reg:CCZ FLAGS_REG) (const_int 0))
			    (match_operand:SWI48 1 "register_operand")
			    (match_dup 0)))
   (set (match_dup 1)
	(if_then_else:SWI48 (ne (reg:CCZ FLAGS_REG) (const_int 0))
			    (match_operand:SWI48 3 "register_operand")
			    (match_dup 1)))]
  "TARGET_CMOVE"
  "operands[4] = GEN_INT (GET_MODE_BITSIZE (<MODE>mode));")

(define_expand "x86_shift<mode>_adj_2"
  [(use (match_operand:SWI48 0 "register_operand"))
   (use (match_operand:SWI48 1 "register_operand"))
   (use (match_operand:QI 2 "register_operand"))]
  ""
{
  rtx_code_label *label = gen_label_rtx ();
  rtx tmp;

  emit_insn (gen_testqi_ccz_1 (operands[2],
			       GEN_INT (GET_MODE_BITSIZE (<MODE>mode))));

  tmp = gen_rtx_REG (CCZmode, FLAGS_REG);
  tmp = gen_rtx_EQ (VOIDmode, tmp, const0_rtx);
  tmp = gen_rtx_IF_THEN_ELSE (VOIDmode, tmp,
			      gen_rtx_LABEL_REF (VOIDmode, label),
			      pc_rtx);
  tmp = emit_jump_insn (gen_rtx_SET (pc_rtx, tmp));
  JUMP_LABEL (tmp) = label;

  emit_move_insn (operands[0], operands[1]);
  ix86_expand_clear (operands[1]);

  emit_label (label);
  LABEL_NUSES (label) = 1;

  DONE;
})

;; Avoid useless masking of count operand.
(define_insn_and_split "*ashl<mode>3_mask"
  [(set (match_operand:SWI48 0 "nonimmediate_operand")
	(ashift:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand")
	  (subreg:QI
	    (and:SI
	      (match_operand:SI 2 "register_operand")
	      (match_operand:SI 3 "const_int_operand")) 0)))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (ASHIFT, <MODE>mode, operands)
   && (INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (ashift:SWI48 (match_dup 1)
			 (match_dup 2)))
      (clobber (reg:CC FLAGS_REG))])]
  "operands[2] = gen_lowpart (QImode, operands[2]);")

(define_insn_and_split "*ashl<mode>3_mask_1"
  [(set (match_operand:SWI48 0 "nonimmediate_operand")
	(ashift:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand")
	  (and:QI
	    (match_operand:QI 2 "register_operand")
	    (match_operand:QI 3 "const_int_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (ASHIFT, <MODE>mode, operands)
   && (INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (ashift:SWI48 (match_dup 1)
			 (match_dup 2)))
      (clobber (reg:CC FLAGS_REG))])])

(define_insn "*bmi2_ashl<mode>3_1"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(ashift:SWI48 (match_operand:SWI48 1 "nonimmediate_operand" "rm")
		      (match_operand:SWI48 2 "register_operand" "r")))]
  "TARGET_BMI2"
  "shlx\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "type" "ishiftx")
   (set_attr "mode" "<MODE>")])

(define_insn "*ashl<mode>3_1"
  [(set (match_operand:SWI48 0 "nonimmediate_operand" "=rm,r,r")
	(ashift:SWI48 (match_operand:SWI48 1 "nonimmediate_operand" "0,l,rm")
		      (match_operand:QI 2 "nonmemory_operand" "c<S>,M,r")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (ASHIFT, <MODE>mode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
    case TYPE_ISHIFTX:
      return "#";

    case TYPE_ALU:
      gcc_assert (operands[2] == const1_rtx);
      gcc_assert (rtx_equal_p (operands[0], operands[1]));
      return "add{<imodesuffix>}\t%0, %0";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "sal{<imodesuffix>}\t%0";
      else
	return "sal{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set_attr "isa" "*,*,bmi2")
   (set (attr "type")
     (cond [(eq_attr "alternative" "1")
	      (const_string "lea")
	    (eq_attr "alternative" "2")
	      (const_string "ishiftx")
            (and (and (match_test "TARGET_DOUBLE_WITH_ADD")
		      (match_operand 0 "register_operand"))
		 (match_operand 2 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift")
		 (and (match_operand 2 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

;; Convert shift to the shiftx pattern to avoid flags dependency.
(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(ashift:SWI48 (match_operand:SWI48 1 "nonimmediate_operand")
		      (match_operand:QI 2 "register_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(ashift:SWI48 (match_dup 1) (match_dup 2)))]
  "operands[2] = gen_lowpart (<MODE>mode, operands[2]);")

(define_insn "*bmi2_ashlsi3_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (ashift:SI (match_operand:SI 1 "nonimmediate_operand" "rm")
		     (match_operand:SI 2 "register_operand" "r"))))]
  "TARGET_64BIT && TARGET_BMI2"
  "shlx\t{%2, %1, %k0|%k0, %1, %2}"
  [(set_attr "type" "ishiftx")
   (set_attr "mode" "SI")])

(define_insn "*ashlsi3_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r,r,r")
	(zero_extend:DI
	  (ashift:SI (match_operand:SI 1 "nonimmediate_operand" "0,l,rm")
		     (match_operand:QI 2 "nonmemory_operand" "cI,M,r"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (ASHIFT, SImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
    case TYPE_ISHIFTX:
      return "#";

    case TYPE_ALU:
      gcc_assert (operands[2] == const1_rtx);
      return "add{l}\t%k0, %k0";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "sal{l}\t%k0";
      else
	return "sal{l}\t{%2, %k0|%k0, %2}";
    }
}
  [(set_attr "isa" "*,*,bmi2")
   (set (attr "type")
     (cond [(eq_attr "alternative" "1")
	      (const_string "lea")
	    (eq_attr "alternative" "2")
	      (const_string "ishiftx")
            (and (match_test "TARGET_DOUBLE_WITH_ADD")
		 (match_operand 2 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift")
		 (and (match_operand 2 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "SI")])

;; Convert shift to the shiftx pattern to avoid flags dependency.
(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (ashift:SI (match_operand:SI 1 "nonimmediate_operand")
		     (match_operand:QI 2 "register_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(zero_extend:DI (ashift:SI (match_dup 1) (match_dup 2))))]
  "operands[2] = gen_lowpart (SImode, operands[2]);")

(define_insn "*ashlhi3_1"
  [(set (match_operand:HI 0 "nonimmediate_operand" "=rm,Yp")
	(ashift:HI (match_operand:HI 1 "nonimmediate_operand" "0,l")
		   (match_operand:QI 2 "nonmemory_operand" "cI,M")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (ASHIFT, HImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
      return "#";

    case TYPE_ALU:
      gcc_assert (operands[2] == const1_rtx);
      return "add{w}\t%0, %0";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "sal{w}\t%0";
      else
	return "sal{w}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "1")
	      (const_string "lea")
            (and (and (match_test "TARGET_DOUBLE_WITH_ADD")
		      (match_operand 0 "register_operand"))
		 (match_operand 2 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift")
		 (and (match_operand 2 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "HI,SI")])

(define_insn "*ashlqi3_1"
  [(set (match_operand:QI 0 "nonimmediate_operand" "=qm,r,Yp")
	(ashift:QI (match_operand:QI 1 "nonimmediate_operand" "0,0,l")
		   (match_operand:QI 2 "nonmemory_operand" "cI,cI,M")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (ASHIFT, QImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_LEA:
      return "#";

    case TYPE_ALU:
      gcc_assert (operands[2] == const1_rtx);
      if (REG_P (operands[1]) && !ANY_QI_REGNO_P (REGNO (operands[1])))
        return "add{l}\t%k0, %k0";
      else
        return "add{b}\t%0, %0";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	{
	  if (get_attr_mode (insn) == MODE_SI)
	    return "sal{l}\t%k0";
	  else
	    return "sal{b}\t%0";
	}
      else
	{
	  if (get_attr_mode (insn) == MODE_SI)
	    return "sal{l}\t{%2, %k0|%k0, %2}";
	  else
	    return "sal{b}\t{%2, %0|%0, %2}";
	}
    }
}
  [(set (attr "type")
     (cond [(eq_attr "alternative" "2")
	      (const_string "lea")
            (and (and (match_test "TARGET_DOUBLE_WITH_ADD")
		      (match_operand 0 "register_operand"))
		 (match_operand 2 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift")
		 (and (match_operand 2 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "QI,SI,SI")
   ;; Potential partial reg stall on alternative 1.
   (set (attr "preferred_for_speed")
     (cond [(eq_attr "alternative" "1")
	      (symbol_ref "!TARGET_PARTIAL_REG_STALL")]
	   (symbol_ref "true")))])

