.text
	.allow_optimizations
	.file	"3"
	.section	.text.__runCodelet_AddF16,"ax",@progbits
	.globl	__runCodelet_AddF16             # -- Begin function __runCodelet_AddF16
	.p2align	2
	.type	__runCodelet_AddF16,@function
__runCodelet_AddF16:                    # @__runCodelet_AddF16
	.worker
.Lfunc_begin0:
# %bb.0:                                # %entry
	setzi $m6, 40
	cmpult $m6, $m6, __worker_stack_max_growth_plus_one
	brz $m6, _stackoverflow
	add $m11, $m12, -40
	get $m0, 1
	ld32 $m10, $m13, $m15, 3
	and $m9, $m0, 7
	ld32 $m1, $m13, $m15, 4
	ld32 $m3, $m13, $m15, 5
	st32 $m9, $m11, $m15, 4                 # 4-byte Folded Spill
	st32 $m10, $m11, $m15, 9                # 4-byte Folded Spill
	brz $m10, .LBB0_23
# %bb.1:                                # %entry
	and $m0, $m10, 3
	brnz $m0, .LBB0_23
# %bb.2:                                # %entry
	and $m0, $m1, 3
	brnz $m0, .LBB0_23
# %bb.3:                                # %entry
	and $m0, $m3, 3
	brnz $m0, .LBB0_23
# %bb.4:                                # %land.lhs.true12.i
	ld32 $m2, $m13, $m15, 0
	ld32 $m6, $m13, $m15, 1
	or $m4, $m6, $m2
	ld32 $m0, $m13, $m15, 2
	or $m4, $m4, $m0
	and $m4, $m4, 6
	brz $m4, .LBB0_5
.LBB0_23:                               # %if.end62.i
	shr $m4, $m10, 1
	brz $m9, .LBB0_25
# %bb.24:                               # %if.end62.i
	cmpult $m0, $m9, $m4
	brnz $m0, .LBB0_25
.LBB0_77:                               # %_ZN6AddF167computeEj.exit
	exitz $m15
.LBB0_25:                               # %if.end68.i
	ld32 $m9, $m11, $m15, 9                 # 4-byte Folded Reload
	and $m5, $m9, 1
	brnz $m5, .LBB0_55
# %bb.26:                               # %if.end68.i
	and $m0, $m1, 1
	brnz $m0, .LBB0_55
# %bb.27:                               # %if.end68.i
	and $m0, $m3, 1
	brnz $m0, .LBB0_55
# %bb.28:                               # %if.then77.i
	cmpne $m2, $m1, $m9
	ld32 $m0, $m13, $m15, 0
	st32 $m0, $m11, $m15, 8                 # 4-byte Folded Spill
	ld32 $m7, $m13, $m15, 1
	ld32 $m0, $m13, $m15, 2
	st32 $m0, $m11, $m15, 7                 # 4-byte Folded Spill
	cmpne $m0, $m3, $m9
	brnz $m2, .LBB0_33
# %bb.29:                               # %if.then77.i
	brnz $m0, .LBB0_33
# %bb.30:                               # %for.cond89.preheader.i
	ld32 $m5, $m11, $m15, 4                 # 4-byte Folded Reload
	cmpult $m0, $m5, $m4
	brz $m0, .LBB0_77
# %bb.31:                               # %for.body93.i.preheader
	xnor $m0, $m5, $m15
	add $m0, $m0, $m4
	setzi $m1, 43691
	sort4x16lo $m2, $m0, $m15
	shr $m0, $m0, 16
	mul $m3, $m2, $m1
	mul $m1, $m0, $m1
	shr $m3, $m3, 16
	add $m1, $m1, $m3
	setzi $m3, 43690
	mul $m2, $m2, $m3
	mul $m0, $m0, $m3
	sort4x16lo $m3, $m1, $m15
	shr $m1, $m1, 16
	add $m2, $m2, $m3
	add $m0, $m0, $m1
	shr $m1, $m2, 16
	add $m0, $m0, $m1
	shr $m0, $m0, 2
	shl $m3, $m5, 2
	ld32 $m1, $m11, $m15, 7                 # 4-byte Folded Reload
	add $m1, $m1, $m3
	add $m2, $m7, $m3
	ld32 $m4, $m11, $m15, 8                 # 4-byte Folded Reload
	add $m3, $m4, $m3
.LBB0_32:                               # %for.body93.i
                                        # =>This Inner Loop Header: Depth=1
	ld32step $a0, $m15, $m3+=, 6
	ld32step $a1, $m15, $m2+=, 6
	f16v2add $a0, $a0, $a1
