name
string
code
string
asm
string
file
string
embree::avx::HeuristicArraySpatialSAH<embree::avx::QuadSplitterFactory, embree::PrimRef, 32ul, 16ul>::parallel_spatial_find(embree::avx::PrimInfoExtRange const&, unsigned long)
parallel_spatial_find(const PrimInfoExtRange& set, const size_t logBlockSize) { SpatialBinner binner(empty); const SpatialBinMapping<SPATIAL_BINS> mapping(set); const SpatialBinMapping<SPATIAL_BINS>& _mapping = mapping; // CLANG 3.4 parser bug workaround binner = parallel_reduce(set.begin(),set.end(),PARALLEL_FIND_BLOCK_SIZE,binner, [&] (const range<size_t>& r) -> SpatialBinner { SpatialBinner binner(empty); binner.bin2(splitterFactory,prims0,r.begin(),r.end(),_mapping); return binner; }, [&] (const SpatialBinner& b0, const SpatialBinner& b1) -> SpatialBinner { return SpatialBinner::reduce(b0,b1); }); /* todo: best spatial split not exceeding the extended range does not provide any benefit ?*/ return binner.best(mapping,logBlockSize); //,set.ext_size()); }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x40, %rsp subq $0x1180, %rsp # imm = 0x1180 movq %rcx, %rbx movq %rdi, %r14 movl $0x700, %eax # imm = 0x700 leaq 0x190(%rsp), %rcx vbroadcastss 0xc1e5b0(%rip), %xmm0 # 0x1eeba20 vbroadcastss 0xc1f70b(%rip), %xmm1 # 0x1eecb84 vxorps %xmm2, %xmm2, %xmm2 vmovaps %xmm0, -0x10(%rcx) vmovaps %xmm1, (%rcx) vmovaps %xmm0, -0x30(%rcx) vmovaps %xmm1, -0x20(%rcx) vmovaps %xmm0, -0x50(%rcx) vmovaps %xmm1, -0x40(%rcx) vmovaps %xmm2, 0x140(%rsp,%rax) vmovaps %xmm2, 0x40(%rsp,%rax) addq $0x60, %rcx addq $0x10, %rax cmpq $0x800, %rax # imm = 0x800 jne 0x12cd47d vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vbroadcastss 0xc539f9(%rip), %xmm2 # 0x1f20ec4 vandps %xmm2, %xmm0, %xmm3 vandps %xmm2, %xmm1, %xmm2 vmaxps %xmm2, %xmm3, %xmm2 vbroadcastss 0xc5ab50(%rip), %xmm3 # 0x1f28030 vmulps %xmm3, %xmm2, %xmm2 vsubps %xmm0, %xmm1, %xmm1 vmaxps %xmm1, %xmm2, %xmm3 vbroadcastss 0xc33c5b(%rip), %xmm4 # 0x1f01150 vdivps %xmm3, %xmm4, %xmm3 vcmpnleps %xmm2, %xmm1, %xmm1 vandps %xmm3, %xmm1, %xmm1 leaq 0x70(%rsp), %rax vmovaps %xmm1, 0x10(%rax) vmovaps %xmm0, (%rax) vbroadcastss 0xc1f1fb(%rip), %xmm0 # 0x1eec714 vdivps %xmm1, %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rax) leaq 0x50(%rsp), %r15 movq %rsi, (%r15) movq %rax, 0x8(%r15) leaq 0xc0(%rsp), %r12 movw $0x401, 0xc(%r12) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r12) movq $0x8, 0x40(%r12) vpermilps $0x4e, 0x40(%rdx), %xmm0 # xmm0 = mem[2,3,0,1] vmovaps %xmm0, 0x40(%rsp) movq %r12, %rdi callq 0x6a660 leaq 0xa0(%rsp), %rsi vmovaps 0x40(%rsp), %xmm0 vmovaps %xmm0, (%rsi) movq $0x400, 0x10(%rsi) # imm = 0x400 leaq 0x3f(%rsp), %r8 leaq 0x60(%rsp), %rcx movq %r8, (%rcx) movq %r15, 0x8(%rcx) leaq 0x940(%rsp), %r13 leaq 0x140(%rsp), %r15 movq %r13, %rdi movq %r15, %rdx movq %r12, %r9 callq 0x12cd6d0 leaq 0xc0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12cd66d leaq 0xc0(%rsp), %rdi callq 0x6aab0 xorl %eax, %eax movl $0x10, %ecx vmovaps -0x10(%r13,%rcx), %xmm0 vmovaps %xmm0, -0x10(%r15,%rcx) vmovaps (%r13,%rcx), %xmm0 vmovaps %xmm0, (%r15,%rcx) addq $0x20, %rcx cmpq $0x70, %rcx jne 0x12cd5d9 incq %rax addq $0x60, %r13 addq $0x60, %r15 cmpq $0x10, %rax jne 0x12cd5d4 leaq 0xf40(%rsp), %rsi movl $0x100, %edx # imm = 0x100 leaq 0x740(%rsp), %rdi callq 0x6a0f0 leaq 0x1040(%rsp), %rsi movl $0x100, %edx # imm = 0x100 leaq 0x840(%rsp), %rdi callq 0x6a0f0 leaq 0x140(%rsp), %rsi leaq 0x70(%rsp), %rdx movq %r14, %rdi movq %rbx, %rcx callq 0x1ef0bc movq %r14, %rax leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xc1e46f(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe57578(%rip), %rsi # 0x2124c08 movq 0xe57331(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12cd6b3 jmp 0x12cd6c8 jmp 0x12cd6b0 movq %rax, %rbx leaq 0xc0(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/heuristic_spatial_array.h
embree::avx::HeuristicArraySpatialSAH<embree::avx::QuadSplitterFactory, embree::PrimRef, 32ul, 16ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)
__noinline std::pair<size_t,size_t> parallel_object_split(const ObjectSplit& split, const PrimInfoExtRange& set, PrimInfoExtRange& lset, PrimInfoExtRange& rset) { const size_t begin = set.begin(); const size_t end = set.end(); PrimInfo left(empty); PrimInfo right(empty); const unsigned int splitPos = split.pos; const unsigned int splitDim = split.dim; const unsigned int splitDimMask = (unsigned int)1 << splitDim; const typename ObjectBinner::vint vSplitPos(splitPos); const typename ObjectBinner::vbool vSplitMask(splitDimMask); auto isLeft = [&] (const PrimRef &ref) { return split.mapping.bin_unsafe(ref,vSplitPos,vSplitMask); }; const size_t center = parallel_partitioning( prims0,begin,end,EmptyTy(),left,right,isLeft, [] (PrimInfo &pinfo,const PrimRef &ref) { pinfo.add_center2(ref,ref.lower.u >> (32-RESERVED_NUM_SPATIAL_SPLITS_GEOMID_BITS)); }, [] (PrimInfo &pinfo0,const PrimInfo &pinfo1) { pinfo0.merge(pinfo1); }, PARALLEL_PARTITION_BLOCK_SIZE); const size_t left_weight = left.end; const size_t right_weight = right.end; left.begin = begin; left.end = center; right.begin = center; right.end = end; new (&lset) PrimInfoExtRange(begin,center,center,left); new (&rset) PrimInfoExtRange(center,end,end,right); assert(area(left.geomBounds) >= 0.0f); assert(area(right.geomBounds) >= 0.0f); return std::pair<size_t,size_t>(left_weight,right_weight); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xf8, %rsp movq %rcx, %r14 movq 0x40(%rdx), %r15 movq 0x48(%rdx), %rbx vbroadcastss 0xc1afe3(%rip), %xmm0 # 0x1eeba20 leaq 0xa0(%rsp), %r13 vmovaps %xmm0, (%r13) vbroadcastss 0xc1c130(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%r13) vmovaps %xmm0, 0x20(%r13) vmovaps %xmm1, 0x30(%r13) vxorps %xmm2, %xmm2, %xmm2 vmovaps %xmm2, 0x40(%r13) leaq 0x50(%rsp), %r12 vmovaps %xmm0, (%r12) vmovaps %xmm1, 0x10(%r12) vmovaps %xmm0, 0x20(%r12) vmovaps %xmm1, 0x30(%r12) vmovaps %xmm2, 0x40(%r12) movb 0x4(%rsi), %cl movl $0x1, %eax shlq %cl, %rax movq %r8, %rbp vpermilps $0x0, 0x8(%rsi), %xmm0 # xmm0 = mem[0,0,0,0] leaq 0x20(%rsp), %rcx vmovaps %xmm0, (%rcx) shlq $0x4, %rax leaq 0xe7f4c0(%rip), %rdx # 0x214ff80 vmovaps (%rdx,%rax), %xmm0 leaq 0x10(%rsp), %rax vmovaps %xmm0, (%rax) leaq 0x38(%rsp), %r10 movq %rsi, (%r10) movq %rcx, 0x8(%r10) movq %rax, 0x10(%r10) movq (%rdi), %rdi leaq 0xd(%rsp), %rax leaq 0xe(%rsp), %r11 leaq 0xf(%rsp), %rcx movq %r15, %rsi movq %rbx, %rdx movq %r13, %r8 movq %r12, %r9 pushq $0x80 pushq %rax pushq %r11 pushq %r10 callq 0x12d2ede addq $0x20, %rsp movq 0x48(%r13), %rcx movq 0x48(%r12), %rdx movq %r15, 0x40(%r13) movq %rax, 0x48(%r13) movq %rax, 0x40(%r12) movq %rbx, 0x48(%r12) vmovaps (%r13), %xmm0 vmovaps %xmm0, (%r14) vmovaps 0x10(%r13), %xmm0 vmovaps %xmm0, 0x10(%r14) vmovaps 0x20(%r13), %xmm0 vmovaps %xmm0, 0x20(%r14) vmovaps 0x30(%r13), %xmm0 vmovaps %xmm0, 0x30(%r14) movq %r15, 0x40(%r14) movq %rax, 0x48(%r14) movq %rax, 0x50(%r14) vmovaps (%r12), %xmm0 vmovaps %xmm0, (%rbp) vmovaps 0x10(%r12), %xmm0 vmovaps %xmm0, 0x10(%rbp) vmovaps 0x20(%r12), %xmm0 vmovaps %xmm0, 0x20(%rbp) vmovaps 0x30(%r12), %xmm0 vmovaps %xmm0, 0x30(%rbp) movq %rax, 0x40(%rbp) movq %rbx, 0x48(%rbp) movq %rbx, 0x50(%rbp) movq %rcx, %rax addq $0xf8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/embree[P]embree/kernels/bvh/../builders/heuristic_spatial_array.h
embree::BVHNodeRecordMB4D<embree::NodeRefPtr<4>> const embree::avx::BVHBuilderMSMBlur::build<embree::NodeRefPtr<4>, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::TriangleMesh, embree::TriangleMi<4>>, embree::Scene::BuildProgressMonitorInterface>(embree::vector_t<embree::PrimRefMB, embree::aligned_monitored_allocator<embree::PrimRefMB, 16ul>>&, embree::PrimInfoMBT<embree::LBBox<embree::Vec3fa>> const&, embree::MemoryMonitorInterface*, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::TriangleMesh, embree::TriangleMi<4>>, embree::Scene::BuildProgressMonitorInterface, embree::avx::BVHBuilderMSMBlur::Settings const&)
static const BVHNodeRecordMB4D<NodeRef> build(mvector<PrimRefMB>& prims, const PrimInfoMB& pinfo, MemoryMonitorInterface* device, const RecalculatePrimRef recalculatePrimRef, const CreateAllocFunc createAlloc, const CreateNodeFunc createNode, const SetNodeFunc setNode, const CreateLeafFunc createLeaf, const ProgressMonitorFunc progressMonitor, const Settings& settings) { typedef BuilderT< NodeRef, RecalculatePrimRef, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, SetNodeFunc, CreateLeafFunc, ProgressMonitorFunc> Builder; Builder builder(device, recalculatePrimRef, createAlloc, createNode, setNode, createLeaf, progressMonitor, settings); return builder(prims,pinfo); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 movq %r9, %r10 movq %r8, %rax movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx movq 0x1b0(%rsp), %r8 movq 0x1c0(%rsp), %rdx movq 0x1b8(%rsp), %rsi leaq 0xe28744(%rip), %rdi # 0x2102970 leaq 0x38(%rsp), %r9 movq %rdi, (%r9) movq 0x8(%rsi), %rsi movq %rsi, 0x8(%r9) movq %rdx, (%rsp) leaq 0x48(%rsp), %r12 movq %r12, %rdi movq %rcx, %rsi movq %rax, %rdx movq %r10, %rcx callq 0x12da2f6 vmovups (%r15), %ymm0 vmovups 0x20(%r15), %ymm1 vmovups 0x40(%r15), %ymm2 vmovups 0x70(%r15), %ymm3 leaq 0xd0(%rsp), %rdx vmovups %ymm3, 0x80(%rdx) movq $0x1, (%rdx) vmovups %ymm0, 0x10(%rdx) vmovups %ymm1, 0x30(%rdx) vmovups %ymm2, 0x50(%rdx) vmovaps 0x60(%r15), %xmm0 vmovaps %xmm0, 0x70(%rdx) movq %r14, 0xa0(%rdx) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) movq $0x0, 0x30(%rsp) movq 0x30(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x20(%rsp), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r12, %rsi movl $0x1, %ecx vzeroupper callq 0x12da410 mfence movq %rbx, %rax addq $0x188, %rsp # imm = 0x188 popq %rbx popq %r12 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, 2ul>::TemporalBinInfo::best(int, embree::BBox<float>, embree::SetMB const&)
Split best(int logBlockSize, BBox1f time_range, const SetMB& set) { float bestSAH = inf; float bestPos = 0.0f; for (int b=0; b<BINS-1; b++) { float t = float(b+1)/float(BINS); float ct = lerp(time_range.lower,time_range.upper,t); const float center_time = set.align_time(ct); if (center_time <= time_range.lower) continue; if (center_time >= time_range.upper) continue; const BBox1f dt0(time_range.lower,center_time); const BBox1f dt1(center_time,time_range.upper); /* calculate sah */ const size_t lCount = (count0[b]+(size_t(1) << logBlockSize)-1) >> int(logBlockSize); const size_t rCount = (count1[b]+(size_t(1) << logBlockSize)-1) >> int(logBlockSize); float sah0 = expectedApproxHalfArea(bounds0[b])*float(lCount)*dt0.size(); float sah1 = expectedApproxHalfArea(bounds1[b])*float(rCount)*dt1.size(); if (unlikely(lCount == 0)) sah0 = 0.0f; // happens for initial splits when objects not alive over entire shutter time if (unlikely(rCount == 0)) sah1 = 0.0f; const float sah = sah0+sah1; if (sah < bestSAH) { bestSAH = sah; bestPos = center_time; } } return Split(bestSAH*MBLUR_TIME_SPLIT_THRESHOLD,(unsigned)Split::SPLIT_TEMPORAL,0,bestPos); }
movq %rcx, %r8 movl %edx, %ecx movq %rdi, %rax vmovshdup %xmm0, %xmm1 # xmm1 = xmm0[1,1,3,3] vmovss 0xc0c3c8(%rip), %xmm2 # 0x1eecb80 vmulss %xmm2, %xmm1, %xmm3 vmulss %xmm2, %xmm0, %xmm4 vaddss %xmm3, %xmm4, %xmm4 vmovss 0x80(%r8), %xmm3 vsubss %xmm3, %xmm4, %xmm5 vmovss 0x84(%r8), %xmm4 vsubss %xmm3, %xmm4, %xmm4 vdivss %xmm4, %xmm5, %xmm5 movq 0x78(%r8), %rdx testq %rdx, %rdx js 0x12e07f2 vcvtsi2ss %rdx, %xmm6, %xmm6 jmp 0x12e0807 movq %rdx, %rdi shrq %rdi andl $0x1, %edx orq %rdi, %rdx vcvtsi2ss %rdx, %xmm6, %xmm6 vaddss %xmm6, %xmm6, %xmm6 vmulss %xmm6, %xmm5, %xmm5 vbroadcastss 0xc406ac(%rip), %xmm7 # 0x1f20ec0 vandps %xmm7, %xmm5, %xmm7 vbroadcastss 0xc436ff(%rip), %xmm8 # 0x1f23f20 vorps %xmm7, %xmm8, %xmm7 vaddss %xmm7, %xmm5, %xmm5 vroundss $0xb, %xmm5, %xmm5, %xmm5 vdivss %xmm6, %xmm5, %xmm5 vmulss %xmm4, %xmm5, %xmm4 movl $0x1, %edi shlq %cl, %rdi vaddss %xmm3, %xmm4, %xmm3 movq 0x8(%rsi), %rdx addq %rdi, %rdx decq %rdx shrq %cl, %rdx testq %rdx, %rdx js 0x12e085c vcvtsi2ss %rdx, %xmm9, %xmm4 jmp 0x12e0875 movq %rdx, %r8 shrq %r8 movl %edx, %r9d andl $0x1, %r9d orq %r8, %r9 vcvtsi2ss %r9, %xmm9, %xmm4 vaddss %xmm4, %xmm4, %xmm4 vucomiss %xmm3, %xmm0 vmovss 0xc0b19f(%rip), %xmm5 # 0x1eeba20 vxorps %xmm6, %xmm6, %xmm6 jae 0x12e09ac vucomiss %xmm1, %xmm3 jae 0x12e09ac movl %ecx, %ecx movq (%rsi), %r8 addq %r8, %rdi decq %rdi shrq %cl, %rdi movq %rdi, %r8 shrq %r8 movl %edi, %ecx andl $0x1, %ecx orq %r8, %rcx testq %rdi, %rdi js 0x12e08bd vcvtsi2ss %rdi, %xmm9, %xmm5 jmp 0x12e08c6 vcvtsi2ss %rcx, %xmm9, %xmm5 vaddss %xmm5, %xmm5, %xmm5 je 0x12e09c4 vmovaps 0x20(%rsi), %xmm6 vsubps 0x10(%rsi), %xmm6, %xmm6 vmovaps 0x40(%rsi), %xmm7 vsubps 0x30(%rsi), %xmm7, %xmm7 vshufpd $0x1, %xmm6, %xmm6, %xmm8 # xmm8 = xmm6[1,0] vinsertps $0x4c, %xmm6, %xmm7, %xmm9 # xmm9 = xmm6[1],xmm7[1],zero,zero vinsertps $0x9c, %xmm7, %xmm8, %xmm8 # xmm8 = xmm8[0],xmm7[2],zero,zero vaddps %xmm8, %xmm9, %xmm10 vinsertps $0x1c, %xmm7, %xmm6, %xmm6 # xmm6 = xmm6[0],xmm7[0],zero,zero vmulps %xmm6, %xmm10, %xmm6 vmulps %xmm8, %xmm9, %xmm7 vaddps %xmm7, %xmm6, %xmm6 vhaddps %xmm6, %xmm6, %xmm6 vsubss %xmm0, %xmm3, %xmm0 vmulss 0xc0c267(%rip), %xmm6, %xmm6 # 0x1eecb80 vmulss %xmm5, %xmm6, %xmm5 vmulss %xmm5, %xmm0, %xmm0 testq %rdx, %rdx je 0x12e09cd vmovaps 0x60(%rsi), %xmm5 vsubps 0x50(%rsi), %xmm5, %xmm5 vmovaps 0x80(%rsi), %xmm6 vsubps 0x70(%rsi), %xmm6, %xmm6 vmovshdup %xmm5, %xmm7 # xmm7 = xmm5[1,1,3,3] vshufpd $0x1, %xmm5, %xmm5, %xmm8 # xmm8 = xmm5[1,0] vmovshdup %xmm6, %xmm9 # xmm9 = xmm6[1,1,3,3] vshufpd $0x1, %xmm6, %xmm6, %xmm10 # xmm10 = xmm6[1,0] vaddss %xmm7, %xmm8, %xmm11 vaddss %xmm10, %xmm9, %xmm12 vmulss %xmm7, %xmm8, %xmm7 vmulss %xmm5, %xmm11, %xmm5 vaddss %xmm7, %xmm5, %xmm5 vmulss %xmm10, %xmm9, %xmm7 vmulss %xmm6, %xmm12, %xmm6 vaddss %xmm7, %xmm6, %xmm6 vaddss %xmm6, %xmm5, %xmm5 vmulss %xmm2, %xmm5, %xmm2 vmulss %xmm4, %xmm2, %xmm2 vsubss %xmm3, %xmm1, %xmm1 vmulss %xmm2, %xmm1, %xmm1 vaddss %xmm1, %xmm0, %xmm0 vmovss 0xc0b08b(%rip), %xmm5 # 0x1eeba20 vucomiss %xmm5, %xmm0 vcmpneq_oqss %xmm5, %xmm0, %xmm1 vandps %xmm3, %xmm1, %xmm6 je 0x12e09ac vmulss 0xc1e6b4(%rip), %xmm0, %xmm5 # 0x1eff060 vmovss %xmm5, (%rax) movl $0x0, 0x4(%rax) vmovss %xmm6, 0x8(%rax) movl $0x2, 0xc(%rax) retq vxorps %xmm0, %xmm0, %xmm0 jmp 0x12e0921 vxorps %xmm1, %xmm1, %xmm1 jmp 0x12e0989 nop
/embree[P]embree/kernels/bvh/../builders/heuristic_timesplit_array.h
embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, 2ul>::TemporalBinInfo::merge2(embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, 2ul>::TemporalBinInfo const&, embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, 2ul>::TemporalBinInfo const&)
static __forceinline const TemporalBinInfo merge2(const TemporalBinInfo& a, const TemporalBinInfo& b) { TemporalBinInfo r = a; r.merge(b); return r; }
movq %rdi, %rax vmovaps (%rsi), %xmm0 vmovaps %xmm0, (%rdi) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, 0x10(%rdi) vmovaps 0x20(%rsi), %xmm1 vmovaps %xmm1, 0x20(%rdi) vmovaps 0x30(%rsi), %xmm2 vmovaps %xmm2, 0x30(%rdi) vmovaps 0x40(%rsi), %xmm2 vmovaps %xmm2, 0x40(%rdi) vmovaps 0x50(%rsi), %xmm2 vmovaps %xmm2, 0x50(%rdi) vmovaps 0x60(%rsi), %xmm3 vmovaps %xmm3, 0x60(%rdi) vmovaps 0x70(%rsi), %xmm3 vmovaps %xmm3, 0x70(%rdi) vmovaps 0x80(%rsi), %xmm3 vmovaps %xmm3, 0x80(%rdi) vmovdqa (%rdi), %xmm3 vpaddq (%rdx), %xmm3, %xmm3 vmovdqa %xmm3, (%rdi) vminps 0x10(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x10(%rdi) vmaxps 0x20(%rdx), %xmm1, %xmm0 vmovaps %xmm0, 0x20(%rdi) vmovaps 0x30(%rdi), %xmm0 vminps 0x30(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x30(%rdi) vmovaps 0x40(%rdi), %xmm0 vmaxps 0x40(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x40(%rdi) vminps 0x50(%rdx), %xmm2, %xmm0 vmovaps %xmm0, 0x50(%rdi) vmovaps 0x60(%rdi), %xmm0 vmaxps 0x60(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rdi) vmovaps 0x70(%rdi), %xmm0 vminps 0x70(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x70(%rdi) vmovaps 0x80(%rdi), %xmm0 vmaxps 0x80(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x80(%rdi) retq
