.version 2.3
	.target sm_20
	.address_size 64
	// compiled with /usr/local/cuda/open64/lib//be
	// nvopencc 4.0 built on 2011-05-12

	//-----------------------------------------------------------
	// Compiling /tmp/tmpxft_0000bfe5_00000000-9_crml_gpu_kernel.cpp3.i (/home/sjplimp/ccBI#.xCE2Bc)
	//-----------------------------------------------------------

	//-----------------------------------------------------------
	// Options:
	//-----------------------------------------------------------
	//  Target:ptx, ISA:sm_20, Endian:little, Pointer Size:64
	//  -O3	(Optimization level)
	//  -g0	(Debug level)
	//  -m2	(Report advisories)
	//-----------------------------------------------------------

	.file	1	"<command-line>"
	.file	2	"/tmp/tmpxft_0000bfe5_00000000-8_crml_gpu_kernel.cudafe2.gpu"
	.file	3	"/usr/lib/gcc/x86_64-redhat-linux/4.4.5/include/stddef.h"
	.file	4	"/usr/local/cuda/include/crt/device_runtime.h"
	.file	5	"/usr/local/cuda/include/host_defines.h"
	.file	6	"/usr/local/cuda/include/builtin_types.h"
	.file	7	"/usr/local/cuda/include/device_types.h"
	.file	8	"/usr/local/cuda/include/driver_types.h"
	.file	9	"/usr/local/cuda/include/surface_types.h"
	.file	10	"/usr/local/cuda/include/texture_types.h"
	.file	11	"/usr/local/cuda/include/vector_types.h"
	.file	12	"/usr/local/cuda/include/device_launch_parameters.h"
	.file	13	"/usr/local/cuda/include/crt/storage_class.h"
	.file	14	"/usr/include/bits/types.h"
	.file	15	"/usr/include/time.h"
	.file	16	"crml_gpu_kernel.cu"
	.file	17	"/usr/local/cuda/include/common_functions.h"
	.file	18	"/usr/local/cuda/include/math_functions.h"
	.file	19	"/usr/local/cuda/include/math_constants.h"
	.file	20	"/usr/local/cuda/include/device_functions.h"
	.file	21	"/usr/local/cuda/include/sm_11_atomic_functions.h"
	.file	22	"/usr/local/cuda/include/sm_12_atomic_functions.h"
	.file	23	"/usr/local/cuda/include/sm_13_double_functions.h"
	.file	24	"/usr/local/cuda/include/sm_20_atomic_functions.h"
	.file	25	"/usr/local/cuda/include/sm_20_intrinsics.h"
	.file	26	"/usr/local/cuda/include/surface_functions.h"
	.file	27	"/usr/local/cuda/include/texture_fetch_functions.h"
	.file	28	"/usr/local/cuda/include/math_functions_dbl_ptx3.h"

	.global .texref pos_tex;
	.global .texref q_tex;

	.entry kernel_pair (
		.param .u64 __cudaparm_kernel_pair_x_,
		.param .u64 __cudaparm_kernel_pair_lj1,
		.param .s32 __cudaparm_kernel_pair_lj_types,
		.param .u64 __cudaparm_kernel_pair_sp_lj_in,
		.param .u64 __cudaparm_kernel_pair_dev_nbor,
		.param .u64 __cudaparm_kernel_pair_dev_packed,
		.param .u64 __cudaparm_kernel_pair_ans,
		.param .u64 __cudaparm_kernel_pair_engv,
		.param .s32 __cudaparm_kernel_pair_eflag,
		.param .s32 __cudaparm_kernel_pair_vflag,
		.param .s32 __cudaparm_kernel_pair_inum,
		.param .s32 __cudaparm_kernel_pair_nbor_pitch,
		.param .u64 __cudaparm_kernel_pair_q_,
		.param .f32 __cudaparm_kernel_pair_cut_coulsq,
		.param .f32 __cudaparm_kernel_pair_qqrd2e,
		.param .f32 __cudaparm_kernel_pair_g_ewald,
		.param .f32 __cudaparm_kernel_pair_denom_lj,
		.param .f32 __cudaparm_kernel_pair_cut_bothsq,
		.param .f32 __cudaparm_kernel_pair_cut_ljsq,
		.param .f32 __cudaparm_kernel_pair_cut_lj_innersq,
		.param .s32 __cudaparm_kernel_pair_t_per_atom)
	{
	.reg .u32 %r<91>;
	.reg .u64 %rd<64>;
	.reg .f32 %f<190>;
	.reg .pred %p<23>;
	.shared .align 16 .b8 __cuda___cuda_local_var_32500_33_non_const_sp_lj120[32];
	.shared .align 4 .b8 __cuda___cuda_local_var_32624_35_non_const_red_acc152[3072];
	// __cuda_local_var_32512_10_non_const_f = 64
	// __cuda_local_var_32516_9_non_const_virial = 16
	// __cuda_local_var_32564_43_non_const_r6inv = 40
	// __cuda_local_var_32564_50_non_const_prefactor = 52
	// __cuda_local_var_32564_61_non_const__erfc = 48
	// __cuda_local_var_32564_68_non_const_switch1 = 44
	.loc	16	110	0
$LDWbegin_kernel_pair:
	.loc	16	118	0
	ld.param.u64 	%rd1, [__cudaparm_kernel_pair_sp_lj_in];
	ldu.global.f32 	%f1, [%rd1+0];
	.loc	16	119	0
	ld.global.f32 	%f2, [%rd1+4];
	.loc	16	120	0
	ld.global.f32 	%f3, [%rd1+8];
	.loc	16	121	0
	ld.global.f32 	%f4, [%rd1+12];
	st.shared.v4.f32 	[__cuda___cuda_local_var_32500_33_non_const_sp_lj120+0], {%f1,%f2,%f3,%f4};
	.loc	16	122	0
	ld.global.f32 	%f5, [%rd1+16];
	.loc	16	123	0
	ld.global.f32 	%f6, [%rd1+20];
	.loc	16	124	0
	ld.global.f32 	%f7, [%rd1+24];
	.loc	16	125	0
	ld.global.f32 	%f8, [%rd1+28];
	st.shared.v4.f32 	[__cuda___cuda_local_var_32500_33_non_const_sp_lj120+16], {%f5,%f6,%f7,%f8};
	.loc	16	135	0
	mov.f32 	%f9, 0f00000000;     	// 0
	mov.f32 	%f10, %f9;
	mov.f32 	%f11, 0f00000000;    	// 0
	mov.f32 	%f12, %f11;
	mov.f32 	%f13, 0f00000000;    	// 0
	mov.f32 	%f14, %f13;
	mov.f32 	%f15, 0f00000000;    	// 0
	mov.f32 	%f16, %f15;
	mov.f32 	%f17, 0f00000000;    	// 0
	mov.f32 	%f18, %f17;
	mov.f32 	%f19, 0f00000000;    	// 0
	mov.f32 	%f20, %f19;
	ld.param.s32 	%r1, [__cudaparm_kernel_pair_t_per_atom];
	cvt.s32.u32 	%r2, %tid.x;
	div.s32 	%r3, %r2, %r1;