	st32step $a0, $m15, $m1+=, 6
	brnzdec $m0, .LBB0_32
	bri .LBB0_77
.LBB0_55:                               # %if.end180.i
	ld32 $m0, $m11, $m15, 4                 # 4-byte Folded Reload
	shl $m4, $m0, 1
	add $m6, $m1, $m4
	mov	$m0, $m4
.LBB0_56:                               # %while.cond.i448.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m2, $m0, $m1
	sub $m0, $m0, $m1
	sub $m6, $m6, $m1
	brz $m2, .LBB0_56
# %bb.57:                               # %while.cond.i453.i.preheader
	st32 $m5, $m11, $m15, 3                 # 4-byte Folded Spill
	add $m7, $m3, $m4
	mov	$m0, $m4
.LBB0_58:                               # %while.cond.i453.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m2, $m0, $m3
	sub $m0, $m0, $m3
	sub $m7, $m7, $m3
	brz $m2, .LBB0_58
# %bb.59:                               # %for.cond188.preheader.i
	or $m0, $m4, 1
	cmpult $m0, $m0, $m9
	brz $m0, .LBB0_70
# %bb.60:                               # %for.body192.lr.ph.i
	shl $m8, $m1, 1
	shl $m9, $m3, 1
	add $m0, $m8, 2
	st32 $m0, $m11, $m15, 8                 # 4-byte Folded Spill
	add $m0, $m9, 2
	st32 $m0, $m11, $m15, 7                 # 4-byte Folded Spill
	add $m0, $m1, 12
	st32 $m0, $m11, $m15, 6                 # 4-byte Folded Spill
	add $m0, $m3, 12
	st32 $m0, $m11, $m15, 5                 # 4-byte Folded Spill
.LBB0_61:                               # %for.body192.i
                                        # =>This Loop Header: Depth=1
                                        #     Child Loop BB0_62 Depth 2
                                        #     Child Loop BB0_64 Depth 2
                                        #     Child Loop BB0_66 Depth 2
                                        #     Child Loop BB0_68 Depth 2
	ld32 $m5, $m13, $m15, 0
	shl $m0, $m6, 1
	ldb16 $a0, $m5, $m15, $m6
	add $m10, $m6, 1
	ld32 $m2, $m11, $m15, 8                 # 4-byte Folded Reload
	add $m0, $m2, $m0
.LBB0_62:                               # %while.cond.i458.i
                                        #   Parent Loop BB0_61 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	cmpult $m2, $m10, $m1
	sub $m10, $m10, $m1
	sub $m0, $m0, $m8
	brz $m2, .LBB0_62
# %bb.63:                               # %_ZN6AddF164wrapEjj.exit459.i
                                        #   in Loop: Header=BB0_61 Depth=1
	ld32 $m10, $m13, $m15, 1
	ldb16 $a1, $m5, $m0, 0
	shl $m2, $m7, 1
	ldb16 $a2, $m10, $m15, $m7
	add $m0, $m7, 1
	ld32 $m5, $m11, $m15, 7                 # 4-byte Folded Reload
	add $m5, $m5, $m2
.LBB0_64:                               # %while.cond.i463.i
                                        #   Parent Loop BB0_61 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	cmpult $m2, $m0, $m3
	sub $m0, $m0, $m3
	sub $m5, $m5, $m9
	brz $m2, .LBB0_64
# %bb.65:                               # %_ZN6AddF164wrapEjj.exit464.i
                                        #   in Loop: Header=BB0_61 Depth=1
	{
		ldb16 $a1, $m10, $m5, 0
		sort4x16lo $a0, $a0, $a1
	}
	{
		ld32 $m0, $m13, $m15, 2
		sort4x16lo $a1, $a2, $a1
	}
	{
		shl $m2, $m4, 1
		f16v2add $a0, $a0, $a1
	}
	st32 $a0, $m0, $m2, 0
	add $m0, $m6, 12
	ld32 $m2, $m11, $m15, 6                 # 4-byte Folded Reload
	add $m6, $m2, $m6
.LBB0_66:                               # %while.cond.i468.i