(define_insn "*ashlqi3_1_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+qm"))
	(ashift:QI (match_dup 0)
		   (match_operand:QI 1 "nonmemory_operand" "cI")))
   (clobber (reg:CC FLAGS_REG))]
  "(optimize_function_for_size_p (cfun)
    || !TARGET_PARTIAL_FLAG_REG_STALL
    || (operands[1] == const1_rtx
	&& (TARGET_SHIFT1
	    || (TARGET_DOUBLE_WITH_ADD && REG_P (operands[0])))))"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ALU:
      gcc_assert (operands[1] == const1_rtx);
      return "add{b}\t%0, %0";

    default:
      if (operands[1] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "sal{b}\t%0";
      else
	return "sal{b}\t{%1, %0|%0, %1}";
    }
}
  [(set (attr "type")
     (cond [(and (and (match_test "TARGET_DOUBLE_WITH_ADD")
		      (match_operand 0 "register_operand"))
		 (match_operand 1 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift1")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift1")
		 (and (match_operand 1 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "QI")])

;; Convert ashift to the lea pattern to avoid flags dependency.
(define_split
  [(set (match_operand:SWI 0 "register_operand")
	(ashift:SWI (match_operand:SWI 1 "index_register_operand")
		    (match_operand 2 "const_0_to_3_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "reload_completed
   && REGNO (operands[0]) != REGNO (operands[1])"
  [(set (match_dup 0)
	(mult:<LEAMODE> (match_dup 1) (match_dup 2)))]
{
  if (<MODE>mode != <LEAMODE>mode)
    {
      operands[0] = gen_lowpart (<LEAMODE>mode, operands[0]);
      operands[1] = gen_lowpart (<LEAMODE>mode, operands[1]);
    }
  operands[2] = GEN_INT (1 << INTVAL (operands[2]));
})

;; Convert ashift to the lea pattern to avoid flags dependency.
(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (ashift:SI (match_operand:SI 1 "index_register_operand")
		     (match_operand 2 "const_0_to_3_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && reload_completed
   && REGNO (operands[0]) != REGNO (operands[1])"
  [(set (match_dup 0)
	(zero_extend:DI (mult:SI (match_dup 1) (match_dup 2))))]
{
  operands[1] = gen_lowpart (SImode, operands[1]);
  operands[2] = GEN_INT (1 << INTVAL (operands[2]));
})

;; This pattern can't accept a variable shift count, since shifts by
;; zero don't affect the flags.  We assume that shifts by constant
;; zero are optimized away.
(define_insn "*ashl<mode>3_cmp"
  [(set (reg FLAGS_REG)
	(compare
	  (ashift:SWI (match_operand:SWI 1 "nonimmediate_operand" "0")
		      (match_operand:QI 2 "<shift_immediate_operand>" "<S>"))
	  (const_int 0)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(ashift:SWI (match_dup 1) (match_dup 2)))]
  "(optimize_function_for_size_p (cfun)
    || !TARGET_PARTIAL_FLAG_REG_STALL
    || (operands[2] == const1_rtx
	&& (TARGET_SHIFT1
	    || (TARGET_DOUBLE_WITH_ADD && REG_P (operands[0])))))
   && ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (ASHIFT, <MODE>mode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ALU:
      gcc_assert (operands[2] == const1_rtx);
      return "add{<imodesuffix>}\t%0, %0";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "sal{<imodesuffix>}\t%0";
      else
	return "sal{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (cond [(and (and (match_test "TARGET_DOUBLE_WITH_ADD")
		      (match_operand 0 "register_operand"))
		 (match_operand 2 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift")
		 (and (match_operand 2 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

(define_insn "*ashlsi3_cmp_zext"
  [(set (reg FLAGS_REG)
	(compare
	  (ashift:SI (match_operand:SI 1 "register_operand" "0")
		     (match_operand:QI 2 "const_1_to_31_operand" "I"))
	  (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (ashift:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT
   && (optimize_function_for_size_p (cfun)
       || !TARGET_PARTIAL_FLAG_REG_STALL
       || (operands[2] == const1_rtx
	   && (TARGET_SHIFT1
	       || TARGET_DOUBLE_WITH_ADD)))
   && ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (ASHIFT, SImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ALU:
      gcc_assert (operands[2] == const1_rtx);
      return "add{l}\t%k0, %k0";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "sal{l}\t%k0";
      else
	return "sal{l}\t{%2, %k0|%k0, %2}";
    }
}
  [(set (attr "type")
     (cond [(and (match_test "TARGET_DOUBLE_WITH_ADD")
		 (match_operand 2 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift")
		 (and (match_operand 2 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "SI")])

(define_insn "*ashl<mode>3_cconly"
  [(set (reg FLAGS_REG)
	(compare
	  (ashift:SWI (match_operand:SWI 1 "register_operand" "0")
		      (match_operand:QI 2 "<shift_immediate_operand>" "<S>"))
	  (const_int 0)))
   (clobber (match_scratch:SWI 0 "=<r>"))]
  "(optimize_function_for_size_p (cfun)
    || !TARGET_PARTIAL_FLAG_REG_STALL
    || (operands[2] == const1_rtx
	&& (TARGET_SHIFT1
	    || TARGET_DOUBLE_WITH_ADD)))
   && ix86_match_ccmode (insn, CCGOCmode)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ALU:
      gcc_assert (operands[2] == const1_rtx);
      return "add{<imodesuffix>}\t%0, %0";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "sal{<imodesuffix>}\t%0";
      else
	return "sal{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set (attr "type")
     (cond [(and (and (match_test "TARGET_DOUBLE_WITH_ADD")
		      (match_operand 0 "register_operand"))
		 (match_operand 2 "const1_operand"))
	      (const_string "alu")
	   ]
	   (const_string "ishift")))
   (set (attr "length_immediate")
     (if_then_else
       (ior (eq_attr "type" "alu")
	    (and (eq_attr "type" "ishift")
		 (and (match_operand 2 "const1_operand")
		      (ior (match_test "TARGET_SHIFT1")
			   (match_test "optimize_function_for_size_p (cfun)")))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

;; See comment above `ashl<mode>3' about how this works.

(define_expand "<shift_insn><mode>3"
  [(set (match_operand:SDWIM 0 "<shift_operand>")
	(any_shiftrt:SDWIM (match_operand:SDWIM 1 "<shift_operand>")
			   (match_operand:QI 2 "nonmemory_operand")))]
  ""
  "ix86_expand_binary_operator (<CODE>, <MODE>mode, operands); DONE;")

;; Avoid useless masking of count operand.
(define_insn_and_split "*<shift_insn><mode>3_mask"
  [(set (match_operand:SWI48 0 "nonimmediate_operand")
	(any_shiftrt:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand")
	  (subreg:QI
	    (and:SI
	      (match_operand:SI 2 "register_operand")
	      (match_operand:SI 3 "const_int_operand")) 0)))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)
   && (INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (any_shiftrt:SWI48 (match_dup 1)
			      (match_dup 2)))
      (clobber (reg:CC FLAGS_REG))])]
  "operands[2] = gen_lowpart (QImode, operands[2]);")

(define_insn_and_split "*<shift_insn><mode>3_mask_1"
  [(set (match_operand:SWI48 0 "nonimmediate_operand")
	(any_shiftrt:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand")
	  (and:QI
	    (match_operand:QI 2 "register_operand")
	    (match_operand:QI 3 "const_int_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)
   && (INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (any_shiftrt:SWI48 (match_dup 1)
			      (match_dup 2)))
      (clobber (reg:CC FLAGS_REG))])])

(define_insn_and_split "*<shift_insn><mode>3_doubleword"
  [(set (match_operand:DWI 0 "register_operand" "=&r")
	(any_shiftrt:DWI (match_operand:DWI 1 "register_operand" "0")
			 (match_operand:QI 2 "nonmemory_operand" "<S>c")))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "#"
  "(optimize && flag_peephole2) ? epilogue_completed : reload_completed"
  [(const_int 0)]
  "ix86_split_<shift_insn> (operands, NULL_RTX, <MODE>mode); DONE;"
  [(set_attr "type" "multi")])

;; By default we don't ask for a scratch register, because when DWImode
;; values are manipulated, registers are already at a premium.  But if
;; we have one handy, we won't turn it away.