	cvt.s32.u32 	%r4, %ntid.x;
	div.s32 	%r5, %r4, %r1;
	rem.s32 	%r6, %r2, %r1;
	cvt.s32.u32 	%r7, %ctaid.x;
	mul.lo.s32 	%r8, %r7, %r5;
	add.s32 	%r9, %r3, %r8;
	ld.param.s32 	%r10, [__cudaparm_kernel_pair_inum];
	setp.lt.s32 	%p1, %r9, %r10;
	@!%p1 bra 	$Lt_0_23810;
	.loc	16	139	0
	cvt.s64.s32 	%rd2, %r9;
	mul.wide.s32 	%rd3, %r9, 4;
	ld.param.u64 	%rd4, [__cudaparm_kernel_pair_dev_nbor];
	add.u64 	%rd5, %rd3, %rd4;
	ld.global.s32 	%r11, [%rd5+0];
	.loc	16	141	0
	ld.param.s32 	%r12, [__cudaparm_kernel_pair_nbor_pitch];
	cvt.s64.s32 	%rd6, %r12;
	mul.wide.s32 	%rd7, %r12, 4;
	add.u64 	%rd8, %rd7, %rd5;
	ld.global.s32 	%r13, [%rd8+0];
	add.u64 	%rd9, %rd7, %rd8;
	ld.param.u64 	%rd10, [__cudaparm_kernel_pair_dev_packed];
	setp.ne.u64 	%p2, %rd10, %rd4;
	@%p2 bra 	$Lt_0_24322;
	.loc	16	147	0
	cvt.s32.s64 	%r14, %rd6;
	mul.lo.s32 	%r15, %r14, %r13;
	cvt.s64.s32 	%rd11, %r15;
	mul.wide.s32 	%rd12, %r15, 4;
	add.u64 	%rd13, %rd9, %rd12;
	.loc	16	148	0
	mul.lo.s32 	%r16, %r6, %r14;
	cvt.s64.s32 	%rd14, %r16;
	mul.wide.s32 	%rd15, %r16, 4;
	add.u64 	%rd16, %rd9, %rd15;
	.loc	16	149	0
	mul.lo.s32 	%r17, %r14, %r1;
	bra.uni 	$Lt_0_24066;
$Lt_0_24322:
	.loc	16	151	0
	ld.global.s32 	%r18, [%rd9+0];
	cvt.s64.s32 	%rd17, %r18;
	mul.wide.s32 	%rd18, %r18, 4;
	add.u64 	%rd19, %rd10, %rd18;
	.loc	16	152	0
	cvt.s64.s32 	%rd20, %r13;
	mul.wide.s32 	%rd21, %r13, 4;
	add.u64 	%rd13, %rd19, %rd21;
	.loc	16	153	0
	mov.s32 	%r17, %r1;
	.loc	16	154	0
	cvt.s64.s32 	%rd22, %r6;
	mul.wide.s32 	%rd23, %r6, 4;
	add.u64 	%rd16, %rd19, %rd23;
$Lt_0_24066:
	.loc	16	157	0
	mov.u32 	%r19, %r11;
	mov.s32 	%r20, 0;
	mov.u32 	%r21, %r20;
	mov.s32 	%r22, 0;
	mov.u32 	%r23, %r22;
	mov.s32 	%r24, 0;
	mov.u32 	%r25, %r24;
	tex.1d.v4.f32.s32 {%f21,%f22,%f23,%f24},[pos_tex,{%r19,%r21,%r23,%r25}];
	mov.f32 	%f25, %f21;
	mov.f32 	%f26, %f22;
	mov.f32 	%f27, %f23;
	mov.f32 	%f28, %f24;
	.loc	16	158	0
	mov.u32 	%r26, %r11;
	mov.s32 	%r27, 0;
	mov.u32 	%r28, %r27;
	mov.s32 	%r29, 0;
	mov.u32 	%r30, %r29;
	mov.s32 	%r31, 0;
	mov.u32 	%r32, %r31;
	tex.1d.v4.f32.s32 {%f29,%f30,%f31,%f32},[q_tex,{%r26,%r28,%r30,%r32}];
	mov.f32 	%f33, %f29;
	setp.ge.u64 	%p3, %rd16, %rd13;
	@%p3 bra 	$Lt_0_34818;
	cvt.s64.s32 	%rd24, %r17;
	ld.param.f32 	%f34, [__cudaparm_kernel_pair_cut_bothsq];
	mov.f32 	%f35, 0f00000000;    	// 0
	mov.f32 	%f36, 0f00000000;    	// 0
	mov.f32 	%f37, 0f00000000;    	// 0
	mov.f32 	%f38, 0f00000000;    	// 0
	mov.f32 	%f39, 0f00000000;    	// 0
	mov.u64 	%rd25, __cuda___cuda_local_var_32500_33_non_const_sp_lj120;
$Lt_0_25090:
 //<loop> Loop body line 158, nesting depth: 1, estimated iterations: unknown
	.loc	16	162	0
	ld.global.s32 	%r33, [%rd16+0];
	.loc	16	165	0
	shr.s32 	%r34, %r33, 30;
	and.b32 	%r35, %r34, 3;
	cvt.s64.s32 	%rd26, %r35;
	mul.wide.s32 	%rd27, %r35, 4;
	add.u64 	%rd28, %rd25, %rd27;
	ld.shared.f32 	%f40, [%rd28+0];
	.loc	16	166	0
	mov.f32 	%f41, 0f3f800000;    	// 1
	ld.shared.f32 	%f42, [%rd28+16];
	sub.ftz.f32 	%f43, %f41, %f42;
	.loc	16	169	0
	and.b32 	%r36, %r33, 1073741823;
	mov.u32 	%r37, %r36;
	mov.s32 	%r38, 0;
	mov.u32 	%r39, %r38;
	mov.s32 	%r40, 0;
	mov.u32 	%r41, %r40;
	mov.s32 	%r42, 0;
	mov.u32 	%r43, %r42;
	tex.1d.v4.f32.s32 {%f44,%f45,%f46,%f47},[pos_tex,{%r37,%r39,%r41,%r43}];
	mov.f32 	%f48, %f44;
	mov.f32 	%f49, %f45;
	mov.f32 	%f50, %f46;
	mov.f32 	%f51, %f47;
	sub.ftz.f32 	%f52, %f26, %f49;
	sub.ftz.f32 	%f53, %f25, %f48;
	sub.ftz.f32 	%f54, %f27, %f50;
	mul.ftz.f32 	%f55, %f52, %f52;
	fma.rn.ftz.f32 	%f56, %f53, %f53, %f55;
	fma.rn.ftz.f32 	%f57, %f54, %f54, %f56;
	setp.lt.ftz.f32 	%p4, %f57, %f34;
	@!%p4 bra 	$Lt_0_28418;
	ld.param.f32 	%f58, [__cudaparm_kernel_pair_cut_ljsq];
	setp.lt.ftz.f32 	%p5, %f57, %f58;
	rcp.approx.ftz.f32 	%f59, %f57;
	@!%p5 bra 	$Lt_0_26114;
	.loc	16	184	0
	mul.ftz.f32 	%f60, %f59, %f59;
	mul.ftz.f32 	%f61, %f59, %f60;
	mov.f32 	%f62, %f61;
	.loc	16	185	0
	cvt.rzi.ftz.s32.f32 	%r44, %f51;
	cvt.rzi.ftz.s32.f32 	%r45, %f28;
	ld.param.u64 	%rd29, [__cudaparm_kernel_pair_lj1];
	ld.param.s32 	%r46, [__cudaparm_kernel_pair_lj_types];
	mul.lo.s32 	%r47, %r46, %r45;
	add.s32 	%r48, %r44, %r47;
	cvt.s64.s32 	%rd30, %r48;
	mul.wide.s32 	%rd31, %r48, 16;
	add.u64 	%rd32, %rd29, %rd31;
	mul.ftz.f32 	%f63, %f61, %f40;
	ld.global.v2.f32 	{%f64,%f65}, [%rd32+0];
	mul.ftz.f32 	%f66, %f64, %f61;
	sub.ftz.f32 	%f67, %f66, %f65;
	mul.ftz.f32 	%f68, %f63, %f67;
	ld.param.f32 	%f69, [__cudaparm_kernel_pair_cut_lj_innersq];
	setp.gt.ftz.f32 	%p6, %f57, %f69;
	@!%p6 bra 	$Lt_0_25858;
	.loc	16	191	0
	add.ftz.f32 	%f70, %f57, %f57;
	sub.ftz.f32 	%f71, %f58, %f57;
	add.ftz.f32 	%f72, %f70, %f58;
	mul.ftz.f32 	%f73, %f71, %f71;
	mov.f32 	%f74, 0f40400000;    	// 3
	mul.ftz.f32 	%f75, %f74, %f69;
	sub.ftz.f32 	%f76, %f72, %f75;
	ld.param.f32 	%f77, [__cudaparm_kernel_pair_denom_lj];
	div.approx.ftz.f32 	%f78, %f76, %f77;
	mul.ftz.f32 	%f79, %f73, %f78;
	mov.f32 	%f80, %f79;
	.loc	16	194	0
	mov.f32 	%f81, 0f41400000;    	// 12