                                        #   Parent Loop BB0_61 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	cmpult $m2, $m0, $m1
	sub $m0, $m0, $m1
	sub $m6, $m6, $m1
	brz $m2, .LBB0_66
# %bb.67:                               # %_ZN6AddF164wrapEjj.exit469.i
                                        #   in Loop: Header=BB0_61 Depth=1
	add $m0, $m7, 12
	ld32 $m2, $m11, $m15, 5                 # 4-byte Folded Reload
	add $m7, $m2, $m7
	ld32 $m5, $m11, $m15, 9                 # 4-byte Folded Reload
.LBB0_68:                               # %while.cond.i473.i
                                        #   Parent Loop BB0_61 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	cmpult $m2, $m0, $m3
	sub $m0, $m0, $m3
	sub $m7, $m7, $m3
	brz $m2, .LBB0_68
# %bb.69:                               # %_ZN6AddF164wrapEjj.exit474.i
                                        #   in Loop: Header=BB0_61 Depth=1
	add $m4, $m4, 12
	or $m0, $m4, 1
	cmpult $m0, $m0, $m5
	brnz $m0, .LBB0_61
.LBB0_70:                               # %for.cond.cleanup191.i
	ld32 $m0, $m11, $m15, 4                 # 4-byte Folded Reload
	brnz $m0, .LBB0_77
# %bb.71:                               # %for.cond.cleanup191.i
	ld32 $m0, $m11, $m15, 3                 # 4-byte Folded Reload
	brz $m0, .LBB0_77
# %bb.72:                               # %if.then224.i
	ld32 $m0, $m11, $m15, 9                 # 4-byte Folded Reload
	shl $m2, $m0, 1
	shl $m4, $m1, 1
	add $m0, $m0, -1
	add $m5, $m2, $m4
	add $m5, $m5, -2
	mov	$m6, $m0
.LBB0_73:                               # %while.cond.i478.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m7, $m6, $m1
	sub $m6, $m6, $m1
	sub $m5, $m5, $m4
	brz $m7, .LBB0_73
# %bb.74:                               # %_ZN6AddF164wrapEjj.exit479.i
	shl $m4, $m3, 1
	add $m1, $m2, $m4
	add $m1, $m1, -2
	ld32 $m2, $m13, $m15, 0
	ldb16 $a0, $m2, $m5, 0
	mov	$m2, $m0
.LBB0_75:                               # %while.cond.i483.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m5, $m2, $m3
	sub $m2, $m2, $m3
	sub $m1, $m1, $m4
	brz $m5, .LBB0_75
# %bb.76:                               # %_ZN6AddF164wrapEjj.exit484.i
	{
		ld32 $m2, $m13, $m15, 1
		setzi $a1, 21844
	}
	{
		shl $m0, $m0, 1
		setzi $a2, 43691
	}
	{
		ld32 $m3, $m13, $m15, 2
		or $a1, $a2, $a1
	}
	ldb16 $a2, $m2, $m1, 0
	{
		andc $m0, $m0, 3
		f16v2add $a0, $a0, $a2
	}
	{
		ld32 $m1, $m3, $m0, 0
		and $a0, $a0, $a1
	}
	setzi $m2, 65535
	andc $m1, $m1, $m2
	mov	$m2, $a0
	or $m1, $m1, $m2
	st32 $m1, $m3, $m0, 0
	exitz $m15
.LBB0_5:                                # %if.then.i
	cmpne $m4, $m1, $m10
	brnz $m4, .LBB0_12
# %bb.6:                                # %if.then.i
	cmpne $m4, $m3, $m10
	brnz $m4, .LBB0_12
# %bb.7:                                # %if.then19.i
	cmpult $m1, $m10, 48
	brnz $m1, .LBB0_10
# %bb.8:                                # %if.then19.i
	and $m1, $m10, 4
	brnz $m1, .LBB0_10
# %bb.9:                                # %if.then.i.i
	shr $m1, $m10, 3
	sub $m1, $m1, $m9
	setzi $m3, 43691
	mul $m1, $m1, $m3
	shl $m4, $m9, 4
	add $m2, $m2, $m4
	add $m5, $m6, $m4
	add $m0, $m0, $m4
	setzi $m4, 5
	mul $m3, $m4, $m3
	add $m1, $m1, $m3
	shr $m1, $m1, 18
	#APP
	{
		rpt $m1, 5
		fnop
	}