(define_peephole2
  [(match_scratch:DWIH 3 "r")
   (parallel [(set (match_operand:<DWI> 0 "register_operand")
		   (any_shiftrt:<DWI>
		     (match_operand:<DWI> 1 "register_operand")
		     (match_operand:QI 2 "nonmemory_operand")))
	      (clobber (reg:CC FLAGS_REG))])
   (match_dup 3)]
  "TARGET_CMOVE"
  [(const_int 0)]
  "ix86_split_<shift_insn> (operands, operands[3], <DWI>mode); DONE;")

(define_insn "x86_64_shrd"
  [(set (match_operand:DI 0 "nonimmediate_operand" "+r*m")
        (ior:DI (lshiftrt:DI (match_dup 0)
		  (match_operand:QI 2 "nonmemory_operand" "Jc"))
		(ashift:DI (match_operand:DI 1 "register_operand" "r")
		  (minus:QI (const_int 64) (match_dup 2)))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT"
  "shrd{q}\t{%s2%1, %0|%0, %1, %2}"
  [(set_attr "type" "ishift")
   (set_attr "prefix_0f" "1")
   (set_attr "mode" "DI")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "vector")
   (set_attr "bdver1_decode" "vector")])

(define_insn "x86_shrd"
  [(set (match_operand:SI 0 "nonimmediate_operand" "+r*m")
        (ior:SI (lshiftrt:SI (match_dup 0)
		  (match_operand:QI 2 "nonmemory_operand" "Ic"))
		(ashift:SI (match_operand:SI 1 "register_operand" "r")
		  (minus:QI (const_int 32) (match_dup 2)))))
   (clobber (reg:CC FLAGS_REG))]
  ""
  "shrd{l}\t{%s2%1, %0|%0, %1, %2}"
  [(set_attr "type" "ishift")
   (set_attr "prefix_0f" "1")
   (set_attr "mode" "SI")
   (set_attr "pent_pair" "np")
   (set_attr "athlon_decode" "vector")
   (set_attr "amdfam10_decode" "vector")
   (set_attr "bdver1_decode" "vector")])

(define_insn "ashrdi3_cvt"
  [(set (match_operand:DI 0 "nonimmediate_operand" "=*d,rm")
	(ashiftrt:DI (match_operand:DI 1 "nonimmediate_operand" "*a,0")
		     (match_operand:QI 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && INTVAL (operands[2]) == 63
   && (TARGET_USE_CLTD || optimize_function_for_size_p (cfun))
   && ix86_binary_operator_ok (ASHIFTRT, DImode, operands)"
  "@
   {cqto|cqo}
   sar{q}\t{%2, %0|%0, %2}"
  [(set_attr "type" "imovx,ishift")
   (set_attr "prefix_0f" "0,*")
   (set_attr "length_immediate" "0,*")
   (set_attr "modrm" "0,1")
   (set_attr "mode" "DI")])

(define_insn "*ashrsi3_cvt_zext"
  [(set (match_operand:DI 0 "register_operand" "=*d,r")
	(zero_extend:DI
	  (ashiftrt:SI (match_operand:SI 1 "register_operand" "*a,0")
		       (match_operand:QI 2 "const_int_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && INTVAL (operands[2]) == 31
   && (TARGET_USE_CLTD || optimize_function_for_size_p (cfun))
   && ix86_binary_operator_ok (ASHIFTRT, SImode, operands)"
  "@
   {cltd|cdq}
   sar{l}\t{%2, %k0|%k0, %2}"
  [(set_attr "type" "imovx,ishift")
   (set_attr "prefix_0f" "0,*")
   (set_attr "length_immediate" "0,*")
   (set_attr "modrm" "0,1")
   (set_attr "mode" "SI")])

(define_insn "ashrsi3_cvt"
  [(set (match_operand:SI 0 "nonimmediate_operand" "=*d,rm")
	(ashiftrt:SI (match_operand:SI 1 "nonimmediate_operand" "*a,0")
		     (match_operand:QI 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "INTVAL (operands[2]) == 31
   && (TARGET_USE_CLTD || optimize_function_for_size_p (cfun))
   && ix86_binary_operator_ok (ASHIFTRT, SImode, operands)"
  "@
   {cltd|cdq}
   sar{l}\t{%2, %0|%0, %2}"
  [(set_attr "type" "imovx,ishift")
   (set_attr "prefix_0f" "0,*")
   (set_attr "length_immediate" "0,*")
   (set_attr "modrm" "0,1")
   (set_attr "mode" "SI")])

(define_expand "x86_shift<mode>_adj_3"
  [(use (match_operand:SWI48 0 "register_operand"))
   (use (match_operand:SWI48 1 "register_operand"))
   (use (match_operand:QI 2 "register_operand"))]
  ""
{
  rtx_code_label *label = gen_label_rtx ();
  rtx tmp;

  emit_insn (gen_testqi_ccz_1 (operands[2],
			       GEN_INT (GET_MODE_BITSIZE (<MODE>mode))));

  tmp = gen_rtx_REG (CCZmode, FLAGS_REG);
  tmp = gen_rtx_EQ (VOIDmode, tmp, const0_rtx);
  tmp = gen_rtx_IF_THEN_ELSE (VOIDmode, tmp,
			      gen_rtx_LABEL_REF (VOIDmode, label),
			      pc_rtx);
  tmp = emit_jump_insn (gen_rtx_SET (pc_rtx, tmp));
  JUMP_LABEL (tmp) = label;

  emit_move_insn (operands[0], operands[1]);
  emit_insn (gen_ashr<mode>3_cvt (operands[1], operands[1],
				  GEN_INT (GET_MODE_BITSIZE (<MODE>mode)-1)));
  emit_label (label);
  LABEL_NUSES (label) = 1;

  DONE;
})

(define_insn "*bmi2_<shift_insn><mode>3_1"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(any_shiftrt:SWI48 (match_operand:SWI48 1 "nonimmediate_operand" "rm")
			   (match_operand:SWI48 2 "register_operand" "r")))]
  "TARGET_BMI2"
  "<shift>x\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "type" "ishiftx")
   (set_attr "mode" "<MODE>")])

(define_insn "*<shift_insn><mode>3_1"
  [(set (match_operand:SWI48 0 "nonimmediate_operand" "=rm,r")
	(any_shiftrt:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand" "0,rm")
	  (match_operand:QI 2 "nonmemory_operand" "c<S>,r")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ISHIFTX:
      return "#";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "<shift>{<imodesuffix>}\t%0";
      else
	return "<shift>{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set_attr "isa" "*,bmi2")
   (set_attr "type" "ishift,ishiftx")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 2 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

;; Convert shift to the shiftx pattern to avoid flags dependency.
(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(any_shiftrt:SWI48 (match_operand:SWI48 1 "nonimmediate_operand")
			   (match_operand:QI 2 "register_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(any_shiftrt:SWI48 (match_dup 1) (match_dup 2)))]
  "operands[2] = gen_lowpart (<MODE>mode, operands[2]);")

(define_insn "*bmi2_<shift_insn>si3_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (any_shiftrt:SI (match_operand:SI 1 "nonimmediate_operand" "rm")
			  (match_operand:SI 2 "register_operand" "r"))))]
  "TARGET_64BIT && TARGET_BMI2"
  "<shift>x\t{%2, %1, %k0|%k0, %1, %2}"
  [(set_attr "type" "ishiftx")
   (set_attr "mode" "SI")])

(define_insn "*<shift_insn>si3_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r,r")
	(zero_extend:DI
	  (any_shiftrt:SI (match_operand:SI 1 "nonimmediate_operand" "0,rm")
			  (match_operand:QI 2 "nonmemory_operand" "cI,r"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (<CODE>, SImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ISHIFTX:
      return "#";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "<shift>{l}\t%k0";
      else
	return "<shift>{l}\t{%2, %k0|%k0, %2}";
    }
}
  [(set_attr "isa" "*,bmi2")
   (set_attr "type" "ishift,ishiftx")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 2 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "SI")])