	mul.ftz.f32 	%f82, %f57, %f81;
	mul.ftz.f32 	%f83, %f71, %f82;
	sub.ftz.f32 	%f84, %f57, %f69;
	mul.ftz.f32 	%f85, %f83, %f84;
	div.approx.ftz.f32 	%f86, %f85, %f77;
	ld.global.v2.f32 	{%f87,%f88}, [%rd32+8];
	mul.ftz.f32 	%f89, %f87, %f61;
	sub.ftz.f32 	%f90, %f89, %f88;
	mul.ftz.f32 	%f91, %f61, %f90;
	mul.ftz.f32 	%f92, %f86, %f91;
	fma.rn.ftz.f32 	%f68, %f68, %f79, %f92;
	bra.uni 	$Lt_0_25858;
$Lt_0_26114:
	.loc	16	197	0
	mov.f32 	%f68, 0f00000000;    	// 0
$Lt_0_25858:
	ld.param.f32 	%f93, [__cudaparm_kernel_pair_cut_coulsq];
	setp.gt.ftz.f32 	%p7, %f93, %f57;
	@!%p7 bra 	$Lt_0_27138;
	.loc	16	204	0
	sqrt.approx.ftz.f32 	%f94, %f57;
	ld.param.f32 	%f95, [__cudaparm_kernel_pair_g_ewald];
	mul.ftz.f32 	%f96, %f95, %f94;
	mul.ftz.f32 	%f97, %f96, %f96;
	mov.f32 	%f98, 0f3f800000;    	// 1
	mov.f32 	%f99, 0f3ea7ba05;    	// 0.327591
	fma.rn.ftz.f32 	%f100, %f99, %f96, %f98;
	neg.ftz.f32 	%f101, %f97;
	rcp.approx.ftz.f32 	%f102, %f100;
	mov.f32 	%f103, 0f3fb8aa3b;   	// 1.4427
	mul.ftz.f32 	%f104, %f101, %f103;
	ex2.approx.ftz.f32 	%f105, %f104;
	mov.f32 	%f106, 0f3e827906;   	// 0.25483
	mov.f32 	%f107, 0fbe91a98e;   	// -0.284497
	mov.f32 	%f108, 0f3fb5f0e3;   	// 1.42141
	mov.f32 	%f109, 0fbfba00e3;   	// -1.45315
	mov.f32 	%f110, 0f3f87dc22;   	// 1.06141
	fma.rn.ftz.f32 	%f111, %f110, %f102, %f109;
	fma.rn.ftz.f32 	%f112, %f102, %f111, %f108;
	fma.rn.ftz.f32 	%f113, %f102, %f112, %f107;
	fma.rn.ftz.f32 	%f114, %f102, %f113, %f106;
	mul.ftz.f32 	%f115, %f102, %f114;
	mul.ftz.f32 	%f116, %f105, %f115;
	mov.f32 	%f117, %f116;
	.loc	16	205	0
	mov.u32 	%r49, %r36;
	mov.s32 	%r50, 0;
	mov.u32 	%r51, %r50;
	mov.s32 	%r52, 0;
	mov.u32 	%r53, %r52;
	mov.s32 	%r54, 0;
	mov.u32 	%r55, %r54;
	tex.1d.v4.f32.s32 {%f118,%f119,%f120,%f121},[q_tex,{%r49,%r51,%r53,%r55}];
	mov.f32 	%f122, %f118;
	ld.param.f32 	%f123, [__cudaparm_kernel_pair_qqrd2e];
	mul.ftz.f32 	%f124, %f123, %f33;
	mul.ftz.f32 	%f125, %f124, %f122;
	div.approx.ftz.f32 	%f126, %f125, %f94;
	mov.f32 	%f127, %f126;
	.loc	16	206	0
	mov.f32 	%f128, 0f3f906ebb;   	// 1.12838
	mul.ftz.f32 	%f129, %f96, %f128;
	fma.rn.ftz.f32 	%f130, %f105, %f129, %f116;
	sub.ftz.f32 	%f131, %f130, %f43;
	mul.ftz.f32 	%f132, %f126, %f131;
	bra.uni 	$Lt_0_26882;
$Lt_0_27138:
	.loc	16	208	0
	mov.f32 	%f132, 0f00000000;   	// 0
$Lt_0_26882:
	.loc	16	212	0
	add.ftz.f32 	%f133, %f132, %f68;
	mul.ftz.f32 	%f134, %f133, %f59;
	fma.rn.ftz.f32 	%f37, %f53, %f134, %f37;
	.loc	16	213	0
	fma.rn.ftz.f32 	%f36, %f52, %f134, %f36;
	.loc	16	214	0
	fma.rn.ftz.f32 	%f35, %f54, %f134, %f35;
	ld.param.s32 	%r56, [__cudaparm_kernel_pair_eflag];
	mov.u32 	%r57, 0;
	setp.le.s32 	%p8, %r56, %r57;
	@%p8 bra 	$Lt_0_27906;
	.loc	16	217	0
	mov.f32 	%f135, %f127;
	mov.f32 	%f136, %f117;
	sub.ftz.f32 	%f137, %f136, %f43;
	fma.rn.ftz.f32 	%f138, %f135, %f137, %f38;
	selp.f32 	%f38, %f138, %f38, %p7;
	@!%p5 bra 	$Lt_0_27906;
	.loc	16	220	0
	cvt.rzi.ftz.s32.f32 	%r58, %f51;
	cvt.rzi.ftz.s32.f32 	%r59, %f28;
	ld.param.u64 	%rd33, [__cudaparm_kernel_pair_lj1];
	ld.param.s32 	%r60, [__cudaparm_kernel_pair_lj_types];
	mul.lo.s32 	%r61, %r60, %r59;
	add.s32 	%r62, %r58, %r61;
	cvt.s64.s32 	%rd34, %r62;
	mul.wide.s32 	%rd35, %r62, 16;
	add.u64 	%rd32, %rd33, %rd35;
	mov.f32 	%f139, %f62;
	ld.global.v2.f32 	{%f140,%f141}, [%rd32+8];
	mul.ftz.f32 	%f142, %f140, %f139;
	sub.ftz.f32 	%f143, %f142, %f141;
	mul.ftz.f32 	%f144, %f139, %f143;
	mov.f32 	%f145, %f80;
	mul.ftz.f32 	%f146, %f145, %f144;
	ld.param.f32 	%f147, [__cudaparm_kernel_pair_cut_lj_innersq];
	setp.lt.ftz.f32 	%p9, %f147, %f57;
	selp.f32 	%f148, %f146, %f144, %p9;
	.loc	16	223	0
	fma.rn.ftz.f32 	%f39, %f40, %f148, %f39;
$Lt_0_27906:
$Lt_0_27394:
	ld.param.s32 	%r63, [__cudaparm_kernel_pair_vflag];
	mov.u32 	%r64, 0;
	setp.le.s32 	%p10, %r63, %r64;
	@%p10 bra 	$Lt_0_28418;
	.loc	16	227	0
	mov.f32 	%f149, %f10;
	mul.ftz.f32 	%f150, %f53, %f53;
	fma.rn.ftz.f32 	%f151, %f134, %f150, %f149;
	mov.f32 	%f10, %f151;
	.loc	16	228	0
	mov.f32 	%f152, %f12;
	fma.rn.ftz.f32 	%f153, %f134, %f55, %f152;
	mov.f32 	%f12, %f153;
	.loc	16	229	0
	mov.f32 	%f154, %f14;
	mul.ftz.f32 	%f155, %f54, %f54;
	fma.rn.ftz.f32 	%f156, %f134, %f155, %f154;
	mov.f32 	%f14, %f156;
	.loc	16	230	0
	mov.f32 	%f157, %f16;
	mul.ftz.f32 	%f158, %f52, %f53;
	fma.rn.ftz.f32 	%f159, %f134, %f158, %f157;
	mov.f32 	%f16, %f159;
	.loc	16	231	0
	mov.f32 	%f160, %f18;
	mul.ftz.f32 	%f161, %f53, %f54;
	fma.rn.ftz.f32 	%f162, %f134, %f161, %f160;
	mov.f32 	%f18, %f162;
	.loc	16	232	0
	mul.ftz.f32 	%f163, %f52, %f54;
	fma.rn.ftz.f32 	%f19, %f134, %f163, %f19;
	mov.f32 	%f20, %f19;
$Lt_0_28418:
$Lt_0_25346:
	.loc	16	161	0
	mul.lo.u64 	%rd36, %rd24, 4;
	add.u64 	%rd16, %rd16, %rd36;
	setp.lt.u64 	%p11, %rd16, %rd13;
	@%p11 bra 	$Lt_0_25090;
	bra.uni 	$Lt_0_23554;
$Lt_0_34818:
	mov.f32 	%f35, 0f00000000;    	// 0
	mov.f32 	%f36, 0f00000000;    	// 0
	mov.f32 	%f37, 0f00000000;    	// 0
	mov.f32 	%f38, 0f00000000;    	// 0
