221 lines
5.9 KiB
Plaintext
221 lines
5.9 KiB
Plaintext
//
|
|
// Generated by NVIDIA NVVM Compiler
|
|
//
|
|
// Compiler Build ID: CL-31968024
|
|
// Cuda compilation tools, release 12.0, V12.0.76
|
|
// Based on NVVM 7.0.1
|
|
//
|
|
|
|
.version 8.0
|
|
.target sm_52
|
|
.address_size 64
|
|
|
|
.const .align 16 .b8 params[1184];
|
|
|
|
.visible .func (.param .align 8 .b8 func_retval0[32]) __direct_callable__oxMain(
|
|
.param .b32 __direct_callable__oxMain_param_0,
|
|
.param .align 8 .b8 __direct_callable__oxMain_param_1[8]
|
|
)
|
|
{
|
|
.reg .pred %p<6>;
|
|
.reg .b16 %rs<41>;
|
|
.reg .f32 %f<119>;
|
|
.reg .b32 %r<31>;
|
|
.reg .b64 %rd<48>;
|
|
|
|
|
|
ld.param.u32 %r5, [__direct_callable__oxMain_param_0];
|
|
ld.param.f32 %f33, [__direct_callable__oxMain_param_1+4];
|
|
ld.param.f32 %f32, [__direct_callable__oxMain_param_1];
|
|
// begin inline asm
|
|
call (%r6), _optix_get_launch_index_x, ();
|
|
// end inline asm
|
|
// begin inline asm
|
|
call (%r7), _optix_get_launch_index_y, ();
|
|
// end inline asm
|
|
ld.const.u64 %rd6, [params+72];
|
|
cvta.to.global.u64 %rd7, %rd6;
|
|
mul.wide.s32 %rd8, %r5, 4;
|
|
add.s64 %rd9, %rd7, %rd8;
|
|
ld.global.u32 %r3, [%rd9];
|
|
setp.eq.s32 %p1, %r3, 0;
|
|
@%p1 bra $L__BB0_2;
|
|
|
|
ld.const.u64 %rd10, [params+480];
|
|
cvta.to.global.u64 %rd11, %rd10;
|
|
ld.const.u32 %r9, [params+472];
|
|
mad.lo.s32 %r10, %r9, %r7, %r6;
|
|
cvt.u64.u32 %rd12, %r10;
|
|
add.s64 %rd13, %rd11, %rd12;
|
|
ld.global.u8 %r11, [%rd13];
|
|
mov.u32 %r12, 1;
|
|
shl.b32 %r13, %r12, %r11;
|
|
and.b32 %r14, %r13, %r3;
|
|
setp.eq.s32 %p2, %r14, 0;
|
|
mov.f32 %f113, 0f00000000;
|
|
mov.f32 %f114, %f113;
|
|
mov.f32 %f115, %f113;
|
|
mov.f32 %f116, %f113;
|
|
mov.f32 %f117, %f113;
|
|
mov.f32 %f118, %f113;
|
|
@%p2 bra $L__BB0_9;
|
|
|
|
$L__BB0_2:
|
|
ld.const.u64 %rd14, [params+24];
|
|
cvta.to.global.u64 %rd15, %rd14;
|
|
mul.wide.u32 %rd16, %r5, 12;
|
|
add.s64 %rd17, %rd15, %rd16;
|
|
ld.global.s32 %rd1, [%rd17];
|
|
ld.global.s32 %rd2, [%rd17+4];
|
|
ld.global.s32 %rd3, [%rd17+8];
|
|
ld.const.u32 %r4, [params+32];
|
|
setp.gt.u32 %p3, %r4, %r5;
|
|
@%p3 bra $L__BB0_4;
|
|
bra.uni $L__BB0_3;
|
|
|
|
$L__BB0_4:
|
|
mov.f32 %f118, 0f3F800000;