;; Convert shift to the shiftx pattern to avoid flags dependency.
(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (any_shiftrt:SI (match_operand:SI 1 "nonimmediate_operand")
			  (match_operand:QI 2 "register_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(zero_extend:DI (any_shiftrt:SI (match_dup 1) (match_dup 2))))]
  "operands[2] = gen_lowpart (SImode, operands[2]);")

(define_insn "*<shift_insn><mode>3_1"
  [(set (match_operand:SWI12 0 "nonimmediate_operand" "=<r>m")
	(any_shiftrt:SWI12
	  (match_operand:SWI12 1 "nonimmediate_operand" "0")
	  (match_operand:QI 2 "nonmemory_operand" "c<S>")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
  if (operands[2] == const1_rtx
      && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
    return "<shift>{<imodesuffix>}\t%0";
  else
    return "<shift>{<imodesuffix>}\t{%2, %0|%0, %2}";
}
  [(set_attr "type" "ishift")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 2 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

(define_insn "*<shift_insn>qi3_1_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+qm"))
	(any_shiftrt:QI (match_dup 0)
			(match_operand:QI 1 "nonmemory_operand" "cI")))
   (clobber (reg:CC FLAGS_REG))]
  "(optimize_function_for_size_p (cfun)
    || !TARGET_PARTIAL_REG_STALL
    || (operands[1] == const1_rtx
	&& TARGET_SHIFT1))"
{
  if (operands[1] == const1_rtx
      && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
    return "<shift>{b}\t%0";
  else
    return "<shift>{b}\t{%1, %0|%0, %1}";
}
  [(set_attr "type" "ishift1")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 1 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "QI")])

;; This pattern can't accept a variable shift count, since shifts by
;; zero don't affect the flags.  We assume that shifts by constant
;; zero are optimized away.
(define_insn "*<shift_insn><mode>3_cmp"
  [(set (reg FLAGS_REG)
	(compare
	  (any_shiftrt:SWI
	    (match_operand:SWI 1 "nonimmediate_operand" "0")
	    (match_operand:QI 2 "<shift_immediate_operand>" "<S>"))
	  (const_int 0)))
   (set (match_operand:SWI 0 "nonimmediate_operand" "=<r>m")
	(any_shiftrt:SWI (match_dup 1) (match_dup 2)))]
  "(optimize_function_for_size_p (cfun)
    || !TARGET_PARTIAL_FLAG_REG_STALL
    || (operands[2] == const1_rtx
	&& TARGET_SHIFT1))
   && ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
  if (operands[2] == const1_rtx
      && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
    return "<shift>{<imodesuffix>}\t%0";
  else
    return "<shift>{<imodesuffix>}\t{%2, %0|%0, %2}";
}
  [(set_attr "type" "ishift")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 2 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

(define_insn "*<shift_insn>si3_cmp_zext"
  [(set (reg FLAGS_REG)
	(compare
	  (any_shiftrt:SI (match_operand:SI 1 "register_operand" "0")
			  (match_operand:QI 2 "const_1_to_31_operand" "I"))
	  (const_int 0)))
   (set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI (any_shiftrt:SI (match_dup 1) (match_dup 2))))]
  "TARGET_64BIT
   && (optimize_function_for_size_p (cfun)
       || !TARGET_PARTIAL_FLAG_REG_STALL
       || (operands[2] == const1_rtx
	   && TARGET_SHIFT1))
   && ix86_match_ccmode (insn, CCGOCmode)
   && ix86_binary_operator_ok (<CODE>, SImode, operands)"
{
  if (operands[2] == const1_rtx
      && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
    return "<shift>{l}\t%k0";
  else
    return "<shift>{l}\t{%2, %k0|%k0, %2}";
}
  [(set_attr "type" "ishift")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 2 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "SI")])

(define_insn "*<shift_insn><mode>3_cconly"
  [(set (reg FLAGS_REG)
	(compare
	  (any_shiftrt:SWI
	    (match_operand:SWI 1 "register_operand" "0")
	    (match_operand:QI 2 "<shift_immediate_operand>" "<S>"))
	  (const_int 0)))
   (clobber (match_scratch:SWI 0 "=<r>"))]
  "(optimize_function_for_size_p (cfun)
    || !TARGET_PARTIAL_FLAG_REG_STALL
    || (operands[2] == const1_rtx
	&& TARGET_SHIFT1))
   && ix86_match_ccmode (insn, CCGOCmode)"
{
  if (operands[2] == const1_rtx
      && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
    return "<shift>{<imodesuffix>}\t%0";
  else
    return "<shift>{<imodesuffix>}\t{%2, %0|%0, %2}";
}
  [(set_attr "type" "ishift")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 2 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

;; Rotate instructions

(define_expand "<rotate_insn>ti3"
  [(set (match_operand:TI 0 "register_operand")
	(any_rotate:TI (match_operand:TI 1 "register_operand")
		       (match_operand:QI 2 "nonmemory_operand")))]
  "TARGET_64BIT"
{
  if (const_1_to_63_operand (operands[2], VOIDmode))
    emit_insn (gen_ix86_<rotate_insn>ti3_doubleword
		(operands[0], operands[1], operands[2]));
  else
    FAIL;

  DONE;
})

(define_expand "<rotate_insn>di3"
  [(set (match_operand:DI 0 "shiftdi_operand")
	(any_rotate:DI (match_operand:DI 1 "shiftdi_operand")
		       (match_operand:QI 2 "nonmemory_operand")))]
 ""
{
  if (TARGET_64BIT)
    ix86_expand_binary_operator (<CODE>, DImode, operands);
  else if (const_1_to_31_operand (operands[2], VOIDmode))
    emit_insn (gen_ix86_<rotate_insn>di3_doubleword
		(operands[0], operands[1], operands[2]));
  else
    FAIL;

  DONE;
})

(define_expand "<rotate_insn><mode>3"
  [(set (match_operand:SWIM124 0 "nonimmediate_operand")
	(any_rotate:SWIM124 (match_operand:SWIM124 1 "nonimmediate_operand")
			    (match_operand:QI 2 "nonmemory_operand")))]
  ""
  "ix86_expand_binary_operator (<CODE>, <MODE>mode, operands); DONE;")

;; Avoid useless masking of count operand.
(define_insn_and_split "*<rotate_insn><mode>3_mask"
  [(set (match_operand:SWI48 0 "nonimmediate_operand")
	(any_rotate:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand")
	  (subreg:QI
	    (and:SI
	      (match_operand:SI 2 "register_operand")
	      (match_operand:SI 3 "const_int_operand")) 0)))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)
   && (INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (any_rotate:SWI48 (match_dup 1)
			     (match_dup 2)))
      (clobber (reg:CC FLAGS_REG))])]
  "operands[2] = gen_lowpart (QImode, operands[2]);")

(define_insn_and_split "*<rotate_insn><mode>3_mask_1"
  [(set (match_operand:SWI48 0 "nonimmediate_operand")
	(any_rotate:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand")
	  (and:QI
	    (match_operand:QI 2 "register_operand")
	    (match_operand:QI 3 "const_int_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)
   && (INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (any_rotate:SWI48 (match_dup 1)
			     (match_dup 2)))
      (clobber (reg:CC FLAGS_REG))])])

;; Implement rotation using two double-precision
;; shift instructions and a scratch register.