	{
		ld64step $ap0, $m15, $m2+=, 1
		fnop
	}

	{
		ld64step $ap1, $m15, $m5+=, 1
		fnop
	}

	{
		ld64step $ap2, $m15, $m2+=, 11
		f16v4add $ap0, $ap0, $ap1
	}

	{
		ld64step $ap3, $m15, $m5+=, 11
		fnop
	}

	{
		st64step $ap0, $m15, $m0+=, 1
		f16v4add $ap2, $ap2, $ap3
	}

	{
		st64step $ap2, $m15, $m0+=, 11
		fnop
	}


	#NO_APP
	exitz $m15
.LBB0_33:                               # %if.else.i
	cmpeq $m2, $m1, $m9
	brnz $m2, .LBB0_35
# %bb.34:                               # %if.else.i
	brnz $m0, .LBB0_44
.LBB0_35:                               # %land.lhs.true103.i
	cmpult $m8, $m1, $m3
	mov	$m0, $m3
	movnz	$m0, $m8, $m1
	sub $m8, 0, $m0
.LBB0_36:                               # %while.cond.i423.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m10, $m9, $m0
	sub $m9, $m9, $m0
	add $m8, $m8, $m0
	brz $m10, .LBB0_36
# %bb.37:                               # %_ZN6AddF164wrapEjj.exit424.i
	ld32 $m5, $m11, $m15, 9                 # 4-byte Folded Reload
	cmpne $m0, $m5, $m8
	brnz $m0, .LBB0_44
# %bb.38:                               # %if.then111.i
	cmpult $m0, $m5, 2
	ld32 $m8, $m11, $m15, 4                 # 4-byte Folded Reload
	brnz $m0, .LBB0_77
# %bb.39:                               # %for.cond138.preheader.lr.ph.i
	mov	$m0, $m15
	movnz	$m1, $m2, $m3
	shr $m1, $m1, 1
	xnor $m3, $m8, $m15
	add $m3, $m3, $m1
	setzi $m5, 43691
	sort4x16lo $m9, $m3, $m15
	shr $m3, $m3, 16
	mul $m10, $m9, $m5
	mul $m5, $m3, $m5
	shr $m10, $m10, 16
	add $m5, $m5, $m10
	setzi $m10, 43690
	mul $m9, $m9, $m10
	mul $m3, $m3, $m10
	sort4x16lo $m10, $m5, $m15
	shr $m5, $m5, 16
	add $m9, $m9, $m10
	add $m3, $m3, $m5
	shr $m5, $m9, 16
	mov	$m9, $m7
	add $m3, $m3, $m5
	ld32 $m5, $m11, $m15, 8                 # 4-byte Folded Reload
	movnz	$m9, $m2, $m5
	movnz	$m5, $m2, $m7
	shl $m2, $m8, 2
	shr $m7, $m3, 2
	add $m3, $m5, $m2
	st32 $m3, $m11, $m15, 9                 # 4-byte Folded Spill
	ld32 $m3, $m11, $m15, 7                 # 4-byte Folded Reload
	add $m3, $m3, $m2
	shl $m5, $m1, 2
	add $m6, $m9, $m2
	add $m7, $m7, 1
	bri .LBB0_40
.LBB0_43:                               # %for.cond.cleanup140.i
                                        #   in Loop: Header=BB0_40 Depth=1
	add $m0, $m0, $m1
	cmpult $m2, $m0, $m4
	add $m3, $m3, $m5
	add $m6, $m6, $m5
	ld32 $m8, $m11, $m15, 4                 # 4-byte Folded Reload
	brz $m2, .LBB0_77
.LBB0_40:                               # %for.cond138.preheader.i
                                        # =>This Loop Header: Depth=1
                                        #     Child Loop BB0_42 Depth 2
	cmpult $m2, $m8, $m1
	brz $m2, .LBB0_43
# %bb.41:                               # %for.body141.i.preheader
                                        #   in Loop: Header=BB0_40 Depth=1
