.text
	.allow_optimizations
	.file	"3"
	.section	.text.__runCodelet_LayerNormApply,"ax",@progbits
	.globl	__runCodelet_LayerNormApply     # -- Begin function __runCodelet_LayerNormApply
	.p2align	2
	.type	__runCodelet_LayerNormApply,@function
__runCodelet_LayerNormApply:            # @__runCodelet_LayerNormApply
	.worker
.Lfunc_begin0:
# %bb.0:                                # %entry
	setzi $m6, 128
	cmpult $m6, $m6, __worker_stack_max_growth_plus_one
	brz $m6, _stackoverflow
	add $m11, $m12, -128
	get $m2, 1
	ld32 $m0, $m13, $m15, 5
	brz $m0, .LBB0_19
# %bb.1:                                # %for.cond9.preheader.lr.ph.i
	mov	$m1, $m15
	and $m3, $m2, 7
	{
		add $m0, $m0, -1
		setzi $a0, 407485
	}
	ld32 $m2, $m13, $m15, 6
	st32 $m2, $m11, $m15, 1
	ld32 $a1, $m11, $m15, 1
	{
		shr $m5, $m2, 1
		f32fromui32 $a5, $a1
	}
	{
		sub $m4, 5, $m3
		or $a6, $a0, 897581056
	}
	shr $m6, $m2, 2
	st32 $m5, $m11, $m15, 23                # 4-byte Folded Spill
	add $m5, $m5, $m4
	st32 $m6, $m11, $m15, 11                # 4-byte Folded Spill
	add $m4, $m6, $m4
	setzi $m6, 43691
	sort4x16lo $m7, $m5, $m15
	shr $m5, $m5, 16
	sort4x16lo $m8, $m4, $m15
	shr $m4, $m4, 16
	mul $m9, $m7, $m6
	shr $m9, $m9, 16
	st32 $m9, $m11, $m15, 29                # 4-byte Folded Spill
	mul $m10, $m8, $m6
	shr $m9, $m10, 16
	st32 $m9, $m11, $m15, 28                # 4-byte Folded Spill
	mul $m9, $m5, $m6
	ld32 $m10, $m11, $m15, 29               # 4-byte Folded Reload
	add $m9, $m9, $m10
	mul $m6, $m4, $m6
	ld32 $m10, $m11, $m15, 28               # 4-byte Folded Reload
	add $m6, $m6, $m10
	setzi $m10, 43690
	mul $m7, $m7, $m10
	mul $m8, $m8, $m10
	mul $m5, $m5, $m10
	mul $m4, $m4, $m10
	sort4x16lo $m10, $m9, $m15
	add $m7, $m7, $m10
	shr $m9, $m9, 16
	add $m5, $m5, $m9
	sort4x16lo $m9, $m6, $m15
	shr $m10, $m6, 16
	add $m8, $m8, $m9
	ld32 $m6, $m13, $m15, 7
	shr $m7, $m7, 16
	add $m4, $m4, $m10
	shr $m8, $m8, 16
	ld32 $m9, $m13, $m15, 1
	st32 $m6, $m11, $m15, 1
	ld32 $a0, $m11, $m15, 1
	{
		add $m5, $m5, $m7
		f32fromui32 $a0, $a0
	}
	ld32 $m10, $m13, $m15, 2
	{
		st32 $a0, $m11, $m15, 21
		f32div $a0, $azero, $a0
	}
	st32 $a0, $m11, $m15, 13                # 4-byte Folded Spill
	add $m4, $m4, $m8
	ld32 $m7, $m13, $m15, 3
	st32 $m7, $m11, $m15, 29                # 4-byte Folded Spill
	shr $m5, $m5, 2
	st32 $m5, $m11, $m15, 17                # 4-byte Folded Spill
	shr $m4, $m4, 2
	st32 $m4, $m11, $m15, 5                 # 4-byte Folded Spill
	mul $m4, $m6, $m2
	st32 $m4, $m11, $m15, 1
	ld32 $a0, $m11, $m15, 1
	{
		shl $m4, $m3, 1
		f32fromui32 $a7, $a0
	}
	shl $m4, $m4, 1
	st32 $m4, $m11, $m15, 16                # 4-byte Folded Spill
	shl $m5, $m3, 2
	shl $m4, $m3, 3
