diff options
Diffstat (limited to 'VRCSDK3Worlds/Assets/Editor/x64/Bakery/trimeshTex.ptx')
| -rw-r--r-- | VRCSDK3Worlds/Assets/Editor/x64/Bakery/trimeshTex.ptx | 490 |
1 files changed, 490 insertions, 0 deletions
diff --git a/VRCSDK3Worlds/Assets/Editor/x64/Bakery/trimeshTex.ptx b/VRCSDK3Worlds/Assets/Editor/x64/Bakery/trimeshTex.ptx new file mode 100644 index 00000000..68139ca5 --- /dev/null +++ b/VRCSDK3Worlds/Assets/Editor/x64/Bakery/trimeshTex.ptx @@ -0,0 +1,490 @@ +// +// Generated by NVIDIA NVVM Compiler +// +// Compiler Build ID: CL-23083092 +// Cuda compilation tools, release 9.1, V9.1.85 +// Based on LLVM 3.4svn +// + +.version 6.1 +.target sm_30 +.address_size 64 + + // .globl _Z9intersecti +.global .align 8 .b8 pixelID[8]; +.global .align 8 .b8 resolution[8]; +.global .align 4 .b8 normal[12]; +.global .align 4 .b8 camPos[12]; +.global .align 4 .b8 root[4]; +.global .align 4 .u32 imageEnabled; +.global .texref lightmap; +.global .align 16 .b8 tileInfo[16]; +.global .align 4 .u32 additive; +.global .align 8 .b8 texCoords[8]; +.global .align 1 .b8 vertex_buffer[1]; +.global .align 1 .b8 index_buffer[1]; +.global .align 4 .u32 firstAlphaTriangle; +.global .align 1 .b8 vertex_buffer_uv[1]; +.global .align 1 .b8 triangleAlphaIDs[1]; +.global .align 1 .b8 alphaTextures[1]; +.global .align 4 .b8 ray[36]; +.global .align 4 .b8 _ZN21rti_internal_typeinfo7pixelIDE[8] = {82, 97, 121, 0, 8, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo10resolutionE[8] = {82, 97, 121, 0, 8, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo6normalE[8] = {82, 97, 121, 0, 12, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo6camPosE[8] = {82, 97, 121, 0, 12, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo4rootE[8] = {82, 97, 121, 0, 4, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo12imageEnabledE[8] = {82, 97, 121, 0, 4, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo8tileInfoE[8] = {82, 97, 121, 0, 16, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo8additiveE[8] = {82, 97, 121, 0, 4, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo9texCoordsE[8] = {82, 97, 121, 0, 8, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo18firstAlphaTriangleE[8] = {82, 97, 121, 0, 4, 0, 0, 0}; +.global .align 4 .b8 _ZN21rti_internal_typeinfo3rayE[8] = {82, 97, 121, 0, 36, 0, 0, 0}; +.global .align 8 .u64 _ZN21rti_internal_register20reg_bitness_detectorE; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail0E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail1E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail2E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail3E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail4E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail5E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail6E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail7E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail8E; +.global .align 8 .u64 _ZN21rti_internal_register24reg_exception_64_detail9E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail0E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail1E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail2E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail3E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail4E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail5E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail6E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail7E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail8E; +.global .align 4 .u32 _ZN21rti_internal_register21reg_exception_detail9E; +.global .align 4 .u32 _ZN21rti_internal_register14reg_rayIndex_xE; +.global .align 4 .u32 _ZN21rti_internal_register14reg_rayIndex_yE; +.global .align 4 .u32 _ZN21rti_internal_register14reg_rayIndex_zE; +.global .align 8 .b8 _ZN21rti_internal_typename7pixelIDE[6] = {117, 105, 110, 116, 50, 0}; +.global .align 8 .b8 _ZN21rti_internal_typename10resolutionE[6] = {117, 105, 110, 116, 50, 0}; +.global .align 8 .b8 _ZN21rti_internal_typename6normalE[7] = {102, 108, 111, 97, 116, 51, 0}; +.global .align 8 .b8 _ZN21rti_internal_typename6camPosE[7] = {102, 108, 111, 97, 116, 51, 0}; +.global .align 16 .b8 _ZN21rti_internal_typename4rootE[9] = {114, 116, 79, 98, 106, 101, 99, 116, 0}; +.global .align 4 .b8 _ZN21rti_internal_typename12imageEnabledE[4] = {105, 110, 116, 0}; +.global .align 8 .b8 _ZN21rti_internal_typename8tileInfoE[6] = {117, 105, 110, 116, 52, 0}; +.global .align 4 .b8 _ZN21rti_internal_typename8additiveE[4] = {105, 110, 116, 0}; +.global .align 8 .b8 _ZN21rti_internal_typename9texCoordsE[7] = {102, 108, 111, 97, 116, 50, 0}; +.global .align 4 .b8 _ZN21rti_internal_typename18firstAlphaTriangleE[4] = {105, 110, 116, 0}; +.global .align 4 .b8 _ZN21rti_internal_typename3rayE[4] = {82, 97, 121, 0}; +.global .align 4 .u32 _ZN21rti_internal_typeenum7pixelIDE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum10resolutionE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum6normalE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum6camPosE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum4rootE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum12imageEnabledE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum8tileInfoE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum8additiveE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum9texCoordsE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum18firstAlphaTriangleE = 4919; +.global .align 4 .u32 _ZN21rti_internal_typeenum3rayE = 4919; +.global .align 16 .b8 _ZN21rti_internal_semantic7pixelIDE[14] = {114, 116, 76, 97, 117, 110, 99, 104, 73, 110, 100, 101, 120, 0}; +.global .align 16 .b8 _ZN21rti_internal_semantic10resolutionE[12] = {114, 116, 76, 97, 117, 110, 99, 104, 68, 105, 109, 0}; +.global .align 16 .b8 _ZN21rti_internal_semantic6normalE[17] = {97, 116, 116, 114, 105, 98, 117, 116, 101, 32, 110, 111, 114, 109, 97, 108, 0}; +.global .align 1 .b8 _ZN21rti_internal_semantic6camPosE[1]; +.global .align 1 .b8 _ZN21rti_internal_semantic4rootE[1]; +.global .align 1 .b8 _ZN21rti_internal_semantic12imageEnabledE[1]; +.global .align 1 .b8 _ZN21rti_internal_semantic8tileInfoE[1]; +.global .align 1 .b8 _ZN21rti_internal_semantic8additiveE[1]; +.global .align 16 .b8 _ZN21rti_internal_semantic9texCoordsE[20] = {97, 116, 116, 114, 105, 98, 117, 116, 101, 32, 116, 101, 120, 67, 111, 111, 114, 100, 115, 0}; +.global .align 1 .b8 _ZN21rti_internal_semantic18firstAlphaTriangleE[1]; +.global .align 16 .b8 _ZN21rti_internal_semantic3rayE[13] = {114, 116, 67, 117, 114, 114, 101, 110, 116, 82, 97, 121, 0}; +.global .align 1 .b8 _ZN23rti_internal_annotation7pixelIDE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation10resolutionE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation6normalE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation6camPosE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation4rootE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation12imageEnabledE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation8tileInfoE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation8additiveE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation9texCoordsE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation18firstAlphaTriangleE[1]; +.global .align 1 .b8 _ZN23rti_internal_annotation3rayE[1]; + +.visible .entry _Z9intersecti( + .param .u32 _Z9intersecti_param_0 +) +{ + .reg .pred %p<14>; + .reg .f32 %f<152>; + .reg .b32 %r<51>; + .reg .b64 %rd<117>; + + + ld.param.u32 %r2, [_Z9intersecti_param_0]; + cvt.s64.s32 %rd7, %r2; + mov.u64 %rd29, index_buffer; + cvta.global.u64 %rd6, %rd29; + mov.u32 %r9, 1; + mov.u32 %r4, 12; + mov.u64 %rd28, 0; + // inline asm + call (%rd5), _rt_buffer_get_64, (%rd6, %r9, %r4, %rd7, %rd28, %rd28, %rd28); + // inline asm + ld.s32 %rd13, [%rd5]; + mov.u64 %rd30, vertex_buffer; + cvta.global.u64 %rd12, %rd30; + ld.s32 %rd19, [%rd5+4]; + ld.s32 %rd25, [%rd5+8]; + mov.u32 %r10, 32; + // inline asm + call (%rd11), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd13, %rd28, %rd28, %rd28); + // inline asm + ld.f32 %f14, [%rd11+8]; + ld.v2.f32 {%f15, %f16}, [%rd11]; + // inline asm + call (%rd17), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd19, %rd28, %rd28, %rd28); + // inline asm + ld.f32 %f19, [%rd17+8]; + ld.v2.f32 {%f20, %f21}, [%rd17]; + // inline asm + call (%rd23), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd25, %rd28, %rd28, %rd28); + // inline asm + sub.f32 %f24, %f20, %f15; + sub.f32 %f25, %f21, %f16; + sub.f32 %f26, %f19, %f14; + ld.f32 %f27, [%rd23+8]; + ld.v2.f32 {%f28, %f29}, [%rd23]; + sub.f32 %f32, %f15, %f28; + sub.f32 %f33, %f16, %f29; + sub.f32 %f34, %f14, %f27; + mul.f32 %f35, %f26, %f33; + mul.f32 %f36, %f25, %f34; + sub.f32 %f37, %f35, %f36; + mul.f32 %f38, %f24, %f34; + mul.f32 %f39, %f26, %f32; + sub.f32 %f40, %f38, %f39; + mul.f32 %f41, %f25, %f32; + mul.f32 %f42, %f24, %f33; + sub.f32 %f43, %f41, %f42; + ld.global.f32 %f44, [ray+12]; + ld.global.f32 %f45, [ray+16]; + mul.f32 %f46, %f45, %f40; + fma.rn.f32 %f47, %f44, %f37, %f46; + ld.global.f32 %f48, [ray+20]; + fma.rn.f32 %f49, %f48, %f43, %f47; + rcp.rn.f32 %f50, %f49; + ld.global.f32 %f51, [ray]; + sub.f32 %f52, %f15, %f51; + ld.global.f32 %f53, [ray+4]; + sub.f32 %f54, %f16, %f53; + ld.global.f32 %f55, [ray+8]; + sub.f32 %f56, %f14, %f55; + mul.f32 %f57, %f50, %f52; + mul.f32 %f58, %f50, %f54; + mul.f32 %f59, %f50, %f56; + mul.f32 %f60, %f45, %f59; + mul.f32 %f61, %f58, %f48; + sub.f32 %f62, %f60, %f61; + mul.f32 %f63, %f57, %f48; + mul.f32 %f64, %f59, %f44; + sub.f32 %f65, %f63, %f64; + mul.f32 %f66, %f58, %f44; + mul.f32 %f67, %f57, %f45; + sub.f32 %f68, %f66, %f67; + mul.f32 %f69, %f33, %f65; + fma.rn.f32 %f70, %f32, %f62, %f69; + fma.rn.f32 %f1, %f34, %f68, %f70; + mul.f32 %f71, %f25, %f65; + fma.rn.f32 %f72, %f24, %f62, %f71; + fma.rn.f32 %f2, %f26, %f68, %f72; + mul.f32 %f73, %f40, %f58; + fma.rn.f32 %f74, %f37, %f57, %f73; + fma.rn.f32 %f3, %f43, %f59, %f74; + ld.global.f32 %f75, [ray+32]; + setp.geu.f32 %p1, %f3, %f75; + ld.global.f32 %f76, [ray+28]; + setp.leu.f32 %p2, %f3, %f76; + or.pred %p3, %p1, %p2; + setp.ltu.f32 %p4, %f1, 0f00000000; + or.pred %p5, %p3, %p4; + setp.ltu.f32 %p6, %f2, 0f00000000; + or.pred %p7, %p5, %p6; + add.f32 %f77, %f1, %f2; + setp.gtu.f32 %p8, %f77, 0f3F800000; + or.pred %p9, %p7, %p8; + @%p9 bra BB0_8; + + ld.global.u32 %r1, [firstAlphaTriangle]; + setp.gt.s32 %p10, %r1, %r2; + @%p10 bra BB0_3; + + sub.s32 %r29, %r2, %r1; + cvt.s64.s32 %rd33, %r29; + mov.u64 %rd76, triangleAlphaIDs; + cvta.global.u64 %rd32, %rd76; + mov.u32 %r25, 4; + // inline asm + call (%rd31), _rt_buffer_get_64, (%rd32, %r9, %r25, %rd33, %rd28, %rd28, %rd28); + // inline asm + ld.u32 %rd77, [%rd31]; + cvt.u32.u64 %r30, %rd77; + shr.u32 %r31, %r30, 16; + mov.u64 %rd78, vertex_buffer_uv; + cvta.global.u64 %rd38, %rd78; + mov.u32 %r18, 8; + // inline asm + call (%rd37), _rt_buffer_get_64, (%rd38, %r9, %r18, %rd13, %rd28, %rd28, %rd28); + // inline asm + ld.v2.f32 {%f78, %f79}, [%rd37]; + // inline asm + call (%rd43), _rt_buffer_get_64, (%rd38, %r9, %r18, %rd19, %rd28, %rd28, %rd28); + // inline asm + ld.v2.f32 {%f82, %f83}, [%rd43]; + // inline asm + call (%rd49), _rt_buffer_get_64, (%rd38, %r9, %r18, %rd25, %rd28, %rd28, %rd28); + // inline asm + mov.f32 %f86, 0f3F800000; + sub.f32 %f87, %f86, %f1; + sub.f32 %f88, %f87, %f2; + mul.f32 %f89, %f1, %f82; + mul.f32 %f90, %f1, %f83; + fma.rn.f32 %f91, %f88, %f78, %f89; + fma.rn.f32 %f92, %f88, %f79, %f90; + ld.v2.f32 {%f93, %f94}, [%rd49]; + fma.rn.f32 %f97, %f2, %f93, %f91; + fma.rn.f32 %f98, %f2, %f94, %f92; + abs.f32 %f99, %f97; + cvt.rmi.f32.f32 %f100, %f99; + sub.f32 %f101, %f99, %f100; + abs.f32 %f102, %f98; + cvt.rmi.f32.f32 %f103, %f102; + sub.f32 %f104, %f102, %f103; + and.b64 %rd57, %rd77, 65535; + mov.u64 %rd79, alphaTextures; + cvta.global.u64 %rd56, %rd79; + // inline asm + call (%rd55), _rt_buffer_get_64, (%rd56, %r9, %r25, %rd57, %rd28, %rd28, %rd28); + // inline asm + ld.u32 %r21, [%rd55]; + mov.u32 %r27, 2; + // inline asm + call (%rd61, %rd62, %rd63, %rd64), _rt_buffer_get_id_size_64, (%r21, %r27, %r9); + // inline