	mov	$m8, $m7
	mov	$m9, $m6
	mov	$m10, $m3
	ld32 $m2, $m11, $m15, 9                 # 4-byte Folded Reload
	add $m8, $m8, -1
.LBB0_42:                               # %for.body141.i
                                        #   Parent Loop BB0_40 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	ld32step $a0, $m15, $m9+=, 6
	ld32step $a1, $m15, $m2+=, 6
	f16v2add $a0, $a0, $a1
	st32step $a0, $m15, $m10+=, 6
	brnzdec $m8, .LBB0_42
	bri .LBB0_43
.LBB0_12:                               # %if.end.i
	cmpeq $m7, $m1, $m10
	brnz $m7, .LBB0_14
# %bb.13:                               # %if.end.i
	cmpne $m4, $m3, $m10
	brnz $m4, .LBB0_23
.LBB0_14:                               # %land.lhs.true28.i
	cmpult $m5, $m1, $m3
	mov	$m4, $m3
	movnz	$m4, $m5, $m1
	sub $m5, 0, $m4
	mov	$m8, $m10
.LBB0_15:                               # %while.cond.i.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m9, $m8, $m4
	sub $m8, $m8, $m4
	add $m5, $m5, $m4
	brz $m9, .LBB0_15
# %bb.16:                               # %_ZN6AddF164wrapEjj.exit.i
	cmpne $m4, $m10, $m5
	ld32 $m9, $m11, $m15, 4                 # 4-byte Folded Reload
	brnz $m4, .LBB0_23
# %bb.17:                               # %if.then32.i
	mov	$m4, $m15
	mov	$m5, $m15
	movnz	$m1, $m7, $m3
	shr $m3, $m1, 2
	sub $m8, 5, $m9
	mov	$m10, $m9
	shr $m9, $m1, 3
	add $m9, $m9, $m8
	st32 $m3, $m11, $m15, 8                 # 4-byte Folded Spill
	add $m8, $m3, $m8
	mov	$m3, $m6
	movnz	$m3, $m7, $m2
	st32 $m3, $m11, $m15, 2                 # 4-byte Folded Spill
	movnz	$m2, $m7, $m6
	setzi $m6, 47
	and $m7, $m1, 4
	cmpult $m6, $m6, $m1
	cmpeq $m7, $m7, 0
	and $m6, $m6, $m7
	setzi $m7, 43691
	mul $m3, $m8, $m7
	mul $m8, $m9, $m7
	shl $m7, $m10, 3
	shr $m3, $m3, 18
	st32 $m3, $m11, $m15, 6                 # 4-byte Folded Spill
	shr $m8, $m8, 18
	shl $m9, $m10, 4
	add $m3, $m2, $m7
	st32 $m3, $m11, $m15, 5                 # 4-byte Folded Spill
	add $m10, $m2, $m9
	add $m2, $m0, $m9
	st32 $m2, $m11, $m15, 7                 # 4-byte Folded Spill
	mov	$m2, $m7
	add $m0, $m0, $m2
	st32 $m0, $m11, $m15, 3                 # 4-byte Folded Spill
	ld32 $m0, $m11, $m15, 2                 # 4-byte Folded Reload
	add $m7, $m0, $m9
	add $m0, $m0, $m2
	st32 $m0, $m11, $m15, 2                 # 4-byte Folded Spill
	shl $m0, $m1, 1
	st32 $m0, $m11, $m15, 1                 # 4-byte Folded Spill
	bri .LBB0_18
.LBB0_19:                               # %if.then.i407.i
                                        #   in Loop: Header=BB0_18 Depth=1
	add $m9, $m7, $m4
	ld32 $m2, $m11, $m15, 7                 # 4-byte Folded Reload
	add $m3, $m2, $m4
	mov	$m2, $m10
	#APP
	{
		rpt $m8, 5
		fnop
	}