(define_insn_and_split "ix86_rotl<dwi>3_doubleword"
 [(set (match_operand:<DWI> 0 "register_operand" "=r")
       (rotate:<DWI> (match_operand:<DWI> 1 "register_operand" "0")
		     (match_operand:QI 2 "<shift_immediate_operand>" "<S>")))
  (clobber (reg:CC FLAGS_REG))
  (clobber (match_scratch:DWIH 3 "=&r"))]
 ""
 "#"
 "reload_completed"
 [(set (match_dup 3) (match_dup 4))
  (parallel
   [(set (match_dup 4)
	 (ior:DWIH (ashift:DWIH (match_dup 4) (match_dup 2))
		   (lshiftrt:DWIH (match_dup 5)
				  (minus:QI (match_dup 6) (match_dup 2)))))
    (clobber (reg:CC FLAGS_REG))])
  (parallel
   [(set (match_dup 5)
	 (ior:DWIH (ashift:DWIH (match_dup 5) (match_dup 2))
		   (lshiftrt:DWIH (match_dup 3)
				  (minus:QI (match_dup 6) (match_dup 2)))))
    (clobber (reg:CC FLAGS_REG))])]
{
  operands[6] = GEN_INT (GET_MODE_BITSIZE (<MODE>mode));

  split_double_mode (<DWI>mode, &operands[0], 1, &operands[4], &operands[5]);
})

(define_insn_and_split "ix86_rotr<dwi>3_doubleword"
 [(set (match_operand:<DWI> 0 "register_operand" "=r")
       (rotatert:<DWI> (match_operand:<DWI> 1 "register_operand" "0")
		       (match_operand:QI 2 "<shift_immediate_operand>" "<S>")))
  (clobber (reg:CC FLAGS_REG))
  (clobber (match_scratch:DWIH 3 "=&r"))]
 ""
 "#"
 "reload_completed"
 [(set (match_dup 3) (match_dup 4))
  (parallel
   [(set (match_dup 4)
	 (ior:DWIH (lshiftrt:DWIH (match_dup 4) (match_dup 2))
		   (ashift:DWIH (match_dup 5)
				(minus:QI (match_dup 6) (match_dup 2)))))
    (clobber (reg:CC FLAGS_REG))])
  (parallel
   [(set (match_dup 5)
	 (ior:DWIH (lshiftrt:DWIH (match_dup 5) (match_dup 2))
		   (ashift:DWIH (match_dup 3)
				(minus:QI (match_dup 6) (match_dup 2)))))
    (clobber (reg:CC FLAGS_REG))])]
{
  operands[6] = GEN_INT (GET_MODE_BITSIZE (<MODE>mode));

  split_double_mode (<DWI>mode, &operands[0], 1, &operands[4], &operands[5]);
})

(define_mode_attr rorx_immediate_operand
	[(SI "const_0_to_31_operand")
	 (DI "const_0_to_63_operand")])

(define_insn "*bmi2_rorx<mode>3_1"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(rotatert:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand" "rm")
	  (match_operand:QI 2 "<rorx_immediate_operand>" "<S>")))]
  "TARGET_BMI2"
  "rorx\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "type" "rotatex")
   (set_attr "mode" "<MODE>")])

(define_insn "*<rotate_insn><mode>3_1"
  [(set (match_operand:SWI48 0 "nonimmediate_operand" "=rm,r")
	(any_rotate:SWI48
	  (match_operand:SWI48 1 "nonimmediate_operand" "0,rm")
	  (match_operand:QI 2 "nonmemory_operand" "c<S>,<S>")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ROTATEX:
      return "#";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "<rotate>{<imodesuffix>}\t%0";
      else
	return "<rotate>{<imodesuffix>}\t{%2, %0|%0, %2}";
    }
}
  [(set_attr "isa" "*,bmi2")
   (set_attr "type" "rotate,rotatex")
   (set (attr "length_immediate")
     (if_then_else
       (and (eq_attr "type" "rotate")
	    (and (match_operand 2 "const1_operand")
		 (ior (match_test "TARGET_SHIFT1")
		      (match_test "optimize_function_for_size_p (cfun)"))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

;; Convert rotate to the rotatex pattern to avoid flags dependency.
(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(rotate:SWI48 (match_operand:SWI48 1 "nonimmediate_operand")
		      (match_operand:QI 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(rotatert:SWI48 (match_dup 1) (match_dup 2)))]
{
  int bitsize = GET_MODE_BITSIZE (<MODE>mode);

  operands[2] = GEN_INT ((bitsize - INTVAL (operands[2])) % bitsize);
})

(define_split
  [(set (match_operand:SWI48 0 "register_operand")
	(rotatert:SWI48 (match_operand:SWI48 1 "nonimmediate_operand")
			(match_operand:QI 2 "const_int_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(rotatert:SWI48 (match_dup 1) (match_dup 2)))])

(define_insn "*bmi2_rorxsi3_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r")
	(zero_extend:DI
	  (rotatert:SI (match_operand:SI 1 "nonimmediate_operand" "rm")
		       (match_operand:QI 2 "const_0_to_31_operand" "I"))))]
  "TARGET_64BIT && TARGET_BMI2"
  "rorx\t{%2, %1, %k0|%k0, %1, %2}"
  [(set_attr "type" "rotatex")
   (set_attr "mode" "SI")])

(define_insn "*<rotate_insn>si3_1_zext"
  [(set (match_operand:DI 0 "register_operand" "=r,r")
	(zero_extend:DI
	  (any_rotate:SI (match_operand:SI 1 "nonimmediate_operand" "0,rm")
			 (match_operand:QI 2 "nonmemory_operand" "cI,I"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && ix86_binary_operator_ok (<CODE>, SImode, operands)"
{
  switch (get_attr_type (insn))
    {
    case TYPE_ROTATEX:
      return "#";

    default:
      if (operands[2] == const1_rtx
	  && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
	return "<rotate>{l}\t%k0";
      else
	return "<rotate>{l}\t{%2, %k0|%k0, %2}";
    }
}
  [(set_attr "isa" "*,bmi2")
   (set_attr "type" "rotate,rotatex")
   (set (attr "length_immediate")
     (if_then_else
       (and (eq_attr "type" "rotate")
	    (and (match_operand 2 "const1_operand")
		 (ior (match_test "TARGET_SHIFT1")
		      (match_test "optimize_function_for_size_p (cfun)"))))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "SI")])

;; Convert rotate to the rotatex pattern to avoid flags dependency.
(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (rotate:SI (match_operand:SI 1 "nonimmediate_operand")
		     (match_operand:QI 2 "const_int_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(zero_extend:DI (rotatert:SI (match_dup 1) (match_dup 2))))]
{
  int bitsize = GET_MODE_BITSIZE (SImode);

  operands[2] = GEN_INT ((bitsize - INTVAL (operands[2])) % bitsize);
})

(define_split
  [(set (match_operand:DI 0 "register_operand")
	(zero_extend:DI
	  (rotatert:SI (match_operand:SI 1 "nonimmediate_operand")
		       (match_operand:QI 2 "const_int_operand"))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && TARGET_BMI2 && reload_completed"
  [(set (match_dup 0)
	(zero_extend:DI (rotatert:SI (match_dup 1) (match_dup 2))))])

(define_insn "*<rotate_insn><mode>3_1"
  [(set (match_operand:SWI12 0 "nonimmediate_operand" "=<r>m")
	(any_rotate:SWI12 (match_operand:SWI12 1 "nonimmediate_operand" "0")
			  (match_operand:QI 2 "nonmemory_operand" "c<S>")))
   (clobber (reg:CC FLAGS_REG))]
  "ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
{
  if (operands[2] == const1_rtx
      && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
    return "<rotate>{<imodesuffix>}\t%0";
  else
    return "<rotate>{<imodesuffix>}\t{%2, %0|%0, %2}";
}
  [(set_attr "type" "rotate")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 2 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "<MODE>")])

(define_insn "*<rotate_insn>qi3_1_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+qm"))
	(any_rotate:QI (match_dup 0)
		       (match_operand:QI 1 "nonmemory_operand" "cI")))
   (clobber (reg:CC FLAGS_REG))]
  "(optimize_function_for_size_p (cfun)
    || !TARGET_PARTIAL_REG_STALL
    || (operands[1] == const1_rtx
	&& TARGET_SHIFT1))"
{
  if (operands[1] == const1_rtx
      && (TARGET_SHIFT1 || optimize_function_for_size_p (cfun)))
    return "<rotate>{b}\t%0";
  else
    return "<rotate>{b}\t{%1, %0|%0, %1}";
}
  [(set_attr "type" "rotate1")
   (set (attr "length_immediate")
     (if_then_else
       (and (match_operand 1 "const1_operand")
	    (ior (match_test "TARGET_SHIFT1")
		 (match_test "optimize_function_for_size_p (cfun)")))
       (const_string "0")
       (const_string "*")))
   (set_attr "mode" "QI")])

(define_split
 [(set (match_operand:HI 0 "register_operand")
       (any_rotate:HI (match_dup 0) (const_int 8)))
  (clobber (reg:CC FLAGS_REG))]
 "reload_completed
  && (TARGET_USE_XCHGB || optimize_function_for_size_p (cfun))"
 [(parallel [(set (strict_low_part (match_dup 0))
		  (bswap:HI (match_dup 0)))
	     (clobber (reg:CC FLAGS_REG))])])

;; Bit set / bit test instructions

;; %%% bts, btr, btc

;; These instructions are *slow* when applied to memory.