	mov.f32 	%f39, 0f00000000;    	// 0
	bra.uni 	$Lt_0_23554;
$Lt_0_23810:
	mov.f32 	%f35, 0f00000000;    	// 0
	mov.f32 	%f36, 0f00000000;    	// 0
	mov.f32 	%f37, 0f00000000;    	// 0
	mov.f32 	%f38, 0f00000000;    	// 0
	mov.f32 	%f39, 0f00000000;    	// 0
$Lt_0_23554:
	mov.u32 	%r65, 1;
	setp.le.s32 	%p12, %r1, %r65;
	@%p12 bra 	$Lt_0_31234;
	.loc	16	243	0
	mov.u64 	%rd37, __cuda___cuda_local_var_32624_35_non_const_red_acc152;
	cvt.s64.s32 	%rd38, %r2;
	mul.wide.s32 	%rd39, %r2, 4;
	add.u64 	%rd40, %rd37, %rd39;
	mov.f32 	%f164, %f37;
	st.shared.f32 	[%rd40+0], %f164;
	.loc	16	244	0
	mov.f32 	%f165, %f36;
	st.shared.f32 	[%rd40+512], %f165;
	.loc	16	245	0
	mov.f32 	%f166, %f35;
	st.shared.f32 	[%rd40+1024], %f166;
	.loc	16	246	0
	mov.f32 	%f167, %f39;
	st.shared.f32 	[%rd40+1536], %f167;
	.loc	16	247	0
	mov.f32 	%f168, %f38;
	st.shared.f32 	[%rd40+2048], %f168;
	.loc	16	249	0
	shr.s32 	%r66, %r1, 31;
	mov.s32 	%r67, 1;
	and.b32 	%r68, %r66, %r67;
	add.s32 	%r69, %r68, %r1;
	shr.s32 	%r70, %r69, 1;
	mov.s32 	%r71, %r70;
	mov.u32 	%r72, 0;
	setp.ne.u32 	%p13, %r70, %r72;
	@!%p13 bra 	$Lt_0_29698;
$Lt_0_30210:
	setp.ge.u32 	%p14, %r6, %r71;
	@%p14 bra 	$Lt_0_30466;
	.loc	16	252	0
	add.u32 	%r73, %r2, %r71;
	cvt.u64.u32 	%rd41, %r73;
	mul.wide.u32 	%rd42, %r73, 4;
	add.u64 	%rd43, %rd37, %rd42;
	ld.shared.f32 	%f169, [%rd43+0];
	add.ftz.f32 	%f164, %f169, %f164;
	st.shared.f32 	[%rd40+0], %f164;
	ld.shared.f32 	%f170, [%rd43+512];
	add.ftz.f32 	%f165, %f170, %f165;
	st.shared.f32 	[%rd40+512], %f165;
	ld.shared.f32 	%f171, [%rd43+1024];
	add.ftz.f32 	%f166, %f171, %f166;
	st.shared.f32 	[%rd40+1024], %f166;
	ld.shared.f32 	%f172, [%rd43+1536];
	add.ftz.f32 	%f167, %f172, %f167;
	st.shared.f32 	[%rd40+1536], %f167;
	ld.shared.f32 	%f173, [%rd43+2048];
	add.ftz.f32 	%f168, %f173, %f168;
	st.shared.f32 	[%rd40+2048], %f168;
$Lt_0_30466:
	.loc	16	249	0
	shr.u32 	%r71, %r71, 1;
	mov.u32 	%r74, 0;
	setp.ne.u32 	%p15, %r71, %r74;
	@%p15 bra 	$Lt_0_30210;
$Lt_0_29698:
	.loc	16	256	0
	mov.f32 	%f37, %f164;
	.loc	16	257	0
	mov.f32 	%f36, %f165;
	.loc	16	258	0
	mov.f32 	%f35, %f166;
	.loc	16	259	0
	mov.f32 	%f39, %f167;
	.loc	16	260	0
	mov.f32 	%f38, %f168;
	ld.param.s32 	%r75, [__cudaparm_kernel_pair_vflag];
	mov.u32 	%r76, 0;
	setp.le.s32 	%p16, %r75, %r76;
	@%p16 bra 	$Lt_0_31234;
	.loc	16	264	0
	mov.f32 	%f164, %f10;
	st.shared.f32 	[%rd40+0], %f164;
	mov.f32 	%f165, %f12;
	st.shared.f32 	[%rd40+512], %f165;
	mov.f32 	%f166, %f14;
	st.shared.f32 	[%rd40+1024], %f166;
	mov.f32 	%f167, %f16;
	st.shared.f32 	[%rd40+1536], %f167;
	mov.f32 	%f168, %f18;
	st.shared.f32 	[%rd40+2048], %f168;
	mov.f32 	%f174, %f20;
	st.shared.f32 	[%rd40+2560], %f174;
	.loc	16	266	0
	mov.s32 	%r77, %r70;
	@!%p13 bra 	$Lt_0_31746;
$Lt_0_32258:
	setp.ge.u32 	%p17, %r6, %r77;
	@%p17 bra 	$Lt_0_32514;
	.loc	16	269	0
	add.u32 	%r78, %r2, %r77;
	cvt.u64.u32 	%rd44, %r78;
	mul.wide.u32 	%rd45, %r78, 4;
	add.u64 	%rd46, %rd37, %rd45;
	ld.shared.f32 	%f175, [%rd46+0];
	add.ftz.f32 	%f164, %f175, %f164;
	st.shared.f32 	[%rd40+0], %f164;
	ld.shared.f32 	%f176, [%rd46+512];
	add.ftz.f32 	%f165, %f176, %f165;
	st.shared.f32 	[%rd40+512], %f165;
	ld.shared.f32 	%f177, [%rd46+1024];
	add.ftz.f32 	%f166, %f177, %f166;
	st.shared.f32 	[%rd40+1024], %f166;
	ld.shared.f32 	%f178, [%rd46+1536];
	add.ftz.f32 	%f167, %f178, %f167;
	st.shared.f32 	[%rd40+1536], %f167;
	ld.shared.f32 	%f179, [%rd46+2048];
	add.ftz.f32 	%f168, %f179, %f168;
	st.shared.f32 	[%rd40+2048], %f168;
	ld.shared.f32 	%f180, [%rd46+2560];
	add.ftz.f32 	%f174, %f180, %f174;
	st.shared.f32 	[%rd40+2560], %f174;
$Lt_0_32514:
	.loc	16	266	0
	shr.u32 	%r77, %r77, 1;
	mov.u32 	%r79, 0;
	setp.ne.u32 	%p18, %r77, %r79;
	@%p18 bra 	$Lt_0_32258;
$Lt_0_31746:
	.loc	16	274	0
	mov.f32 	%f10, %f164;
	mov.f32 	%f12, %f165;
	mov.f32 	%f14, %f166;
	mov.f32 	%f16, %f167;
	mov.f32 	%f18, %f168;
	mov.f32 	%f20, %f174;
$Lt_0_31234:
$Lt_0_29186:
	selp.s32 	%r80, 1, 0, %p1;
	mov.s32 	%r81, 0;
	set.eq.u32.s32 	%r82, %r6, %r81;
	neg.s32 	%r83, %r82;
	and.b32 	%r84, %r80, %r83;
	mov.u32 	%r85, 0;
	setp.eq.s32 	%p19, %r84, %r85;
	@%p19 bra 	$Lt_0_33282;
	.loc	16	280	0
	cvt.s64.s32 	%rd47, %r9;
	ld.param.u64 	%rd48, [__cudaparm_kernel_pair_engv];
	mul.wide.s32 	%rd49, %r9, 4;
	add.u64 	%rd50, %rd48, %rd49;
	ld.param.s32 	%r86, [__cudaparm_kernel_pair_eflag];
	mov.u32 	%r87, 0;
	setp.le.s32 	%p20, %r86, %r87;
	@%p20 bra 	$Lt_0_33794;
	.loc	16	282	0
	st.global.f32 	[%rd50+0], %f39;
	.loc	16	283	0
	cvt.s64.s32 	%rd51, %r10;
	mul.wide.s32 	%rd52, %r10, 4;
	add.u64 	%rd53, %rd52, %rd50;
	.loc	16	284	0
	st.global.f32 	[%rd53+0], %f38;
	.loc	16	285	0
	add.u64 	%rd50, %rd52, %rd53;
$Lt_0_33794:
	ld.param.s32 	%r88, [__cudaparm_kernel_pair_vflag];
	mov.u32 	%r89, 0;
	setp.le.s32 	%p21, %r88, %r89;
	@%p21 bra 	$Lt_0_34306;
	.loc	16	289	0
	mov.f32 	%f181, %f10;
	st.global.f32 	[%rd50+0], %f181;
	.loc	16	290	0
	cvt.s64.s32 	%rd54, %r10;
	mul.wide.s32 	%rd55, %r10, 4;