/embree[P]embree/kernels/bvh/../builders/heuristic_timesplit_array.h
unsigned long embree::parallel_filter<embree::PrimRefMB, unsigned long, embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, 2ul>::split(embree::avx::BinSplit<32ul> const&, embree::SetMB const&, embree::SetMB&, embree::SetMB&)::'lambda'(embree::PrimRefMB const&)>(embree::PrimRefMB*, unsigned long, unsigned long, unsigned long, embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, 2ul>::split(embree::avx::BinSplit<32ul> const&, embree::SetMB const&, embree::SetMB&, embree::SetMB&)::'lambda'(embree::PrimRefMB const&) const&)
inline Index parallel_filter( Ty* data, const Index begin, const Index end, const Index minStepSize, const Predicate& predicate) { /* sequential fallback */ if (end-begin <= minStepSize) return sequential_filter(data,begin,end,predicate); /* calculate number of tasks to use */ enum { MAX_TASKS = 64 }; const Index numThreads = TaskScheduler::threadCount(); const Index numBlocks = (end-begin+minStepSize-1)/minStepSize; const Index taskCount = min(numThreads,numBlocks,(Index)MAX_TASKS); /* filter blocks */ Index nused[MAX_TASKS]; Index nfree[MAX_TASKS]; parallel_for(taskCount, [&](const Index taskIndex) { const Index i0 = begin+(taskIndex+0)*(end-begin)/taskCount; const Index i1 = begin+(taskIndex+1)*(end-begin)/taskCount; const Index i2 = sequential_filter(data,i0,i1,predicate); nused[taskIndex] = i2-i0; nfree[taskIndex] = i1-i2; }); /* calculate offsets */ Index sused=0; Index sfree=0; Index pfree[MAX_TASKS]; for (Index i=0; i<taskCount; i++) { sused+=nused[i]; Index cfree = nfree[i]; pfree[i] = sfree; sfree+=cfree; } /* return if we did not filter out any element */ assert(sfree <= end-begin); assert(sused <= end-begin); if (sused == end-begin) return end; /* otherwise we have to copy misplaced elements around */ parallel_for(taskCount, [&](const Index taskIndex) { /* destination to write elements to */ Index dst = begin+(taskIndex+0)*(end-begin)/taskCount+nused[taskIndex]; Index dst_end = min(dst+nfree[taskIndex],begin+sused); if (dst_end <= dst) return; /* range of misplaced elements to copy to destination */ Index r0 = pfree[taskIndex]; Index r1 = r0+dst_end-dst; /* find range in misplaced elements in back to front order */ Index k0=0; for (Index i=taskCount-1; i>0; i--) { if (k0 > r1) break; Index k1 = k0+nused[i]; Index src = begin+(i+0)*(end-begin)/taskCount+nused[i]; for (Index i=max(r0,k0); i<min(r1,k1); i++) { Index isrc = src-i+k0-1; assert(dst >= begin && dst < end); assert(isrc >= begin && isrc < end); data[dst++] = data[isrc]; } k0 = k1; } }); return begin+sused; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x708, %rsp # imm = 0x708 movq %r8, %rbx movq %rdi, 0x28(%rsp) movq %rsi, 0x10(%rsp) movq %rdx, 0x20(%rsp) movq %rdx, %rax subq %rsi, %rax cmpq %rcx, %rax jbe 0x12e2740 movq %rcx, %r14 xorl %edi, %edi callq 0x6ab80 movslq %eax, %rcx leaq 0x10(%rsp), %rsi movq (%rsi), %rax leaq 0x20(%rsp), %r13 movq (%r13), %rdx addq %r14, %rdx notq %rax addq %rdx, %rax xorl %edx, %edx divq %r14 cmpq %rax, %rcx cmovaeq %rax, %rcx cmpq $0x40, %rcx movl $0x40, %r14d cmovbq %rcx, %r14 leaq 0x30(%rsp), %rbp movq %r14, (%rbp) leaq 0x80(%rsp), %r12 movq %rsi, (%r12) movq %r13, 0x8(%r12) movq %rbp, 0x10(%r12) leaq 0x28(%rsp), %rax movq %rax, 0x18(%r12) movq %rbx, 0x20(%r12) leaq 0x500(%rsp), %rax movq %rax, 0x28(%r12) leaq 0x300(%rsp), %rax movq %rax, 0x30(%r12) leaq 0x100(%rsp), %r15 movw $0x401, 0xc(%r15) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r15) movq $0x8, 0x40(%r15) movq %r15, %rdi callq 0x6a660 leaq 0x40(%rsp), %rbx movq %r12, (%rbx) leaq 0x38(%rsp), %r8 movl $0x1, %edx xorl %edi, %edi movq %r14, %rsi movq %rbx, %rcx movq %r15, %r9 callq 0x12e3f00 leaq 0x100(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12e28c6 leaq 0x100(%rsp), %r15 movq %r15, %rdi callq 0x6aab0 movq $0x0, 0x18(%rsp) movq 0x30(%rsp), %r14 testq %r14, %r14 je 0x12e27dd xorl %ecx, %ecx xorl %edx, %edx xorl %eax, %eax addq 0x500(%rsp,%rcx,8), %rax movq %rdx, 0x100(%rsp,%rcx,8) addq 0x300(%rsp,%rcx,8), %rdx incq %rcx cmpq %rcx, %r14 jne 0x12e271b jmp 0x12e27df cmpq %rsi, %rdx jbe 0x12e28b1 leaq (%rsi,%rsi,4), %rcx shlq $0x4, %rcx addq %rdi, %rcx addq $0x44, %rcx vmovss 0xc1c904(%rip), %xmm0 # 0x1eff064 vmovss 0xc1c900(%rip), %xmm1 # 0x1eff068 movq (%rbx), %rdx vmovss (%rcx), %xmm2 vmulss %xmm0, %xmm2, %xmm3 vmovss (%rdx), %xmm4 vucomiss %xmm3, %xmm4 jae 0x12e27cf vmovss -0x4(%rcx), %xmm3 vmulss %xmm1, %xmm3, %xmm4 vucomiss 0x4(%rdx), %xmm4 jae 0x12e27cf leaq (%rsi,%rsi,4), %rdx incq %rsi shlq $0x4, %rdx vmovaps -0x44(%rcx), %xmm4 vmovaps %xmm4, (%rdi,%rdx) vmovaps -0x34(%rcx), %xmm4 vmovaps %xmm4, 0x10(%rdi,%rdx) vmovaps -0x24(%rcx), %xmm4 vmovaps %xmm4, 0x20(%rdi,%rdx) vmovaps -0x14(%rcx), %xmm4 vmovaps %xmm4, 0x30(%rdi,%rdx) vmovss %xmm3, 0x40(%rdi,%rdx) vmovss %xmm2, 0x44(%rdi,%rdx) addq $0x50, %rcx decq %rax jne 0x12e2768 jmp 0x12e28b1 xorl %eax, %eax movq %rax, 0x18(%rsp) movq 0x20(%rsp), %rsi movq %rsi, %rcx subq 0x10(%rsp), %rcx cmpq %rcx, %rax je 0x12e28b1 leaq 0x10(%rsp), %rax movq %rax, 0x40(%rsp) movq %r13, 0x48(%rsp) movq %rbp, 0x50(%rsp) leaq 0x500(%rsp), %rax movq %rax, 0x58(%rsp) leaq 0x300(%rsp), %rax movq %rax, 0x60(%rsp) leaq 0x18(%rsp), %rax movq %rax, 0x68(%rsp) movq %r15, 0x70(%rsp) leaq 0x28(%rsp), %rax movq %rax, 0x78(%rsp) leaq 0x80(%rsp), %r15 movw $0x401, 0xc(%r15) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r15) movq $0x8, 0x40(%r15) movq %r15, %rdi callq 0x6a660 leaq 0x38(%rsp), %rcx movq %rbx, (%rcx) leaq 0xf(%rsp), %r8 movl $0x1, %edx xorl %edi, %edi movq %r14, %rsi movq %r15, %r9 callq 0x12e4794 leaq 0x80(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12e28f8 leaq 0x80(%rsp), %rdi callq 0x6aab0 movq 0x18(%rsp), %rsi addq 0x10(%rsp), %rsi movq %rsi, %rax addq $0x708, %rsp # imm = 0x708 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xc09216(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe4231f(%rip), %rsi # 0x2124c08 movq 0xe420d8(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xc091e4(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe422ed(%rip), %rsi # 0x2124c08 movq 0xe420a6(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12e294b jmp 0x12e2978 jmp 0x12e2948 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12e2963 movq %rax, %rbx leaq 0x80(%rsp), %rdi callq 0x6aab0 jmp 0x12e2970 jmp 0x12e2978 jmp 0x12e2978 jmp 0x12e2960 movq %rax, %rbx leaq 0x100(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/../../common/algorithms/parallel_filter.h
embree::avx::BVHNBuilderMBlurSAH<4, embree::TriangleMesh, embree::TriangleMvMB<4>>::buildMultiSegment(unsigned long)
void buildMultiSegment(size_t numPrimitives) { /* create primref array */ mvector<PrimRefMB> prims(scene->device,numPrimitives); PrimInfoMB pinfo = createPrimRefArrayMSMBlur(scene,gtype_,numPrimitives,prims,bvh->scene->progressInterface); /* early out if no valid primitives */ if (pinfo.size() == 0) { bvh->clear(); return; } /* estimate acceleration structure size */ const size_t node_bytes = pinfo.num_time_segments*sizeof(AABBNodeMB)/(4*N); const size_t leaf_bytes = size_t(1.2*Primitive::blocks(pinfo.num_time_segments)*sizeof(Primitive)); bvh->alloc.init_estimate(node_bytes+leaf_bytes); /* settings for BVH build */ BVHBuilderMSMBlur::Settings settings; settings.branchingFactor = N; settings.maxDepth = BVH::maxDepth; settings.logBlockSize = bsr(sahBlockSize); settings.minLeafSize = min(minLeafSize,maxLeafSize); settings.maxLeafSize = maxLeafSize; settings.travCost = travCost; settings.intCost = intCost; settings.singleLeafTimeSegment = Primitive::singleTimeSegment; settings.singleThreadThreshold = bvh->alloc.fixSingleThreadThreshold(N,DEFAULT_SINGLE_THREAD_THRESHOLD,pinfo.size(),node_bytes+leaf_bytes); /* build hierarchy */ auto root = BVHBuilderMSMBlur::build<NodeRef>(prims,pinfo,scene->device, RecalculatePrimRef<Mesh>(scene), typename BVH::CreateAlloc(bvh), typename BVH::AABBNodeMB4D::Create(), typename BVH::AABBNodeMB4D::Set(), CreateMSMBlurLeaf<N,Mesh,Primitive>(bvh), bvh->scene->progressInterface, settings); bvh->set(root.ref,root.lbounds,pinfo.num_time_segments); }
pushq %r15 pushq %r14 pushq %rbx subq $0x170, %rsp # imm = 0x170 movq %rsi, %r14 movq %rdi, %rbx movq 0x18(%rdi), %rax movq 0x188(%rax), %rax leaq 0x550(%rax), %rdi testq %rax, %rax cmoveq %rax, %rdi movq %rdi, 0x8(%rsp) movb $0x0, 0x10(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x18(%rsp) movq $0x0, 0x28(%rsp) testq %rsi, %rsi je 0x12e8b76 movq %r14, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %r15 movq (%rdi), %rax movq %r15, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %r15 # imm = 0x1C00000 jb 0x12e8b5a leaq 0x10(%rsp), %rsi movq %r15, %rdi callq 0x1ee665d jmp 0x12e8b67 movl $0x10, %esi movq %r15, %rdi callq 0x1ee60ac movq %rax, 0x28(%rsp) movq %r14, 0x18(%rsp) movq %r14, 0x20(%rsp) leaq 0x18(%rsp), %r15 movq 0x10(%rbx), %rax movq 0x18(%rbx), %rsi movq 0x40(%rbx), %rdx movl $0x260, %r9d # imm = 0x260 addq 0x68(%rax), %r9 vmovsd 0xc03b57(%rip), %xmm0 # 0x1eec6f0 leaq 0x80(%rsp), %rdi leaq 0x8(%rsp), %r8 movq %r14, %rcx callq 0x14878ca movq 0xe8(%rsp), %rax cmpq 0xe0(%rsp), %rax jne 0x12e8bcf movq 0x10(%rbx), %rdi movq (%rdi), %rax callq *0x28(%rax) jmp 0x12e8d6e movq 0xf0(%rsp), %rax imulq $0xe0, %rax, %rcx addq $0x3, %rax shrq $0x2, %rax vcvtsi2sd %rax, %xmm1, %xmm0 vmulsd 0xc1b565(%rip), %xmm0, %xmm0 # 0x1f04158 vmulsd 0xc23d5d(%rip), %xmm0, %xmm0 # 0x1f0c958 vcvttsd2si %xmm0, %rax shrq $0x4, %rcx movq %rax, %rdx vsubsd 0xc073d1(%rip), %xmm0, %xmm0 # 0x1eeffe0 sarq $0x3f, %rdx vcvttsd2si %xmm0, %r14 andq %rdx, %r14 orq %rax, %r14 addq %rcx, %r14 movq 0x10(%rbx), %rdi addq $0x78, %rdi movq %r14, %rsi callq 0x90acb8 vmovaps 0xc391e7(%rip), %ymm0 # 0x1f21e20 vmovups %ymm0, 0x30(%rsp) movq $0x8, 0x50(%rsp) movl $0x3f800000, 0x58(%rsp) # imm = 0x3F800000 movb $0x0, 0x60(%rsp) movq $0x400, 0x68(%rsp) # imm = 0x400 movq $0x4, 0x30(%rsp) movq $0x50, 0x38(%rsp) movq 0x20(%rbx), %rax bsrq %rax, %rax movq %rax, 0x40(%rsp) movq 0x30(%rbx), %rax movq 0x38(%rbx), %rcx cmpq %rcx, %rax cmovaeq %rcx, %rax movq %rax, 0x48(%rsp) movq %rcx, 0x50(%rsp) vbroadcastss 0xc03a75(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x58(%rsp) vmovss 0x28(%rbx), %xmm0 vmovss %xmm0, 0x5c(%rsp) movb $0x1, 0x60(%rsp) movq 0x10(%rbx), %rdi movq 0xe8(%rsp), %rcx subq 0xe0(%rsp), %rcx addq $0x78, %rdi movl $0x4, %esi movl $0x400, %edx # imm = 0x400 movq %r14, %r8 vzeroupper callq 0xbdde9e leaq 0x30(%rsp), %r10 movq %rax, 0x38(%r10) movq 0x10(%rbx), %rax movq 0x18(%rbx), %r8 movq 0x188(%r8), %rdx leaq 0x78(%rax), %r9 leaq 0x550(%rdx), %rcx testq %rdx, %rdx cmoveq %rdx, %rcx movq 0x68(%rax), %rdx leaq 0xe19c59(%rip), %rsi # 0x2102970 leaq 0x70(%rsp), %r11 movq %rsi, (%r11) movq 0x268(%rdx), %rdx movq %rdx, 0x8(%r11) subq $0x8, %rsp leaq 0x118(%rsp), %rdi leaq 0x10(%rsp), %rsi leaq 0x88(%rsp), %rdx pushq %r10 pushq %r11 pushq %rax callq 0x12e8e62 addq $0x20, %rsp movq 0x10(%rbx), %rdi leaq 0x120(%rsp), %rdx movq -0x10(%rdx), %rsi movq 0xf0(%rsp), %rcx callq 0xb5ccb2 movq 0x20(%rsp), %rbx movq 0x28(%rsp), %rdi testq %rdi, %rdi je 0x12e8da2 movq %rbx, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x12e8d9d movzbl 0x10(%rsp), %edx callq 0x1ee67ca jmp 0x12e8da2 callq 0x1ee612d testq %rbx, %rbx je 0x12e8dc1 movq 0x8(%rsp), %rdi shlq $0x4, %rbx leaq (%rbx,%rbx,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, (%r15) movq $0x0, 0x10(%r15) addq $0x170, %rsp # imm = 0x170 popq %rbx popq %r14 popq %r15 retq jmp 0x12e8de7 jmp 0x12e8de7 jmp 0x12e8de7 jmp 0x12e8de7 movq %rax, %rbx movq 0x20(%rsp), %r14 movq 0x28(%rsp), %rdi testq %rdi, %rdi jne 0x12e8e31 testq %r14, %r14 je 0x12e8e18 movq 0x8(%rsp), %rdi shlq $0x4, %r14 leaq (%r14,%r14,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, (%r15) movq $0x0, 0x10(%r15) movq %rbx, %rdi callq 0x6a600 movq %r14, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x12e8e51 movzbl 0x10(%rsp), %edx callq 0x1ee67ca jmp 0x12e8df9 callq 0x1ee612d jmp 0x12e8df9 jmp 0x12e8e5a movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_sah_mb.cpp
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<4>, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::TriangleMesh, embree::TriangleMvMB<4>>, embree::Scene::BuildProgressMonitorInterface>::find(embree::SetMB const&)
const Split find(const SetMB& set) { /* first try standard object split */ const Split object_split = heuristicObjectSplit.find(set,cfg.logBlockSize); const float object_split_sah = object_split.splitSAH(); /* test temporal splits only when object split was bad */ const float leaf_sah = set.leafSAH(cfg.logBlockSize); if (object_split_sah < 0.50f*leaf_sah) return object_split; /* do temporal splits only if the time range is big enough */ if (set.time_range.size() > 1.01f/float(set.max_num_time_segments)) { const Split temporal_split = heuristicTemporalSplit.find(set,cfg.logBlockSize); const float temporal_split_sah = temporal_split.splitSAH(); /* take temporal split if it improved SAH */ if (temporal_split_sah < object_split_sah) return temporal_split; } return object_split; }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %rbx addq $0x40, %rsi movq 0x10(%r15), %rcx leaq 0x10(%rsp), %r12 movq %r12, %rdi callq 0x12551d0 vmovss (%r12), %xmm6 vmovss 0x8c(%r14), %xmm0 vsubss 0x88(%r14), %xmm0, %xmm0 vmovaps 0x10(%r14), %xmm1 vmovaps 0x30(%r14), %xmm2 vsubps (%r14), %xmm1, %xmm1 vsubps 0x20(%r14), %xmm2, %xmm2 movq 0x10(%r15), %rcx vinsertps $0x4c, %xmm1, %xmm2, %xmm3 # xmm3 = xmm1[1],xmm2[1],zero,zero vshufpd $0x1, %xmm1, %xmm1, %xmm4 # xmm4 = xmm1[1,0] vinsertps $0x9c, %xmm2, %xmm4, %xmm4 # xmm4 = xmm4[0],xmm2[2],zero,zero vaddps %xmm4, %xmm3, %xmm5 vmulps %xmm4, %xmm3, %xmm3 vinsertps $0x1c, %xmm2, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm2[0],zero,zero vmulps %xmm5, %xmm1, %xmm1 vaddps %xmm3, %xmm1, %xmm1 vhaddps %xmm1, %xmm1, %xmm2 vmovss 0xc0088f(%rip), %xmm1 # 0x1eecb80 vmulss %xmm1, %xmm2, %xmm2 vmulss %xmm2, %xmm0, %xmm2 movq $-0x1, %rax shlq %cl, %rax notq %rax addq 0x70(%r14), %rax shrq %cl, %rax testq %rax, %rax js 0x12ec319 vcvtsi2ss %rax, %xmm7, %xmm3 jmp 0x12ec32e movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm7, %xmm3 vaddss %xmm3, %xmm3, %xmm3 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm1, %xmm2, %xmm1 vucomiss %xmm6, %xmm1 ja 0x12ec397 movq 0x78(%r14), %rax testq %rax, %rax js 0x12ec34c vcvtsi2ss %rax, %xmm7, %xmm1 jmp 0x12ec361 movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm7, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmovss 0xc12cf3(%rip), %xmm2 # 0x1eff05c vdivss %xmm1, %xmm2, %xmm1 vucomiss %xmm1, %xmm0 jbe 0x12ec397 addq $0x48, %r15 movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx vmovss %xmm6, 0xc(%rsp) callq 0x12e0502 vmovss 0xc(%rsp), %xmm0 vucomiss (%rbx), %xmm0 ja 0x12ec3c0 vmovaps 0x10(%rsp), %xmm0 vmovaps %xmm0, (%rbx) movq 0x20(%rsp), %rax movq %rax, 0x10(%rbx) vmovaps 0x30(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rbx) vmovaps 0x40(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rbx) movq %rbx, %rax addq $0x58, %rsp popq %rbx popq %r12 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::BVHNBuilderMBlurSAH<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()