	ld32 $m7, $m13, $m15, 0
	st32 $m7, $m11, $m15, 24                # 4-byte Folded Spill
	ld32 $m8, $m13, $m15, 4
	st32 $m8, $m11, $m15, 27                # 4-byte Folded Spill
	st32 $m2, $m11, $m15, 28                # 4-byte Folded Spill
	and $m2, $m2, 3
	st32 $m2, $m11, $m15, 26                # 4-byte Folded Spill
	or $m2, $m10, $m9
	st32 $m2, $m11, $m15, 12                # 4-byte Folded Spill
	add $m2, $m9, $m5
	st32 $m2, $m11, $m15, 15                # 4-byte Folded Spill
	add $m2, $m9, $m4
	st32 $m2, $m11, $m15, 3                 # 4-byte Folded Spill
	add $m2, $m10, $m4
	st32 $m2, $m11, $m15, 2                 # 4-byte Folded Spill
	ld32 $m4, $m11, $m15, 29                # 4-byte Folded Reload
	st32 $m5, $m11, $m15, 4                 # 4-byte Folded Spill
	add $m2, $m10, $m5
	st32 $m2, $m11, $m15, 14                # 4-byte Folded Spill
	shl $m2, $m6, 3
	st32 $m2, $m11, $m15, 25                # 4-byte Folded Spill
	add $m10, $m4, 4
	st32 $m3, $m11, $m15, 20                # 4-byte Folded Spill
	st32 $a6, $m11, $m15, 19                # 4-byte Folded Spill
	st32 $a7, $m11, $m15, 18                # 4-byte Folded Spill
	st32 $a5, $m11, $m15, 6                 # 4-byte Folded Spill
.LBB0_2:                                # %for.cond9.preheader.i
                                        # =>This Loop Header: Depth=1
                                        #     Child Loop BB0_10 Depth 2
                                        #     Child Loop BB0_13 Depth 2
	brz $m6, .LBB0_3
# %bb.8:                                # %for.body12.lr.ph.i
                                        #   in Loop: Header=BB0_2 Depth=1
	{
		setzi $m9, 1
		mov	$a0, $azero
	}
	mov	$m5, $m4
	cmpslt $m8, $m9, $m6
	ld32step $a1, $m15, $m5+=, 2
	add $m9, $m6, -1
	brz $m8, .LBB0_11
# %bb.9:                                # %for.body12.i
                                        #   in Loop: Header=BB0_2 Depth=1
	add $m9, $m9, -1
.LBB0_10:                               # %for.body12.i
                                        #   Parent Loop BB0_2 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	{
		ld32step $a1, $m15, $m5+=, 2
		f32add $a0, $a0, $a1
	}
	brnzdec $m9, .LBB0_10
.LBB0_11:                               #   in Loop: Header=BB0_2 Depth=1
	f32add $a0, $a0, $a1
	{
		ld32 $a2, $m11, $m15, 21
		mov	$a1, $azero
	}
	{
		add $m2, $m15, -4
		f32div $a0, $a0, $a2
	}
	ld64 $ap1, $m10, $m2, 0
	{
		setzi $m2, 1
		f32sub $a2, $a2, $a0
	}
	{
		cmpslt $m2, $m2, $m6
		f32mul $a4, $a2, $a5
	}
	{
		add $m9, $m10, 8
		f32mul $a2, $a4, $a2
	}
	add $m5, $m6, -1
	brz $m2, .LBB0_14
# %bb.12:                               # %for.body18.i
                                        #   in Loop: Header=BB0_2 Depth=1