(define_code_attr btsc [(ior "bts") (xor "btc")])

(define_insn "*<btsc><mode>"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(any_or:SWI48
	  (ashift:SWI48 (const_int 1)
			(match_operand:QI 2 "register_operand" "r"))
	  (match_operand:SWI48 1 "register_operand" "0")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_BT"
  "<btsc>{<imodesuffix>}\t{%<k>2, %0|%0, %<k>2}"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "<MODE>")])

;; Avoid useless masking of count operand.
(define_insn_and_split "*<btsc><mode>_mask"
  [(set (match_operand:SWI48 0 "register_operand")
	(any_or:SWI48
	  (ashift:SWI48
	    (const_int 1)
	    (subreg:QI
	      (and:SI
		(match_operand:SI 1 "register_operand")
		(match_operand:SI 2 "const_int_operand")) 0))
	  (match_operand:SWI48 3 "register_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_BT
   && (INTVAL (operands[2]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (any_or:SWI48
	     (ashift:SWI48 (const_int 1)
			   (match_dup 1))
	     (match_dup 3)))
      (clobber (reg:CC FLAGS_REG))])]
  "operands[1] = gen_lowpart (QImode, operands[1]);")

(define_insn_and_split "*<btsc><mode>_mask_1"
  [(set (match_operand:SWI48 0 "register_operand")
	(any_or:SWI48
	  (ashift:SWI48
	    (const_int 1)
	    (and:QI
	      (match_operand:QI 1 "register_operand")
	      (match_operand:QI 2 "const_int_operand")))
	  (match_operand:SWI48 3 "register_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_BT
   && (INTVAL (operands[2]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (any_or:SWI48
	     (ashift:SWI48 (const_int 1)
			   (match_dup 1))
	     (match_dup 3)))
      (clobber (reg:CC FLAGS_REG))])])

(define_insn "*btr<mode>"
  [(set (match_operand:SWI48 0 "register_operand" "=r")
	(and:SWI48
	  (rotate:SWI48 (const_int -2)
			(match_operand:QI 2 "register_operand" "r"))
	(match_operand:SWI48 1 "register_operand" "0")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_BT"
  "btr{<imodesuffix>}\t{%<k>2, %0|%0, %<k>2}"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "<MODE>")])

;; Avoid useless masking of count operand.
(define_insn_and_split "*btr<mode>_mask"
  [(set (match_operand:SWI48 0 "register_operand")
	(and:SWI48
	  (rotate:SWI48
	    (const_int -2)
	    (subreg:QI
	      (and:SI
		(match_operand:SI 1 "register_operand")
		(match_operand:SI 2 "const_int_operand")) 0))
	  (match_operand:SWI48 3 "register_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_BT
   && (INTVAL (operands[2]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (and:SWI48
	     (rotate:SWI48 (const_int -2)
			   (match_dup 1))
	     (match_dup 3)))
      (clobber (reg:CC FLAGS_REG))])]
  "operands[1] = gen_lowpart (QImode, operands[1]);")

(define_insn_and_split "*btr<mode>_mask_1"
  [(set (match_operand:SWI48 0 "register_operand")
	(and:SWI48
	  (rotate:SWI48
	    (const_int -2)
	    (and:QI
	      (match_operand:QI 1 "register_operand")
	      (match_operand:QI 2 "const_int_operand")))
	  (match_operand:SWI48 3 "register_operand")))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_USE_BT
   && (INTVAL (operands[2]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(parallel
     [(set (match_dup 0)
	   (and:SWI48
	     (rotate:SWI48 (const_int -2)
			   (match_dup 1))
	     (match_dup 3)))
      (clobber (reg:CC FLAGS_REG))])])

;; These instructions are never faster than the corresponding
;; and/ior/xor operations when using immediate operand, so with
;; 32-bit there's no point.  But in 64-bit, we can't hold the
;; relevant immediates within the instruction itself, so operating
;; on bits in the high 32-bits of a register becomes easier.
;;
;; These are slow on Nocona, but fast on Athlon64.  We do require the use
;; of btrq and btcq for corner cases of post-reload expansion of absdf and
;; negdf respectively, so they can never be disabled entirely.

(define_insn "*btsq_imm"
  [(set (zero_extract:DI (match_operand:DI 0 "nonimmediate_operand" "+rm")
			 (const_int 1)
			 (match_operand 1 "const_0_to_63_operand" "J"))
	(const_int 1))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && (TARGET_USE_BT || reload_completed)"
  "bts{q}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "DI")])

(define_insn "*btrq_imm"
  [(set (zero_extract:DI (match_operand:DI 0 "nonimmediate_operand" "+rm")
			 (const_int 1)
			 (match_operand 1 "const_0_to_63_operand" "J"))
	(const_int 0))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && (TARGET_USE_BT || reload_completed)"
  "btr{q}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "DI")])

(define_insn "*btcq_imm"
  [(set (zero_extract:DI (match_operand:DI 0 "nonimmediate_operand" "+rm")
			 (const_int 1)
			 (match_operand 1 "const_0_to_63_operand" "J"))
	(not:DI (zero_extract:DI (match_dup 0) (const_int 1) (match_dup 1))))
   (clobber (reg:CC FLAGS_REG))]
  "TARGET_64BIT && (TARGET_USE_BT || reload_completed)"
  "btc{q}\t{%1, %0|%0, %1}"
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set_attr "znver1_decode" "double")
   (set_attr "mode" "DI")])

;; Allow Nocona to avoid these instructions if a register is available.

(define_peephole2
  [(match_scratch:DI 2 "r")
   (parallel [(set (zero_extract:DI
		     (match_operand:DI 0 "nonimmediate_operand")
		     (const_int 1)
		     (match_operand 1 "const_0_to_63_operand"))
		   (const_int 1))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_64BIT && !TARGET_USE_BT"
  [(parallel [(set (match_dup 0)
		   (ior:DI (match_dup 0) (match_dup 3)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  int i = INTVAL (operands[1]);

  operands[3] = gen_int_mode (HOST_WIDE_INT_1U << i, DImode);

  if (!x86_64_immediate_operand (operands[3], DImode))
    {
      emit_move_insn (operands[2], operands[3]);
      operands[3] = operands[2];
    }
})

(define_peephole2
  [(match_scratch:DI 2 "r")
   (parallel [(set (zero_extract:DI
		     (match_operand:DI 0 "nonimmediate_operand")
		     (const_int 1)
		     (match_operand 1 "const_0_to_63_operand"))
		   (const_int 0))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_64BIT && !TARGET_USE_BT"
  [(parallel [(set (match_dup 0)
		   (and:DI (match_dup 0) (match_dup 3)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  int i = INTVAL (operands[1]);

  operands[3] = gen_int_mode (~(HOST_WIDE_INT_1U << i), DImode);
 
  if (!x86_64_immediate_operand (operands[3], DImode))
    {
      emit_move_insn (operands[2], operands[3]);
      operands[3] = operands[2];
    }
})

(define_peephole2
  [(match_scratch:DI 2 "r")
   (parallel [(set (zero_extract:DI
		     (match_operand:DI 0 "nonimmediate_operand")
		     (const_int 1)
		     (match_operand 1 "const_0_to_63_operand"))
	      (not:DI (zero_extract:DI
			(match_dup 0) (const_int 1) (match_dup 1))))
	      (clobber (reg:CC FLAGS_REG))])]
  "TARGET_64BIT && !TARGET_USE_BT"
  [(parallel [(set (match_dup 0)
		   (xor:DI (match_dup 0) (match_dup 3)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  int i = INTVAL (operands[1]);

  operands[3] = gen_int_mode (HOST_WIDE_INT_1U << i, DImode);

  if (!x86_64_immediate_operand (operands[3], DImode))
    {
      emit_move_insn (operands[2], operands[3]);
      operands[3] = operands[2];
    }
})