void build() { /* skip build for empty scene */ const size_t numPrimitives = scene->getNumPrimitives(gtype_,true); if (numPrimitives == 0) { bvh->clear(); return; } double t0 = bvh->preBuild(TOSTRING(isa) "::BVH" + toString(N) + "BuilderMBlurSAH"); #if PROFILE profile(2,PROFILE_RUNS,numPrimitives,[&] (ProfileTimer& timer) { #endif //const size_t numTimeSteps = scene->getNumTimeSteps<typename Mesh::type_t,true>(); //const size_t numTimeSegments = numTimeSteps-1; assert(numTimeSteps > 1); /*if (numTimeSegments == 1) buildSingleSegment(numPrimitives); else*/ buildMultiSegment(numPrimitives); #if PROFILE }); #endif /* clear temporary data for static geometry */ bvh->cleanup(); bvh->postBuild(t0); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x68, %rsp movq %rdi, %rbx movq 0x18(%rdi), %rax movq 0x40(%rdi), %rcx btl $0x14, %ecx jb 0x12f0252 xorl %r14d, %r14d jmp 0x12f0259 movq 0x298(%rax), %r14 btl $0x15, %ecx jae 0x12f0266 addq 0x2a8(%rax), %r14 testb $0xf, %cl je 0x12f0272 addq 0x2c8(%rax), %r14 testl $0x77770, %ecx # imm = 0x77770 je 0x12f0281 addq 0x2b8(%rax), %r14 testl $0xe000000, %ecx # imm = 0xE000000 je 0x12f0290 addq 0x348(%rax), %r14 btl $0x17, %ecx jae 0x12f029d addq 0x2d8(%rax), %r14 btl $0x1d, %ecx jae 0x12f02aa addq 0x2e8(%rax), %r14 btl $0x1e, %ecx jae 0x12f02b7 addq 0x2f8(%rax), %r14 testl %ecx, %ecx jns 0x12f02c2 addq 0x308(%rax), %r14 btl $0x18, %ecx jae 0x12f02cf addq 0x318(%rax), %r14 btl $0x16, %ecx jae 0x12f02dc addq 0x328(%rax), %r14 movq 0x10(%rbx), %r15 testq %r14, %r14 je 0x12f034d leaq 0x58(%rsp), %r13 movq %r13, -0x10(%r13) leaq 0x48(%rsp), %r12 movl $0x1, %esi movq %r12, %rdi movl $0x2d, %edx callq 0x6a580 movq (%r12), %rax movb $0x38, (%rax) leaq 0xc31bda(%rip), %rcx # 0x1f21eed movl $0x8, %r8d movq %r12, %rdi xorl %esi, %esi xorl %edx, %edx callq 0x6a1a0 leaq 0x18(%rsp), %r12 movq %r12, -0x10(%r12) movq (%rax), %rdx movq %rax, %rcx addq $0x10, %rcx cmpq %rcx, %rdx je 0x12f0364 movq %rdx, 0x8(%rsp) movq (%rcx), %rdx movq %rdx, 0x18(%rsp) jmp 0x12f036e movq (%r15), %rax movq %r15, %rdi addq $0x68, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp jmpq *0x28(%rax) vmovups (%rcx), %xmm0 vmovups %xmm0, (%r12) movq 0x8(%rax), %rdx leaq 0x8(%rsp), %rdi movq %rdx, 0x8(%rdi) movq %rcx, (%rax) movq $0x0, 0x8(%rax) movb $0x0, 0x10(%rax) leaq 0xc1c621(%rip), %rsi # 0x1f0c9b2 callq 0x6a620 leaq 0x38(%rsp), %rbp movq %rbp, -0x10(%rbp) movq (%rax), %rdx movq %rax, %rcx addq $0x10, %rcx cmpq %rcx, %rdx je 0x12f03bd movq %rdx, 0x28(%rsp) movq (%rcx), %rdx movq %rdx, 0x38(%rsp) jmp 0x12f03c6 vmovups (%rcx), %xmm0 vmovups %xmm0, (%rbp) movq 0x8(%rax), %rdx leaq 0x28(%rsp), %rsi movq %rdx, 0x8(%rsi) movq %rcx, (%rax) movq $0x0, 0x8(%rax) movb $0x0, 0x10(%rax) movq %r15, %rdi callq 0x14cf1d6 vmovsd %xmm0, (%rsp) movq 0x28(%rsp), %rdi cmpq %rbp, %rdi je 0x12f03fe callq 0x6a4f0 movq 0x8(%rsp), %rdi cmpq %r12, %rdi je 0x12f040d callq 0x6a4f0 movq 0x48(%rsp), %rdi cmpq %r13, %rdi je 0x12f041c callq 0x6a4f0 movq %rbx, %rdi movq %r14, %rsi callq 0x12f0492 movq 0x10(%rbx), %rdi callq 0x1242658 movq 0x10(%rbx), %rdi vmovsd (%rsp), %xmm0 callq 0x14cf3c4 addq $0x68, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rax, %rbx movq 0x28(%rsp), %rdi cmpq %rbp, %rdi je 0x12f0464 callq 0x6a4f0 jmp 0x12f0464 movq %rax, %rbx movq 0x8(%rsp), %rdi cmpq %r12, %rdi je 0x12f0478 callq 0x6a4f0 jmp 0x12f0478 movq %rax, %rbx movq 0x48(%rsp), %rdi cmpq %r13, %rdi je 0x12f0487 callq 0x6a4f0 movq %rbx, %rdi callq 0x6a600 nop
/embree[P]embree/kernels/bvh/bvh_builder_sah_mb.cpp
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<8>, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeaf<8, embree::TriangleMesh, embree::TriangleMi<4>>, embree::Scene::BuildProgressMonitorInterface>::recurse(embree::avx::BVHBuilderMSMBlur::BuildRecord const&, embree::FastAllocator::CachedAllocator, bool)
const NodeRecordMB4D recurse(const BuildRecord& current, Allocator alloc, bool toplevel) { /* get thread local allocator */ if (!alloc) alloc = createAlloc(); /* call memory monitor function to signal progress */ if (toplevel && current.size() <= cfg.singleThreadThreshold) progressMonitor(current.size()); /*! find best split */ const Split csplit = find(current.prims); /*! compute leaf and split cost */ const float leafSAH = cfg.intCost*current.prims.leafSAH(cfg.logBlockSize); const float splitSAH = cfg.travCost*current.prims.halfArea()+cfg.intCost*csplit.splitSAH(); assert((current.size() == 0) || ((leafSAH >= 0) && (splitSAH >= 0))); /*! create a leaf node when threshold reached or SAH tells us to stop */ if (current.size() <= cfg.minLeafSize || current.depth+MIN_LARGE_LEAF_LEVELS >= cfg.maxDepth || (current.size() <= cfg.maxLeafSize && leafSAH <= splitSAH)) { current.prims.deterministic_order(); return createLargeLeaf(current,alloc); } /*! perform initial split */ SetMB lprims,rprims; std::unique_ptr<mvector<PrimRefMB>> new_vector = split(csplit,current.prims,lprims,rprims); bool hasTimeSplits = new_vector != nullptr; NodeRecordMB4D values[MAX_BRANCHING_FACTOR]; LocalChildList children(current); { BuildRecord lrecord(lprims,current.depth+1); BuildRecord rrecord(rprims,current.depth+1); children.split(0,lrecord,rrecord,std::move(new_vector)); } /*! split until node is full or SAH tells us to stop */ while (children.size() < cfg.branchingFactor) { /*! find best child to split */ float bestArea = neg_inf; ssize_t bestChild = -1; for (size_t i=0; i<children.size(); i++) { if (children[i].size() <= cfg.minLeafSize) continue; if (expectedApproxHalfArea(children[i].prims.geomBounds) > bestArea) { bestChild = i; bestArea = expectedApproxHalfArea(children[i].prims.geomBounds); } } if (bestChild == -1) break; /* perform split */ BuildRecord& brecord = children[bestChild]; BuildRecord lrecord(current.depth+1); BuildRecord rrecord(current.depth+1); Split csplit = find(brecord.prims); std::unique_ptr<mvector<PrimRefMB>> new_vector = split(csplit,brecord.prims,lrecord.prims,rrecord.prims); hasTimeSplits |= new_vector != nullptr; children.split(bestChild,lrecord,rrecord,std::move(new_vector)); } /* detect time_ranges that have shrunken */ for (size_t i=0; i<children.size(); i++) { const BBox1f c = children[i].prims.time_range; const BBox1f p = current.prims.time_range; hasTimeSplits |= c.lower > p.lower || c.upper < p.upper; } /* sort buildrecords for simpler shadow ray traversal */ //std::sort(&children[0],&children[children.size()],std::greater<BuildRecord>()); // FIXME: reduces traversal performance of bvh8.triangle4 (need to verified) !! /*! create an inner node */ auto node = createNode(children.children.data(), children.numChildren, alloc, hasTimeSplits); LBBox3fa gbounds = empty; /* spawn tasks */ if (unlikely(current.size() > cfg.singleThreadThreshold)) { /*! parallel_for is faster than spawning sub-tasks */ parallel_for(size_t(0), children.size(), [&] (const range<size_t>& r) { for (size_t i=r.begin(); i<r.end(); i++) { values[i] = recurse(children[i],nullptr,true); _mm_mfence(); // to allow non-temporal stores during build } }); /*! merge bounding boxes */ for (size_t i=0; i<children.size(); i++) gbounds.extend(values[i].lbounds); } /* recurse into each child */ else { //for (size_t i=0; i<children.size(); i++) for (ssize_t i=children.size()-1; i>=0; i--) { values[i] = recurse(children[i],alloc,false); gbounds.extend(values[i].lbounds); } } setNode(current,children.children.data(),node,values,children.numChildren); /* calculate geometry bounds of this node */ if (unlikely(hasTimeSplits)) return NodeRecordMB4D(node,current.prims.linearBounds(recalculatePrimRef),current.prims.time_range); else return NodeRecordMB4D(node,gbounds,current.prims.time_range); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1ab8, %rsp # imm = 0x1AB8 movl %ecx, %ebp movq %rdx, (%rsp) movq %rsi, %r14 movq %rdi, %r15 leaq 0x1af0(%rsp), %rax cmpq $0x0, (%rax) movq %rsi, 0x8(%rsp) jne 0x12f0b76 movq 0x60(%r14), %r14 leaq 0xe342c6(%rip), %rdi # 0x2124d38 callq 0x6a480 movq (%rax), %r12 testq %r12, %r12 jne 0x12f0b49 movq %rax, %rbx movl $0xc0, %edi movl $0x40, %esi callq 0x1ee60ac movq %rax, %r12 movq %rax, %rdi callq 0x1ee7b24 movq $0x0, 0x8(%r12) movq %r12, 0x40(%r12) vxorpd %xmm0, %xmm0, %xmm0 vmovupd %ymm0, 0x48(%r12) vmovupd %ymm0, 0x58(%r12) movq %r12, 0x80(%r12) vmovupd %ymm0, 0x88(%r12) vmovupd %ymm0, 0x98(%r12) movq %r12, (%rbx) leaq 0xe35d8f(%rip), %rdi # 0x2126878 movq %rdi, 0x720(%rsp) movb $0x1, 0x728(%rsp) vzeroupper callq 0x1ee7bb6 leaq 0x14b0(%rsp), %rsi movq %r12, (%rsi) leaq 0xe35d6d(%rip), %rdi # 0x2126880 callq 0x90d91e movq 0x14b0(%rsp), %rsi testq %rsi, %rsi je 0x12f0b32 leaq 0x14b0(%rsp), %rdi callq 0x90da94 cmpb $0x1, 0x728(%rsp) jne 0x12f0b49 movq 0x720(%rsp), %rdi callq 0x1ee7c24 leaq 0x40(%r12), %rax subq $-0x80, %r12 cmpb $0x0, 0x10a(%r14) cmovneq %rax, %r12 leaq 0x1af0(%rsp), %rcx movq %r14, (%rcx) movq %rax, 0x8(%rcx) movq %r12, 0x10(%rcx) movq 0x8(%rsp), %r14 testb %bpl, %bpl je 0x12f0bb2 movq (%rsp), %rcx movq 0x78(%rcx), %rax subq 0x70(%rcx), %rax cmpq 0x38(%r14), %rax ja 0x12f0bb2 vmovq %rax, %xmm0 vpunpckldq 0xbfbf26(%rip), %xmm0, %xmm0 # xmm0 = xmm0[0],mem[0],xmm0[1],mem[1] vsubpd 0xbfbf2e(%rip), %xmm0, %xmm0 # 0x1eecad0 movq 0x80(%r14), %rdi vhaddpd %xmm0, %xmm0, %xmm0 callq 0x91839e movq (%rsp), %rbx leaq 0x10(%rbx), %rdx leaq 0x6e0(%rsp), %r12 movq %r12, %rdi movq %r14, %rsi movq %rdx, 0x48(%rsp) callq 0x12f3cf0 movb 0x10(%r14), %cl movq $-0x1, %rax shlq %cl, %rax notq %rax addq 0x80(%rbx), %rax shrq %cl, %rax testq %rax, %rax movq %r15, 0x148(%rsp) js 0x12f0c05 vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rax, %xmm1, %xmm0 jmp 0x12f0c1e movq %rax, %rcx shrq %rcx andl $0x1, %eax orq %rcx, %rax vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rax, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 movq (%rsp), %rbx movq 0x78(%rbx), %r15 movq 0x70(%rbx), %rbp movq %r15, %r13 subq %rbp, %r13 cmpq 0x18(%r14), %r13 jbe 0x12f0cee movq (%rsp), %rax movq (%rax), %rax addq $0x8, %rax cmpq 0x8(%r14), %rax movq (%rsp), %rbx jae 0x12f0cee vmovss 0x98(%rbx), %xmm1 cmpq 0x20(%r14), %r13 ja 0x12f0d48 vmovaps 0x20(%rbx), %xmm2 vmovaps 0x40(%rbx), %xmm3 vsubps 0x10(%rbx), %xmm2, %xmm2 vsubps 0x30(%rbx), %xmm3, %xmm3 vmovshdup %xmm2, %xmm4 # xmm4 = xmm2[1,1,3,3] vshufpd $0x1, %xmm2, %xmm2, %xmm5 # xmm5 = xmm2[1,0] vmovshdup %xmm3, %xmm6 # xmm6 = xmm3[1,1,3,3] vshufpd $0x1, %xmm3, %xmm3, %xmm7 # xmm7 = xmm3[1,0] vaddss %xmm5, %xmm4, %xmm8 vaddss %xmm7, %xmm6, %xmm9 vmulss %xmm5, %xmm4, %xmm4 vmulss %xmm2, %xmm8, %xmm2 vaddss %xmm4, %xmm2, %xmm2 vmulss %xmm7, %xmm6, %xmm4 vmulss %xmm3, %xmm9, %xmm3 vaddss %xmm4, %xmm3, %xmm3 vaddss %xmm3, %xmm2, %xmm2 vmovss 0x9c(%rbx), %xmm3 vmulss 0xbfbec1(%rip), %xmm2, %xmm2 # 0x1eecb80 vsubss %xmm1, %xmm3, %xmm3 vmulss %xmm2, %xmm3, %xmm2 vmovss 0x2c(%r14), %xmm3 vmulss %xmm0, %xmm2, %xmm0 vmulss 0x6e0(%rsp), %xmm3, %xmm4 vmulss 0x28(%r14), %xmm2, %xmm2 vmulss %xmm0, %xmm3, %xmm0 vaddss %xmm4, %xmm2, %xmm2 vucomiss %xmm0, %xmm2 jb 0x12f0d48 movq 0x48(%rsp), %rdi callq 0xba71d2 subq $0x20, %rsp leaq 0x1b10(%rsp), %rcx movq 0x10(%rcx), %rax movq %rax, 0x10(%rsp) vmovups (%rcx), %xmm0 vmovups %xmm0, (%rsp) movq 0x168(%rsp), %r12 movq %r12, %rdi movq %r14, %rsi movq %rbx, %rdx callq 0x12f3e5e addq $0x20, %rsp movq %r12, %rax addq $0x1ab8, %rsp # imm = 0x1AB8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movl 0x6ec(%rsp), %eax testl %eax, %eax jne 0x12f2d0e vbroadcastss 0xbfacc0(%rip), %xmm0 # 0x1eeba20 leaq 0x720(%rsp), %r14 vmovaps %xmm0, (%r14) vbroadcastss 0xbfbe0e(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%r14) vmovaps %xmm0, 0x20(%r14) vmovaps %xmm1, 0x30(%r14) vmovaps %xmm0, 0x40(%r14) vmovaps %xmm1, 0x50(%r14) vxorps %xmm2, %xmm2, %xmm2 vmovups %ymm2, 0x60(%r14) xorl %eax, %eax movl %eax, 0x80(%r14) vbroadcastss 0xbfb964(%rip), %xmm3 # 0x1eec714 vmovlps %xmm3, 0x84(%r14) movl %eax, 0x8c(%r14) leaq 0x14b0(%rsp), %rbx vmovaps %xmm0, (%rbx) vmovaps %xmm1, 0x10(%rbx) vmovaps %xmm0, 0x20(%rbx) vmovaps %xmm1, 0x30(%rbx) vmovaps %xmm0, 0x40(%rbx) vmovaps %xmm1, 0x50(%rbx) vmovups %ymm2, 0x60(%rbx) movl %eax, 0x80(%rbx) vmovlps %xmm3, 0x84(%rbx) movl %eax, 0x8c(%rbx) vpermilps $0x0, 0x6e8(%rsp), %xmm0 # xmm0 = mem[0,0,0,0] leaq 0x2b0(%rsp), %rax vmovaps %xmm0, (%rax) movb 0x6e4(%rsp), %cl movl $0x1, %edx shll %cl, %edx movslq %edx, %rcx shlq $0x4, %rcx leaq 0xe5f14f(%rip), %rdx # 0x214ff80 vmovaps (%rdx,%rcx), %xmm0 leaq 0x220(%rsp), %rcx vmovaps %xmm0, (%rcx) leaq 0x90(%rsp), %r10 movq %r12, (%r10) movq %rax, 0x8(%r10) movq %rcx, 0x10(%r10) movq (%rsp), %r13 movq 0xa0(%r13), %rax movq 0x20(%rax), %rdi subq $0x8, %rsp movl $0x80, %eax leaq 0x3a8(%rsp), %r11 leaq 0x188(%rsp), %r12 leaq 0x5a8(%rsp), %rcx movq %rbp, %rsi movq %r15, %rdx movq %r14, %r8 movq %rbx, %r9 pushq $0xc00 # imm = 0xC00 pushq %rax pushq %r11 pushq %r12 pushq %r10 vzeroupper callq 0x125bcbe addq $0x30, %rsp movq 0xa0(%r13), %rcx vmovsd 0x98(%r13), %xmm0 vmovups (%r14), %ymm1 vmovups 0x20(%r14), %ymm2 vmovups 0x40(%r14), %ymm3 vmovups %ymm1, 0x500(%rsp) vmovups %ymm2, 0x520(%rsp) vmovups %ymm3, 0x540(%rsp) vmovaps 0x60(%r14), %xmm1 vmovaps %xmm1, 0x560(%rsp) vmovups 0x70(%r14), %ymm1 vmovups %ymm1, 0x570(%rsp) movq %rcx, 0x590(%rsp) movq %rbp, 0x560(%rsp) movq %rax, 0x568(%rsp) vmovsd 0x588(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm1, %xmm1 vmovlps %xmm1, 0x588(%rsp) vmovaps (%rbx), %xmm1 vmovaps %xmm1, 0x460(%rsp) vmovaps 0x10(%rbx), %xmm1 vmovaps %xmm1, 0x470(%rsp) vmovaps 0x20(%rbx), %xmm1 vmovaps %xmm1, 0x480(%rsp) vmovaps 0x30(%rbx), %xmm1 vmovaps %xmm1, 0x490(%rsp) vmovaps 0x40(%rbx), %xmm1 vmovaps %xmm1, 0x4a0(%rsp) vmovaps 0x50(%rbx), %xmm1 vmovaps %xmm1, 0x4b0(%rsp) vmovaps 0x60(%rbx), %xmm1 vmovaps %xmm1, 0x4c0(%rsp) vmovups 0x70(%rbx), %ymm1 vmovups %ymm1, 0x4d0(%rsp) movq %rcx, 0x4f0(%rsp) movq %rax, 0x4c0(%rsp) movq %r15, 0x4c8(%rsp) vmovsd 0x4e8(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovlps %xmm0, 0x4e8(%rsp) movq $0x0, 0x10(%rsp) cmpq $0x0, 0x10(%rsp) setne %r15b movl $0x1, %eax leaq 0x12a8(%rsp), %rdx movq %rax, -0x8(%rdx) movq %rax, 0x200(%rdx) movq (%rsp), %rax movq (%rax), %r14 movq %r14, -0xb88(%rdx) vmovaps 0x10(%rax), %xmm0 vmovaps %xmm0, -0xb78(%rdx) vmovaps 0x20(%rax), %xmm0 vmovaps %xmm0, -0xb68(%rdx) vmovaps 0x30(%rax), %xmm0 vmovaps %xmm0, -0xb58(%rdx) vmovaps 0x40(%rax), %xmm0 vmovaps %xmm0, -0xb48(%rdx) vmovaps 0x50(%rax), %xmm0 vmovaps %xmm0, -0xb38(%rdx) vmovaps 0x60(%rax), %xmm0 vmovaps %xmm0, -0xb28(%rdx) vmovups 0x70(%rax), %ymm0 vmovups %ymm0, -0xb18(%rdx) vmovaps 0x90(%rax), %xmm0 vmovaps %xmm0, -0xaf8(%rdx) movq 0xa0(%rax), %rax movq %rax, -0xae8(%rdx) movq %rax, (%rdx) movq $0x2, 0x8(%rdx) movq %rdx, -0x88(%rdx) vmovaps 0x500(%rsp), %xmm10 vmovaps 0x510(%rsp), %xmm11 vmovaps 0x520(%rsp), %xmm12 vmovaps 0x530(%rsp), %xmm13 vmovaps 0x540(%rsp), %xmm14 vmovaps 0x550(%rsp), %xmm15 vmovaps 0x560(%rsp), %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x570(%rsp), %xmm0 vmovaps %xmm0, 0x90(%rsp) vmovaps 0x580(%rsp), %xmm9 movq 0x590(%rsp), %r13 vmovaps 0x460(%rsp), %xmm2 vmovaps 0x470(%rsp), %xmm3 vmovaps 0x480(%rsp), %xmm4 vmovaps 0x490(%rsp), %xmm5 vmovaps 0x4a0(%rsp), %xmm6 vmovaps 0x4b0(%rsp), %xmm7 vmovaps 0x4c0(%rsp), %xmm8 vmovaps 0x4d0(%rsp), %xmm0 vmovaps %xmm0, 0x2b0(%rsp) vmovaps 0x4e0(%rsp), %xmm1 movq 0x4f0(%rsp), %r12 movq $0x0, 0x10(%rsp) cmpq %rax, %r13 je 0x12f11b7 leaq 0x12b8(%rsp), %rax movq $0x2, 0x1f0(%rax) movq %r13, (%rax) movq $0x1, 0x8(%rax) movq %rax, -0x98(%rax) jmp 0x12f11cb movq %rdx, 0x1220(%rsp) movq $0x3, 0x12b0(%rsp) cmpq 0x12a8(%rsp), %r12 