	{
		ld64step $ap0, $m15, $m9+=, 1
		fnop
	}

	{
		ld64step $ap1, $m15, $m2+=, 1
		fnop
	}

	{
		ld64step $ap2, $m15, $m9+=, 11
		f16v4add $ap0, $ap0, $ap1
	}

	{
		ld64step $ap3, $m15, $m2+=, 11
		fnop
	}

	{
		st64step $ap0, $m15, $m3+=, 1
		f16v4add $ap2, $ap2, $ap3
	}

	{
		st64step $ap2, $m15, $m3+=, 11
		fnop
	}


	#NO_APP
.LBB0_22:                               # %_ZL8addQuadsPKDhS0_PDhjj.exit419.i
                                        #   in Loop: Header=BB0_18 Depth=1
	add $m5, $m5, $m1
	ld32 $m2, $m11, $m15, 9                 # 4-byte Folded Reload
	cmpult $m2, $m5, $m2
	add $m4, $m4, $m0
	brz $m2, .LBB0_77
.LBB0_18:                               # %for.body.i
                                        # =>This Inner Loop Header: Depth=1
	brnz $m6, .LBB0_19
# %bb.20:                               # %if.end.i409.i
                                        #   in Loop: Header=BB0_18 Depth=1
	ld32 $m2, $m11, $m15, 4                 # 4-byte Folded Reload
	ld32 $m3, $m11, $m15, 8                 # 4-byte Folded Reload
	cmpult $m2, $m2, $m3
	brz $m2, .LBB0_22
# %bb.21:                               # %if.end10.i418.i
                                        #   in Loop: Header=BB0_18 Depth=1
	ld32 $m0, $m11, $m15, 2                 # 4-byte Folded Reload
	add $m2, $m0, $m4
	ld32 $m3, $m11, $m15, 3                 # 4-byte Folded Reload
	add $m3, $m3, $m4
	ld32 $m9, $m11, $m15, 5                 # 4-byte Folded Reload
	mov	$m0, $m6
	mov	$m6, $m7
	mov	$m7, $m10
	mov	$m10, $m8
	ld32 $m8, $m11, $m15, 6                 # 4-byte Folded Reload
	#APP
	{
		rpt $m8, 3
		fnop
	}