;; %%% bt

(define_insn "*bt<mode>"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extract:SWI48
	    (match_operand:SWI48 0 "nonimmediate_operand" "r,m")
	    (const_int 1)
	    (match_operand:SI 1 "nonmemory_operand" "r<S>,<S>"))
	  (const_int 0)))]
  ""
{
  switch (get_attr_mode (insn))
    {
    case MODE_SI:
      return "bt{l}\t{%1, %k0|%k0, %1}";

    case MODE_DI:
      return "bt{q}\t{%q1, %0|%0, %q1}";

    default:
      gcc_unreachable ();
    }
}
  [(set_attr "type" "alu1")
   (set_attr "prefix_0f" "1")
   (set (attr "mode")
	(if_then_else
	  (and (match_test "CONST_INT_P (operands[1])")
	       (match_test "INTVAL (operands[1]) < 32"))
	  (const_string "SI")
	  (const_string "<MODE>")))])

(define_insn_and_split "*jcc_bt<mode>"
  [(set (pc)
  	(if_then_else (match_operator 0 "bt_comparison_operator"
			[(zero_extract:SWI48
			   (match_operand:SWI48 1 "nonimmediate_operand")
			   (const_int 1)
			   (match_operand:SI 2 "nonmemory_operand"))
			 (const_int 0)])
		      (label_ref (match_operand 3))
		      (pc)))
   (clobber (reg:CC FLAGS_REG))]
  "(TARGET_USE_BT || optimize_function_for_size_p (cfun))
   && (CONST_INT_P (operands[2])
       ? (INTVAL (operands[2]) < GET_MODE_BITSIZE (<MODE>mode)
	  && INTVAL (operands[2])
	       >= (optimize_function_for_size_p (cfun) ? 8 : 32))
       : !memory_operand (operands[1], <MODE>mode))
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extract:SWI48
	    (match_dup 1)
	    (const_int 1)
	    (match_dup 2))
	  (const_int 0)))
   (set (pc)
	(if_then_else (match_op_dup 0 [(reg:CCC FLAGS_REG) (const_int 0)])
		      (label_ref (match_dup 3))
		      (pc)))]
{
  operands[0] = shallow_copy_rtx (operands[0]);
  PUT_CODE (operands[0], reverse_condition (GET_CODE (operands[0])));
})

(define_insn_and_split "*jcc_bt<mode>_1"
  [(set (pc)
  	(if_then_else (match_operator 0 "bt_comparison_operator"
			[(zero_extract:SWI48
			   (match_operand:SWI48 1 "register_operand")
			   (const_int 1)
			   (zero_extend:SI
			     (match_operand:QI 2 "register_operand")))
			 (const_int 0)])
		      (label_ref (match_operand 3))
		      (pc)))
   (clobber (reg:CC FLAGS_REG))]
  "(TARGET_USE_BT || optimize_function_for_size_p (cfun))
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extract:SWI48
	    (match_dup 1)
	    (const_int 1)
	    (match_dup 2))
	  (const_int 0)))
   (set (pc)
	(if_then_else (match_op_dup 0 [(reg:CCC FLAGS_REG) (const_int 0)])
		      (label_ref (match_dup 3))
		      (pc)))]
{
  operands[2] = lowpart_subreg (SImode, operands[2], QImode);
  operands[0] = shallow_copy_rtx (operands[0]);
  PUT_CODE (operands[0], reverse_condition (GET_CODE (operands[0])));
})

;; Avoid useless masking of bit offset operand.
(define_insn_and_split "*jcc_bt<mode>_mask"
  [(set (pc)
  	(if_then_else (match_operator 0 "bt_comparison_operator"
			[(zero_extract:SWI48
			   (match_operand:SWI48 1 "register_operand")
			   (const_int 1)
			   (and:SI
			     (match_operand:SI 2 "register_operand")
			     (match_operand 3 "const_int_operand")))])
		      (label_ref (match_operand 4))
		      (pc)))
   (clobber (reg:CC FLAGS_REG))]
  "(TARGET_USE_BT || optimize_function_for_size_p (cfun))
   && (INTVAL (operands[3]) & (GET_MODE_BITSIZE (<MODE>mode)-1))
      == GET_MODE_BITSIZE (<MODE>mode)-1
   && can_create_pseudo_p ()"
  "#"
  "&& 1"
  [(set (reg:CCC FLAGS_REG)
	(compare:CCC
	  (zero_extract:SWI48
	    (match_dup 1)
	    (const_int 1)
	    (match_dup 2))
	  (const_int 0)))
   (set (pc)
	(if_then_else (match_op_dup 0 [(reg:CCC FLAGS_REG) (const_int 0)])
		      (label_ref (match_dup 4))
		      (pc)))]
{
  operands[0] = shallow_copy_rtx (operands[0]);
  PUT_CODE (operands[0], reverse_condition (GET_CODE (operands[0])));
})

;; Store-flag instructions.

;; For all sCOND expanders, also expand the compare or test insn that
;; generates cc0.  Generate an equality comparison if `seq' or `sne'.

(define_insn_and_split "*setcc_di_1"
  [(set (match_operand:DI 0 "register_operand" "=q")
	(match_operator:DI 1 "ix86_comparison_operator"
	  [(reg FLAGS_REG) (const_int 0)]))]
  "TARGET_64BIT && !TARGET_PARTIAL_REG_STALL"
  "#"
  "&& reload_completed"
  [(set (match_dup 2) (match_dup 1))
   (set (match_dup 0) (zero_extend:DI (match_dup 2)))]
{
  operands[1] = shallow_copy_rtx (operands[1]);
  PUT_MODE (operands[1], QImode);
  operands[2] = gen_lowpart (QImode, operands[0]);
})

(define_insn_and_split "*setcc_si_1_and"
  [(set (match_operand:SI 0 "register_operand" "=q")
	(match_operator:SI 1 "ix86_comparison_operator"
	  [(reg FLAGS_REG) (const_int 0)]))
   (clobber (reg:CC FLAGS_REG))]
  "!TARGET_PARTIAL_REG_STALL
   && TARGET_ZERO_EXTEND_WITH_AND && optimize_function_for_speed_p (cfun)"
  "#"
  "&& reload_completed"
  [(set (match_dup 2) (match_dup 1))
   (parallel [(set (match_dup 0) (zero_extend:SI (match_dup 2)))
	      (clobber (reg:CC FLAGS_REG))])]
{
  operands[1] = shallow_copy_rtx (operands[1]);
  PUT_MODE (operands[1], QImode);
  operands[2] = gen_lowpart (QImode, operands[0]);
})

(define_insn_and_split "*setcc_si_1_movzbl"
  [(set (match_operand:SI 0 "register_operand" "=q")
	(match_operator:SI 1 "ix86_comparison_operator"
	  [(reg FLAGS_REG) (const_int 0)]))]
  "!TARGET_PARTIAL_REG_STALL
   && (!TARGET_ZERO_EXTEND_WITH_AND || optimize_function_for_size_p (cfun))"
  "#"
  "&& reload_completed"
  [(set (match_dup 2) (match_dup 1))
   (set (match_dup 0) (zero_extend:SI (match_dup 2)))]
{
  operands[1] = shallow_copy_rtx (operands[1]);
  PUT_MODE (operands[1], QImode);
  operands[2] = gen_lowpart (QImode, operands[0]);
})

(define_insn "*setcc_qi"
  [(set (match_operand:QI 0 "nonimmediate_operand" "=qm")
	(match_operator:QI 1 "ix86_comparison_operator"
	  [(reg FLAGS_REG) (const_int 0)]))]
  ""
  "set%C1\t%0"
  [(set_attr "type" "setcc")
   (set_attr "mode" "QI")])

(define_insn "*setcc_qi_slp"
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand" "+qm"))
	(match_operator:QI 1 "ix86_comparison_operator"
	  [(reg FLAGS_REG) (const_int 0)]))]
  ""
  "set%C1\t%0"
  [(set_attr "type" "setcc")
   (set_attr "mode" "QI")])

;; In general it is not safe to assume too much about CCmode registers,
;; so simplify-rtx stops when it sees a second one.  Under certain
;; conditions this is safe on x86, so help combine not create
;;
;;	seta	%al
;;	testb	%al, %al
;;	sete	%al

(define_split
  [(set (match_operand:QI 0 "nonimmediate_operand")
	(ne:QI (match_operator 1 "ix86_comparison_operator"
	         [(reg FLAGS_REG) (const_int 0)])
	    (const_int 0)))]
  ""
  [(set (match_dup 0) (match_dup 1))]
{
  operands[1] = shallow_copy_rtx (operands[1]);
  PUT_MODE (operands[1], QImode);
})