je 0x12f1210 movq 0x14a8(%rsp), %rax leaq 0x1(%rax), %rcx movq %rcx, 0x14a8(%rsp) shlq $0x4, %rax leaq (%rdx,%rax), %rcx movq %r12, (%rdx,%rax) movq $0x1, 0x8(%rdx,%rax) movq 0x12a0(%rsp), %rax movq %rcx, 0x1220(%rsp,%rax,8) jmp 0x12f1228 movq 0x12a0(%rsp), %rax movq %rdx, 0x1220(%rsp,%rax,8) incq 0x12b0(%rsp) decq 0x12b0(%rsp) jne 0x12f13c0 movq 0x12a8(%rsp), %rbx testq %rbx, %rbx je 0x12f13c0 vmovaps %xmm15, 0x660(%rsp) vmovaps %xmm14, 0x670(%rsp) vmovaps %xmm13, 0x680(%rsp) vmovaps %xmm12, 0x690(%rsp) vmovaps %xmm11, 0x6a0(%rsp) vmovaps %xmm10, 0x6b0(%rsp) vmovaps %xmm9, 0x6c0(%rsp) vmovaps %xmm8, 0x6d0(%rsp) vmovaps %xmm7, 0x390(%rsp) vmovaps %xmm6, 0x170(%rsp) vmovaps %xmm5, 0x210(%rsp) vmovaps %xmm4, 0x80(%rsp) vmovaps %xmm3, 0x30(%rsp) vmovaps %xmm2, 0x20(%rsp) vmovaps %xmm1, 0x50(%rsp) movq 0x18(%rbx), %rbp movq 0x20(%rbx), %rdi testq %rdi, %rdi je 0x12f12fc movq %rbp, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x12f12f4 movzbl 0x8(%rbx), %edx vzeroupper callq 0x1ee67ca jmp 0x12f12fc vzeroupper callq 0x1ee612d testq %rbp, %rbp je 0x12f1323 movq (%rbx), %rdi shlq $0x4, %rbp leaq (,%rbp,4), %rsi addq %rbp, %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %rbx, %rdi vzeroupper callq 0x6a4f0 vmovaps 0x50(%rsp), %xmm1 vmovaps 0x20(%rsp), %xmm2 vmovaps 0x30(%rsp), %xmm3 vmovaps 0x80(%rsp), %xmm4 vmovaps 0x210(%rsp), %xmm5 vmovaps 0x170(%rsp), %xmm6 vmovaps 0x390(%rsp), %xmm7 vmovaps 0x6d0(%rsp), %xmm8 vmovaps 0x6c0(%rsp), %xmm9 vmovaps 0x6b0(%rsp), %xmm10 vmovaps 0x6a0(%rsp), %xmm11 vmovaps 0x690(%rsp), %xmm12 vmovaps 0x680(%rsp), %xmm13 vmovaps 0x670(%rsp), %xmm14 vmovaps 0x660(%rsp), %xmm15 incq %r14 leaq 0x798(%rsp), %rcx movq %r14, -0x78(%rcx) vmovaps %xmm10, -0x68(%rcx) vmovaps %xmm11, -0x58(%rcx) vmovaps %xmm12, -0x48(%rcx) vmovaps %xmm13, -0x38(%rcx) vmovaps %xmm14, -0x28(%rcx) vmovaps %xmm15, -0x18(%rcx) vmovaps 0x60(%rsp), %xmm0 vmovaps %xmm0, -0x8(%rcx) leaq 0xa0(%rsp), %rax vmovaps -0x10(%rax), %xmm0 vmovaps %xmm0, 0x8(%rcx) vmovaps %xmm9, 0x18(%rcx) movq %r13, 0x28(%rcx) imulq $0xb0, 0xb08(%rcx), %rax movq %r14, 0x720(%rsp,%rax) vmovaps %xmm2, 0x730(%rsp,%rax) vmovaps %xmm3, 0x740(%rsp,%rax) vmovaps %xmm4, 0x750(%rsp,%rax) vmovaps %xmm5, 0x760(%rsp,%rax) vmovaps %xmm6, 0x770(%rsp,%rax) vmovaps %xmm7, 0x780(%rsp,%rax) vmovaps %xmm8, 0x790(%rsp,%rax) leaq 0x2c0(%rsp), %rdx vmovaps -0x10(%rdx), %xmm0 vmovaps %xmm0, 0x7a0(%rsp,%rax) vmovaps %xmm1, 0x7b0(%rsp,%rax) movq %r12, 0x7c0(%rsp,%rax) incq 0xb08(%rcx) movq 0x8(%rsp), %rsi leaq 0x48(%rsi), %rax movq %rax, 0x210(%rsp) leaq 0x1e0(%rsp), %r12 movl $0x80, %eax leaq 0x90(%rsp), %rcx addq %rax, %rcx movq %rcx, 0x30(%rsp) leaq 0x2b0(%rsp), %rcx addq %rax, %rcx movq %rcx, 0x80(%rsp) vbroadcastss 0xbfa543(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x60(%rsp) vbroadcastss 0xbfb698(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x50(%rsp) movq 0x12a0(%rsp), %rax cmpq (%rsi), %rax jae 0x12f2338 testq %rax, %rax je 0x12f159b movq 0x18(%rsi), %rcx movq $-0x1, %r13 leaq 0x798(%rsp), %rdx vmovss 0xbfb65d(%rip), %xmm0 # 0x1eecb84 xorl %r8d, %r8d movq (%rdx), %rdi subq -0x8(%rdx), %rdi cmpq %rcx, %rdi jbe 0x12f158a vmovaps -0x58(%rdx), %xmm1 vmovaps -0x38(%rdx), %xmm2 vsubps -0x68(%rdx), %xmm1, %xmm1 vsubps -0x48(%rdx), %xmm2, %xmm2 vinsertps $0x4c, %xmm1, %xmm2, %xmm3 # xmm3 = xmm1[1],xmm2[1],zero,zero vshufpd $0x1, %xmm1, %xmm1, %xmm4 # xmm4 = xmm1[1,0] vinsertps $0x9c, %xmm2, %xmm4, %xmm4 # xmm4 = xmm4[0],xmm2[2],zero,zero vaddps %xmm4, %xmm3, %xmm5 vmulps %xmm4, %xmm3, %xmm3 vinsertps $0x1c, %xmm2, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm2[0],zero,zero vmulps %xmm5, %xmm1, %xmm1 vaddps %xmm3, %xmm1, %xmm1 vhaddps %xmm1, %xmm1, %xmm1 vmulss 0xbfb603(%rip), %xmm1, %xmm1 # 0x1eecb80 vucomiss %xmm0, %xmm1 jbe 0x12f158a movq %r8, %r13 vmovaps %xmm1, %xmm0 incq %r8 addq $0xb0, %rdx cmpq %r8, %rax jne 0x12f152a jmp 0x12f15a2 movq $-0x1, %r13 cmpq $-0x1, %r13 je 0x12f1c0a imulq $0xb0, %r13, %rbx movq (%rsp), %rax movq (%rax), %rax incq %rax movq %rax, 0x90(%rsp) movq %rax, 0x2b0(%rsp) leaq (%rsp,%rbx), %rbp addq $0x730, %rbp # imm = 0x730 leaq 0x620(%rsp), %rdi movq %rbp, %rdx vzeroupper callq 0x12f3cf0 leaq (%rsp,%rbx), %r14 addq $0x720, %r14 # imm = 0x720 movl 0x62c(%rsp), %eax testl %eax, %eax movb %r15b, 0x20(%rsp) jne 0x12f1c19 movq 0x70(%r14), %rbx movq %r12, %rbp movq 0x78(%r14), %r12 vbroadcastss 0xbfa402(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x220(%rsp) vbroadcastss 0xbfb554(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x230(%rsp) vmovaps %xmm0, 0x240(%rsp) vmovaps %xmm1, 0x250(%rsp) vmovaps %xmm0, 0x260(%rsp) vmovaps %xmm1, 0x270(%rsp) leaq 0x290(%rsp), %r15 movl $0x0, %eax movl %eax, 0x10(%r15) vxorps %xmm2, %xmm2, %xmm2 vmovups %ymm2, -0x10(%r15) vbroadcastss 0xbfb093(%rip), %xmm3 # 0x1eec714 vmovlps %xmm3, 0x2a4(%rsp) movl %eax, 0x2ac(%rsp) vmovaps %xmm0, 0x180(%rsp) vmovaps %xmm1, 0x190(%rsp) vmovaps %xmm0, 0x1a0(%rsp) vmovaps %xmm1, 0x1b0(%rsp) vmovaps %xmm0, 0x1c0(%rsp) vmovaps %xmm1, 0x1d0(%rsp) movl %eax, 0x20(%rbp) vmovups %ymm2, (%rbp) vmovlps %xmm3, 0x204(%rsp) movl %eax, 0x20c(%rsp) vpermilps $0x0, 0x628(%rsp), %xmm0 # xmm0 = mem[0,0,0,0] vmovaps %xmm0, 0x5a0(%rsp) movb 0x624(%rsp), %cl movl $0x1, %eax shll %cl, %eax cltq shlq $0x4, %rax leaq 0xe5e872(%rip), %rcx # 0x214ff80 vmovaps (%rcx,%rax), %xmm0 vmovaps %xmm0, 0x360(%rsp) leaq 0x620(%rsp), %rax movq %rax, 0x3a0(%rsp) leaq 0x5a0(%rsp), %rax movq %rax, 0x3a8(%rsp) leaq 0x360(%rsp), %rax movq %rax, 0x3b0(%rsp) movq 0xa0(%r14), %rax movq 0x20(%rax), %rdi subq $0x8, %rsp movq %rbx, %rsi movq %r12, %rdx leaq 0x458(%rsp), %rcx leaq 0x228(%rsp), %r8 leaq 0x188(%rsp), %r9 pushq $0xc00 # imm = 0xC00 pushq $0x80 leaq 0x450(%rsp), %rax pushq %rax leaq 0x178(%rsp), %rax pushq %rax leaq 0x3c8(%rsp), %rax pushq %rax vzeroupper callq 0x125bcbe addq $0x30, %rsp movq 0xa0(%r14), %rcx vmovsd 0x98(%r14), %xmm0 vmovups 0x220(%rsp), %ymm1 vmovups 0x240(%rsp), %ymm2 vmovups 0x260(%rsp), %ymm3 vmovups %ymm1, 0xa0(%rsp) vmovups %ymm2, 0xc0(%rsp) vmovups %ymm3, 0xe0(%rsp) vmovaps 0x280(%rsp), %xmm1 vmovaps %xmm1, 0x100(%rsp) vmovups (%r15), %ymm1 movq 0x30(%rsp), %rdx vmovups %ymm1, (%rdx) movq %rcx, 0x130(%rsp) movq %rbx, 0x100(%rsp) movq %rax, 0x108(%rsp) vmovsd 0x128(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm1, %xmm1 vmovlps %xmm1, 0x128(%rsp) vmovaps 0x180(%rsp), %xmm1 vmovaps %xmm1, 0x2c0(%rsp) vmovaps 0x190(%rsp), %xmm1 vmovaps %xmm1, 0x2d0(%rsp) vmovaps 0x1a0(%rsp), %xmm1 vmovaps %xmm1, 0x2e0(%rsp) vmovaps 0x1b0(%rsp), %xmm1 vmovaps %xmm1, 0x2f0(%rsp) vmovaps 0x1c0(%rsp), %xmm1 vmovaps %xmm1, 0x300(%rsp) vmovaps 0x1d0(%rsp), %xmm1 vmovaps %xmm1, 0x310(%rsp) vmovaps 0x1e0(%rsp), %xmm1 vmovaps %xmm1, 0x320(%rsp) vmovups 0x10(%rbp), %ymm1 movq 0x80(%rsp), %rdx vmovups %ymm1, (%rdx) movq %rcx, 0x350(%rsp) movq %rax, 0x320(%rsp) movq %r12, 0x328(%rsp) vmovsd 0x348(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovlps %xmm0, 0x348(%rsp) movq $0x0, 0x18(%rsp) leaq 0x12a8(%rsp), %rbx movb 0x20(%rsp), %r15b movq 0x18(%rsp), %r12 movq $0x0, 0x18(%rsp) movq 0x1220(%rsp,%r13,8), %rax movq 0x130(%rsp), %rcx cmpq (%rax), %rcx je 0x12f198d movq 0x14a8(%rsp), %rdx leaq 0x1(%rdx), %rsi movq %rsi, 0x14a8(%rsp) shlq $0x4, %rdx leaq (%rbx,%rdx), %rsi movq %rcx, (%rbx,%rdx) movq $0x1, 0x8(%rbx,%rdx) movq %rsi, 0x1220(%rsp,%r13,8) jmp 0x12f1991 incq 0x8(%rax) movq 0x350(%rsp), %rcx cmpq (%rax), %rcx je 0x12f19d9 movq 0x14a8(%rsp), %rdx leaq 0x1(%rdx), %rsi movq %rsi, 0x14a8(%rsp) shlq $0x4, %rdx leaq (%rbx,%rdx), %rsi movq %rcx, (%rbx,%rdx) movq $0x1, 0x8(%rbx,%rdx) movq 0x12a0(%rsp), %rcx movq %rsi, 0x1220(%rsp,%rcx,8) jmp 0x12f19ed movq 0x12a0(%rsp), %rcx movq %rax, 0x1220(%rsp,%rcx,8) incq 0x8(%rax) decq 0x8(%rax) jne 0x12f1a7c movq (%rax), %rbx testq %rbx, %rbx je 0x12f1a7c movq 0x18(%rbx), %rbp movq 0x20(%rbx), %rdi testq %rdi, %rdi je 0x12f1a36 movq %rbp, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x12f1a2e movzbl 0x8(%rbx), %edx vzeroupper callq 0x1ee67ca jmp 0x12f1a36 vzeroupper callq 0x1ee612d testq %rbp, %rbp je 0x12f1a5d movq (%rbx), %rdi shlq $0x4, %rbp leaq (,%rbp,4), %rsi addq %rbp, %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %rbx, %rdi vzeroupper callq 0x6a4f0 testq %r12, %r12 setne %bl movq 0x90(%rsp), %rax movq %rax, (%r14) vmovaps 0xa0(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r14) vmovaps 0xb0(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r14) vmovaps 0xc0(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r14) vmovaps 0xd0(%rsp), %xmm0 vmovaps %xmm0, 0x40(%r14) vmovaps 0xe0(%rsp), %xmm0 vmovaps %xmm0, 0x50(%r14) vmovaps 0xf0(%rsp), %xmm0 vmovaps %xmm0, 0x60(%r14) leaq 0xa0(%rsp), %rax vmovups 0x60(%rax), %ymm0 vmovups %ymm0, 0x70(%r14) vmovaps 0x120(%rsp), %xmm0 vmovaps %xmm0, 0x90(%r14) movq 0x130(%rsp), %rax movq %rax, 0xa0(%r14) imulq $0xb0, 0x12a0(%rsp), %rax movq 0x2b0(%rsp), %rcx movq %rcx, 0x720(%rsp,%rax) vmovaps 0x2c0(%rsp), %xmm0 vmovaps %xmm0, 0x730(%rsp,%rax) vmovaps 0x2d0(%rsp), %xmm0 vmovaps %xmm0, 0x740(%rsp,%rax) vmovaps 0x2e0(%rsp), %xmm0 vmovaps %xmm0, 0x750(%rsp,%rax) vmovaps 0x2f0(%rsp), %xmm0 vmovaps %xmm0, 0x760(%rsp,%rax) vmovaps 0x300(%rsp), %xmm0 vmovaps %xmm0, 0x770(%rsp,%rax) vmovaps 0x310(%rsp), %xmm0 vmovaps %xmm0, 0x780(%rsp,%rax) leaq 0x2c0(%rsp), %rcx vmovups 0x60(%rcx), %ymm0 vmovups %ymm0, 0x790(%rsp,%rax) vmovaps 0x340(%rsp), %xmm0 vmovaps %xmm0, 0x7b0(%rsp,%rax) movq 0x350(%rsp), %rcx movq %rcx, 0x7c0(%rsp,%rax) incq 0x12a0(%rsp) movq 0x18(%rsp), %rsi testq %rsi, %rsi je 0x12f1bfa leaq 0x18(%rsp), %rdi vzeroupper callq 0xbbb88e orb %bl, %r15b movq 0x8(%rsp), %rsi leaq 0x1e0(%rsp), %r12 cmpq $-0x1, %r13 jne 0x12f14f2 jmp 0x12f2338 cmpl $0x2, %eax jne 0x12f22d6 vmovss 0x628(%rsp), %xmm0 vmovss 0x98(%r14), %xmm1 vmovss %xmm1, 0x78(%rsp) vmovss %xmm0, 0x7c(%rsp) vmovss %xmm0, 0x70(%rsp) vmovss 0x9c(%r14), %xmm0 vmovss %xmm0, 0x74(%rsp) movq 0xa0(%r14), %rax movq %rax, 0x170(%rsp) movl $0x28, %edi callq 0x6a170 movq %rax, %r12 movq 0x210(%rsp), %rax movq (%rax), %rdi movq 0x78(%r14), %rax movq %rdi, (%r12) movb $0x0, 0x8(%r12) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%r12) movq $0x0, 0x20(%r12) subq 0x70(%r14), %rax movb 0x20(%rsp), %r15b je 0x12f1d01 movq %rax, 0x390(%rsp) shlq $0x4, %rax leaq (%rax,%rax,4), %rbx movq (%rdi), %rax movq %rbx, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %rbx # imm = 0x1C00000 jb 0x12f1cdd movq %r12, %rsi addq $0x8, %rsi movq %rbx, %rdi callq 0x1ee665d jmp 0x12f1cea movl $0x10, %esi movq %rbx, %rdi callq 0x1ee60ac movq %rax, 0x20(%r12) movq 0x390(%rsp), %rax movq %rax, 0x10(%r12) movq %rax, 0x18(%r12) movq %r12, 0x18(%rsp) movq %r12, 0x150(%rsp) movq 0x170(%rsp), %rax movq %rax, 0x360(%rsp) leaq 0x78(%rsp), %rax movq %rax, 0x368(%rsp) movq 0x210(%rsp), %rax movq %rax, 0x370(%rsp) leaq 0x150(%rsp), %rax movq %rax, 0x378(%rsp) movq %rbp, 0x380(%rsp) movq 0x70(%r14), %rbx movq 0x78(%r14), %r12 vbroadcastss 0xbf9cbc(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x180(%rsp) vbroadcastss 0xbfae0e(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x190(%rsp) vmovaps %xmm0, 0x1a0(%rsp) vmovaps %xmm1, 0x1b0(%rsp) vmovaps %xmm0, 0x1c0(%rsp) vmovaps %xmm1, 0x1d0(%rsp) leaq 0x1e0(%rsp), %rax xorl %ecx, %ecx movl %ecx, 0x20(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, (%rax) vbroadcastss 0xbfa953(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x204(%rsp) movl %ecx, 0x20c(%rsp) movq %r12, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x12f212e movq %rbx, 0x3a0(%rsp) movq %r12, 0x3a8(%rsp) leaq 0x220(%rsp), %rdi leaq 0x360(%rsp), %rsi leaq 0x3a0(%rsp), %rdx vzeroupper callq 0x12e2d2e movq 0x170(%rsp), %rbp movq 0x288(%rsp), %rax subq 0x280(%rsp), %rax movq 0x78(%r14), %rdx subq 0x70(%r14), %rdx cmpq %rdx, %rax leaq 0x290(%rsp), %rbx je 0x12f1e75 movq 0x150(%rsp), %rax movq 0x20(%rax), %rdi leaq 0x78(%rsp), %rax movq %rax, 0x180(%rsp) movl $0x400, %ecx # imm = 0x400 xorl %esi, %esi leaq 0x180(%rsp), %r8 callq 0x12e25d3 movq %rax, 0x288(%rsp) movq 0x150(%rsp), %rax vmovsd 0x78(%rsp), %xmm0 vmovups 0x220(%rsp), %ymm1 vmovups 0x240(%rsp), %ymm2 vmovups 0x260(%rsp), %ymm3 vmovaps (%rbx), %xmm4 movq 0x30(%rsp), %rcx vmovaps %xmm4, (%rcx) vmovups %ymm1, 0xa0(%rsp) vmovups %ymm2, 0xc0(%rsp) vmovups %ymm3, 0xe0(%rsp) movq 0x280(%rsp), %rcx movq %rcx, 0x100(%rsp) movq 0x288(%rsp), %rcx movq %rcx, 0x108(%rsp) vmovsd 0x2a0(%rsp), %xmm1 vmovsd 0x2a8(%rsp), %xmm2 vcmpltps %xmm0, %xmm2, %xmm3 vinsertps $0x50, %xmm2, %xmm0, %xmm4 # xmm4 = xmm0[0],xmm2[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm2, %xmm0 # xmm0 = xmm2[0],xmm0[1],xmm2[2,3] vblendvps %xmm3, %xmm4, %xmm0, %xmm0 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x120(%rsp) movq %rax, 0x130(%rsp) movq %rbp, 0x438(%rsp) leaq 0x70(%rsp), %rax movq %rax, 0x440(%rsp) movq 0x210(%rsp), %rax movq %rax, 0x448(%rsp) movq 0x70(%r14), %rbx movq 0x78(%r14), %r12 vbroadcastss 0xbf9ac6(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x3a0(%rsp) vbroadcastss 0xbfac18(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x3b0(%rsp) vmovaps %xmm0, 0x3c0(%rsp) vmovaps %xmm1, 0x3d0(%rsp) vmovaps %xmm0, 0x3e0(%rsp) vmovaps %xmm1, 0x3f0(%rsp) leaq 0x3c0(%rsp), %rax xorl %ecx, %ecx movl %ecx, 0x60(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x40(%rax) vbroadcastss 0xbfa75c(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x424(%rsp) movl %ecx, 0x42c(%rsp) movq %r12, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x12f21fe movq %rbx, 0x5a0(%rsp) movq %r12, 0x5a8(%rsp) leaq 0x180(%rsp), %rdi leaq 0x438(%rsp), %rsi leaq 0x5a0(%rsp), %rdx vzeroupper callq 0x12e5072 movq 0x1e8(%rsp), %rax subq 0x1e0(%rsp), %rax movq 0x70(%r14), %rsi leaq (%rax,%rsi), %rcx movq %rsi, 0x1e0(%rsp) movq %rcx, 0x1e8(%rsp) movq 0x78(%r14), %rdx movq %rdx, %rcx subq %rsi, %rcx cmpq %rcx, %rax leaq 0x12a8(%rsp), %rbx leaq 0x1e0(%rsp), %r12 je 0x12f207c movq 0x20(%rbp), %rdi leaq 0x70(%rsp), %rax movq %rax, 0x3a0(%rsp) movl $0x400, %ecx # imm = 0x400 leaq 0x3a0(%rsp), %r8 callq 0x12e2980 movq %rax, 0x1e8(%rsp) vmovsd 0x70(%rsp), %xmm0 vmovups 0x180(%rsp), %ymm1 vmovups 0x1a0(%rsp), %ymm2 vmovups 0x1c0(%rsp), %ymm3 vmovaps 0x10(%r12), %xmm4 movq 0x80(%rsp), %rax vmovaps %xmm4, (%rax) vmovups %ymm1, 0x2c0(%rsp) vmovups %ymm2, 0x2e0(%rsp) vmovups %ymm3, 0x300(%rsp) movq 0x1e0(%rsp), %rax movq %rax, 0x320(%rsp) movq 0x1e8(%rsp), %rax movq %rax, 0x328(%rsp) vmovsd 0x208(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovsd 0x200(%rsp), %xmm1 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x340(%rsp) movq %rbp, 0x350(%rsp) jmp 0x12f1937 movb $0x1, 0x3ac(%rsp) leaq 0x3c0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0x3e0(%rsp) movb $0x4, 0x3ad(%rsp) leaq 0x3a0(%rsp), %rdi vzeroupper callq 0x6a660 movq %r12, 0x5a0(%rsp) movq %rbx, 0x5a8(%rsp) movq $0x80, 0x5b0(%rsp) leaq -0x742041(%rip), %r8 # 0xbb014c movq %r8, 0x158(%rsp) leaq 0x360(%rsp), %rax movq %rax, 0x160(%rsp) leaq 0x220(%rsp), %rdi leaq 0x5a0(%rsp), %rsi leaq 0x180(%rsp), %rdx leaq 0x158(%rsp), %rcx leaq 0x3a0(%rsp), %r9 callq 0x12e3461 leaq 0x3a0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12f38b9 leaq 0x3a0(%rsp), %rdi callq 0x6aab0 movb 0x20(%rsp), %r15b jmp 0x12f1e13 movb $0x1, 0x5ac(%rsp) leaq 0x5c0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0x5e0(%rsp) movb $0x4, 0x5ad(%rsp) leaq 0x5a0(%rsp), %rdi vzeroupper callq 0x6a660 movq %r12, 0x158(%rsp) movq %rbx, 0x160(%rsp) movq $0x80, 0x168(%rsp) leaq -0x742111(%rip), %r8 # 0xbb014c movq %r8, 0x450(%rsp) leaq 0x438(%rsp), %rax movq %rax, 0x458(%rsp) leaq 0x180(%rsp), %rdi leaq 0x158(%rsp), %rsi leaq 0x3a0(%rsp), %rdx leaq 0x450(%rsp), %rcx