	add.u64 	%rd56, %rd55, %rd50;
	.loc	16	289	0
	mov.f32 	%f182, %f12;
	st.global.f32 	[%rd56+0], %f182;
	.loc	16	290	0
	add.u64 	%rd57, %rd55, %rd56;
	.loc	16	289	0
	mov.f32 	%f183, %f14;
	st.global.f32 	[%rd57+0], %f183;
	.loc	16	290	0
	add.u64 	%rd58, %rd55, %rd57;
	.loc	16	289	0
	mov.f32 	%f184, %f16;
	st.global.f32 	[%rd58+0], %f184;
	.loc	16	290	0
	add.u64 	%rd50, %rd55, %rd58;
	.loc	16	289	0
	mov.f32 	%f185, %f18;
	st.global.f32 	[%rd50+0], %f185;
	mov.f32 	%f186, %f20;
	add.u64 	%rd59, %rd55, %rd50;
	st.global.f32 	[%rd59+0], %f186;
$Lt_0_34306:
	.loc	16	293	0
	ld.param.u64 	%rd60, [__cudaparm_kernel_pair_ans];
	mul.lo.u64 	%rd61, %rd47, 16;
	add.u64 	%rd62, %rd60, %rd61;
	mov.f32 	%f187, %f188;
	st.global.v4.f32 	[%rd62+0], {%f37,%f36,%f35,%f187};
$Lt_0_33282:
	.loc	16	295	0
	exit;
$LDWend_kernel_pair:
	} // kernel_pair

	.entry kernel_pair_fast (
		.param .u64 __cudaparm_kernel_pair_fast_x_,
		.param .u64 __cudaparm_kernel_pair_fast_ljd_in,
		.param .u64 __cudaparm_kernel_pair_fast_sp_lj_in,
		.param .u64 __cudaparm_kernel_pair_fast_dev_nbor,
		.param .u64 __cudaparm_kernel_pair_fast_dev_packed,
		.param .u64 __cudaparm_kernel_pair_fast_ans,
		.param .u64 __cudaparm_kernel_pair_fast_engv,
		.param .s32 __cudaparm_kernel_pair_fast_eflag,
		.param .s32 __cudaparm_kernel_pair_fast_vflag,
		.param .s32 __cudaparm_kernel_pair_fast_inum,
		.param .s32 __cudaparm_kernel_pair_fast_nbor_pitch,
		.param .u64 __cudaparm_kernel_pair_fast_q_,
		.param .f32 __cudaparm_kernel_pair_fast_cut_coulsq,
		.param .f32 __cudaparm_kernel_pair_fast_qqrd2e,
		.param .f32 __cudaparm_kernel_pair_fast_g_ewald,
		.param .f32 __cudaparm_kernel_pair_fast_denom_lj,
		.param .f32 __cudaparm_kernel_pair_fast_cut_bothsq,
		.param .f32 __cudaparm_kernel_pair_fast_cut_ljsq,
		.param .f32 __cudaparm_kernel_pair_fast_cut_lj_innersq,
		.param .s32 __cudaparm_kernel_pair_fast_t_per_atom)
	{
	.reg .u32 %r<86>;
	.reg .u64 %rd<71>;
	.reg .f32 %f<196>;
	.reg .pred %p<25>;
	.shared .align 4 .b8 __cuda___cuda_local_var_32697_33_non_const_sp_lj3336[32];
	.shared .align 8 .b8 __cuda___cuda_local_var_32696_34_non_const_ljd3368[1024];
	.shared .align 4 .b8 __cuda___cuda_local_var_32826_35_non_const_red_acc4392[3072];
	// __cuda_local_var_32706_10_non_const_f = 64
	// __cuda_local_var_32710_9_non_const_virial = 16
	// __cuda_local_var_32759_43_non_const_prefactor = 56
	// __cuda_local_var_32759_54_non_const__erfc = 52
	// __cuda_local_var_32759_61_non_const_switch1 = 48
	// __cuda_local_var_32760_15_non_const_lj3 = 44
	// __cuda_local_var_32760_20_non_const_lj4 = 40
	.loc	16	307	0
$LDWbegin_kernel_pair_fast:
	cvt.s32.u32 	%r1, %tid.x;
	cvt.s64.s32 	%rd1, %r1;
	mov.u32 	%r2, 7;
	setp.gt.s32 	%p1, %r1, %r2;
	@%p1 bra 	$Lt_1_25090;
	.loc	16	316	0
	mov.u64 	%rd2, __cuda___cuda_local_var_32697_33_non_const_sp_lj3336;
	mul.lo.u64 	%rd3, %rd1, 4;
	ld.param.u64 	%rd4, [__cudaparm_kernel_pair_fast_sp_lj_in];
	add.u64 	%rd5, %rd4, %rd3;
	ld.global.f32 	%f1, [%rd5+0];
	add.u64 	%rd6, %rd3, %rd2;
	st.shared.f32 	[%rd6+0], %f1;
$Lt_1_25090:
	mov.u64 	%rd2, __cuda___cuda_local_var_32697_33_non_const_sp_lj3336;
	mov.u64 	%rd7, __cuda___cuda_local_var_32696_34_non_const_ljd3368;
	.loc	16	317	0
	mul.lo.u64 	%rd8, %rd1, 8;
	ld.param.u64 	%rd9, [__cudaparm_kernel_pair_fast_ljd_in];
	add.u64 	%rd10, %rd9, %rd8;
	add.u64 	%rd11, %rd8, %rd7;
	ld.global.v2.f32 	{%f2,%f3}, [%rd10+0];
	st.shared.v2.f32 	[%rd11+0], {%f2,%f3};
	add.s32 	%r3, %r1, 128;
	mov.u32 	%r4, 127;
	setp.gt.s32 	%p2, %r3, %r4;
	@%p2 bra 	$Lt_1_25602;
	ld.global.v2.f32 	{%f4,%f5}, [%rd10+1024];
	st.shared.v2.f32 	[%rd11+1024], {%f4,%f5};
$Lt_1_25602:
	.loc	16	329	0
	mov.f32 	%f6, 0f00000000;     	// 0
	mov.f32 	%f7, %f6;
	mov.f32 	%f8, 0f00000000;     	// 0
	mov.f32 	%f9, %f8;
	mov.f32 	%f10, 0f00000000;    	// 0
	mov.f32 	%f11, %f10;
	mov.f32 	%f12, 0f00000000;    	// 0
	mov.f32 	%f13, %f12;
	mov.f32 	%f14, 0f00000000;    	// 0
	mov.f32 	%f15, %f14;
	mov.f32 	%f16, 0f00000000;    	// 0
	mov.f32 	%f17, %f16;
	.loc	16	331	0
	bar.sync 	0;
	ld.param.s32 	%r5, [__cudaparm_kernel_pair_fast_t_per_atom];
	div.s32 	%r6, %r1, %r5;
	cvt.s32.u32 	%r7, %ntid.x;
	div.s32 	%r8, %r7, %r5;
	rem.s32 	%r9, %r1, %r5;
	cvt.s32.u32 	%r10, %ctaid.x;
	mul.lo.s32 	%r11, %r10, %r8;
	add.s32 	%r12, %r6, %r11;
	ld.param.s32 	%r13, [__cudaparm_kernel_pair_fast_inum];
	setp.lt.s32 	%p3, %r12, %r13;
	@!%p3 bra 	$Lt_1_26370;
	.loc	16	335	0
	cvt.s64.s32 	%rd12, %r12;
	mul.wide.s32 	%rd13, %r12, 4;
	ld.param.u64 	%rd14, [__cudaparm_kernel_pair_fast_dev_nbor];
	add.u64 	%rd15, %rd13, %rd14;
	ld.global.s32 	%r14, [%rd15+0];
	.loc	16	337	0
	ld.param.s32 	%r15, [__cudaparm_kernel_pair_fast_nbor_pitch];
	cvt.s64.s32 	%rd16, %r15;
	mul.wide.s32 	%rd17, %r15, 4;
	add.u64 	%rd18, %rd17, %rd15;
	ld.global.s32 	%r16, [%rd18+0];
	add.u64 	%rd19, %rd17, %rd18;
	ld.param.u64 	%rd20, [__cudaparm_kernel_pair_fast_dev_packed];
	setp.ne.u64 	%p4, %rd20, %rd14;
	@%p4 bra 	$Lt_1_26882;
