MOVQ productTable+0(FP), pTbl
	MOVQ data_base+8(FP), aut
	MOVQ data_len+16(FP), autLen
	MOVQ T+32(FP), tPtr

	PXOR ACC0, ACC0
	MOVOU bswapMask<>(SB), BSWAP
	MOVOU gcmPoly<>(SB), POLY

	TESTQ autLen, autLen
	JEQ dataBail

	CMPQ autLen, $13	// optimize the TLS case
	JE dataTLS
	CMPQ autLen, $128
	JB startSinglesLoop
	JMP dataOctaLoop

dataTLS:
	MOVOU (16*14)(pTbl), T1

}

func UintLtZero(a uint8, b uint16, c uint32, d uint64) int {
	// amd64: -`(TESTB|TESTW|TESTL|TESTQ|JCC|JCS)`
	// arm64: -`(CMPW|CMP|BHS|BLO)`
	if a < 0 || b < 0 || c < 0 || d < 0 {
		return 1
	}
	return 0
}

func UintGeqZero(a uint8, b uint16, c uint32, d uint64) int {
	// amd64: -`(TESTB|TESTW|TESTL|TESTQ|JCS|JCC)`
	// arm64: -`(CMPW|CMP|BLO|BHS)`
	if a >= 0 || b >= 0 || c >= 0 || d >= 0 {
		return 1
	}

	// amd64:"TESTL"
	_ = x << v32
	// amd64:"TESTQ"
	_ = x << v64
}

// We want to avoid generating a test + panicshift for these cases.
func lshSignedMasked(v8 int8, v16 int16, v32 int32, v64 int64, x int) {
	// amd64:-"TESTB"
	_ = x << (v8 & 7)
	// amd64:-"TESTW"
	_ = x << (v16 & 15)
	// amd64:-"TESTL"
	_ = x << (v32 & 31)
	// amd64:-"TESTQ"
	_ = x << (v64 & 63)
}

// ------------------ //

	// 386:"TESTL\t[$]63",-"DIVL",-"SHRL"
	// amd64:"TESTQ\t[$]63",-"DIVQ",-"SHRQ"
	// arm:"AND\t[$]63",-".*udiv",-"SRA"
	// arm64:"TST\t[$]63",-"UDIV",-"ASR",-"AND"
	// ppc64x:"ANDCC",-"RLDICL",-"SRAD",-"CMP"
	a := n1%64 == 0 // signed divisible

	// 386:"TESTL\t[$]63",-"DIVL",-"SHRL"
	// amd64:"TESTQ\t[$]63",-"DIVQ",-"SHRQ"
	// arm:"AND\t[$]63",-".*udiv",-"SRA"

	MOVQ (8*2)(res_ptr), y_ptr
	MOVQ (8*3)(res_ptr), t1
	// Speculatively subtract
	SUBQ acc5, acc0
	SBBQ x_ptr, acc1
	SBBQ y_ptr, acc2
	SBBQ t1, acc3
	// If condition is 0, keep original value
	TESTQ t0, t0
	CMOVQEQ acc5, acc0
	CMOVQEQ x_ptr, acc1
	CMOVQEQ y_ptr, acc2
	CMOVQEQ t1, acc3
	// Store result
	MOVQ acc0, (8*0)(res_ptr)
	MOVQ acc1, (8*1)(res_ptr)
	MOVQ acc2, (8*2)(res_ptr)