leaq 0x5a0(%rsp), %r9 callq 0x12e5713 leaq 0x5a0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12f38eb leaq 0x5a0(%rsp), %rdi callq 0x6aab0 movb 0x20(%rsp), %r15b movq 0x170(%rsp), %rbp jmp 0x12f200a cmpl $0x1, %eax je 0x12f230e cmpl $0x3, %eax jne 0x12f1921 movq %rbp, %rdi callq 0xba71d2 movq 0x8(%rsp), %rdi movq %rbp, %rsi leaq 0xa0(%rsp), %rdx leaq 0x2c0(%rsp), %rcx callq 0x1f1a30 jmp 0x12f1921 movq %rbp, %rdi callq 0xba71d2 movq 0x8(%rsp), %rdi movq %rbp, %rsi leaq 0xa0(%rsp), %rdx leaq 0x2c0(%rsp), %rcx callq 0x12f6e60 jmp 0x12f1921 movq 0x12a0(%rsp), %rax testq %rax, %rax je 0x12f238b movq (%rsp), %rcx vmovss 0x98(%rcx), %xmm0 vmovss 0x9c(%rcx), %xmm1 leaq 0x7bc(%rsp), %rcx vmovss (%rcx), %xmm2 vmovss -0x4(%rcx), %xmm3 vcmpltps %xmm3, %xmm0, %xmm3 vcmpltps %xmm1, %xmm2, %xmm2 vorps %xmm2, %xmm3, %xmm2 vmovd %xmm2, %edx orb %dl, %r15b addq $0xb0, %rcx decq %rax jne 0x12f2361 leaq 0x1af0(%rsp), %rax movq (%rax), %rbp movq 0x8(%rax), %r14 testb $0x1, %r15b je 0x12f2476 movq $0x200, 0x180(%rsp) # imm = 0x200 movq (%r14), %rbx movq 0x8(%rbx), %rax cmpq %rax, %rbp je 0x12f2654 movq %rbx, 0x2b0(%rsp) movb $0x1, 0x2b8(%rsp) movq %rbx, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%rbx), %rax testq %rax, %rax je 0x12f2437 movq 0xa8(%rbx), %rax addq 0x68(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%rbx), %rax addq 0x98(%rbx), %rax movq 0x50(%rbx), %rcx addq 0x90(%rbx), %rcx subq %rcx, %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%rbx), %rax addq 0x70(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%rbx) vmovups %ymm0, 0x48(%rbx) testq %rbp, %rbp je 0x12f2545 movq 0x10(%rbp), %rax movq %rax, 0x60(%rbx) vmovups %ymm0, 0x88(%rbx) vmovups %ymm0, 0x98(%rbx) movq 0x10(%rbp), %rax movq %rax, 0xa0(%rbx) jmp 0x12f2555 movq $0x1c0, 0x180(%rsp) # imm = 0x1C0 movq (%r14), %rbx movq 0x8(%rbx), %rax cmpq %rax, %rbp je 0x12f2786 movq %rbx, 0x2b0(%rsp) movb $0x1, 0x2b8(%rsp) movq %rbx, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%rbx), %rax testq %rax, %rax je 0x12f2509 movq 0xa8(%rbx), %rax addq 0x68(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%rbx), %rax addq 0x98(%rbx), %rax movq 0x50(%rbx), %rcx addq 0x90(%rbx), %rcx subq %rcx, %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%rbx), %rax addq 0x70(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%rbx) vmovups %ymm0, 0x48(%rbx) testq %rbp, %rbp je 0x12f25a8 movq 0x10(%rbp), %rax movq %rax, 0x60(%rbx) vmovups %ymm0, 0x88(%rbx) vmovups %ymm0, 0x98(%rbx) movq 0x10(%rbp), %rax movq %rax, 0xa0(%rbx) jmp 0x12f25b8 vmovups %ymm0, 0x98(%rbx) vmovups %ymm0, 0x88(%rbx) movq %rbp, %rax xchgq %rax, 0x8(%rbx) movq %rbx, 0x220(%rsp) leaq 0xe3430d(%rip), %rdi # 0x2126878 movq %rdi, 0x90(%rsp) movb $0x1, 0x98(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%rbp), %rsi cmpq 0x140(%rbp), %rsi je 0x12f2612 movq 0x220(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%rbp) jmp 0x12f2626 vmovups %ymm0, 0x98(%rbx) vmovups %ymm0, 0x88(%rbx) movq %rbp, %rax xchgq %rax, 0x8(%rbx) movq %rbx, 0x220(%rsp) leaq 0xe342aa(%rip), %rdi # 0x2126878 movq %rdi, 0x90(%rsp) movb $0x1, 0x98(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%rbp), %rsi cmpq 0x140(%rbp), %rsi je 0x12f2744 movq 0x220(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%rbp) jmp 0x12f2758 leaq 0x130(%rbp), %rdi leaq 0x220(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x98(%rsp) jne 0x12f263d movq 0x90(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x2b8(%rsp) jne 0x12f2654 movq 0x2b0(%rsp), %rdi callq 0x1ee7c24 movq 0x180(%rsp), %rax addq %rax, 0x28(%r14) movq 0x10(%r14), %rcx movl %ecx, %edx negl %edx andl $0x1f, %edx leaq (%rcx,%rax), %r13 addq %rdx, %r13 movq %r13, 0x10(%r14) cmpq 0x18(%r14), %r13 ja 0x12f3027 addq %rdx, 0x30(%r14) subq %rax, %r13 addq 0x8(%r14), %r13 vbroadcastss 0xbf938c(%rip), %ymm0 # 0x1eeba20 vmovaps %ymm0, 0x1c0(%r13) vbroadcastss 0xbfa4de(%rip), %ymm1 # 0x1eecb84 vmovaps %ymm1, 0x1e0(%r13) vmovaps %ymm0, 0xc0(%r13) vmovaps %ymm0, 0x80(%r13) vmovaps %ymm0, 0x40(%r13) vmovaps %ymm1, 0xe0(%r13) vmovaps %ymm1, 0xa0(%r13) vmovaps %ymm1, 0x60(%r13) vxorps %xmm0, %xmm0, %xmm0 vmovaps %ymm0, 0x100(%r13) vmovaps %ymm0, 0x120(%r13) vmovaps %ymm0, 0x140(%r13) vmovaps %ymm0, 0x160(%r13) vmovaps %ymm0, 0x180(%r13) vmovaps %ymm0, 0x1a0(%r13) xorl %eax, %eax vbroadcastsd 0xc2f754(%rip), %ymm0 # 0x1f21e78 movq 0x8(%rsp), %rdx vmovups %ymm0, (%r13,%rax,8) addq $0x4, %rax cmpq $0x8, %rax jne 0x12f2729 movl $0x6, %eax jmp 0x12f285f leaq 0x130(%rbp), %rdi leaq 0x220(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x98(%rsp) jne 0x12f276f movq 0x90(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x2b8(%rsp) jne 0x12f2786 movq 0x2b0(%rsp), %rdi callq 0x1ee7c24 movq 0x180(%rsp), %rax addq %rax, 0x28(%r14) movq 0x10(%r14), %rcx movl %ecx, %edx negl %edx andl $0x1f, %edx leaq (%rcx,%rax), %r13 addq %rdx, %r13 movq %r13, 0x10(%r14) cmpq 0x18(%r14), %r13 ja 0x12f305e addq %rdx, 0x30(%r14) subq %rax, %r13 addq 0x8(%r14), %r13 vbroadcastss 0xbf925a(%rip), %ymm0 # 0x1eeba20 vmovaps %ymm0, 0xc0(%r13) vmovaps %ymm0, 0x80(%r13) vmovaps %ymm0, 0x40(%r13) vbroadcastss 0xbfa39d(%rip), %ymm0 # 0x1eecb84 vmovaps %ymm0, 0xe0(%r13) vmovaps %ymm0, 0xa0(%r13) vmovaps %ymm0, 0x60(%r13) vxorps %xmm0, %xmm0, %xmm0 vmovaps %ymm0, 0x100(%r13) vmovaps %ymm0, 0x120(%r13) vmovaps %ymm0, 0x140(%r13) vmovaps %ymm0, 0x160(%r13) vmovaps %ymm0, 0x180(%r13) vmovaps %ymm0, 0x1a0(%r13) xorl %eax, %eax vbroadcastsd 0xc2f634(%rip), %ymm0 # 0x1f21e78 movq 0x8(%rsp), %rdx vmovups %ymm0, (%r13,%rax,8) addq $0x4, %rax cmpq $0x8, %rax jne 0x12f2849 movl $0x1, %eax movq %rax, 0x80(%rsp) movq (%rsp), %rcx movq 0x78(%rcx), %rax subq 0x70(%rcx), %rax movq 0x12a0(%rsp), %rbp cmpq 0x38(%rdx), %rax ja 0x12f2dbd movq %rbp, %rax decq %rax vbroadcastss 0xbfa2f0(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x20(%rsp) vbroadcastss 0xbf917d(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x30(%rsp) js 0x12f2a0d leaq (,%rbp,2), %rax addq %rbp, %rax shlq $0x5, %rax leaq 0x14b0(%rsp), %rcx leaq (%rax,%rcx), %r12 addq $-0x10, %r12 imulq $0xb0, %rbp, %rax leaq 0x720(%rsp), %rcx leaq (%rax,%rcx), %rbx addq $-0xb0, %rbx vbroadcastss 0xbfa293(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xbf9126(%rip), %xmm1 # 0x1eeba20 leaq 0x90(%rsp), %r14 vmovaps %xmm1, 0x30(%rsp) vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x20(%rsp) vmovaps %xmm0, 0x50(%rsp) subq $0x20, %rsp leaq 0x1b10(%rsp), %rcx movq 0x10(%rcx), %rax movq %rax, 0x10(%rsp) vmovups (%rcx), %xmm0 vmovups %xmm0, (%rsp) movq %r14, %rdi movq 0x28(%rsp), %rsi movq %rbx, %rdx xorl %ecx, %ecx vzeroupper callq 0x12f0a32 addq $0x20, %rsp movq 0x90(%rsp), %rax movq %rax, -0x50(%r12) vmovaps 0xa0(%rsp), %xmm0 vmovaps %xmm0, -0x40(%r12) vmovaps 0xb0(%rsp), %xmm1 vmovaps %xmm1, -0x30(%r12) vmovaps 0xc0(%rsp), %xmm1 vmovaps %xmm1, -0x20(%r12) vmovaps 0xd0(%rsp), %xmm1 vmovaps %xmm1, -0x10(%r12) vmovsd 0xe0(%rsp), %xmm1 vmovsd %xmm1, (%r12) vmovaps 0x30(%rsp), %xmm1 vminps %xmm0, %xmm1, %xmm1 vmovaps %xmm1, 0x30(%rsp) vmovaps 0x20(%rsp), %xmm0 vmaxps -0x30(%r12), %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) vmovaps 0x60(%rsp), %xmm0 vminps -0x20(%r12), %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x50(%rsp), %xmm0 vmaxps -0x10(%r12), %xmm0, %xmm0 vmovaps %xmm0, 0x50(%rsp) decq %rbp addq $-0x60, %r12 addq $-0xb0, %rbx testq %rbp, %rbp jg 0x12f291a movq 0x80(%rsp), %rax orq %r13, %rax movq 0x12a0(%rsp), %rcx movq %rax, %rbp andl $0xf, %eax cmpl $0x1, %eax jne 0x12f2f23 testq %rcx, %rcx je 0x12f2be5 andq $-0x10, %r13 leaq 0x1504(%rsp), %rax xorl %edx, %edx vmovss 0xbf9cc6(%rip), %xmm0 # 0x1eec714 vbroadcastss 0xc2e469(%rip), %xmm1 # 0x1f20ec0 vbroadcastss 0xc2f084(%rip), %xmm2 # 0x1f21ae4 vbroadcastss 0xc2f077(%rip), %xmm3 # 0x1f21ae0 vbroadcastss 0xc2e452(%rip), %xmm4 # 0x1f20ec4 vbroadcastss 0xbfe569(%rip), %xmm5 # 0x1ef0fe4 movq -0x54(%rax), %rsi movq %rsi, (%r13,%rdx,8) vmovss (%rax), %xmm6 vmovss -0x4(%rax), %xmm7 vsubss %xmm7, %xmm6, %xmm6 vdivss %xmm6, %xmm0, %xmm6 vxorps %xmm1, %xmm7, %xmm8 vmulss %xmm6, %xmm8, %xmm8 vsubss %xmm8, %xmm0, %xmm9 vshufps $0x0, %xmm8, %xmm8, %xmm8 # xmm8 = xmm8[0,0,0,0] vmovaps -0x44(%rax), %xmm10 vmovaps -0x34(%rax), %xmm11 vmovaps -0x24(%rax), %xmm12 vmovaps -0x14(%rax), %xmm13 vmulps %xmm8, %xmm12, %xmm14 vshufps $0x0, %xmm9, %xmm9, %xmm9 # xmm9 = xmm9[0,0,0,0] vmulps %xmm9, %xmm10, %xmm15 vaddps %xmm15, %xmm14, %xmm14 vmulps %xmm8, %xmm13, %xmm8 vmulps %xmm9, %xmm11, %xmm9 vaddps %xmm9, %xmm8, %xmm8 vsubss %xmm7, %xmm0, %xmm7 vmulss %xmm6, %xmm7, %xmm6 vsubss %xmm6, %xmm0, %xmm7 vshufps $0x0, %xmm6, %xmm6, %xmm6 # xmm6 = xmm6[0,0,0,0] vmulps %xmm6, %xmm12, %xmm9 vshufps $0x0, %xmm7, %xmm7, %xmm7 # xmm7 = xmm7[0,0,0,0] vmulps %xmm7, %xmm10, %xmm10 vaddps %xmm10, %xmm9, %xmm9 vmulps %xmm6, %xmm13, %xmm6 vmulps %xmm7, %xmm11, %xmm7 vaddps %xmm7, %xmm6, %xmm6 vminps %xmm2, %xmm14, %xmm7 vmaxps %xmm3, %xmm8, %xmm8 vminps %xmm2, %xmm9, %xmm9 vmaxps %xmm3, %xmm6, %xmm6 vandps %xmm4, %xmm7, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vsubps %xmm10, %xmm7, %xmm7 vandps %xmm4, %xmm8, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vaddps %xmm10, %xmm8, %xmm8 vandps %xmm4, %xmm9, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vsubps %xmm10, %xmm9, %xmm9 vandps %xmm4, %xmm6, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vmovss %xmm7, 0x40(%r13,%rdx,4) vextractps $0x1, %xmm7, 0x80(%r13,%rdx,4) vaddps %xmm6, %xmm10, %xmm6 vextractps $0x2, %xmm7, 0xc0(%r13,%rdx,4) vmovss %xmm8, 0x60(%r13,%rdx,4) vextractps $0x1, %xmm8, 0xa0(%r13,%rdx,4) vsubps %xmm7, %xmm9, %xmm7 vextractps $0x2, %xmm8, 0xe0(%r13,%rdx,4) vmovss %xmm7, 0x100(%r13,%rdx,4) vextractps $0x1, %xmm7, 0x140(%r13,%rdx,4) vsubps %xmm8, %xmm6, %xmm6 vextractps $0x2, %xmm7, 0x180(%r13,%rdx,4) vmovss %xmm6, 0x120(%r13,%rdx,4) vextractps $0x1, %xmm6, 0x160(%r13,%rdx,4) vextractps $0x2, %xmm6, 0x1a0(%r13,%rdx,4) incq %rdx addq $0x60, %rax cmpq %rdx, %rcx jne 0x12f2a7b testb $0x1, %r15b jne 0x12f2f44 movq 0x148(%rsp), %r12 movq %rbp, (%r12) vmovaps 0x30(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r12) vmovaps 0x20(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r12) vmovaps 0x60(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r12) vmovaps 0x50(%rsp), %xmm0 vmovaps %xmm0, 0x40(%r12) movq (%rsp), %rax movq 0x98(%rax), %rax movq %rax, 0x50(%r12) cmpq $0x0, 0x12a0(%rsp) je 0x12f2cee xorl %r14d, %r14d movq 0x1220(%rsp,%r14,8), %rax decq 0x8(%rax) jne 0x12f2cdd movq (%rax), %rbx testq %rbx, %rbx je 0x12f2cdd movq 0x18(%rbx), %r15 movq 0x20(%rbx), %rdi testq %rdi, %rdi je 0x12f2c9e movq %r15, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x12f2c96 movzbl 0x8(%rbx), %edx vzeroupper callq 0x1ee67ca jmp 0x12f2c9e vzeroupper callq 0x1ee612d testq %r15, %r15 je 0x12f2cbe movq (%rbx), %rdi shlq $0x4, %r15 leaq (%r15,%r15,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %rbx, %rdi vzeroupper callq 0x6a4f0 incq %r14 cmpq 0x12a0(%rsp), %r14 jb 0x12f2c51 movq 0x10(%rsp), %rsi testq %rsi, %rsi je 0x12f0d30 leaq 0x10(%rsp), %rdi vzeroupper callq 0xbbb88e jmp 0x12f0d30 cmpl $0x2, %eax jne 0x12f375a vmovss 0x6e8(%rsp), %xmm0 vmovss %xmm1, 0x620(%rsp) vmovss %xmm0, 0x624(%rsp) vmovss %xmm0, 0x360(%rsp) movq (%rsp), %rax vmovss 0x9c(%rax), %xmm0 vmovss %xmm0, 0x364(%rsp) movq 0xa0(%rax), %r15 movl $0x28, %edi callq 0x6a170 movq %rax, %r14 movq 0x8(%rsp), %rax movq 0x48(%rax), %rdi movq %rdi, (%r14) movb $0x0, 0x8(%r14) movq %r13, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rbx vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x10(%r14) movq $0x0, 0x20(%r14) movq (%rdi), %rax movq %rbx, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %rbx # imm = 0x1C00000 jb 0x12f3095 movq %r14, %rsi addq $0x8, %rsi movq %rbx, %rdi callq 0x1ee665d jmp 0x12f30a2 leaq 0x14b0(%rsp), %rax movq %rax, 0x220(%rsp) movq %rdx, 0x228(%rsp) leaq 0x720(%rsp), %rax movq %rax, 0x230(%rsp) leaq 0x90(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x2b0(%rsp), %rdi movq %rbp, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x180(%rsp), %rsi leaq 0x220(%rsp), %rax movq %rax, (%rsi) leaq 0x3a0(%rsp), %rdx leaq 0x90(%rsp), %rcx callq 0x12f729a leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12f3887 leaq 0x90(%rsp), %rdi callq 0x6aab0 movq 0x12a0(%rsp), %rax vbroadcastss 0xbf9d02(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x20(%rsp) vbroadcastss 0xbf8b8f(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x30(%rsp) testq %rax, %rax je 0x12f2a0d vbroadcastss 0xbf9cdb(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xbf8b6e(%rip), %xmm1 # 0x1eeba20 leaq 0x14f0(%rsp), %rcx vmovaps %xmm1, 0x30(%rsp) vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x20(%rsp) vmovaps %xmm0, 0x50(%rsp) vmovaps 0x30(%rsp), %xmm0 vminps -0x30(%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x30(%rsp) vmovaps 0x20(%rsp), %xmm0 vmaxps -0x20(%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) vmovaps 0x60(%rsp), %xmm0 vminps -0x10(%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x50(%rsp), %xmm0 vmaxps (%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x50(%rsp) addq $0x60, %rcx decq %rax jne 0x12f2ed2 jmp 0x12f2a0d xorl %edi, %edi testq %rcx, %rcx sete %dil leaq 0x14b0(%rsp), %rdx movq %r13, %rsi vzeroupper callq 0x1f7206 jmp 0x12f2be5 movq 0x8(%rsp), %rax addq $0x58, %rax movq 0x48(%rsp), %rcx movq %rcx, 0x5a0(%rsp) movq %rax, 0x5a8(%rsp) movq (%rsp), %rax movq 0x70(%rax), %rbx movq 0x78(%rax), %r14 vbroadcastss 0xbf8aa9(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x2b0(%rsp) vbroadcastss 0xbf9bfb(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x2c0(%rsp) vmovaps %xmm0, 0x2d0(%rsp) vmovaps %xmm1, 0x2e0(%rsp) movq %r14, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x12f3579 leaq 0x90(%rsp), %rdx movq %rbx, (%rdx) movq %r14, 0x8(%rdx) leaq 0x220(%rsp), %rdi leaq 0x5a0(%rsp), %rsi vzeroupper callq 0x12e7ac2 movq 0x148(%rsp), %r12 movq %rbp, (%r12) vmovaps 0x220(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r12) vmovaps 0x230(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r12) vmovaps 0x240(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r12) vmovaps 0x250(%rsp), %xmm0 jmp 0x12f2c28 movq %rcx, 0x10(%r14) shlq $0x2, %rax movq 0x20(%r14), %rcx cmpq %rcx, %rax jbe 0x12f34a7 leaq 0x180(%rsp), %rsi movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 jmp 0x12f268b movq %rcx, 0x10(%r14) shlq $0x2, %rax movq 0x20(%r14), %rcx cmpq %rcx, %rax jbe 0x12f3510 leaq 0x180(%rsp), %rsi movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 jmp 0x12f27bd movl $0x10, %esi movq %rbx, %rdi callq 0x1ee60ac movq 0x8(%rsp), %rcx leaq 0x48(%rcx), %rbx movq %rax, 0x20(%r14) movq %r13, 0x10(%r14) movq %r13, 0x18(%r14) movq %r14, 0x10(%rsp) leaq 0x158(%rsp), %rax movq %r14, (%rax) movq %r15, 0x220(%rsp) leaq 0x620(%rsp), %r14 movq %r14, 0x228(%rsp) movq %rbx, 0x230(%rsp) movq %rax, 0x238(%rsp) movq 0x48(%rsp), %rax movq %rax, 0x240(%rsp) movq (%rsp), %rax movq 0x70(%rax), %r12 movq 0x78(%rax), %r13 vbroadcastss 0xbf890f(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x14b0(%rsp) vbroadcastss 0xbf9a61(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x14c0(%rsp) vmovaps %xmm0, 0x14d0(%rsp) vmovaps %xmm1, 0x14e0(%rsp) vmovaps %xmm0, 0x14f0(%rsp) vmovaps %xmm1, 0x1500(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x1510(%rsp) xorl %eax, %eax movl %eax, 0x1530(%rsp) vbroadcastss 0xbf95a5(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x1534(%rsp) movl %eax, 0x153c(%rsp) movq %r13, %rax subq %r12, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x12f361a leaq 0x90(%rsp), %rdx