	.loc	16	343	0
	cvt.s32.s64 	%r17, %rd16;
	mul.lo.s32 	%r18, %r17, %r16;
	cvt.s64.s32 	%rd21, %r18;
	mul.wide.s32 	%rd22, %r18, 4;
	add.u64 	%rd23, %rd19, %rd22;
	.loc	16	344	0
	mul.lo.s32 	%r19, %r9, %r17;
	cvt.s64.s32 	%rd24, %r19;
	mul.wide.s32 	%rd25, %r19, 4;
	add.u64 	%rd26, %rd19, %rd25;
	.loc	16	345	0
	mul.lo.s32 	%r20, %r17, %r5;
	bra.uni 	$Lt_1_26626;
$Lt_1_26882:
	.loc	16	347	0
	ld.global.s32 	%r21, [%rd19+0];
	cvt.s64.s32 	%rd27, %r21;
	mul.wide.s32 	%rd28, %r21, 4;
	add.u64 	%rd29, %rd20, %rd28;
	.loc	16	348	0
	cvt.s64.s32 	%rd30, %r16;
	mul.wide.s32 	%rd31, %r16, 4;
	add.u64 	%rd23, %rd29, %rd31;
	.loc	16	349	0
	mov.s32 	%r20, %r5;
	.loc	16	350	0
	cvt.s64.s32 	%rd32, %r9;
	mul.wide.s32 	%rd33, %r9, 4;
	add.u64 	%rd26, %rd29, %rd33;
$Lt_1_26626:
	.loc	16	353	0
	mov.u32 	%r22, %r14;
	mov.s32 	%r23, 0;
	mov.u32 	%r24, %r23;
	mov.s32 	%r25, 0;
	mov.u32 	%r26, %r25;
	mov.s32 	%r27, 0;
	mov.u32 	%r28, %r27;
	tex.1d.v4.f32.s32 {%f18,%f19,%f20,%f21},[pos_tex,{%r22,%r24,%r26,%r28}];
	mov.f32 	%f22, %f18;
	mov.f32 	%f23, %f19;
	mov.f32 	%f24, %f20;
	mov.f32 	%f25, %f21;
	.loc	16	354	0
	mov.u32 	%r29, %r14;
	mov.s32 	%r30, 0;
	mov.u32 	%r31, %r30;
	mov.s32 	%r32, 0;
	mov.u32 	%r33, %r32;
	mov.s32 	%r34, 0;
	mov.u32 	%r35, %r34;
	tex.1d.v4.f32.s32 {%f26,%f27,%f28,%f29},[q_tex,{%r29,%r31,%r33,%r35}];
	mov.f32 	%f30, %f26;
	setp.ge.u64 	%p5, %rd26, %rd23;
	@%p5 bra 	$Lt_1_37378;
	cvt.rzi.ftz.s32.f32 	%r36, %f25;
	cvt.s64.s32 	%rd34, %r20;
	ld.param.f32 	%f31, [__cudaparm_kernel_pair_fast_cut_bothsq];
	mov.f32 	%f32, 0f00000000;    	// 0
	mov.f32 	%f33, 0f00000000;    	// 0
	mov.f32 	%f34, 0f00000000;    	// 0
	mov.f32 	%f35, 0f00000000;    	// 0
	mov.f32 	%f36, 0f00000000;    	// 0
$Lt_1_27650:
 //<loop> Loop body line 354, nesting depth: 1, estimated iterations: unknown
	.loc	16	358	0
	ld.global.s32 	%r37, [%rd26+0];
	.loc	16	361	0
	shr.s32 	%r38, %r37, 30;
	and.b32 	%r39, %r38, 3;
	cvt.s64.s32 	%rd35, %r39;
	mul.wide.s32 	%rd36, %r39, 4;
	add.u64 	%rd37, %rd2, %rd36;
	ld.shared.f32 	%f37, [%rd37+0];
	.loc	16	362	0
	mov.f32 	%f38, 0f3f800000;    	// 1
	ld.shared.f32 	%f39, [%rd37+16];
	sub.ftz.f32 	%f40, %f38, %f39;
	.loc	16	365	0
	and.b32 	%r40, %r37, 1073741823;
	mov.u32 	%r41, %r40;
	mov.s32 	%r42, 0;
	mov.u32 	%r43, %r42;
	mov.s32 	%r44, 0;
	mov.u32 	%r45, %r44;
	mov.s32 	%r46, 0;
	mov.u32 	%r47, %r46;
	tex.1d.v4.f32.s32 {%f41,%f42,%f43,%f44},[pos_tex,{%r41,%r43,%r45,%r47}];
	mov.f32 	%f45, %f41;
	mov.f32 	%f46, %f42;
	mov.f32 	%f47, %f43;
	mov.f32 	%f48, %f44;
	sub.ftz.f32 	%f49, %f23, %f46;
	sub.ftz.f32 	%f50, %f22, %f45;
	sub.ftz.f32 	%f51, %f24, %f47;
	mul.ftz.f32 	%f52, %f49, %f49;
	fma.rn.ftz.f32 	%f53, %f50, %f50, %f52;
	fma.rn.ftz.f32 	%f54, %f51, %f51, %f53;
	setp.lt.ftz.f32 	%p6, %f54, %f31;
	@!%p6 bra 	$Lt_1_30978;
	ld.param.f32 	%f55, [__cudaparm_kernel_pair_fast_cut_ljsq];
	setp.lt.ftz.f32 	%p7, %f54, %f55;
	rcp.approx.ftz.f32 	%f56, %f54;
	@!%p7 bra 	$Lt_1_28674;
	.loc	16	380	0
	cvt.rzi.ftz.s32.f32 	%r48, %f48;
	cvt.s64.s32 	%rd38, %r36;
	mul.wide.s32 	%rd39, %r36, 8;
	add.u64 	%rd40, %rd7, %rd39;
	cvt.s64.s32 	%rd41, %r48;
	mul.wide.s32 	%rd42, %r48, 8;
	add.u64 	%rd43, %rd7, %rd42;
	ld.shared.v2.f32 	{%f57,%f58}, [%rd40+0];
	ld.shared.v2.f32 	{%f59,%f60}, [%rd43+0];
	mul.ftz.f32 	%f61, %f57, %f59;
	.loc	16	381	0
	add.ftz.f32 	%f62, %f58, %f60;
	mov.f32 	%f63, 0f3f000000;    	// 0.5
	mul.ftz.f32 	%f64, %f62, %f63;
	.loc	16	385	0
	mul.ftz.f32 	%f65, %f64, %f64;
	sqrt.approx.ftz.f32 	%f66, %f61;
	mov.f32 	%f67, 0f40800000;    	// 4
	mul.ftz.f32 	%f68, %f66, %f67;
	mul.ftz.f32 	%f69, %f65, %f56;
	mul.ftz.f32 	%f70, %f69, %f69;
	mul.ftz.f32 	%f71, %f69, %f70;
	mul.ftz.f32 	%f72, %f68, %f71;
	mov.f32 	%f73, %f72;
	.loc	16	386	0
	mul.ftz.f32 	%f74, %f71, %f72;
	mov.f32 	%f75, %f74;
	.loc	16	387	0
	mov.f32 	%f76, 0f40c00000;    	// 6
	mul.ftz.f32 	%f77, %f72, %f76;
	mov.f32 	%f78, 0f41400000;    	// 12
	mul.ftz.f32 	%f79, %f78, %f74;
	sub.ftz.f32 	%f80, %f79, %f77;
	mul.ftz.f32 	%f81, %f37, %f80;
	ld.param.f32 	%f82, [__cudaparm_kernel_pair_fast_cut_lj_innersq];
	setp.gt.ftz.f32 	%p8, %f54, %f82;
	@!%p8 bra 	$Lt_1_28418;
	.loc	16	393	0
	add.ftz.f32 	%f83, %f54, %f54;
	sub.ftz.f32 	%f84, %f55, %f54;
	add.ftz.f32 	%f85, %f83, %f55;
	mul.ftz.f32 	%f86, %f84, %f84;
	mov.f32 	%f87, 0f40400000;    	// 3
	mul.ftz.f32 	%f88, %f87, %f82;
	sub.ftz.f32 	%f89, %f85, %f88;
	ld.param.f32 	%f90, [__cudaparm_kernel_pair_fast_denom_lj];
	div.approx.ftz.f32 	%f91, %f89, %f90;
	mul.ftz.f32 	%f92, %f86, %f91;
	mov.f32 	%f93, %f92;
	.loc	16	396	0
	mov.f32 	%f94, 0f41400000;    	// 12
	mul.ftz.f32 	%f95, %f54, %f94;
	mul.ftz.f32 	%f96, %f84, %f95;
	sub.ftz.f32 	%f97, %f54, %f82;
	mul.ftz.f32 	%f98, %f96, %f97;
	div.approx.ftz.f32 	%f99, %f98, %f90;
	sub.ftz.f32 	%f100, %f74, %f72;
	mul.ftz.f32 	%f101, %f99, %f100;
	fma.rn.ftz.f32 	%f81, %f81, %f92, %f101;
	bra.uni 	$Lt_1_28418;
$Lt_1_28674:
	.loc	16	399	0
	mov.f32 	%f81, 0f00000000;    	// 0
