TEXT cmpbody<>(SB),NOSPLIT,$0-0
	CMPQ	SI, DI
	JEQ	allsame
	CMPQ	BX, DX
	MOVQ	DX, R8
	CMOVQLT	BX, R8 // R8 = min(alen, blen) = # of bytes to compare
	CMPQ	R8, $8
	JB	small

	CMPQ	R8, $63
	JBE	loop
#ifndef hasAVX2
	CMPB	internal∕cpu·X86+const_offsetX86HasAVX2(SB), $1
	JEQ     big_loop_avx2
	JMP	big_loop
#else
	JMP	big_loop_avx2
#endif
loop:
	CMPQ	R8, $16
	JBE	_0through16
	MOVOU	(SI), X0

		t.Fatalf("Child's process group is not the child's process ID")
	}
}

func TestPgid(t *testing.T) {
	ppid, ppgrp := parent()

	cmd1 := create(t)

	cmd1.proc.SysProcAttr = &syscall.SysProcAttr{Setpgid: true}
	cmd1.Start()
	defer cmd1.Stop()

	cpid1, cpgrp1 := cmd1.Info()

	if cpid1 == ppid {
		t.Fatalf("Parent and child 1 have the same process ID")
	}

	if cpgrp1 == ppgrp {

TEXT runtime·asmstdcall(SB),NOSPLIT,$0
	MOVL	fn+0(FP), BX
	MOVL	SP, BP	// save stack pointer

	// SetLastError(0).
	MOVL	$0, 0x34(FS)

	MOVL	libcall_n(BX), CX

	// Fast version, do not store args on the stack.
	CMPL	CX, $0
	JE	docall

	// Copy args to the stack.
	MOVL	CX, AX
	SALL	$2, AX
	SUBL	AX, SP			// room for args
	MOVL	SP, DI
	MOVL	libcall_args(BX), SI
	CLD
	REP; MOVSL

docall:

#include "textflag.h"

TEXT ·Count(SB),NOSPLIT,$0-40
#ifndef hasPOPCNT
	CMPB	internal∕cpu·X86+const_offsetX86HasPOPCNT(SB), $1
	JEQ	2(PC)
	JMP	·countGeneric(SB)
#endif
	MOVQ	b_base+0(FP), SI
	MOVQ	b_len+8(FP), BX
	MOVB	c+24(FP), AL
	LEAQ	ret+32(FP), R8
	JMP	countbody<>(SB)

TEXT ·CountString(SB),NOSPLIT,$0-32
#ifndef hasPOPCNT
	CMPB	internal∕cpu·X86+const_offsetX86HasPOPCNT(SB), $1
	JEQ	2(PC)

                    artifact(name: expectedVariant)
                }
            }
        }

        where:
        capability | expectedVariant
        'cap1'     | 'first'
        'cap2'     | 'second'
    }

	MOVL	dig+0(FP),	BP
	MOVL	p+4(FP),	SI
	MOVL	p_len+8(FP), DX
	SHRL	$6,		DX
	SHLL	$6,		DX

	LEAL	(SI)(DX*1),	DI
	MOVL	(0*4)(BP),	AX
	MOVL	(1*4)(BP),	BX
	MOVL	(2*4)(BP),	CX
	MOVL	(3*4)(BP),	DX

	CMPL	SI,		DI
	JEQ	end

	MOVL	DI,		16(SP)

loop:
	MOVL	AX,		0(SP)
	MOVL	BX,		4(SP)
	MOVL	CX,		8(SP)
	MOVL	DX,		12(SP)

	MOVL	(0*4)(SI),	DI
	MOVL	DX,		BP

	ROUND1(AX,BX,CX,DX, 1,0xd76aa478, 7);

	MOVQ DI, 40(SP)
	MOVQ R8, 48(SP)
	MOVQ R9, 56(SP)
	MOVQ R10, 64(SP)
	MOVQ R11, 72(SP)
	MOVQ R12, 80(SP)
	MOVQ R13, 88(SP)
	MOVQ R14, 96(SP)
	MOVQ R15, 104(SP)
	#ifdef GOOS_darwin
	#ifndef hasAVX
	CMPB internal∕cpu·X86+const_offsetX86HasAVX(SB), $0
	JE 2(PC)
	#endif
	VZEROUPPER
	#endif
	MOVUPS X0, 112(SP)
	MOVUPS X1, 128(SP)
	MOVUPS X2, 144(SP)
	MOVUPS X3, 160(SP)
	MOVUPS X4, 176(SP)
	MOVUPS X5, 192(SP)

	addMulVVW(1536)
	addMulVVW(2048)

	Generate()
}

func addMulVVW(bits int) {
	if bits%64 != 0 {
		panic("bit size unsupported")
	}

	Implement("addMulVVW" + strconv.Itoa(bits))

	CMPB(Mem{Symbol: Symbol{Name: "·supportADX"}, Base: StaticBase}, Imm(1))
	JEQ(LabelRef("adx"))

	z := Mem{Base: Load(Param("z"), GP64())}
	x := Mem{Base: Load(Param("x"), GP64())}
	y := Load(Param("y"), GP64())

(CMP(Q|L|W|B)loadidx1 {sym} [off] ptr idx x mem) => (CMP(Q|L|W|B) (MOV(Q|L|W|B)loadidx1 {sym} [off] ptr idx mem) x)
(CMPQloadidx8 {sym} [off] ptr idx x mem) => (CMPQ (MOVQloadidx8 {sym} [off] ptr idx mem) x)
(CMPLloadidx4 {sym} [off] ptr idx x mem) => (CMPL (MOVLloadidx4 {sym} [off] ptr idx mem) x)
(CMPWloadidx2 {sym} [off] ptr idx x mem) => (CMPW (MOVWloadidx2 {sym} [off] ptr idx mem) x)

	SHA256ROUND(index, const, a, b, c, d, e, f, g, h)

TEXT ·block(SB),0,$296-16
	MOVL	p_base+4(FP), SI
	MOVL	p_len+8(FP), DX
	SHRL	$6, DX
	SHLL	$6, DX

	LEAL	(SI)(DX*1), DI
	MOVL	DI, 288(SP)
	CMPL	SI, DI
	JEQ	end

	LEAL	256(SP), DI		// variables

	MOVL	dig+0(FP), BP
	MOVL	(0*4)(BP), AX		// a = H0
	MOVL	AX, (0*4)(DI)
	MOVL	(1*4)(BP), BX		// b = H1
	MOVL	BX, (1*4)(DI)