movq %r12, (%rdx) movq %r13, 0x8(%rdx) leaq 0x720(%rsp), %rdi leaq 0x220(%rsp), %rsi vzeroupper callq 0x12e2d2e movq 0x788(%rsp), %rax subq 0x780(%rsp), %rax movq (%rsp), %rcx movq 0x78(%rcx), %rdx subq 0x70(%rcx), %rdx cmpq %rdx, %rax je 0x12f3204 movq 0x158(%rsp), %rax movq 0x20(%rax), %rdi leaq 0x14b0(%rsp), %r8 movq %r14, (%r8) movl $0x400, %ecx # imm = 0x400 xorl %esi, %esi callq 0x12e25d3 movq %rax, 0x788(%rsp) movq 0x158(%rsp), %rax vmovsd 0x620(%rsp), %xmm0 vmovups 0x720(%rsp), %ymm1 vmovups 0x740(%rsp), %ymm2 vmovups 0x760(%rsp), %ymm3 vmovaps 0x790(%rsp), %xmm4 vmovaps %xmm4, 0x570(%rsp) vmovups %ymm1, 0x500(%rsp) vmovups %ymm2, 0x520(%rsp) vmovups %ymm3, 0x540(%rsp) movq 0x780(%rsp), %rcx movq %rcx, 0x560(%rsp) movq 0x788(%rsp), %rcx movq %rcx, 0x568(%rsp) vmovsd 0x7a0(%rsp), %xmm1 vmovsd 0x7a8(%rsp), %xmm2 vcmpltps %xmm0, %xmm2, %xmm3 vinsertps $0x50, %xmm2, %xmm0, %xmm4 # xmm4 = xmm0[0],xmm2[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm2, %xmm0 # xmm0 = xmm2[0],xmm0[1],xmm2[2,3] vblendvps %xmm3, %xmm4, %xmm0, %xmm0 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x580(%rsp) movq %rax, 0x590(%rsp) movq %r15, 0x3a0(%rsp) leaq 0x360(%rsp), %r14 movq %r14, 0x3a8(%rsp) movq %rbx, 0x3b0(%rsp) movq (%rsp), %rax movq 0x70(%rax), %rbx movq 0x78(%rax), %r12 vbroadcastss 0xbf8730(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x90(%rsp) vbroadcastss 0xbf9882(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0xa0(%rsp) vmovaps %xmm0, 0xb0(%rsp) vmovaps %xmm1, 0xc0(%rsp) vmovaps %xmm0, 0xd0(%rsp) vmovaps %xmm1, 0xe0(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0xf0(%rsp) xorl %eax, %eax movl %eax, 0x110(%rsp) vbroadcastss 0xbf93c6(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x114(%rsp) movl %eax, 0x11c(%rsp) movq %r12, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x12f36ba leaq 0x2b0(%rsp), %rdx movq %rbx, (%rdx) movq %r12, 0x8(%rdx) leaq 0x14b0(%rsp), %rdi leaq 0x3a0(%rsp), %rsi vzeroupper callq 0x12e5072 movq 0x1518(%rsp), %rax subq 0x1510(%rsp), %rax movq (%rsp), %rdx movq 0x70(%rdx), %rsi leaq (%rax,%rsi), %rcx movq %rsi, 0x1510(%rsp) movq %rcx, 0x1518(%rsp) movq 0x78(%rdx), %rdx movq %rdx, %rcx subq %rsi, %rcx cmpq %rcx, %rax je 0x12f33f3 movq 0x20(%r15), %rdi leaq 0x90(%rsp), %r8 movq %r14, (%r8) movl $0x400, %ecx # imm = 0x400 callq 0x12e2980 movq %rax, 0x1518(%rsp) vmovsd 0x360(%rsp), %xmm0 vmovups 0x14b0(%rsp), %ymm1 vmovups 0x14d0(%rsp), %ymm2 vmovups 0x14f0(%rsp), %ymm3 vmovaps 0x1520(%rsp), %xmm4 vmovaps %xmm4, 0x4d0(%rsp) vmovups %ymm1, 0x460(%rsp) vmovups %ymm2, 0x480(%rsp) vmovups %ymm3, 0x4a0(%rsp) movq 0x1510(%rsp), %rax movq %rax, 0x4c0(%rsp) movq 0x1518(%rsp), %rax movq %rax, 0x4c8(%rsp) vmovsd 0x1538(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovsd 0x1530(%rsp), %xmm1 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x4e0(%rsp) movq %r15, 0x4f0(%rsp) jmp 0x12f0ffd leaq 0x90(%rsp), %rsi movq %rcx, (%rsi) movl $0x40, %edx movq %rbp, %rdi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx ja 0x12f379b movq %rax, 0x30(%r14) jmp 0x12f268b leaq 0x90(%rsp), %rsi movq %rcx, (%rsi) movl $0x40, %edx movq %rbp, %rdi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx ja 0x12f3811 movq %rax, 0x30(%r14) jmp 0x12f27bd leaq 0x90(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x180(%rsp), %rsi movq %r14, (%rsi) movq %rbx, 0x8(%rsi) movq $0x400, 0x10(%rsi) # imm = 0x400 leaq 0x3a0(%rsp), %rcx leaq 0x620(%rsp), %r8 movq %r8, (%rcx) leaq 0x5a0(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0x220(%rsp), %rdi leaq 0x2b0(%rsp), %rdx leaq 0x90(%rsp), %r9 callq 0x12e7f58 leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12f391d leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x12f2fdd leaq 0x90(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x2b0(%rsp), %rsi movq %r13, (%rsi) movq %r12, 0x8(%rsi) movq $0x80, 0x10(%rsi) leaq -0x743513(%rip), %r8 # 0xbb014c leaq 0x180(%rsp), %rcx movq %r8, (%rcx) leaq 0x220(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0x720(%rsp), %rdi leaq 0x14b0(%rsp), %rdx leaq 0x90(%rsp), %r9 callq 0x12e3461 leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12f394f leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x12f31b8 leaq 0x2b0(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x180(%rsp), %rsi movq %r12, (%rsi) movq %rbx, 0x8(%rsi) movq $0x80, 0x10(%rsi) leaq -0x7435b3(%rip), %r8 # 0xbb014c leaq 0x5a0(%rsp), %rcx movq %r8, (%rcx) leaq 0x3a0(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0x14b0(%rsp), %rdi leaq 0x90(%rsp), %rdx leaq 0x2b0(%rsp), %r9 callq 0x12e5713 leaq 0x2b0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x12f3981 leaq 0x2b0(%rsp), %rdi callq 0x6aab0 jmp 0x12f3397 cmpl $0x1, %eax je 0x12f39b3 cmpl $0x3, %eax jne 0x12f0ff4 movq 0x48(%rsp), %rbx movq %rbx, %rdi callq 0xba71d2 leaq 0x500(%rsp), %rdx leaq 0x460(%rsp), %rcx movq 0x8(%rsp), %rdi movq %rbx, %rsi callq 0x1f1a30 jmp 0x12f0ff4 movq $0x0, 0x10(%r14) movq 0x20(%r14), %rax leaq 0x90(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx jbe 0x12f3507 movq $0x0, 0x10(%r14) xorl %r13d, %r13d jmp 0x12f268b movq $0x0, 0x10(%r14) movq 0x20(%r14), %rax leaq 0x90(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx jbe 0x12f3570 movq $0x0, 0x10(%r14) xorl %r13d, %r13d jmp 0x12f27bd movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbf8255(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe3135e(%rip), %rsi # 0x2124c08 movq 0xe31117(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbf8223(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe3132c(%rip), %rsi # 0x2124c08 movq 0xe310e5(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbf81f1(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe312fa(%rip), %rsi # 0x2124c08 movq 0xe310b3(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbf81bf(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe312c8(%rip), %rsi # 0x2124c08 movq 0xe31081(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbf818d(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe31296(%rip), %rsi # 0x2124c08 movq 0xe3104f(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbf815b(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe31264(%rip), %rsi # 0x2124c08 movq 0xe3101d(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq 0x48(%rsp), %rbx movq %rbx, %rdi callq 0xba71d2 leaq 0x500(%rsp), %rdx leaq 0x460(%rsp), %rcx movq 0x8(%rsp), %rdi movq %rbx, %rsi callq 0x12f6e60 jmp 0x12f0ff4 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12f3a21 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12f3a3b movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12f3a55 jmp 0x12f3ce7 jmp 0x12f3a1e jmp 0x12f3ce7 jmp 0x12f3a38 jmp 0x12f3ce7 jmp 0x12f3a52 movq %rax, %rbx leaq 0x2b0(%rsp), %rdi callq 0x6aab0 jmp 0x12f3ad5 jmp 0x12f3ce7 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x12f3ad5 jmp 0x12f3ce7 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x12f3c27 jmp 0x12f3ce7 jmp 0x12f3aa2 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12f3aa5 jmp 0x12f3abe movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12f3ac1 jmp 0x12f3ad2 jmp 0x12f3ad2 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x12f3b15 jmp 0x12f3ce7 jmp 0x12f3aa2 movq %rax, %rbx leaq 0x5a0(%rsp), %rdi callq 0x6aab0 jmp 0x12f3b45 jmp 0x12f3ce7 jmp 0x12f3abe movq %rax, %rbx leaq 0x3a0(%rsp), %rdi callq 0x6aab0 jmp 0x12f3b45 jmp 0x12f3ad2 movq %rax, %rbx movq 0x10(%rsp), %rsi testq %rsi, %rsi je 0x12f3ae9 leaq 0x10(%rsp), %rdi callq 0xbbb88e movq $0x0, 0x10(%rsp) jmp 0x12f3cdf jmp 0x12f3c24 jmp 0x12f3ce7 jmp 0x12f3b12 jmp 0x12f3c24 jmp 0x12f3ce7 jmp 0x12f3ce7 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x12f3c27 jmp 0x12f3ce7 movq %rax, %rbx movq %r14, %rdi callq 0x6a4f0 jmp 0x12f3cdf jmp 0x12f3b42 jmp 0x12f3b42 jmp 0x12f3b42 movq %rax, %rbx movq 0x18(%rsp), %rsi testq %rsi, %rsi je 0x12f3b59 leaq 0x18(%rsp), %rdi callq 0xbbb88e movq $0x0, 0x18(%rsp) jmp 0x12f3c27 movq %rax, %rbx movq %r12, %rdi callq 0x6a4f0 jmp 0x12f3c27 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x8d6eda jmp 0x12f3bbc movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x8d6eda jmp 0x12f3bbc jmp 0x12f3ce7 jmp 0x12f3ce7 jmp 0x12f3ce7 jmp 0x12f3ce7 jmp 0x12f3ce7 movq %rax, %rbx jmp 0x12f3bbc movq %rax, %rbx leaq 0x2b0(%rsp), %rdi callq 0x8d6eda jmp 0x12f3c27 jmp 0x12f3ce7 jmp 0x12f3c24 movq %rax, %rbx movq 0x14b0(%rsp), %rsi testq %rsi, %rsi je 0x12f3bef leaq 0x14b0(%rsp), %rdi callq 0x90da94 leaq 0x720(%rsp), %rdi callq 0x8d6eda jmp 0x12f3cdf movq %rax, %rbx movq %r12, %rdi callq 0x1ee612d jmp 0x12f3cdf jmp 0x12f3ce7 jmp 0x12f3ce7 jmp 0x12f3ce7 jmp 0x12f3c24 jmp 0x12f3c24 movq %rax, %rbx cmpq $0x0, 0x12a0(%rsp) je 0x12f3ccb xorl %r15d, %r15d movq 0x1220(%rsp,%r15,8), %rax decq 0x8(%rax) jne 0x12f3cba movq (%rax), %r14 testq %r14, %r14 je 0x12f3cba movq 0x18(%r14), %r12 movq 0x20(%r14), %rdi testq %rdi, %rdi je 0x12f3c81 movq %r12, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x12f3c7c movzbl 0x8(%r14), %edx callq 0x1ee67ca jmp 0x12f3c81 callq 0x1ee612d testq %r12, %r12 je 0x12f3c9e movq (%r14), %rdi shlq $0x4, %r12 leaq (%r12,%r12,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) leaq 0x10(%r14), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %r14, %rdi callq 0x6a4f0 incq %r15 cmpq 0x12a0(%rsp), %r15 jb 0x12f3c39 movq 0x10(%rsp), %rsi testq %rsi, %rsi je 0x12f3cdf leaq 0x10(%rsp), %rdi callq 0xbbb88e movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<8>, embree::avx::RecalculatePrimRef<embree::TriangleMesh>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeaf<8, embree::TriangleMesh, embree::TriangleMi<4>>, embree::Scene::BuildProgressMonitorInterface>::splitFallback(embree::SetMB const&, embree::SetMB&, embree::SetMB&)
void splitFallback(const SetMB& set, SetMB& lset, SetMB& rset) { mvector<PrimRefMB>& prims = *set.prims; const size_t begin = set.begin(); const size_t end = set.end(); const size_t center = (begin + end + 1) / 2; PrimInfoMB linfo = empty; for (size_t i=begin; i<center; i++) linfo.add_primref(prims[i]); PrimInfoMB rinfo = empty; for (size_t i=center; i<end; i++) rinfo.add_primref(prims[i]); new (&lset) SetMB(linfo,set.prims,range<size_t>(begin,center),set.time_range); new (&rset) SetMB(rinfo,set.prims,range<size_t>(center,end ),set.time_range); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x28, %rsp movq 0x90(%rsi), %r8 movq 0x60(%rsi), %r9 movq 0x68(%rsi), %rax leaq (%r9,%rax), %rdi incq %rdi shrq %rdi vmovss 0xbf5885(%rip), %xmm0 # 0x1eec714 vmovaps %xmm0, -0x60(%rsp) vmovsd 0xbf5853(%rip), %xmm0 # 0x1eec6f0 vmovaps %xmm0, -0x70(%rsp) vbroadcastss 0xbf4b74(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, -0x80(%rsp) vbroadcastss 0xbf5cc9(%rip), %xmm4 # 0x1eecb84 movq %rdi, %rbx subq %r9, %rbx jbe 0x12f6fd1 movq 0x20(%r8), %r10 leaq (%r9,%r9,4), %r11 shlq $0x4, %r11 leaq (%r11,%r10), %r14 addq $0x40, %r14 vmovss 0xbf5831(%rip), %xmm9 # 0x1eec714 vmovsd 0xbf5805(%rip), %xmm6 # 0x1eec6f0 vbroadcastss 0xbf5c90(%rip), %xmm8 # 0x1eecb84 vbroadcastss 0xbf4b23(%rip), %xmm10 # 0x1eeba20 xorl %r10d, %r10d vmovaps %xmm10, %xmm7 vmovaps %xmm8, %xmm5 vmovaps %xmm10, %xmm13 vmovaps %xmm8, %xmm11 xorl %r11d, %r11d movq %r11, %r15 vmovaps -0x40(%r14), %xmm0 vmovaps %xmm5, %xmm3 vmovaps %xmm13, %xmm5 vmovaps -0x30(%r14), %xmm13 vmovaps -0x20(%r14), %xmm14 vmovaps -0x10(%r14), %xmm15 vbroadcastss 0xbf5c3f(%rip), %xmm2 # 0x1eecb80 vmulps %xmm2, %xmm14, %xmm12 vmulps %xmm2, %xmm0, %xmm1 vaddps %xmm1, %xmm12, %xmm1 vmulps %xmm2, %xmm15, %xmm12 vmulps %xmm2, %xmm13, %xmm2 vaddps %xmm2, %xmm12, %xmm2 vaddps %xmm2, %xmm1, %xmm1 vminps %xmm0, %xmm10, %xmm10 vmaxps %xmm13, %xmm8, %xmm8 vmovaps %xmm5, %xmm13 vminps %xmm14, %xmm7, %xmm7 vmaxps %xmm15, %xmm3, %xmm5 vminps %xmm1, %xmm13, %xmm13 vmovsd (%r14), %xmm0 vcmpltps %xmm0, %xmm9, %xmm2 vinsertps $0x50, %xmm0, %xmm9, %xmm12 # xmm12 = xmm9[0],xmm0[1],xmm9[2,3] vblendps $0x2, %xmm9, %xmm0, %xmm9 # xmm9 = xmm0[0],xmm9[1],xmm0[2,3] vblendvps %xmm2, %xmm12, %xmm9, %xmm9 vmaxps %xmm1, %xmm11, %xmm11 movl -0x14(%r14), %r11d addq %r11, %r10 movl -0x4(%r14), %r11d cmpq %r11, %r15 cmovaq %r15, %r11 setb %bpl vmovd %ebp, %xmm1 vpshufd $0x50, %xmm1, %xmm1 # xmm1 = xmm1[0,0,1,1] vpslld $0x1f, %xmm1, %xmm1 vblendvps %xmm1, %xmm0, %xmm6, %xmm6 addq $0x50, %r14 decq %rbx jne 0x12f6f15 jmp 0x12f7005 xorl %r11d, %r11d xorl %r10d, %r10d vmovaps %xmm4, %xmm11 vbroadcastss 0xbf4a3c(%rip), %xmm10 # 0x1eeba20 vmovaps %xmm10, %xmm13 vmovaps %xmm4, %xmm5 vmovaps %xmm10, %xmm7 vmovaps %xmm4, %xmm8 vmovsd 0xbf56f3(%rip), %xmm6 # 0x1eec6f0 vmovss 0xbf570f(%rip), %xmm9 # 0x1eec714 vmovaps %xmm13, 0x10(%rsp) movq %rax, %r15 subq %rdi, %r15 jbe 0x12f7192 vmovaps %xmm7, -0x50(%rsp) vmovaps %xmm5, -0x40(%rsp) vmovaps %xmm11, -0x30(%rsp) vmovaps %xmm10, -0x20(%rsp) vmovaps %xmm8, -0x10(%rsp) vmovaps %xmm6, (%rsp) movq 0x20(%r8), %rbx leaq (%rdi,%rdi,4), %r14 shlq $0x4, %r14 leaq (%r14,%rbx), %r12 addq $0x40, %r12 vmovss 0xbf56be(%rip), %xmm0 # 0x1eec714 vmovaps %xmm0, -0x60(%rsp) vmovsd 0xbf568c(%rip), %xmm0 # 0x1eec6f0 vmovaps %xmm0, -0x70(%rsp) vbroadcastss 0xbf5b11(%rip), %xmm12 # 0x1eecb84 vbroadcastss 0xbf49a4(%rip), %xmm13 # 0x1eeba20 xorl %ebx, %ebx vbroadcastss 0xbf5af9(%rip), %xmm8 # 0x1eecb80 vmovaps %xmm13, %xmm15 vmovaps %xmm12, %xmm14 vmovaps %xmm13, -0x80(%rsp) vmovaps %xmm12, %xmm4 xorl %r14d, %r14d movq %r14, %r13 vmovaps -0x40(%r12), %xmm1 vmovaps -0x30(%r12), %xmm2 vmovaps -0x20(%r12), %xmm0 vmovaps -0x10(%r12), %xmm3 vmulps %xmm0, %xmm8, %xmm10 vmovaps -0x60(%rsp), %xmm6 vmulps %xmm1, %xmm8, %xmm5 vaddps %xmm5, %xmm10, %xmm11 vmulps %xmm3, %xmm8, %xmm5 vmulps %xmm2, %xmm8, %xmm7 vaddps %xmm5, %xmm7, %xmm5 vmovaps -0x80(%rsp), %xmm7 vaddps %xmm5, %xmm11, %xmm10 vminps %xmm1, %xmm13, %xmm13 vmaxps %xmm2, %xmm12, %xmm12 vminps %xmm0, %xmm15, %xmm15 vmaxps %xmm3, %xmm14, %xmm14 vminps %xmm10, %xmm7, %xmm7 vmovaps %xmm7, -0x80(%rsp) vmovsd (%r12), %xmm0 vcmpltps %xmm0, %xmm6, %xmm1 vinsertps $0x50, %xmm0, %xmm6, %xmm2 # xmm2 = xmm6[0],xmm0[1],xmm6[2,3] vblendps $0x2, %xmm6, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm6[1],xmm0[2,3] vblendvps %xmm1, %xmm2, %xmm3, %xmm1 vmovaps %xmm1, -0x60(%rsp) vmovaps -0x70(%rsp), %xmm2 vmaxps %xmm10, %xmm4, %xmm4 movl -0x14(%r12), %r14d addq %r14, %rbx movl -0x4(%r12), %r14d cmpq %r14, %r13 setb %bpl vmovd %ebp, %xmm1 vpshufd $0x50, %xmm1, %xmm1 # xmm1 = xmm1[0,0,1,1] vpslld $0x1f, %xmm1, %xmm1 cmovaq %r13, %r14 vblendvps %xmm1, %xmm0, %xmm2, %xmm2 vmovaps %xmm2, -0x70(%rsp) addq $0x50, %r12 decq %r15 jne 0x12f709e vmovaps (%rsp), %xmm6 vmovaps -0x10(%rsp), %xmm8 vmovaps -0x20(%rsp), %xmm10 vmovaps -0x30(%rsp), %xmm11 vmovaps -0x40(%rsp), %xmm5 vmovaps -0x50(%rsp), %xmm7 jmp 0x12f71ad xorl %r14d, %r14d xorl %ebx, %ebx vmovaps %xmm4, %xmm14 vbroadcastss 0xbf487c(%rip), %xmm13 # 0x1eeba20 vmovaps %xmm13, %xmm15 vmovaps %xmm4, %xmm12 vmovsd 0x88(%rsi), %xmm0 vcmpltps %xmm0, %xmm9, %xmm1 vblendps $0x2, %xmm9, %xmm0, %xmm2 # xmm2 = xmm0[0],xmm9[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm9, %xmm0 # xmm0 = xmm9[0],xmm0[1],xmm9[2,3] vblendvps %xmm1, %xmm2, %xmm0, %xmm0 vmovaps %xmm10, (%rdx) vmovaps %xmm8, 0x10(%rdx) vmovaps %xmm7, 0x20(%rdx) vmovaps %xmm5, 0x30(%rdx) vmovaps 0x10(%rsp), %xmm1 vmovaps %xmm1, 0x40(%rdx) vmovaps %xmm11, 0x50(%rdx) movq %r10, 0x70(%rdx) movq %r11, 0x78(%rdx) vmovlhps %xmm0, %xmm6, %xmm0 # xmm0 = xmm6[0],xmm0[0] vmovaps %xmm0, 0x80(%rdx) movq %r8, 0x90(%rdx) movq %r9, 0x60(%rdx) movq %rdi, 0x68(%rdx) vmovsd 0x88(%rsi), %xmm0 vmovaps -0x60(%rsp), %xmm5 vcmpltps %xmm0, %xmm5, %xmm1 vblendps $0x2, %xmm5, %xmm0, %xmm2 # xmm2 = xmm0[0],xmm5[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm5, %xmm0 # xmm0 = xmm5[0],xmm0[1],xmm5[2,3] vblendvps %xmm1, %xmm2, %xmm0, %xmm0 movq 0x90(%rsi), %rdx vmovaps %xmm13, (%rcx) vmovaps %xmm12, 0x10(%rcx) vmovaps %xmm15, 0x20(%rcx) vmovaps %xmm14, 0x30(%rcx) vmovaps -0x80(%rsp), %xmm1 vmovaps %xmm1, 0x40(%rcx) vmovaps %xmm4, 0x50(%rcx) vmovaps -0x70(%rsp), %xmm1 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] movq %rbx, 0x70(%rcx) movq %r14, 0x78(%rcx) vmovaps %xmm0, 0x80(%rcx) movq %rdx, 0x90(%rcx) movq %rdi, 0x60(%rcx) movq %rax, 0x68(%rcx) addq $0x28, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<4>, embree::avx::RecalculatePrimRef<embree::QuadMesh>, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::QuadMesh, embree::QuadMi<4>>, embree::Scene::BuildProgressMonitorInterface>::createLargeLeaf(embree::avx::BVHBuilderMSMBlur::BuildRecord const&, embree::FastAllocator::CachedAllocator)