$Lt_1_28418:
	ld.param.f32 	%f102, [__cudaparm_kernel_pair_fast_cut_coulsq];
	setp.gt.ftz.f32 	%p9, %f102, %f54;
	@!%p9 bra 	$Lt_1_29698;
	.loc	16	406	0
	sqrt.approx.ftz.f32 	%f103, %f54;
	ld.param.f32 	%f104, [__cudaparm_kernel_pair_fast_g_ewald];
	mul.ftz.f32 	%f105, %f104, %f103;
	mul.ftz.f32 	%f106, %f105, %f105;
	mov.f32 	%f107, 0f3f800000;   	// 1
	mov.f32 	%f108, 0f3ea7ba05;   	// 0.327591
	fma.rn.ftz.f32 	%f109, %f108, %f105, %f107;
	neg.ftz.f32 	%f110, %f106;
	rcp.approx.ftz.f32 	%f111, %f109;
	mov.f32 	%f112, 0f3fb8aa3b;   	// 1.4427
	mul.ftz.f32 	%f113, %f110, %f112;
	ex2.approx.ftz.f32 	%f114, %f113;
	mov.f32 	%f115, 0f3e827906;   	// 0.25483
	mov.f32 	%f116, 0fbe91a98e;   	// -0.284497
	mov.f32 	%f117, 0f3fb5f0e3;   	// 1.42141
	mov.f32 	%f118, 0fbfba00e3;   	// -1.45315
	mov.f32 	%f119, 0f3f87dc22;   	// 1.06141
	fma.rn.ftz.f32 	%f120, %f119, %f111, %f118;
	fma.rn.ftz.f32 	%f121, %f111, %f120, %f117;
	fma.rn.ftz.f32 	%f122, %f111, %f121, %f116;
	fma.rn.ftz.f32 	%f123, %f111, %f122, %f115;
	mul.ftz.f32 	%f124, %f111, %f123;
	mul.ftz.f32 	%f125, %f114, %f124;
	mov.f32 	%f126, %f125;
	.loc	16	407	0
	mov.u32 	%r49, %r40;
	mov.s32 	%r50, 0;
	mov.u32 	%r51, %r50;
	mov.s32 	%r52, 0;
	mov.u32 	%r53, %r52;
	mov.s32 	%r54, 0;
	mov.u32 	%r55, %r54;
	tex.1d.v4.f32.s32 {%f127,%f128,%f129,%f130},[q_tex,{%r49,%r51,%r53,%r55}];
	mov.f32 	%f131, %f127;
	ld.param.f32 	%f132, [__cudaparm_kernel_pair_fast_qqrd2e];
	mul.ftz.f32 	%f133, %f132, %f30;
	mul.ftz.f32 	%f134, %f133, %f131;
	div.approx.ftz.f32 	%f135, %f134, %f103;
	mov.f32 	%f136, %f135;
	.loc	16	408	0
	mov.f32 	%f137, 0f3f906ebb;   	// 1.12838
	mul.ftz.f32 	%f138, %f105, %f137;
	fma.rn.ftz.f32 	%f139, %f114, %f138, %f125;
	sub.ftz.f32 	%f140, %f139, %f40;
	mul.ftz.f32 	%f141, %f135, %f140;
	bra.uni 	$Lt_1_29442;
$Lt_1_29698:
	.loc	16	410	0
	mov.f32 	%f141, 0f00000000;   	// 0
$Lt_1_29442:
	.loc	16	414	0
	add.ftz.f32 	%f142, %f141, %f81;
	mul.ftz.f32 	%f143, %f142, %f56;
	fma.rn.ftz.f32 	%f34, %f50, %f143, %f34;
	.loc	16	415	0
	fma.rn.ftz.f32 	%f33, %f49, %f143, %f33;
	.loc	16	416	0
	fma.rn.ftz.f32 	%f32, %f51, %f143, %f32;
	ld.param.s32 	%r56, [__cudaparm_kernel_pair_fast_eflag];
	mov.u32 	%r57, 0;
	setp.le.s32 	%p10, %r56, %r57;
	@%p10 bra 	$Lt_1_30466;
	.loc	16	419	0
	mov.f32 	%f144, %f136;
	mov.f32 	%f145, %f126;
	sub.ftz.f32 	%f146, %f145, %f40;
	fma.rn.ftz.f32 	%f147, %f144, %f146, %f35;
	selp.f32 	%f35, %f147, %f35, %p9;
	@!%p7 bra 	$Lt_1_30466;
	.loc	16	425	0
	mov.f32 	%f148, %f75;
	mov.f32 	%f149, %f73;
	sub.ftz.f32 	%f150, %f148, %f149;
	mov.f32 	%f151, %f93;
	mul.ftz.f32 	%f152, %f151, %f150;
	ld.param.f32 	%f153, [__cudaparm_kernel_pair_fast_cut_lj_innersq];
	setp.lt.ftz.f32 	%p11, %f153, %f54;
	selp.f32 	%f154, %f152, %f150, %p11;
	fma.rn.ftz.f32 	%f36, %f37, %f154, %f36;
$Lt_1_30466:
$Lt_1_29954:
	ld.param.s32 	%r58, [__cudaparm_kernel_pair_fast_vflag];
	mov.u32 	%r59, 0;
	setp.le.s32 	%p12, %r58, %r59;
	@%p12 bra 	$Lt_1_30978;
	.loc	16	429	0
	mov.f32 	%f155, %f7;
	mul.ftz.f32 	%f156, %f50, %f50;
	fma.rn.ftz.f32 	%f157, %f143, %f156, %f155;
	mov.f32 	%f7, %f157;
	.loc	16	430	0
	mov.f32 	%f158, %f9;
	fma.rn.ftz.f32 	%f159, %f143, %f52, %f158;
	mov.f32 	%f9, %f159;
	.loc	16	431	0
	mov.f32 	%f160, %f11;
	mul.ftz.f32 	%f161, %f51, %f51;
	fma.rn.ftz.f32 	%f162, %f143, %f161, %f160;
	mov.f32 	%f11, %f162;
	.loc	16	432	0
	mov.f32 	%f163, %f13;
	mul.ftz.f32 	%f164, %f49, %f50;
	fma.rn.ftz.f32 	%f165, %f143, %f164, %f163;
	mov.f32 	%f13, %f165;
	.loc	16	433	0
	mov.f32 	%f166, %f15;
	mul.ftz.f32 	%f167, %f50, %f51;
	fma.rn.ftz.f32 	%f168, %f143, %f167, %f166;
	mov.f32 	%f15, %f168;
	.loc	16	434	0
	mul.ftz.f32 	%f169, %f49, %f51;
	fma.rn.ftz.f32 	%f16, %f143, %f169, %f16;
	mov.f32 	%f17, %f16;
$Lt_1_30978:
$Lt_1_27906:
	.loc	16	357	0
	mul.lo.u64 	%rd44, %rd34, 4;
	add.u64 	%rd26, %rd26, %rd44;
	setp.lt.u64 	%p13, %rd26, %rd23;
	@%p13 bra 	$Lt_1_27650;
	bra.uni 	$Lt_1_26114;
$Lt_1_37378:
	mov.f32 	%f32, 0f00000000;    	// 0
	mov.f32 	%f33, 0f00000000;    	// 0
	mov.f32 	%f34, 0f00000000;    	// 0
	mov.f32 	%f35, 0f00000000;    	// 0
	mov.f32 	%f36, 0f00000000;    	// 0
	bra.uni 	$Lt_1_26114;
$Lt_1_26370:
	mov.f32 	%f32, 0f00000000;    	// 0
	mov.f32 	%f33, 0f00000000;    	// 0
	mov.f32 	%f34, 0f00000000;    	// 0
	mov.f32 	%f35, 0f00000000;    	// 0
	mov.f32 	%f36, 0f00000000;    	// 0
$Lt_1_26114:
	mov.u32 	%r60, 1;
	setp.le.s32 	%p14, %r5, %r60;
	@%p14 bra 	$Lt_1_33794;
	.loc	16	445	0
	mov.u64 	%rd45, __cuda___cuda_local_var_32826_35_non_const_red_acc4392;
	mul.lo.u64 	%rd46, %rd1, 4;
	add.u64 	%rd47, %rd45, %rd46;
	mov.f32 	%f170, %f34;
	st.shared.f32 	[%rd47+0], %f170;
	.loc	16	446	0
	mov.f32 	%f171, %f33;
	st.shared.f32 	[%rd47+512], %f171;
	.loc	16	447	0
	mov.f32 	%f172, %f32;
	st.shared.f32 	[%rd47+1024], %f172;
	.loc	16	448	0
	mov.f32 	%f173, %f36;
	st.shared.f32 	[%rd47+1536], %f173;
	.loc	16	449	0
	mov.f32 	%f174, %f35;
	st.shared.f32 	[%rd47+2048], %f174;