	add $m5, $m5, -1
.LBB0_13:                               # %for.body18.i
                                        #   Parent Loop BB0_2 Depth=1
                                        # =>  This Inner Loop Header: Depth=2
	{
		add $m2, $m15, -4
		f32add $a4, $a2, $a3
	}
	ld64 $ap1, $m9, $m2, 0
	{
		add $m9, $m9, 8
		f32sub $a2, $a2, $a0
	}
	f32add $a1, $a1, $a4
	f32mul $a4, $a2, $a5
	{
		brnzdec $m5, .LBB0_13
		f32mul $a2, $a4, $a2
	}
.LBB0_14:                               #   in Loop: Header=BB0_2 Depth=1
	f32add $a2, $a2, $a3
	{
		bri .LBB0_4
		f32add $a1, $a1, $a2
	}
.LBB0_3:                                #   in Loop: Header=BB0_2 Depth=1
	{
		ld32 $a0, $m11, $m15, 13
		mov	$a1, $azero
	}
.LBB0_4:                                # %for.cond.cleanup17.i
                                        #   in Loop: Header=BB0_2 Depth=1
	{
		ld32 $m2, $m11, $m15, 28
		f32div $a1, $a1, $a7
	}
	{
		mul $m5, $m1, $m2
		f32add $a1, $a1, $a6
	}
	shl $m5, $m5, 1
	#APP
	f32oorx $a1, $a1
	#NO_APP
	ld32 $m2, $m11, $m15, 24                # 4-byte Folded Reload
	add $m9, $m2, $m5
	ld32 $m2, $m11, $m15, 27                # 4-byte Folded Reload
	add $m5, $m2, $m5
	ld32 $m2, $m11, $m15, 26                # 4-byte Folded Reload
	brnz $m2, .LBB0_15
# %bb.5:                                # %land.lhs.true.i.i
                                        #   in Loop: Header=BB0_2 Depth=1
	ld32 $m2, $m11, $m15, 12                # 4-byte Folded Reload
	or $m8, $m2, $m9
	or $m8, $m8, $m5
	and $m8, $m8, 6
	brz $m8, .LBB0_6
.LBB0_15:                               # %if.end18.i.i
                                        #   in Loop: Header=BB0_2 Depth=1
	ld32 $m2, $m11, $m15, 23                # 4-byte Folded Reload
	cmpult $m2, $m3, $m2
	brz $m2, .LBB0_18
# %bb.16:                               # %if.end22.i.i
                                        #   in Loop: Header=BB0_2 Depth=1
	{
		mov	$m2, $m15
		f32sub $a0, $azero, $a0
	}
	st32 $m4, $m11, $m15, 29                # 4-byte Folded Spill
	ld32 $m4, $m11, $m15, 16                # 4-byte Folded Reload
	add $m3, $m9, $m4
	add $m5, $m5, $m4
	st32 $a0, $m11, $m15, 30
	st32 $a1, $m11, $m15, 31
	add $m4, $m11, 120
	st32 $m4, $m11, $m15, 22                # 4-byte Folded Spill
	ld32 $m9, $m11, $m15, 14                # 4-byte Folded Reload
	ld32 $m4, $m11, $m15, 15                # 4-byte Folded Reload
	mov	$m7, $m0
	mov	$m0, $m1
	mov	$m1, $m6
	ld32 $m6, $m11, $m15, 17                # 4-byte Folded Reload
	ld32 $m8, $m11, $m15, 22                # 4-byte Folded Reload
	#APP
	ld32 $a6, $m8, $m15, 0
	ld32 $a7, $m8, $m15, 1
	{
		rpt $m6, 9
		fnop
	}

