TEXT _rt0_amd64_lib_go(SB),NOSPLIT,$0
	MOVQ	_rt0_amd64_lib_argc<>(SB), DI
	MOVQ	_rt0_amd64_lib_argv<>(SB), SI
	JMP	runtime·rt0_go(SB)

DATA _rt0_amd64_lib_argc<>(SB)/8, $0
GLOBL _rt0_amd64_lib_argc<>(SB),NOPTR, $8
DATA _rt0_amd64_lib_argv<>(SB)/8, $0
GLOBL _rt0_amd64_lib_argv<>(SB),NOPTR, $8

#ifdef GOAMD64_v2
DATA bad_cpu_msg<>+0x00(SB)/84, $"This program can only be run on AMD64 processors with v2 microarchitecture support.\n"
#endif

DATA p256mul<>+0x90(SB)/8, $0xffffffff00000000 // (1*2^256)%P256
DATA p256mul<>+0x98(SB)/8, $0x0000000000000001 // (1*2^256)%P256
GLOBL p256ordK0<>(SB), 8, $4
GLOBL p256ord<>(SB), 8, $32
GLOBL p256<>(SB), 8, $96
GLOBL p256mul<>(SB), 8, $160

// func p256OrdLittleToBig(res *[32]byte, in *p256OrdElement)
TEXT ·p256OrdLittleToBig(SB), NOSPLIT, $0
	JMP ·p256BigToLittle(SB)

DATA p256mul<>+0x90(SB)/8, $0x00000000fffffffe // (1*2^256)%P256
DATA p256mul<>+0x98(SB)/8, $0xffffffffffffffff // (1*2^256)%P256

// External declarations for constants
GLOBL p256ord<>(SB), 8, $32
GLOBL p256<>(SB), 8, $80
GLOBL p256mul<>(SB), 8, $160

// The following macros are used to implement the ppc64le
// equivalent function from the corresponding s390x
// instruction for vector multiply high, low, and add,

func TestConcurrentAsm(t *testing.T) {
	skipIfGccgo(t, "gccgo does not use cmd/asm")
	tg := testgo(t)
	defer tg.cleanup()
	tg.parallel()
	asm := `DATA ·constants<>+0x0(SB)/8,$0
GLOBL ·constants<>(SB),8,$8
`
	tg.tempFile("go/src/p/a.s", asm)
	tg.tempFile("go/src/p/b.s", asm)
	tg.tempFile("go/src/p/p.go", `package p`)
	tg.setenv("GOPATH", tg.path("go"))
	tg.run("build", "p")
}