|
|
sub.f32 %f71, %f118, %f32;
|
|
sub.f32 %f109, %f71, %f33;
|
|
bra.uni $L__BB0_5;
|
|
|
|
$L__BB0_3:
|
|
sub.s32 %r15, %r5, %r4;
|
|
ld.const.v2.u64 {%rd18, %rd19}, [params+48];
|
|
cvta.to.global.u64 %rd22, %rd18;
|
|
mul.wide.u32 %rd23, %r15, 4;
|
|
add.s64 %rd24, %rd22, %rd23;
|
|
ld.const.u64 %rd25, [params+40];
|
|
cvta.to.global.u64 %rd26, %rd25;
|
|
shl.b64 %rd27, %rd1, 3;
|
|
add.s64 %rd28, %rd26, %rd27;
|
|
ld.global.v2.f32 {%f40, %f41}, [%rd28];
|
|
shl.b64 %rd29, %rd2, 3;
|
|
add.s64 %rd30, %rd26, %rd29;
|
|
ld.global.v2.f32 {%f44, %f45}, [%rd30];
|
|
shl.b64 %rd31, %rd3, 3;
|
|
add.s64 %rd32, %rd26, %rd31;
|
|
ld.global.v2.f32 {%f48, %f49}, [%rd32];
|
|
mov.f32 %f52, 0f3F800000;
|
|
sub.f32 %f53, %f52, %f32;
|
|
sub.f32 %f109, %f53, %f33;
|
|
mul.f32 %f54, %f32, %f44;
|
|
mul.f32 %f55, %f32, %f45;
|
|
fma.rn.f32 %f56, %f109, %f40, %f54;
|
|
fma.rn.f32 %f57, %f109, %f41, %f55;
|
|
fma.rn.f32 %f58, %f33, %f48, %f56;
|
|
fma.rn.f32 %f59, %f33, %f49, %f57;
|
|
ld.global.u32 %r16, [%rd24];
|
|
shr.u32 %r17, %r16, 16;
|
|
abs.f32 %f60, %f58;
|
|
cvt.rmi.f32.f32 %f61, %f60;
|
|
sub.f32 %f62, %f60, %f61;
|
|
abs.f32 %f63, %f59;
|
|
cvt.rmi.f32.f32 %f64, %f63;
|
|
sub.f32 %f65, %f63, %f64;
|
|
cvta.to.global.u64 %rd33, %rd19;
|
|
shl.b32 %r18, %r16, 4;
|
|
cvt.u64.u32 %rd34, %r18;
|
|
and.b64 %rd35, %rd34, 1048560;
|
|
add.s64 %rd36, %rd33, %rd35;
|
|
ld.global.v2.u32 {%r19, %r20}, [%rd36];
|
|
cvt.rn.f32.u32 %f66, %r19;
|
|
mul.f32 %f67, %f62, %f66;
|
|
cvt.rzi.u32.f32 %r23, %f67;
|
|
cvt.rn.f32.u32 %f68, %r20;
|
|
mul.f32 %f69, %f65, %f68;
|
|
cvt.rzi.u32.f32 %r24, %f69;
|
|
mad.lo.s32 %r25, %r19, %r24, %r23;
|
|
cvt.u64.u32 %rd37, %r25;
|
|
ld.global.u64 %rd38, [%rd36+8];
|
|
add.s64 %rd39, %rd38, %rd37;
|
|
ld.u8 %r26, [%rd39];
|
|
and.b32 %r27, %r17, %r26;
|
|
setp.eq.s32 %p4, %r27, 0;
|
|
selp.f32 %f118, 0f00000000, 0f3F800000, %p4;
|
|
|
|
$L__BB0_5:
|
|
cvt.u32.u64 %r28, %rd1;
|
|
cvt.u32.u64 %r29, %rd2;
|
|
cvt.u32.u64 %r30, %rd3;
|
|
ld.const.u64 %rd40, [params+16];
|
|
cvta.to.global.u64 %rd41, %rd40;
|
|
mul.wide.s32 %rd42, %r28, 32;
|
|
add.s64 %rd43, %rd41, %rd42;
|
|
mul.wide.s32 %rd44, %r29, 32;
|
|
add.s64 %rd45, %rd41, %rd44;
|
|