	{
		ld32step $a0, $m15, $m3+=, 6
		fnop
	}

	{
		nop
		f16v2tof32 $ap0, $a0
	}

	{
		nop
		f32v2add $ap0, $a6:B, $ap0
	}

	{
		ld32step $a2, $m15, $m4+=, 6
		f32v2mul $ap0, $a7:B, $ap0
	}

	{
		nop
		f16v2tof32 $ap1, $a2
	}

	{
		ld32step $a4, $m15, $m9+=, 6
		f32v2mul $ap0, $ap0, $ap1
	}

	{
		nop
		f16v2tof32 $ap1, $a4
	}

	{
		nop
		f32v2add $ap0, $ap0, $ap1
	}

	{
		nop
		f32v2tof16 $a0, $ap0
	}

	{
		st32step $a0, $m15, $m5+=, 6
		fnop
	}


	#NO_APP
	ld32 $m4, $m11, $m15, 29                # 4-byte Folded Reload
	ld32 $a7, $m11, $m15, 18                # 4-byte Folded Reload
	mov	$m6, $m1
	mov	$m1, $m0
	mov	$m0, $m7
	ld32 $a6, $m11, $m15, 19                # 4-byte Folded Reload
.LBB0_17:                               # %_ZL9normApplyPKDhS0_S0_S0_PDhjjff.exit.i
                                        #   in Loop: Header=BB0_2 Depth=1
	ld32 $m3, $m11, $m15, 20                # 4-byte Folded Reload
.LBB0_18:                               # %_ZL9normApplyPKDhS0_S0_S0_PDhjjff.exit.i
                                        #   in Loop: Header=BB0_2 Depth=1
	add $m1, $m1, 1
	ld32 $m2, $m11, $m15, 25                # 4-byte Folded Reload
	add $m4, $m4, $m2
	add $m10, $m10, $m2
	brnzdec $m0, .LBB0_2
	bri .LBB0_19
.LBB0_6:                                # %if.then.i.i
                                        #   in Loop: Header=BB0_2 Depth=1
	ld32 $m2, $m11, $m15, 11                # 4-byte Folded Reload
	cmpult $m8, $m3, $m2
	brz $m8, .LBB0_18
# %bb.7:                                # %if.end.i.i
                                        #   in Loop: Header=BB0_2 Depth=1
	st32 $m15, $m11, $m15, 10               # 4-byte Folded Spill
	ld32 $m2, $m11, $m15, 4                 # 4-byte Folded Reload
	{
		shl $m2, $m2, 1
		f32sub $a0, $azero, $a0
	}
	add $m3, $m9, $m2
	st32 $m3, $m11, $m15, 9                 # 4-byte Folded Spill
	st32 $a0, $m11, $m15, 30
	add $m2, $m5, $m2
	st32 $m2, $m11, $m15, 8                 # 4-byte Folded Spill
	st32 $a1, $m11, $m15, 31
	add $m3, $m11, 120
	st32 $m3, $m11, $m15, 22                # 4-byte Folded Spill
	ld32 $m2, $m11, $m15, 2                 # 4-byte Folded Reload
	st32 $m2, $m11, $m15, 7                 # 4-byte Folded Spill
	ld32 $m5, $m11, $m15, 3                 # 4-byte Folded Reload
	st32 $m4, $m11, $m15, 29                # 4-byte Folded Spill
	ld32 $m4, $m11, $m15, 5                 # 4-byte Folded Reload
	ld32 $m8, $m11, $m15, 22                # 4-byte Folded Reload
	ld32 $m9, $m11, $m15, 10                # 4-byte Folded Reload
	ld32 $m2, $m11, $m15, 9                 # 4-byte Folded Reload
	ld32 $m3, $m11, $m15, 8                 # 4-byte Folded Reload
	ld32 $m7, $m11, $m15, 7                 # 4-byte Folded Reload
	#APP
	ld32 $a6, $m8, $m15, 0
	ld32 $a7, $m8, $m15, 1
	{
		rpt $m4, 11
		fnop
	}

	{
		ld64step $ap2, $m15, $m2+=, 6
		fnop
	}

	{
		nop
		f16v2tof32 $ap0, $a4
	}

	{
		nop
		f16v2tof32 $ap1, $a5
	}

	{
		nop
		f32v2add $ap0, $a6:B, $ap0
	}

	{
		nop
		f32v2add $ap1, $a6:B, $ap1
	}

	{
		nop
		f32v2mul $ap0, $a7:B, $ap0
	}

	{
		nop
		f32v2mul $ap1, $a7:B, $ap1
	}

	{
		ld64step $ap2, $m15, $m5+=, 6
		f32v2tof16 $a0, $ap0
	}

	{
		ld64step $ap1, $m15, $m7+=, 6
		f32v2tof16 $a1, $ap1
	}

	{
		nop
		f16v4mul $ap0, $ap0, $ap2
	}

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

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


	#NO_APP
	ld32 $m4, $m11, $m15, 29                # 4-byte Folded Reload
	ld32 $a7, $m11, $m15, 18                # 4-byte Folded Reload
	ld32 $a6, $m11, $m15, 19                # 4-byte Folded Reload
	ld32 $a5, $m11, $m15, 6                 # 4-byte Folded Reload
	bri .LBB0_17
.LBB0_19:                               # %_ZN14LayerNormApply7computeEj.exit
	exitz $m15
.Lfunc_end0:
	.size	__runCodelet_LayerNormApply, .Lfunc_end0-__runCodelet_LayerNormApply
	.section	.stack_sizes,"o",@progbits,.text.__runCodelet_LayerNormApply
	.long	.Lfunc_begin0
	.ascii	"\200\001"
	.section	.text.__runCodelet_LayerNormApply,"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