asm + cvt.rn.f32.u64 %f105, %rd61; + mul.f32 %f106, %f101, %f105; + cvt.rzi.u32.f32 %r32, %f106; + cvt.rn.f32.u64 %f107, %rd62; + mul.f32 %f108, %f104, %f107; + cvt.rzi.u32.f32 %r33, %f108; + // inline asm + call (%rd65), _rt_buffer_get_64, (%rd56, %r9, %r25, %rd57, %rd28, %rd28, %rd28); + // inline asm + ld.u32 %r26, [%rd65]; + cvt.u64.u32 %rd72, %r32; + cvt.u64.u32 %rd73, %r33; + // inline asm + call (%rd71), _rt_buffer_get_id_64, (%r26, %r27, %r9, %rd72, %rd73, %rd28, %rd28); + // inline asm + ld.u8 %r34, [%rd71]; + and.b32 %r35, %r34, %r31; + setp.eq.s32 %p11, %r35, 0; + @%p11 bra BB0_8; + +BB0_3: + // inline asm + call (%r36), _rt_potential_intersection, (%f3); + // inline asm + setp.eq.s32 %p12, %r36, 0; + @%p12 bra BB0_8; + + // inline asm + call (%rd80), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd13, %rd28, %rd28, %rd28); + // inline asm + ld.f32 %f110, [%rd80+20]; + ld.f32 %f111, [%rd80+16]; + ld.f32 %f112, [%rd80+12]; + // inline asm + call (%rd86), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd19, %rd28, %rd28, %rd28); + // inline asm + ld.f32 %f113, [%rd86+20]; + ld.f32 %f114, [%rd86+16]; + ld.f32 %f115, [%rd86+12]; + // inline asm + call (%rd92), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd25, %rd28, %rd28, %rd28); + // inline asm + mov.f32 %f116, 0f3F800000; + sub.f32 %f117, %f116, %f1; + sub.f32 %f4, %f117, %f2; + mul.f32 %f118, %f1, %f115; + mul.f32 %f119, %f1, %f114; + mul.f32 %f120, %f1, %f113; + fma.rn.f32 %f121, %f4, %f112, %f118; + fma.rn.f32 %f122, %f4, %f111, %f119; + fma.rn.f32 %f123, %f4, %f110, %f120; + ld.f32 %f124, [%rd92+20]; + ld.f32 %f125, [%rd92+16]; + ld.f32 %f126, [%rd92+12]; + fma.rn.f32 %f127, %f2, %f126, %f121; + fma.rn.f32 %f128, %f2, %f125, %f122; + fma.rn.f32 %f129, %f2, %f124, %f123; + mul.f32 %f130, %f128, %f128; + fma.rn.f32 %f131, %f127, %f127, %f130; + fma.rn.f32 %f132, %f129, %f129, %f131; + sqrt.rn.f32 %f133, %f132; + rcp.rn.f32 %f134, %f133; + mul.f32 %f135, %f134, %f127; + mul.f32 %f136, %f134, %f128; + mul.f32 %f137, %f134, %f129; + st.global.f32 [normal], %f135; + st.global.f32 [normal+4], %f136; + st.global.f32 [normal+8], %f137; + // inline asm + call (%rd98), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd13, %rd28, %rd28, %rd28); + // inline asm + ld.v2.f32 {%f150, %f139}, [%rd98+24]; + // inline asm + call (%rd104), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd19, %rd28, %rd28, %rd28); + // inline asm + ld.v2.f32 {%f140, %f141}, [%rd104+24]; + // inline asm + call (%rd110), _rt_buffer_get_64, (%rd12, %r9, %r10, %rd25, %rd28, %rd28, %rd28); + // inline asm + setp.lt.f32 %p13, %f139, 0f00000000; + @%p13 bra BB0_6; + bra.uni BB0_5; + +BB0_6: + add.f32 %f151, %f139, 0f3F800000; + bra.uni BB0_7; + +BB0_5: + ld.v2.f32 {%f142, %f143}, [%rd110+24]; + mul.f32 %f146, %f1, %f140; + mul.f32 %f147, %f1, %f141; + fma.rn.f32 %f148, %f4, %f150, %f146; + fma.rn.f32 %f149, %f4, %f139, %f147; + fma.rn.f32 %f150, %f2, %f142, %f148; + fma.rn.f32 %f151, %f2, %f143, %f149; + +BB0_7: + st.global.v2.f32 [texCoords], {%f150, %f151}; + mov.u32 %r50, 0; + // inline asm + call (%r49), _rt_report_intersection, (%r50); + // inline asm + +BB0_8: + ret; +} + + // .globl _Z4bboxiPN5optix4AabbE +.visible .entry _Z4bboxiPN5optix4AabbE( + .param .u32 _Z4bboxiPN5optix4AabbE_param_0, + .param .u64 _Z4bboxiPN5optix4AabbE_param_1 +) +{ + .reg .pred %p<6>; + .reg .f32 %f<42>; + .reg .b32 %r<11>; + .reg .b64 %rd<29>; + + + ld.param.u64 %rd2, [_Z4bboxiPN5optix4AabbE_param_1]; + ld.param.s32 %rd5, [_Z4bboxiPN5optix4AabbE_param_0]; + mov.u64 %rd27, index_buffer; + cvta.global.u64 %rd4, %rd27; + mov.u32 %r7, 1; + mov.u32 %r2, 12; + mov.u64 %rd26, 0; + // inline asm + call (%rd3), _rt_buffer_get_64, (%rd4, %r7, %r2, %rd5, %rd26, %rd26, %rd26); + // inline asm + ld.s32 %rd11, [%rd3]; + mov.u64 %rd28, vertex_buffer; + cvta.global.u64 %rd10, %rd28; + ld.s32 %rd17, [%rd3+4]; + ld.s32 %rd23, [%rd3+8]; + mov.u32 %r8, 32; + // inline asm + call (%rd9), _rt_buffer_get_64, (%rd10, %r7, %r8, %rd11, %rd26, %rd26, %rd26); + // inline asm + ld.f32 %f3, [%rd9+8]; + ld.f32 %f2, [%rd9+4]; + ld.f32 %f1, [%rd9]; + // inline asm + call (%rd15), _rt_buffer_get_64, (%rd10, %r7, %r8, %rd17, %rd26, %rd26, %rd26); + // inline asm + ld.f32 %f6, [%rd15+8]; + ld.f32 %f5, [%rd15+4]; + ld.f32 %f4, [%rd15]; + // inline asm + call (%rd21), _rt_buffer_get_64, (%rd10, %r7, %r8, %rd23, %rd26, %rd26, %rd26); + // inline asm + sub.f32 %f11, %f4, %f1; + sub.f32 %f12, %f5, %f2; + sub.f32 %f13, %f6, %f3; + ld.f32 %f9, [%rd21+8]; + ld.f32 %f8, [%rd21+4]; + ld.f32 %f7, [%rd21]; + sub.f32 %f14, %f7, %f1; + sub.f32 %f15, %f8, %f2; + sub.f32 %f16, %f9, %f3; + mul.f32 %f17, %f12, %f16; + mul.f32 %f18, %f13, %f15; + sub.f32 %f19, %f17, %f18; + mul.f32 %f20, %f13, %f14; + mul.f32 %f21, %f11, %f16; + sub.f32 %f22, %f20, %f21; + mul.f32 %f23, %f11, %f15; + mul.f32 %f24, %f12, %f14; + sub.f32 %f25, %f23, %f24; + mul.f32 %f26, %f22, %f22; + fma.rn.f32 %f27, %f19, %f19, %f26; + fma.rn.f32 %f28, %f25, %f25, %f27; + sqrt.rn.f32 %f10, %f28; + mov.pred %p5, 0; + setp.leu.f32 %p4, %f10, 0f00000000; + @%p4 bra BB1_2; + + abs.f32 %f29, %f10; + setp.neu.f32 %p5, %f29, 0f7F800000; + +BB1_2: + cvta.to.global.u64 %rd1, %rd2; + @%p5 bra BB1_4; + bra.uni BB1_3; + +BB1_4: + min.f32 %f30, %f1, %f4; + min.f32 %f31, %f30, %f7; + min.f32 %f32, %f2, %f5; + min.f32 %f33, %f32, %f8; + min.f32 %f34, %f3, %f6; + min.f32 %f35, %f34, %f9; + st.global.f32 [%rd1], %f31; + st.global.f32 [%rd1+4], %f33; + st.global.f32 [%rd1+8], %f35; + max.f32 %f36, %f1, %f4; + max.f32 %f37, %f36, %f7; + max.f32 %f38, %f2, %f5; + max.f32 %f39, %f38, %f8; + max.f32 %f40, %f3, %f6; + max.f32 %f41, %f40, %f9; + st.global.f32 [%rd1+12], %f37; + st.global.f32 [%rd1+16], %f39; + st.global.f32 [%rd1+20], %f41; + bra.uni BB1_5; + +BB1_3: + mov.u32 %r9, 2096152002; + st.global.u32 [%rd1], %r9; + st.global.u32 [%rd1+8], %r9; + st.global.u32 [%rd1+4], %r9; + mov.u32 %r10, -51331646; + st.global.u32 [%rd1+12], %r10; + st.global.u32 [%rd1+16], %r10; + st.global.u32 [%rd1+20], %r10; + +BB1_5: + ret; +} + + |