add.s64 %rd4, %rd45, 12;
|
|
mul.wide.s32 %rd46, %r30, 32;
|
|
add.s64 %rd47, %rd41, %rd46;
|
|
add.s64 %rd5, %rd47, 12;
|
|
ld.global.f32 %f72, [%rd43+12];
|
|
ld.global.f32 %f73, [%rd43+16];
|
|
ld.global.f32 %f74, [%rd43+20];
|
|
ld.global.f32 %f75, [%rd45+12];
|
|
mul.f32 %f76, %f32, %f75;
|
|
ld.global.f32 %f77, [%rd45+16];
|
|
mul.f32 %f78, %f32, %f77;
|
|
ld.global.f32 %f79, [%rd45+20];
|
|
mul.f32 %f80, %f32, %f79;
|
|
fma.rn.f32 %f81, %f109, %f72, %f76;
|
|
fma.rn.f32 %f82, %f109, %f73, %f78;
|
|
fma.rn.f32 %f83, %f109, %f74, %f80;
|
|
ld.global.f32 %f84, [%rd47+12];
|
|
ld.global.f32 %f85, [%rd47+16];
|
|
ld.global.f32 %f86, [%rd47+20];
|
|
fma.rn.f32 %f87, %f33, %f84, %f81;
|
|
fma.rn.f32 %f88, %f33, %f85, %f82;
|
|
fma.rn.f32 %f89, %f33, %f86, %f83;
|
|
mul.f32 %f90, %f88, %f88;
|
|
fma.rn.f32 %f91, %f87, %f87, %f90;
|
|
fma.rn.f32 %f92, %f89, %f89, %f91;
|
|
sqrt.rn.f32 %f93, %f92;
|
|
rcp.rn.f32 %f94, %f93;
|
|
mul.f32 %f113, %f87, %f94;
|
|
mul.f32 %f114, %f88, %f94;
|
|
mul.f32 %f115, %f89, %f94;
|
|
ld.global.v2.f32 {%f116, %f96}, [%rd43+24];
|
|
setp.lt.f32 %p5, %f96, 0f00000000;
|
|
@%p5 bra $L__BB0_7;
|
|
bra.uni $L__BB0_6;
|
|
|
|
$L__BB0_7:
|
|
add.f32 %f117, %f96, 0f3F800000;
|
|
bra.uni $L__BB0_9;
|
|
|
|
$L__BB0_6:
|
|
ld.global.v2.f32 {%f97, %f98}, [%rd4+12];
|
|
ld.global.v2.f32 {%f99, %f100}, [%rd5+12];
|
|
mul.f32 %f102, %f32, %f97;
|
|
mul.f32 %f104, %f32, %f98;
|
|
fma.rn.f32 %f105, %f109, %f116, %f102;
|
|
fma.rn.f32 %f106, %f109, %f96, %f104;
|
|
fma.rn.f32 %f116, %f33, %f99, %f105;
|
|
fma.rn.f32 %f117, %f33, %f100, %f106;
|
|
|
|
$L__BB0_9:
|
|
st.param.f32 [func_retval0+0], %f113;
|
|
st.param.f32 [func_retval0+4], %f114;
|
|
st.param.f32 [func_retval0+8], %f115;
|
|
st.param.v4.b8 [func_retval0+12], {%rs9, %rs10, %rs11, %rs12};
|
|
st.param.f32 [func_retval0+16], %f116;
|
|
st.param.f32 [func_retval0+20], %f117;
|
|
st.param.f32 [func_retval0+24], %f118;
|
|
st.param.v4.b8 [func_retval0+28], {%rs13, %rs14, %rs15, %rs16};
|
|
ret;
|
|
|
|
}
|
|
// .globl oxMain
|
|
.visible .entry oxMain()
|
|
{
|
|
.reg .b64 %rd<2>;
|
|
|
|
|
|
mov.u64 %rd1, __direct_callable__oxMain;
|
|
// begin inline asm
|
|
// end inline asm
|
|
ret;
|
|
|
|
}
|
|
|