const NodeRecordMB4D createLargeLeaf(const BuildRecord& in, Allocator alloc) { /* this should never occur but is a fatal error */ if (in.depth > cfg.maxDepth) throw_RTCError(RTC_ERROR_UNKNOWN,"depth limit reached"); /* replace already found split by fallback split */ const BuildRecordSplit current(BuildRecord(in.prims,in.depth),findFallback(in.prims)); /* special case when directly creating leaf without any splits that could shrink time_range */ bool force_split = false; if (current.depth == 1 && current.size() > 0) { BBox1f c = empty; BBox1f p = current.prims.time_range; for (size_t i=current.prims.begin(); i<current.prims.end(); i++) { mvector<PrimRefMB>& prims = *current.prims.prims; c.extend(prims[i].time_range); } force_split = c.lower > p.lower || c.upper < p.upper; } /* create leaf for few primitives */ if (current.size() <= cfg.maxLeafSize && current.split.data < Split::SPLIT_ENFORCE && !force_split) return createLeaf(current,alloc); /* fill all children by always splitting the largest one */ bool hasTimeSplits = false; NodeRecordMB4D values[MAX_BRANCHING_FACTOR]; LocalChildListSplit children(current); do { /* find best child with largest bounding box area */ size_t bestChild = -1; size_t bestSize = 0; for (size_t i=0; i<children.size(); i++) { /* ignore leaves as they cannot get split */ if (children[i].size() <= cfg.maxLeafSize && children[i].split.data < Split::SPLIT_ENFORCE && !force_split) continue; force_split = false; /* remember child with largest size */ if (children[i].size() > bestSize) { bestSize = children[i].size(); bestChild = i; } } if (bestChild == -1) break; /* perform best found split */ BuildRecordSplit& brecord = children[bestChild]; BuildRecordSplit lrecord(current.depth+1); BuildRecordSplit rrecord(current.depth+1); std::unique_ptr<mvector<PrimRefMB>> new_vector = split(brecord.split,brecord.prims,lrecord.prims,rrecord.prims); hasTimeSplits |= new_vector != nullptr; /* find new splits */ lrecord.split = findFallback(lrecord.prims); rrecord.split = findFallback(rrecord.prims); children.split(bestChild,lrecord,rrecord,std::move(new_vector)); } while (children.size() < cfg.branchingFactor); /* detect time_ranges that have shrunken */ for (size_t i=0; i<children.size(); i++) { const BBox1f c = children[i].prims.time_range; const BBox1f p = in.prims.time_range; hasTimeSplits |= c.lower > p.lower || c.upper < p.upper; } /* create node */ auto node = createNode(children.children.data(),children.numChildren,alloc,hasTimeSplits); /* recurse into each child and perform reduction */ LBBox3fa gbounds = empty; for (size_t i=0; i<children.size(); i++) { values[i] = createLargeLeaf(children[i],alloc); gbounds.extend(values[i].lbounds); } setNode(current,children.children.data(),node,values,children.numChildren); /* calculate geometry bounds of this node */ if (hasTimeSplits) return NodeRecordMB4D(node,current.prims.linearBounds(recalculatePrimRef),current.prims.time_range); else return NodeRecordMB4D(node,gbounds,current.prims.time_range); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1e38, %rsp # imm = 0x1E38 movq %rdi, 0x70(%rsp) movq (%rdx), %r12 cmpq 0x8(%rsi), %r12 ja 0x1305a4d movq %rdx, %rax movq %rsi, %rbx addq $0x10, %rdx vmovaps 0x10(%rax), %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x20(%rax), %xmm0 vmovaps %xmm0, 0x10(%rsp) vmovaps 0x30(%rax), %xmm0 vmovaps %xmm0, 0x30(%rsp) vmovaps 0x40(%rax), %xmm0 vmovaps %xmm0, 0xc0(%rsp) vmovaps 0x50(%rax), %xmm0 vmovaps %xmm0, 0xb0(%rsp) vmovaps 0x60(%rax), %xmm0 vmovaps %xmm0, 0xa0(%rsp) vmovups 0x80(%rax), %ymm0 vmovups %ymm0, 0x1830(%rsp) movq 0xa0(%rax), %r15 movq %rax, 0xd0(%rsp) vmovaps 0x70(%rax), %xmm0 vmovaps %xmm0, 0x90(%rsp) leaq 0x6a0(%rsp), %rbp movq %rbp, %rdi vzeroupper callq 0x1307b0e vmovaps 0xa0(%rsp), %xmm11 vmovaps 0xb0(%rsp), %xmm10 vmovaps 0xc0(%rsp), %xmm9 vmovaps 0x30(%rsp), %xmm8 vmovaps 0x10(%rsp), %xmm7 vmovaps 0x60(%rsp), %xmm6 movq %r12, 0x5a0(%rsp) vmovaps %xmm6, 0x5b0(%rsp) vmovaps %xmm7, 0x5c0(%rsp) vmovaps %xmm8, 0x5d0(%rsp) vmovaps %xmm9, 0x5e0(%rsp) vmovaps %xmm10, 0x5f0(%rsp) vmovaps %xmm11, 0x600(%rsp) vmovaps 0x90(%rsp), %xmm0 vmovaps %xmm0, 0x610(%rsp) vmovups 0x1830(%rsp), %ymm0 vmovups %ymm0, 0x620(%rsp) movq %r15, 0x640(%rsp) vmovaps (%rbp), %xmm0 vmovaps %xmm0, 0x650(%rsp) movq 0x10(%rbp), %rax movq %rax, 0x660(%rsp) vmovups 0x20(%rbp), %ymm0 vmovups %ymm0, 0x670(%rsp) cmpq $0x1, %r12 movq %rbx, 0x8(%rsp) jne 0x130324a movq 0x610(%rsp), %rcx movq 0x618(%rsp), %rax cmpq %rcx, %rax jne 0x130324e xorl %ebx, %ebx jmp 0x13032c7 vmovss 0x638(%rsp), %xmm0 vmovss 0x63c(%rsp), %xmm1 subq %rcx, %rax jbe 0x13032a6 movq 0x20(%r15), %rdx leaq (%rcx,%rcx,4), %rcx shlq $0x4, %rcx addq %rdx, %rcx addq $0x40, %rcx vmovsd 0xc1dc30(%rip), %xmm2 # 0x1f20eb0 vmovsd (%rcx), %xmm3 vcmpltps %xmm3, %xmm2, %xmm4 vinsertps $0x50, %xmm3, %xmm2, %xmm5 # xmm5 = xmm2[0],xmm3[1],xmm2[2,3] vblendps $0x2, %xmm2, %xmm3, %xmm2 # xmm2 = xmm3[0],xmm2[1],xmm3[2,3] vblendvps %xmm4, %xmm5, %xmm2, %xmm2 addq $0x50, %rcx decq %rax jne 0x1303280 jmp 0x13032ae vmovsd 0xc1dc02(%rip), %xmm2 # 0x1f20eb0 vmovshdup %xmm2, %xmm3 # xmm3 = xmm2[1,1,3,3] vcmpltps %xmm1, %xmm3, %xmm1 vcmpltps %xmm2, %xmm0, %xmm0 vorps %xmm1, %xmm0, %xmm0 vmovd %xmm0, %ebx andb $0x1, %bl movq 0x610(%rsp), %r14 movq 0x618(%rsp), %r13 subq %r14, %r13 movq 0x8(%rsp), %rax cmpq 0x20(%rax), %r13 ja 0x13032f7 cmpl $0x1, 0x65c(%rsp) ja 0x13032f7 testb %bl, %bl je 0x1304674 leaq 0x5b0(%rsp), %rcx movl $0xbc, %eax vmovsd 0xbfab94(%rip), %xmm0 # 0x1efdea0 vmovaps %xmm0, 0x694(%rsp,%rax) addq $0xf0, %rax cmpq $0xfbc, %rax # imm = 0xFBC jne 0x130330c movl $0x1, %eax leaq 0x1628(%rsp), %rdx movq %rax, -0x8(%rdx) movq %rax, 0x200(%rdx) movq %r12, -0xf88(%rdx) vmovaps %xmm6, -0xf78(%rdx) vmovaps %xmm7, -0xf68(%rdx) vmovaps %xmm8, -0xf58(%rdx) vmovaps %xmm9, -0xf48(%rdx) vmovaps %xmm10, -0xf38(%rdx) vmovaps %xmm11, -0xf28(%rdx) leaq 0x710(%rsp), %rsi vmovups 0x60(%rcx), %ymm0 vmovups %ymm0, (%rsi) vmovaps 0x630(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rsi) movq %r15, 0x30(%rsi) vmovaps 0xa0(%rcx), %xmm0 vmovaps %xmm0, 0x40(%rsi) movq 0x660(%rsp), %rax movq %rax, 0x50(%rsi) vmovaps 0x670(%rsp), %xmm0 vmovaps %xmm0, 0x60(%rsi) vmovaps 0x680(%rsp), %xmm0 vmovaps %xmm0, 0x70(%rsi) movq %r15, (%rdx) movq $0x2, 0x8(%rdx) movq %rdx, 0xe90(%rsi) movl $0xb0, %eax leaq 0x230(%rsp), %rdx leaq (%rdx,%rax), %rcx movq %rcx, 0xb0(%rsp) leaq 0x330(%rsp), %rsi addq %rsi, %rax movq %rax, 0xa0(%rsp) movq 0x8(%rsp), %rcx leaq 0x48(%rcx), %rax movq %rax, 0x28(%rsp) movl $0x80, %eax addq %rax, %rdx movq %rdx, 0x78(%rsp) addq %rax, %rsi movq %rsi, 0x320(%rsp) vbroadcastss 0xbe85e9(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x60(%rsp) vbroadcastss 0xbe973e(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x90(%rsp) vmovsd 0xbfaa49(%rip), %xmm0 # 0x1efdea0 leaq 0x180(%rsp), %r15 xorl %r9d, %r9d movq 0x1620(%rsp), %rax testq %rax, %rax je 0x13034bb movq 0x20(%rcx), %rcx movq $-0x1, %r14 leaq 0x710(%rsp), %rdx xorl %esi, %esi xorl %edi, %edi movq 0x8(%rdx), %r8 subq (%rdx), %r8 cmpq %rcx, %r8 ja 0x130349d cmpl $0x1, 0x4c(%rdx) ja 0x130349d testb $0x1, %bl je 0x13034aa cmpq %rsi, %r8 cmovaq %r8, %rsi cmovaq %rdi, %r14 xorl %ebx, %ebx incq %rdi addq $0xf0, %rdx cmpq %rdi, %rax jne 0x1303486 jmp 0x13034c2 movq $-0x1, %r14 cmpq $-0x1, %r14 je 0x1303c89 imulq $0xf0, %r14, %rax leaq (%rsp,%rax), %rdx addq $0x6a0, %rdx # imm = 0x6A0 movq 0x5a0(%rsp), %rcx incq %rcx movq %rcx, 0x230(%rsp) vmovaps %xmm0, 0x2e0(%rsp) movq %rcx, 0x330(%rsp) vmovaps %xmm0, 0x3e0(%rsp) movl $0xb0, %ecx leaq (%rdx,%rcx), %rsi movl 0xbc(%rdx), %ecx testl %ecx, %ecx movq %rdx, 0x10(%rsp) movq %r9, 0xc0(%rsp) movq %rsi, 0x30(%rsp) jne 0x1303cae movq 0x70(%rdx), %r13 movq 0x78(%rdx), %r12 vbroadcastss 0xbe84db(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x180(%rsp) vbroadcastss 0xbe962d(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x190(%rsp) vmovaps %xmm0, 0x1a0(%rsp) vmovaps %xmm1, 0x1b0(%rsp) vmovaps %xmm0, 0x1c0(%rsp) vmovaps %xmm1, 0x1d0(%rsp) movq %r15, %r8 leaq 0x1f0(%rsp), %r15 movl $0x0, %eax movl %eax, 0x10(%r15) vxorps %xmm2, %xmm2, %xmm2 vmovups %ymm2, -0x10(%r15) vbroadcastss 0xbe9169(%rip), %xmm3 # 0x1eec714 vmovlps %xmm3, 0x204(%rsp) movl %eax, 0x20c(%rsp) vmovaps %xmm0, 0xf0(%rsp) vmovaps %xmm1, 0x100(%rsp) vmovaps %xmm0, 0x110(%rsp) vmovaps %xmm1, 0x120(%rsp) vmovaps %xmm0, 0x130(%rsp) vmovaps %xmm1, 0x140(%rsp) leaq 0x150(%rsp), %rbp movl %eax, 0x20(%rbp) vmovups %ymm2, (%rbp) vmovlps %xmm3, 0x174(%rsp) movl %eax, 0x17c(%rsp) vbroadcastss 0xb8(%rdx), %xmm0 vmovaps %xmm0, 0x520(%rsp) movb 0xb4(%rdx), %cl movl $0x1, %eax shll %cl, %eax cltq shlq $0x4, %rax leaq 0xe4c943(%rip), %rcx # 0x214ff80 vmovaps (%rcx,%rax), %xmm0 vmovaps %xmm0, 0x420(%rsp) movq %rsi, 0x490(%rsp) leaq 0x520(%rsp), %rax movq %rax, 0x498(%rsp) leaq 0x420(%rsp), %rax movq %rax, 0x4a0(%rsp) movq 0xa0(%rdx), %rax movq 0x20(%rax), %rdi subq $0x8, %rsp movq %r13, %rsi movq %rdx, %r15 movq %r12, %rdx leaq 0x488(%rsp), %rcx leaq 0xf8(%rsp), %r9 pushq $0xc00 # imm = 0xC00 pushq $0x80 leaq 0x480(%rsp), %rax pushq %rax leaq 0x238(%rsp), %rax pushq %rax leaq 0x4b8(%rsp), %rax pushq %rax vzeroupper callq 0x125bcbe addq $0x30, %rsp movq 0xa0(%r15), %rcx vmovsd 0x98(%r15), %xmm0 vmovups 0x180(%rsp), %ymm1 vmovups 0x1a0(%rsp), %ymm2 vmovups 0x1c0(%rsp), %ymm3 vmovups %ymm1, 0x240(%rsp) vmovups %ymm2, 0x260(%rsp) vmovups %ymm3, 0x280(%rsp) vmovaps 0x1e0(%rsp), %xmm1 vmovaps %xmm1, 0x2a0(%rsp) leaq 0x1f0(%rsp), %rdx vmovups (%rdx), %ymm1 movq 0x78(%rsp), %rdx vmovups %ymm1, (%rdx) movq %rcx, 0x2d0(%rsp) movq %r13, 0x2a0(%rsp) movq %rax, 0x2a8(%rsp) vmovsd 0x2c8(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm1, %xmm1 vmovlps %xmm1, 0x2c8(%rsp) vmovaps 0xf0(%rsp), %xmm1 vmovaps %xmm1, 0x340(%rsp) vmovaps 0x100(%rsp), %xmm1 vmovaps %xmm1, 0x350(%rsp) vmovaps 0x110(%rsp), %xmm1 vmovaps %xmm1, 0x360(%rsp) vmovaps 0x120(%rsp), %xmm1 vmovaps %xmm1, 0x370(%rsp) vmovaps 0x130(%rsp), %xmm1 vmovaps %xmm1, 0x380(%rsp) vmovaps 0x140(%rsp), %xmm1 vmovaps %xmm1, 0x390(%rsp) vmovaps 0x150(%rsp), %xmm1 vmovaps %xmm1, 0x3a0(%rsp) vmovups 0x10(%rbp), %ymm1 movq 0x320(%rsp), %rdx vmovups %ymm1, (%rdx) movq %rcx, 0x3d0(%rsp) movq %rax, 0x3a0(%rsp) movq %r12, 0x3a8(%rsp) vmovsd 0x3c8(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovlps %xmm0, 0x3c8(%rsp) movq $0x0, (%rsp) leaq 0x6a0(%rsp), %rbp leaq 0x180(%rsp), %r15 cmpq $0x0, (%rsp) setne %r13b movq %r15, %rdi movq 0x8(%rsp), %r12 movq %r12, %rsi leaq 0x240(%rsp), %rdx vzeroupper callq 0x1307b0e vmovaps 0x180(%rsp), %xmm0 movq 0xb0(%rsp), %rax vmovaps %xmm0, (%rax) movq 0x190(%rsp), %rax movq %rax, 0x2f0(%rsp) vmovups 0x1a0(%rsp), %ymm0 vmovups %ymm0, 0x300(%rsp) movq %r15, %rdi movq %r12, %rsi leaq 0x340(%rsp), %rdx vzeroupper callq 0x1307b0e vmovaps 0x180(%rsp), %xmm0 movq 0xa0(%rsp), %rax vmovaps %xmm0, (%rax) movq 0x190(%rsp), %rax movq %rax, 0x3f0(%rsp) vmovups 0x1a0(%rsp), %ymm0 vmovups %ymm0, 0x400(%rsp) movq $0x0, (%rsp) movq 0x15a0(%rsp,%r14,8), %rax movq 0x2d0(%rsp), %rcx cmpq (%rax), %rcx je 0x1303962 movq 0x1828(%rsp), %rdx leaq 0x1(%rdx), %rsi movq %rsi, 0x1828(%rsp) shlq $0x4, %rdx leaq 0x1628(%rsp), %rdi leaq (%rdi,%rdx), %rsi movq %rcx, (%rdi,%rdx) movq $0x1, 0x8(%rdi,%rdx) movq %rsi, 0x15a0(%rsp,%r14,8) jmp 0x1303966 incq 0x8(%rax) movq 0x3d0(%rsp), %rcx cmpq (%rax), %rcx je 0x13039b6 movq 0x1828(%rsp), %rdx leaq 0x1(%rdx), %rsi movq %rsi, 0x1828(%rsp) shlq $0x4, %rdx leaq 0x1628(%rsp), %rdi leaq (%rdi,%rdx), %rsi movq %rcx, (%rdi,%rdx) movq $0x1, 0x8(%rdi,%rdx) movq 0x1620(%rsp), %rcx movq %rsi, 0x15a0(%rsp,%rcx,8) jmp 0x13039ca movq 0x1620(%rsp), %rcx movq %rax, 0x15a0(%rsp,%rcx,8) incq 0x8(%rax) movq 0x10(%rsp), %rcx movq 0x30(%rsp), %rdx decq 0x8(%rax) jne 0x1303a7f movq (%rax), %r12 testq %r12, %r12 je 0x1303a7f movq 0x18(%r12), %rbp movq 0x20(%r12), %rdi testq %rdi, %rdi je 0x1303a25 movq %rbp, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1303a1d movzbl 0x8(%r12), %edx vzeroupper callq 0x1ee67ca jmp 0x1303a25 vzeroupper callq 0x1ee612d testq %rbp, %rbp je 0x1303a4d movq (%r12), %rdi shlq $0x4, %rbp leaq (,%rbp,4), %rsi addq %rbp, %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%r12), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %r12, %rdi vzeroupper callq 0x6a4f0 leaq 0x6a0(%rsp), %rbp movq 0x10(%rsp), %rcx movq 0x30(%rsp), %rdx movq 0x230(%rsp), %rax movq %rax, (%rcx) vmovaps 0x240(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rcx) vmovaps 0x250(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rcx) vmovaps 0x260(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rcx) vmovaps 0x270(%rsp), %xmm0 vmovaps %xmm0, 0x40(%rcx) vmovaps 0x280(%rsp), %xmm0 vmovaps %xmm0, 0x50(%rcx) vmovaps 0x290(%rsp), %xmm0 vmovaps %xmm0, 0x60(%rcx) leaq 0x240(%rsp), %rax vmovups 0x60(%rax), %ymm0 vmovups %ymm0, 0x70(%rcx) vmovaps 0x2c0(%rsp), %xmm0 vmovaps %xmm0, 0x90(%rcx) movq 0x2d0(%rsp), %rax movq %rax, 0xa0(%rcx) movq 0xb0(%rsp), %rax vmovaps (%rax), %xmm0 vmovaps %xmm0, (%rdx) movq 0x2f0(%rsp), %rax movq %rax, 0xc0(%rcx) vmovaps 0x300(%rsp), %xmm0 vmovaps %xmm0, 0xd0(%rcx) vmovaps 0x310(%rsp), %xmm0 vmovaps %xmm0, 0xe0(%rcx) imulq $0xf0, 0x1620(%rsp), %rax movq 0x330(%rsp), %rcx movq %rcx, 0x6a0(%rsp,%rax) vmovaps 0x340(%rsp), %xmm0 vmovaps %xmm0, 0x6b0(%rsp,%rax) vmovaps 0x350(%rsp), %xmm0 vmovaps %xmm0, 0x6c0(%rsp,%rax) vmovaps 0x360(%rsp), %xmm0 vmovaps %xmm0, 0x6d0(%rsp,%rax) vmovaps 0x370(%rsp), %xmm0 vmovaps %xmm0, 0x6e0(%rsp,%rax) vmovaps 0x380(%rsp), %xmm0 vmovaps %xmm0, 0x6f0(%rsp,%rax) vmovaps 0x390(%rsp), %xmm0 vmovaps %xmm0, 0x700(%rsp,%rax) leaq 0x340(%rsp), %rcx vmovups 0x60(%rcx), %ymm0 vmovups %ymm0, 0x710(%rsp,%rax) vmovaps 0x3c0(%rsp), %xmm0 vmovaps %xmm0, 0x730(%rsp,%rax) movq 0x3d0(%rsp), %rcx movq %rcx, 0x740(%rsp,%rax) movq 0xa0(%rsp), %rcx vmovaps (%rcx), %xmm0 vmovaps %xmm0, 0x750(%rsp,%rax) movq 0x3f0(%rsp), %rcx movq %rcx, 0x760(%rsp,%rax) vmovaps 0x400(%rsp), %xmm0 vmovaps %xmm0, 0x770(%rsp,%rax) vmovaps 0x410(%rsp), %xmm0 vmovaps %xmm0, 0x780(%rsp,%rax) incq 0x1620(%rsp) movq (%rsp), %rsi testq %rsi, %rsi je 0x1303c76 movq %rsp, %rdi vzeroupper callq 0xbbb88e movq 0xc0(%rsp), %r9 orb %r13b, %r9b vmovsd 0xbfa217(%rip), %xmm0 # 0x1efdea0 cmpq $-0x1, %r14 movq 0x8(%rsp), %rcx je 0x130442a movq 0x1620(%rsp), %rax cmpq (%rcx), %rax jb 0x1303462 jmp 0x130442a addq %rsp, %rax addq $0x6b0, %rax # imm = 0x6B0 movq %rax, 0x40(%rsp) cmpl $0x2, %ecx jne 0x13043ba vmovss 0xb8(%rdx), %xmm0 vmovss 0x98(%rdx), %xmm1 vmovss %xmm1, 0x88(%rsp) vmovss %xmm0, 0x8c(%rsp) vmovss %xmm0, 0x80(%rsp) vmovss 0x9c(%rdx), %xmm0 vmovss %xmm0, 0x84(%rsp) movq 0xa0(%rdx), %rax movq %rax, 0x50(%rsp) movl $0x28, %edi movq %rdx, %r13 vzeroupper callq 0x6a170 movq %rax, %r12 