	.loc	16	451	0
	shr.s32 	%r61, %r5, 31;
	mov.s32 	%r62, 1;
	and.b32 	%r63, %r61, %r62;
	add.s32 	%r64, %r63, %r5;
	shr.s32 	%r65, %r64, 1;
	mov.s32 	%r66, %r65;
	mov.u32 	%r67, 0;
	setp.ne.u32 	%p15, %r65, %r67;
	@!%p15 bra 	$Lt_1_32258;
$Lt_1_32770:
	setp.ge.u32 	%p16, %r9, %r66;
	@%p16 bra 	$Lt_1_33026;
	.loc	16	454	0
	add.u32 	%r68, %r1, %r66;
	cvt.u64.u32 	%rd48, %r68;
	mul.wide.u32 	%rd49, %r68, 4;
	add.u64 	%rd50, %rd45, %rd49;
	ld.shared.f32 	%f175, [%rd50+0];
	add.ftz.f32 	%f170, %f175, %f170;
	st.shared.f32 	[%rd47+0], %f170;
	ld.shared.f32 	%f176, [%rd50+512];
	add.ftz.f32 	%f171, %f176, %f171;
	st.shared.f32 	[%rd47+512], %f171;
	ld.shared.f32 	%f177, [%rd50+1024];
	add.ftz.f32 	%f172, %f177, %f172;
	st.shared.f32 	[%rd47+1024], %f172;
	ld.shared.f32 	%f178, [%rd50+1536];
	add.ftz.f32 	%f173, %f178, %f173;
	st.shared.f32 	[%rd47+1536], %f173;
	ld.shared.f32 	%f179, [%rd50+2048];
	add.ftz.f32 	%f174, %f179, %f174;
	st.shared.f32 	[%rd47+2048], %f174;
$Lt_1_33026:
	.loc	16	451	0
	shr.u32 	%r66, %r66, 1;
	mov.u32 	%r69, 0;
	setp.ne.u32 	%p17, %r66, %r69;
	@%p17 bra 	$Lt_1_32770;
$Lt_1_32258:
	.loc	16	458	0
	mov.f32 	%f34, %f170;
	.loc	16	459	0
	mov.f32 	%f33, %f171;
	.loc	16	460	0
	mov.f32 	%f32, %f172;
	.loc	16	461	0
	mov.f32 	%f36, %f173;
	.loc	16	462	0
	mov.f32 	%f35, %f174;
	ld.param.s32 	%r70, [__cudaparm_kernel_pair_fast_vflag];
	mov.u32 	%r71, 0;
	setp.le.s32 	%p18, %r70, %r71;
	@%p18 bra 	$Lt_1_33794;
	.loc	16	466	0
	mov.f32 	%f170, %f7;
	st.shared.f32 	[%rd47+0], %f170;
	mov.f32 	%f171, %f9;
	st.shared.f32 	[%rd47+512], %f171;
	mov.f32 	%f172, %f11;
	st.shared.f32 	[%rd47+1024], %f172;
	mov.f32 	%f173, %f13;
	st.shared.f32 	[%rd47+1536], %f173;
	mov.f32 	%f174, %f15;
	st.shared.f32 	[%rd47+2048], %f174;
	mov.f32 	%f180, %f17;
	st.shared.f32 	[%rd47+2560], %f180;
	.loc	16	468	0
	mov.s32 	%r72, %r65;
	@!%p15 bra 	$Lt_1_34306;
$Lt_1_34818:
	setp.ge.u32 	%p19, %r9, %r72;
	@%p19 bra 	$Lt_1_35074;
	.loc	16	471	0
	add.u32 	%r73, %r1, %r72;
	cvt.u64.u32 	%rd51, %r73;
	mul.wide.u32 	%rd52, %r73, 4;
	add.u64 	%rd53, %rd45, %rd52;
	ld.shared.f32 	%f181, [%rd53+0];
	add.ftz.f32 	%f170, %f181, %f170;
	st.shared.f32 	[%rd47+0], %f170;
	ld.shared.f32 	%f182, [%rd53+512];
	add.ftz.f32 	%f171, %f182, %f171;
	st.shared.f32 	[%rd47+512], %f171;
	ld.shared.f32 	%f183, [%rd53+1024];
	add.ftz.f32 	%f172, %f183, %f172;
	st.shared.f32 	[%rd47+1024], %f172;
	ld.shared.f32 	%f184, [%rd53+1536];
	add.ftz.f32 	%f173, %f184, %f173;
	st.shared.f32 	[%rd47+1536], %f173;
	ld.shared.f32 	%f185, [%rd53+2048];
	add.ftz.f32 	%f174, %f185, %f174;
	st.shared.f32 	[%rd47+2048], %f174;
	ld.shared.f32 	%f186, [%rd53+2560];
	add.ftz.f32 	%f180, %f186, %f180;
	st.shared.f32 	[%rd47+2560], %f180;
$Lt_1_35074:
	.loc	16	468	0
	shr.u32 	%r72, %r72, 1;
	mov.u32 	%r74, 0;
	setp.ne.u32 	%p20, %r72, %r74;
	@%p20 bra 	$Lt_1_34818;
$Lt_1_34306:
	.loc	16	476	0
	mov.f32 	%f7, %f170;
	mov.f32 	%f9, %f171;
	mov.f32 	%f11, %f172;
	mov.f32 	%f13, %f173;
	mov.f32 	%f15, %f174;
	mov.f32 	%f17, %f180;
$Lt_1_33794:
$Lt_1_31746:
	selp.s32 	%r75, 1, 0, %p3;
	mov.s32 	%r76, 0;
	set.eq.u32.s32 	%r77, %r9, %r76;
	neg.s32 	%r78, %r77;
	and.b32 	%r79, %r75, %r78;
	mov.u32 	%r80, 0;
	setp.eq.s32 	%p21, %r79, %r80;
	@%p21 bra 	$Lt_1_35842;
	.loc	16	482	0
	cvt.s64.s32 	%rd54, %r12;
	ld.param.u64 	%rd55, [__cudaparm_kernel_pair_fast_engv];
	mul.wide.s32 	%rd56, %r12, 4;
	add.u64 	%rd57, %rd55, %rd56;
	ld.param.s32 	%r81, [__cudaparm_kernel_pair_fast_eflag];
	mov.u32 	%r82, 0;
	setp.le.s32 	%p22, %r81, %r82;
	@%p22 bra 	$Lt_1_36354;
	.loc	16	484	0
	st.global.f32 	[%rd57+0], %f36;
	.loc	16	485	0
	cvt.s64.s32 	%rd58, %r13;
	mul.wide.s32 	%rd59, %r13, 4;
	add.u64 	%rd60, %rd59, %rd57;
	.loc	16	486	0
	st.global.f32 	[%rd60+0], %f35;
	.loc	16	487	0
	add.u64 	%rd57, %rd59, %rd60;
$Lt_1_36354:
	ld.param.s32 	%r83, [__cudaparm_kernel_pair_fast_vflag];
	mov.u32 	%r84, 0;
	setp.le.s32 	%p23, %r83, %r84;
	@%p23 bra 	$Lt_1_36866;
	.loc	16	491	0
	mov.f32 	%f187, %f7;
	st.global.f32 	[%rd57+0], %f187;
	.loc	16	492	0
	cvt.s64.s32 	%rd61, %r13;
	mul.wide.s32 	%rd62, %r13, 4;
	add.u64 	%rd63, %rd62, %rd57;
	.loc	16	491	0
	mov.f32 	%f188, %f9;
	st.global.f32 	[%rd63+0], %f188;
	.loc	16	492	0
	add.u64 	%rd64, %rd62, %rd63;
	.loc	16	491	0
	mov.f32 	%f189, %f11;
	st.global.f32 	[%rd64+0], %f189;
	.loc	16	492	0
	add.u64 	%rd65, %rd62, %rd64;
	.loc	16	491	0
	mov.f32 	%f190, %f13;
	st.global.f32 	[%rd65+0], %f190;
	.loc	16	492	0
	add.u64 	%rd57, %rd62, %rd65;
	.loc	16	491	0
	mov.f32 	%f191, %f15;
	st.global.f32 	[%rd57+0], %f191;
	mov.f32 	%f192, %f17;
	add.u64 	%rd66, %rd62, %rd57;
	st.global.f32 	[%rd66+0], %f192;
$Lt_1_36866:
	.loc	16	495	0
	ld.param.u64 	%rd67, [__cudaparm_kernel_pair_fast_ans];
	mul.lo.u64 	%rd68, %rd54, 16;
	add.u64 	%rd69, %rd67, %rd68;
	mov.f32 	%f193, %f194;
	st.global.v4.f32 	[%rd69+0], {%f34,%f33,%f32,%f193};
$Lt_1_35842:
	.loc	16	497	0
	exit;
$LDWend_kernel_pair_fast:
	} // kernel_pair_fast