(define_split
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand"))
	(ne:QI (match_operator 1 "ix86_comparison_operator"
	         [(reg FLAGS_REG) (const_int 0)])
	    (const_int 0)))]
  ""
  [(set (match_dup 0) (match_dup 1))]
{
  operands[1] = shallow_copy_rtx (operands[1]);
  PUT_MODE (operands[1], QImode);
})

(define_split
  [(set (match_operand:QI 0 "nonimmediate_operand")
	(eq:QI (match_operator 1 "ix86_comparison_operator"
	         [(reg FLAGS_REG) (const_int 0)])
	    (const_int 0)))]
  ""
  [(set (match_dup 0) (match_dup 1))]
{
  operands[1] = shallow_copy_rtx (operands[1]);
  PUT_MODE (operands[1], QImode);
  PUT_CODE (operands[1],
	    ix86_reverse_condition (GET_CODE (operands[1]),
				    GET_MODE (XEXP (operands[1], 0))));

  /* Make sure that (a) the CCmode we have for the flags is strong
     enough for the reversed compare or (b) we have a valid FP compare.  */
  if (! ix86_comparison_operator (operands[1], VOIDmode))
    FAIL;
})

(define_split
  [(set (strict_low_part (match_operand:QI 0 "nonimmediate_operand"))
	(eq:QI (match_operator 1 "ix86_comparison_operator"
	         [(reg FLAGS_REG) (const_int 0)])
	    (const_int 0)))]
  ""
  [(set (match_dup 0) (match_dup 1))]
{
  operands[1] = shallow_copy_rtx (operands[1]);
  PUT_MODE (operands[1], QImode);
  PUT_CODE (operands[1],
  	    ix86_reverse_condition (GET_CODE (operands[1]),
				    GET_MODE (XEXP (operands[1], 0))));

  /* Make sure that (a) the CCmode we have for the flags is strong
     enough for the reversed compare or (b) we have a valid FP compare.  */
  if (! ix86_comparison_operator (operands[1], VOIDmode))
    FAIL;
})

;; The SSE store flag instructions saves 0 or 0xffffffff to the result.
;; subsequent logical operations are used to imitate conditional moves.
;; 0xffffffff is NaN, but not in normalized form, so we can't represent
;; it directly.

(define_insn "setcc_<mode>_sse"
  [(set (match_operand:MODEF 0 "register_operand" "=x,x")
	(match_operator:MODEF 3 "sse_comparison_operator"
	  [(match_operand:MODEF 1 "register_operand" "0,x")
	   (match_operand:MODEF 2 "nonimmediate_operand" "xm,xm")]))]
  "SSE_FLOAT_MODE_P (<MODE>mode)"
  "@
   cmp%D3<ssemodesuffix>\t{%2, %0|%0, %2}
   vcmp%D3<ssemodesuffix>\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "isa" "noavx,avx")
   (set_attr "type" "ssecmp")
   (set_attr "length_immediate" "1")
   (set_attr "prefix" "orig,vex")
   (set_attr "mode" "<MODE>")])

;; Basic conditional jump instructions.
;; We ignore the overflow flag for signed branch instructions.

(define_insn "*jcc"
  [(set (pc)
	(if_then_else (match_operator 1 "ix86_comparison_operator"
				      [(reg FLAGS_REG) (const_int 0)])
		      (label_ref (match_operand 0))
		      (pc)))]
  ""
  "%!%+j%C1\t%l0"
  [(set_attr "type" "ibr")
   (set_attr "modrm" "0")
   (set (attr "length")
	(if_then_else
	  (and (ge (minus (match_dup 0) (pc))
		   (const_int -126))
	       (lt (minus (match_dup 0) (pc))
		   (const_int 128)))
	  (const_int 2)
	  (const_int 6)))
   (set_attr "maybe_prefix_bnd" "1")])

;; In general it is not safe to assume too much about CCmode registers,
;; so simplify-rtx stops when it sees a second one.  Under certain
;; conditions this is safe on x86, so help combine not create
;;
;;	seta	%al
;;	testb	%al, %al
;;	je	Lfoo

(define_split
  [(set (pc)
	(if_then_else (ne (match_operator 0 "ix86_comparison_operator"
				      [(reg FLAGS_REG) (const_int 0)])
			  (const_int 0))
		      (label_ref (match_operand 1))
		      (pc)))]
  ""
  [(set (pc)
	(if_then_else (match_dup 0)
		      (label_ref (match_dup 1))
		      (pc)))]
{
  operands[0] = shallow_copy_rtx (operands[0]);
  PUT_MODE (operands[0], VOIDmode);
})

(define_split
  [(set (pc)
	(if_then_else (eq (match_operator 0 "ix86_comparison_operator"
				      [(reg FLAGS_REG) (const_int 0)])
			  (const_int 0))
		      (label_ref (match_operand 1))
		      (pc)))]
  ""
  [(set (pc)
	(if_then_else (match_dup 0)
		      (label_ref (match_dup 1))
		      (pc)))]
{
  operands[0] = shallow_copy_rtx (operands[0]);
  PUT_MODE (operands[0], VOIDmode);
  PUT_CODE (operands[0],
  	    ix86_reverse_condition (GET_CODE (operands[0]),
				    GET_MODE (XEXP (operands[0], 0))));

  /* Make sure that (a) the CCmode we have for the flags is strong
     enough for the reversed compare or (b) we have a valid FP compare.  */
  if (! ix86_comparison_operator (operands[0], VOIDmode))
    FAIL;
})

;; Unconditional and other jump instructions

(define_insn "jump"
  [(set (pc)
	(label_ref (match_operand 0)))]
  ""
  "%!jmp\t%l0"
  [(set_attr "type" "ibr")
   (set_attr "modrm" "0")
   (set (attr "length")
	(if_then_else
	  (and (ge (minus (match_dup 0) (pc))
		   (const_int -126))
	       (lt (minus (match_dup 0) (pc))
		   (const_int 128)))
	  (const_int 2)
	  (const_int 5)))
   (set_attr "maybe_prefix_bnd" "1")])

(define_expand "indirect_jump"
  [(set (pc) (match_operand 0 "indirect_branch_operand"))]
  ""
{
  if (TARGET_X32)
    operands[0] = convert_memory_address (word_mode, operands[0]);
})

(define_insn "*indirect_jump"
  [(set (pc) (match_operand:W 0 "indirect_branch_operand" "rBw"))]
  ""
  "%!jmp\t%A0"
  [(set_attr "type" "ibr")
   (set_attr "length_immediate" "0")
   (set_attr "maybe_prefix_bnd" "1")])

(define_expand "tablejump"
  [(parallel [(set (pc) (match_operand 0 "indirect_branch_operand"))
	      (use (label_ref (match_operand 1)))])]
  ""
{
  /* In PIC mode, the table entries are stored GOT (32-bit) or PC (64-bit)
     relative.  Convert the relative address to an absolute address.  */
  if (flag_pic)
    {
      rtx op0, op1;
      enum rtx_code code;

      /* We can't use @GOTOFF for text labels on VxWorks;
	 see gotoff_operand.  */
      if (TARGET_64BIT || TARGET_VXWORKS_RTP)
	{
	  code = PLUS;
	  op0 = operands[0];
	  op1 = gen_rtx_LABEL_REF (Pmode, operands[1]);
	}
      else if (TARGET_MACHO || HAVE_AS_GOTOFF_IN_DATA)
	{
	  code = PLUS;
	  op0 = operands[0];
	  op1 = pic_offset_table_rtx;
	}
      else
	{
	  code = MINUS;
	  op0 = pic_offset_table_rtx;
	  op1 = operands[0];
	}

      operands[0] = expand_simple_binop (Pmode, code, op0, op1, NULL_RTX, 0,
					 OPTAB_DIRECT);
    }

  if (TARGET_X32)
    operands[0] = convert_memory_address (word_mode, operands[0]);
})

(define_insn "*tablejump_1"
  [(set (pc) (match_operand:W 0 "indirect_branch_operand" "rBw"))
   (use (label_ref (match_operand 1)))]
  ""
  "%!jmp\t%A0"
  [(set_attr "type" "ibr")
   (set_attr "length_immediate" "0")
   (set_attr "maybe_prefix_bnd" "1")])

;; Convert setcc + movzbl to xor + setcc if operands don't overlap.

(define_peephole2
  [(set (reg FLAGS_REG) (match_operand 0))