	{
		ld64step $ap0, $m15, $m2+=, 6
		fnop
	}

	{
		ld64step $ap1, $m15, $m9+=, 6
		fnop
	}

	{
		nop
		f16v4add $ap0, $ap0, $ap1
	}

	{
		st64step $ap0, $m15, $m3+=, 6
		fnop
	}


	#NO_APP
	mov	$m8, $m10
	mov	$m10, $m7
	mov	$m7, $m6
	mov	$m6, $m0
	ld32 $m0, $m11, $m15, 1                 # 4-byte Folded Reload
	bri .LBB0_22
.LBB0_44:                               # %if.else154.i
	shr $m1, $m1, 1
	ld32 $m6, $m11, $m15, 4                 # 4-byte Folded Reload
	add $m2, $m1, $m6
	mov	$m0, $m6
.LBB0_45:                               # %while.cond.i428.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m8, $m0, $m1
	sub $m0, $m0, $m1
	sub $m2, $m2, $m1
	brz $m8, .LBB0_45
# %bb.46:                               # %_ZN6AddF164wrapEjj.exit429.i
	shr $m3, $m3, 1
	add $m8, $m3, $m6
	mov	$m0, $m6
.LBB0_47:                               # %while.cond.i433.i
                                        # =>This Inner Loop Header: Depth=1
	cmpult $m9, $m0, $m3
	sub $m0, $m0, $m3
	sub $m8, $m8, $m3
	brz $m9, .LBB0_47
# %bb.48:                               # %for.cond160.preheader.i
	ld32 $m6, $m11, $m15, 4                 # 4-byte Folded Reload
	cmpult $m0, $m6, $m4
	brz $m0, .LBB0_77
# %bb.49:                               # %for.body164.i.preheader
	xnor $m0, $m6, $m15
	add $m0, $m0, $m4
	setzi $m4, 43691
	sort4x16lo $m9, $m0, $m15
	shr $m0, $m0, 16
	mul $m10, $m9, $m4
	mul $m4, $m0, $m4
	shr $m10, $m10, 16
	add $m4, $m4, $m10
	setzi $m10, 43690
	sort4x16lo $m5, $m4, $m15
	mul $m9, $m9, $m10
	shr $m4, $m4, 16
	mul $m0, $m0, $m10
	add $m9, $m9, $m5
	add $m0, $m0, $m4
	shr $m4, $m9, 16
	add $m0, $m0, $m4
	shr $m0, $m0, 2
	add $m4, $m1, 6
	add $m9, $m3, 6
.LBB0_50:                               # %for.body164.i
                                        # =>This Loop Header: Depth=1
                                        #     Child Loop BB0_51 Depth 2
                                        #     Child Loop BB0_53 Depth 2
	ld32 $m5, $m11, $m15, 8                 # 4-byte Folded Reload
	ld32 $a0, $m5, $m15, $m2
	ld32 $a1, $m7, $m15, $m8
	{
		mov	$m5, $m6
		f16v2add $a0, $a0, $a1
	}
	ld32 $m10, $m11, $m15, 7                # 4-byte Folded Reload
	st32 $a0, $m10, $m15, $m6
	add $m10, $m2, 6
	add $m2, $m4, $m2
.LBB0_51:                               # %while.cond.i438.i
                                        #   Parent Loop BB0_50 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	cmpult $m6, $m10, $m1
	sub $m10, $m10, $m1
	sub $m2, $m2, $m1
	brz $m6, .LBB0_51
# %bb.52:                               # %_ZN6AddF164wrapEjj.exit439.i
                                        #   in Loop: Header=BB0_50 Depth=1
	add $m10, $m8, 6
	add $m8, $m9, $m8
.LBB0_53:                               # %while.cond.i443.i
                                        #   Parent Loop BB0_50 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	cmpult $m6, $m10, $m3
	sub $m10, $m10, $m3
	sub $m8, $m8, $m3
	brz $m6, .LBB0_53
# %bb.54:                               # %_ZN6AddF164wrapEjj.exit444.i
                                        #   in Loop: Header=BB0_50 Depth=1
	mov	$m6, $m5
	add $m6, $m6, 6
	brnzdec $m0, .LBB0_50
	bri .LBB0_77
.LBB0_10:                               # %if.end.i.i
	shr $m1, $m10, 2
	cmpult $m3, $m9, $m1
	brz $m3, .LBB0_77
# %bb.11:                               # %if.end10.i.i
	sub $m1, $m1, $m9
	setzi $m3, 43691
	mul $m1, $m1, $m3
	shl $m4, $m9, 3
	add $m2, $m2, $m4
	add $m5, $m6, $m4
	add $m0, $m0, $m4
	setzi $m4, 5
	mul $m3, $m4, $m3
	add $m1, $m1, $m3
	shr $m1, $m1, 18
	#APP
	{
		rpt $m1, 3
		fnop
	}

	{
		ld64step $ap0, $m15, $m2+=, 6
		fnop
	}

	{
		ld64step $ap1, $m15, $m5+=, 6
		fnop
	}

	{
		nop
		f16v4add $ap0, $ap0, $ap1
	}

	{
		st64step $ap0, $m15, $m0+=, 6
		fnop
	}


	#NO_APP
	exitz $m15
.Lfunc_end0:
	.size	__runCodelet_AddF16, .Lfunc_end0-__runCodelet_AddF16
	.section	.stack_sizes,"o",@progbits,.text.__runCodelet_AddF16
	.long	.Lfunc_begin0
	.byte	40
	.section	.text.__runCodelet_AddF16,"ax",@progbits
                                        # -- End function
	.ident	"clang version 16.0.0 (git@github.com:Software-GCAI/llvm-project.git 6343a843cd5af31429c62a83b6ba249a74d36969)"
	.section	".note.GNU-stack","",@progbits
	.addrsig