movq 0x28(%rsp), %rax movq (%rax), %rdi movq 0x78(%r13), %rax movq %rdi, (%r12) movb $0x0, 0x8(%r12) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%r12) movq $0x0, 0x20(%r12) subq 0x70(%r13), %rax leaq 0x180(%rsp), %r15 je 0x1303db5 movq %rax, 0xe0(%rsp) shlq $0x4, %rax leaq (%rax,%rax,4), %r13 movq (%rdi), %rax movq %r13, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %r13 # imm = 0x1C00000 jb 0x1303d8c movq %r12, %rsi addq $0x8, %rsi movq %r13, %rdi callq 0x1ee665d jmp 0x1303d99 movl $0x10, %esi movq %r13, %rdi callq 0x1ee60ac movq %rax, 0x20(%r12) movq 0xe0(%rsp), %rax movq %rax, 0x10(%r12) movq %rax, 0x18(%r12) movq 0x10(%rsp), %r13 movq %r12, (%rsp) movq %r12, 0xd8(%rsp) movq 0x50(%rsp), %rax movq %rax, 0x420(%rsp) leaq 0x88(%rsp), %rax movq %rax, 0x428(%rsp) movq 0x28(%rsp), %rax movq %rax, 0x430(%rsp) leaq 0xd8(%rsp), %rax movq %rax, 0x438(%rsp) movq 0x40(%rsp), %rax movq %rax, 0x440(%rsp) movq 0x70(%r13), %r12 movq 0x78(%r13), %r13 vbroadcastss 0xbe7c07(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0xf0(%rsp) vbroadcastss 0xbe8d59(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x100(%rsp) vmovaps %xmm0, 0x110(%rsp) vmovaps %xmm1, 0x120(%rsp) vmovaps %xmm0, 0x130(%rsp) vmovaps %xmm1, 0x140(%rsp) leaq 0x150(%rsp), %rax xorl %ecx, %ecx movl %ecx, 0x20(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, (%rax) vbroadcastss 0xbe889e(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x174(%rsp) movl %ecx, 0x17c(%rsp) movq %r13, %rax subq %r12, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x1304201 movq %r12, 0x490(%rsp) movq %r13, 0x498(%rsp) movq %r15, %rdi leaq 0x420(%rsp), %rsi leaq 0x490(%rsp), %rdx vzeroupper callq 0x130882e leaq 0x1f0(%rsp), %rcx movq 0x10(%rsp), %rdx movq 0x1e8(%rsp), %rax subq 0x1e0(%rsp), %rax movq %rdx, %rsi movq 0x78(%rdx), %rdx subq 0x70(%rsi), %rdx cmpq %rdx, %rax je 0x1303f30 movq 0xd8(%rsp), %rax movq 0x20(%rax), %rdi leaq 0x88(%rsp), %rax movq %rax, 0xf0(%rsp) movl $0x400, %ecx # imm = 0x400 xorl %esi, %esi leaq 0xf0(%rsp), %r8 callq 0x13080d3 movq %rax, 0x1e8(%rsp) leaq 0x1f0(%rsp), %rcx movq 0xd8(%rsp), %rax vmovsd 0x88(%rsp), %xmm0 vmovups 0x180(%rsp), %ymm1 vmovups 0x1a0(%rsp), %ymm2 vmovups 0x1c0(%rsp), %ymm3 vmovaps (%rcx), %xmm4 movq 0x78(%rsp), %rcx vmovaps %xmm4, (%rcx) vmovups %ymm1, 0x240(%rsp) vmovups %ymm2, 0x260(%rsp) vmovups %ymm3, 0x280(%rsp) movq 0x1e0(%rsp), %rcx movq %rcx, 0x2a0(%rsp) movq 0x1e8(%rsp), %rcx movq %rcx, 0x2a8(%rsp) vmovsd 0x200(%rsp), %xmm1 vmovsd 0x208(%rsp), %xmm2 vcmpltps %xmm0, %xmm2, %xmm3 vinsertps $0x50, %xmm2, %xmm0, %xmm4 # xmm4 = xmm0[0],xmm2[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm2, %xmm0 # xmm0 = xmm2[0],xmm0[1],xmm2[2,3] vblendvps %xmm3, %xmm4, %xmm0, %xmm0 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x2c0(%rsp) movq %rax, 0x2d0(%rsp) movq 0x50(%rsp), %rax movq %rax, 0x468(%rsp) leaq 0x80(%rsp), %rax movq %rax, 0x470(%rsp) movq 0x28(%rsp), %rax movq %rax, 0x478(%rsp) movq 0x10(%rsp), %rax movq 0x70(%rax), %r12 movq 0x78(%rax), %r13 vbroadcastss 0xbe79fe(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x490(%rsp) vbroadcastss 0xbe8b50(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x4a0(%rsp) vmovaps %xmm0, 0x4b0(%rsp) vmovaps %xmm1, 0x4c0(%rsp) vmovaps %xmm0, 0x4d0(%rsp) vmovaps %xmm1, 0x4e0(%rsp) leaq 0x4b0(%rsp), %rax xorl %ecx, %ecx movl %ecx, 0x60(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x40(%rax) vbroadcastss 0xbe8694(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x514(%rsp) movl %ecx, 0x51c(%rsp) movq %r13, %rax subq %r12, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x13042df movq %r12, 0x520(%rsp) movq %r13, 0x528(%rsp) leaq 0xf0(%rsp), %rdi leaq 0x468(%rsp), %rsi leaq 0x520(%rsp), %rdx vzeroupper callq 0x130abf6 movq 0x10(%rsp), %rdx movq 0x158(%rsp), %rax subq 0x150(%rsp), %rax movq 0x70(%rdx), %rsi leaq (%rax,%rsi), %rcx movq %rsi, 0x150(%rsp) movq %rcx, 0x158(%rsp) movq 0x78(%rdx), %rdx movq %rdx, %rcx subq %rsi, %rcx cmpq %rcx, %rax je 0x1304141 movq 0x50(%rsp), %rax movq 0x20(%rax), %rdi leaq 0x80(%rsp), %rax movq %rax, 0x490(%rsp) movl $0x400, %ecx # imm = 0x400 leaq 0x490(%rsp), %r8 callq 0x1308480 movq %rax, 0x158(%rsp) vmovsd 0x80(%rsp), %xmm0 vmovups 0xf0(%rsp), %ymm1 vmovups 0x110(%rsp), %ymm2 vmovups 0x130(%rsp), %ymm3 leaq 0x150(%rsp), %rax vmovaps 0x10(%rax), %xmm4 movq 0x320(%rsp), %rax vmovaps %xmm4, (%rax) vmovups %ymm1, 0x340(%rsp) vmovups %ymm2, 0x360(%rsp) vmovups %ymm3, 0x380(%rsp) movq 0x150(%rsp), %rax movq %rax, 0x3a0(%rsp) movq 0x158(%rsp), %rax movq %rax, 0x3a8(%rsp) vmovsd 0x178(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovsd 0x170(%rsp), %xmm1 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x3c0(%rsp) movq 0x50(%rsp), %rax movq %rax, 0x3d0(%rsp) jmp 0x1303862 movb $0x1, 0x49c(%rsp) leaq 0x4b0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0x4d0(%rsp) movb $0x4, 0x49d(%rsp) leaq 0x490(%rsp), %rdi vzeroupper callq 0x6a660 movq %r13, 0x520(%rsp) movq %r12, 0x528(%rsp) movq $0x80, 0x530(%rsp) leaq -0x754114(%rip), %r8 # 0xbb014c movq %r8, 0x218(%rsp) leaq 0x420(%rsp), %rax movq %rax, 0x220(%rsp) movq %r15, %rdi leaq 0x520(%rsp), %rsi leaq 0xf0(%rsp), %rdx leaq 0x218(%rsp), %rcx leaq 0x490(%rsp), %r9 callq 0x1308fe5 leaq 0x490(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1305c2b leaq 0x490(%rsp), %rdi callq 0x6aab0 leaq 0x6a0(%rsp), %rbp leaq 0x1f0(%rsp), %rcx leaq 0x180(%rsp), %r15 jmp 0x1303ecb movb $0x1, 0x52c(%rsp) leaq 0x540(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0x560(%rsp) movb $0x4, 0x52d(%rsp) leaq 0x520(%rsp), %rdi vzeroupper callq 0x6a660 movq %r13, 0x218(%rsp) movq %r12, 0x220(%rsp) movq $0x80, 0x228(%rsp) leaq -0x7541f2(%rip), %r8 # 0xbb014c movq %r8, 0x480(%rsp) leaq 0x468(%rsp), %rax movq %rax, 0x488(%rsp) leaq 0xf0(%rsp), %rdi leaq 0x218(%rsp), %rsi leaq 0x490(%rsp), %rdx leaq 0x480(%rsp), %rcx leaq 0x520(%rsp), %r9 callq 0x130b326 leaq 0x520(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1305c5d leaq 0x520(%rsp), %rdi callq 0x6aab0 leaq 0x6a0(%rsp), %rbp leaq 0x180(%rsp), %r15 jmp 0x13040d2 cmpl $0x1, %ecx je 0x13043f9 cmpl $0x3, %ecx jne 0x130384a movq 0x40(%rsp), %rdi vzeroupper callq 0xba71d2 movq 0x8(%rsp), %rdi movq 0x40(%rsp), %rsi leaq 0x240(%rsp), %rdx leaq 0x340(%rsp), %rcx callq 0x1f2794 jmp 0x130384a movq 0x40(%rsp), %rdi vzeroupper callq 0xba71d2 movq 0x8(%rsp), %rdi movq 0x40(%rsp), %rsi leaq 0x240(%rsp), %rdx leaq 0x340(%rsp), %rcx callq 0x1307c9a jmp 0x130384a movq 0x1620(%rsp), %rax testq %rax, %rax je 0x1304481 movq 0xd0(%rsp), %rcx vmovss 0x98(%rcx), %xmm0 vmovss 0x9c(%rcx), %xmm1 leaq 0x73c(%rsp), %rcx vmovss (%rcx), %xmm2 vmovss -0x4(%rcx), %xmm3 vcmpltps %xmm3, %xmm0, %xmm3 vcmpltps %xmm1, %xmm2, %xmm2 vorps %xmm2, %xmm3, %xmm2 vmovd %xmm2, %edx orb %dl, %r9b addq $0xf0, %rcx decq %rax jne 0x1304457 leaq 0x1e70(%rsp), %rax movq (%rax), %rbx movq 0x8(%rax), %r14 movq %r9, 0xc0(%rsp) testb $0x1, %r9b je 0x130458b movq $0x100, 0xf0(%rsp) # imm = 0x100 movq (%r14), %r12 movq 0x8(%r12), %rax cmpq %rax, %rbx je 0x130488c movq %r12, 0x330(%rsp) movb $0x1, 0x338(%rsp) movq %r12, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%r12), %rax testq %rax, %rax je 0x1304542 movq 0xa8(%r12), %rax addq 0x68(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%r12), %rax addq 0x98(%r12), %rax movq 0x50(%r12), %rcx addq 0x90(%r12), %rcx subq %rcx, %rax movq 0x8(%r12), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%r12), %rax addq 0x70(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%r12) vmovups %ymm0, 0x48(%r12) testq %rbx, %rbx je 0x130476c movq 0x10(%rbx), %rax movq %rax, 0x60(%r12) vmovups %ymm0, 0x88(%r12) vmovups %ymm0, 0x98(%r12) movq 0x10(%rbx), %rax movq %rax, 0xa0(%r12) jmp 0x1304780 movq $0xe0, 0xf0(%rsp) movq (%r14), %r12 movq 0x8(%r12), %rax cmpq %rax, %rbx je 0x1304933 movq %r12, 0x330(%rsp) movb $0x1, 0x338(%rsp) movq %r12, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%r12), %rax testq %rax, %rax je 0x130462b movq 0xa8(%r12), %rax addq 0x68(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%r12), %rax addq 0x98(%r12), %rax movq 0x50(%r12), %rcx addq 0x90(%r12), %rcx subq %rcx, %rax movq 0x8(%r12), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%r12), %rax addq 0x70(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%r12) vmovups %ymm0, 0x48(%r12) testq %rbx, %rbx je 0x13047db movq 0x10(%rbx), %rax movq %rax, 0x60(%r12) vmovups %ymm0, 0x88(%r12) vmovups %ymm0, 0x98(%r12) movq 0x10(%rbx), %rax movq %rax, 0xa0(%r12) jmp 0x13047ef addq $0x3, %r13 movq %r13, %rax shrq $0x2, %rax movq %rax, %r12 shlq $0x5, %rax leaq (%rax,%rax,2), %rax leaq 0x1e70(%rsp), %rcx movq (%rcx), %rbx movq 0x10(%rcx), %rbp movq %rax, 0x330(%rsp) movq (%rbp), %r15 movq 0x8(%r15), %rax cmpq %rax, %rbx je 0x1304f74 movq %r15, 0x1830(%rsp) movb $0x1, 0x1838(%rsp) movq %r15, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%r15), %rax testq %rax, %rax je 0x1304729 movq 0xa8(%r15), %rax addq 0x68(%r15), %rax movq 0x8(%r15), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%r15), %rax addq 0x98(%r15), %rax movq 0x50(%r15), %rcx addq 0x90(%r15), %rcx subq %rcx, %rax movq 0x8(%r15), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%r15), %rax addq 0x70(%r15), %rax movq 0x8(%r15), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%r15) vmovups %ymm0, 0x48(%r15) testq %rbx, %rbx je 0x1304ecd movq 0x10(%rbx), %rax movq %rax, 0x60(%r15) vmovups %ymm0, 0x88(%r15) vmovups %ymm0, 0x98(%r15) movq 0x10(%rbx), %rax movq %rax, 0xa0(%r15) jmp 0x1304edf vmovups %ymm0, 0x98(%r12) vmovups %ymm0, 0x88(%r12) movq %rbx, %rax xchgq %rax, 0x8(%r12) movq %r12, 0x180(%rsp) leaq 0xe220e1(%rip), %rdi # 0x2126878 movq %rdi, 0x230(%rsp) movb $0x1, 0x238(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%rbx), %rsi cmpq 0x140(%rbx), %rsi je 0x130484a movq 0x180(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%rbx) jmp 0x130485e vmovups %ymm0, 0x98(%r12) vmovups %ymm0, 0x88(%r12) movq %rbx, %rax xchgq %rax, 0x8(%r12) movq %r12, 0x180(%rsp) leaq 0xe22072(%rip), %rdi # 0x2126878 movq %rdi, 0x230(%rsp) movb $0x1, 0x238(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%rbx), %rsi cmpq 0x140(%rbx), %rsi je 0x13048f1 movq 0x180(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%rbx) jmp 0x1304905 leaq 0x130(%rbx), %rdi leaq 0x180(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x238(%rsp) jne 0x1304875 movq 0x230(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x338(%rsp) jne 0x130488c movq 0x330(%rsp), %rdi callq 0x1ee7c24 movq 0xf0(%rsp), %rax addq %rax, 0x28(%r14) movq 0x10(%r14), %rcx movl %ecx, %edx negl %edx andl $0xf, %edx leaq (%rcx,%rax), %r12 addq %rdx, %r12 movq %r12, 0x10(%r14) cmpq 0x18(%r14), %r12 ja 0x13057be addq %rdx, 0x30(%r14) subq %rax, %r12 addq 0x8(%r14), %r12 vbroadcastss 0xbe7154(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0xe0(%r12) vbroadcastss 0xbe82a5(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0xf0(%r12) movl $0x6, %r14d jmp 0x1304970 leaq 0x130(%rbx), %rdi leaq 0x180(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x238(%rsp) jne 0x130491c movq 0x230(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x338(%rsp) jne 0x1304933 movq 0x330(%rsp), %rdi callq 0x1ee7c24 movq 0xf0(%rsp), %rax addq %rax, 0x28(%r14) movq 0x10(%r14), %rcx movl %ecx, %edx negl %edx andl $0xf, %edx leaq (%rcx,%rax), %r12 addq %rdx, %r12 movq %r12, 0x10(%r14) cmpq 0x18(%r14), %r12 ja 0x13057f5 addq %rdx, 0x30(%r14) subq %rax, %r12 addq 0x8(%r14), %r12 movl $0x1, %r14d vbroadcastss 0xbe70a7(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, 0x60(%r12) vmovaps %xmm1, 0x40(%r12) vmovaps %xmm1, 0x20(%r12) vbroadcastss 0xbe81ed(%rip), %xmm2 # 0x1eecb84 vmovaps %xmm2, 0x70(%r12) vmovaps %xmm2, 0x50(%r12) vmovaps %xmm2, 0x30(%r12) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x80(%r12) vmovups %ymm0, 0xa0(%r12) vmovups %ymm0, 0xc0(%r12) vbroadcastsd 0xc1d4a1(%rip), %ymm0 # 0x1f21e78 vmovups %ymm0, (%r12) movq 0x1620(%rsp), %rcx vmovaps %xmm2, 0x10(%rsp) vmovaps %xmm1, 0x30(%rsp) testq %rcx, %rcx je 0x1304b30 vbroadcastss 0xbe8181(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xbe7014(%rip), %xmm1 # 0x1eeba20 leaq 0x1880(%rsp), %r15 xorl %r13d, %r13d leaq 0x230(%rsp), %rbx vmovaps %xmm1, 0x30(%rsp) vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x10(%rsp) vmovaps %xmm0, 0x90(%rsp) subq $0x20, %rsp leaq 0x1e90(%rsp), %rcx movq 0x10(%rcx), %rax movq %rax, 0x10(%rsp) vmovups (%rcx), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq 0x28(%rsp), %rsi movq %rbp, %rdx vzeroupper callq 0x13030b2 addq $0x20, %rsp movq 0x230(%rsp), %rax movq %rax, -0x50(%r15) vmovaps 0x240(%rsp), %xmm0 vmovaps %xmm0, -0x40(%r15) vmovaps 0x250(%rsp), %xmm1 vmovaps %xmm1, -0x30(%r15) vmovaps 0x260(%rsp), %xmm1 vmovaps %xmm1, -0x20(%r15) vmovaps 0x270(%rsp), %xmm1 vmovaps %xmm1, -0x10(%r15) vmovsd 0x280(%rsp), %xmm1 vmovsd %xmm1, (%r15) vmovaps 0x10(%rsp), %xmm1 vmaxps -0x30(%r15), %xmm1, %xmm1 vmovaps %xmm1, 0x10(%rsp) vmovaps 0x60(%rsp), %xmm1 vminps -0x20(%r15), %xmm1, %xmm1 vmovaps %xmm1, 0x60(%rsp) vmovaps 0x30(%rsp), %xmm1 vminps %xmm0, %xmm1, %xmm1 vmovaps %xmm1, 0x30(%rsp) vmovaps 0x90(%rsp), %xmm0 vmaxps -0x10(%r15), %xmm0, %xmm0 vmovaps %xmm0, 0x90(%rsp) incq %r13 movq 0x1620(%rsp), %rcx addq $0x60, %r15 addq $0xf0, %rbp cmpq %rcx, %r13 jb 0x1304a3a orq %r12, %r14 movl %r14d, %eax andl $0xf, %eax cmpl $0x1, %eax jne 0x130579d testq %rcx, %rcx je 0x1304ceb andq $-0x10, %r12 leaq 0x1884(%rsp), %rax xorl %edx, %edx vmovss 0xbe7bb3(%rip), %xmm0 # 0x1eec714 vbroadcastss 0xc1c356(%rip), %xmm1 # 0x1f20ec0 vbroadcastss 0xc1cf71(%rip), %xmm2 # 0x1f21ae4 vbroadcastss 0xc1cf64(%rip), %xmm3 # 0x1f21ae0 vbroadcastss 0xc1c33f(%rip), %xmm4 # 0x1f20ec4 vbroadcastss 0xbec456(%rip), %xmm5 # 0x1ef0fe4 movq -0x54(%rax), %rsi movq %rsi, (%r12,%rdx,8) vmovss (%rax), %xmm6 vmovss -0x4(%rax), %xmm7 vsubss %xmm7, %xmm6, %xmm6 vdivss %xmm6, %xmm0, %xmm6 vxorps %xmm1, %xmm7, %xmm8 vmulss %xmm6, %xmm8, %xmm8 vsubss %xmm8, %xmm0, %xmm9 vshufps $0x0, %xmm8, %xmm8, %xmm8 # xmm8 = xmm8[0,0,0,0] vmovaps -0x44(%rax), %xmm10 vmovaps -0x34(%rax), %xmm11 vmovaps -0x24(%rax), %xmm12 vmovaps -0x14(%rax), %xmm13 vmulps %xmm8, %xmm12, %xmm14 vshufps $0x0, %xmm9, %xmm9, %xmm9 # xmm9 = xmm9[0,0,0,0] vmulps %xmm9, %xmm10, %xmm15 vaddps %xmm15, %xmm14, %xmm14 vmulps %xmm8, %xmm13, %xmm8 vmulps %xmm9, %xmm11, %xmm9 vaddps %xmm9, %xmm8, %xmm8 vsubss %xmm7, %xmm0, %xmm7 vmulss %xmm6, %xmm7, %xmm6 vsubss %xmm6, %xmm0, %xmm7 vshufps $0x0, %xmm6, %xmm6, %xmm6 # xmm6 = xmm6[0,0,0,0] vmulps %xmm6, %xmm12, %xmm9 vshufps $0x0, %xmm7, %xmm7, %xmm7 # xmm7 = xmm7[0,0,0,0] vmulps %xmm7, %xmm10, %xmm10 vaddps %xmm10, %xmm9, %xmm9 vmulps %xmm6, %xmm13, %xmm6 vmulps %xmm7, %xmm11, %xmm7 vaddps %xmm7, %xmm6, %xmm6 vminps %xmm2, %xmm14, %xmm7 vmaxps %xmm3, %xmm8, %xmm8 vminps %xmm2, %xmm9, %xmm9 vmaxps %xmm3, %xmm6, %xmm6 vandps %xmm4, %xmm7, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vsubps %xmm10, %xmm7, %xmm7 vandps %xmm4, %xmm8, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vaddps %xmm10, %xmm8, %xmm8 vandps %xmm4, %xmm9, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vsubps %xmm10, %xmm9, %xmm9 vandps %xmm4, %xmm6, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vmovss %xmm7, 0x20(%r12,%rdx,4) vextractps $0x1, %xmm7, 0x40(%r12,%rdx,4) vaddps %xmm6, %xmm10, %xmm6 vextractps $0x2, %xmm7, 0x60(%r12,%rdx,4) vmovss %xmm8, 0x30(%r12,%rdx,4) vextractps $0x1, %xmm8, 0x50(%r12,%rdx,4) vsubps %xmm7, %xmm9, %xmm7 vextractps $0x2, %xmm8, 0x70(%r12,%rdx,4) vmovss %xmm7, 0x80(%r12,%rdx,4) vextractps $0x1, %xmm7, 0xa0(%r12,%rdx,4) vsubps %xmm8, %xmm6, %xmm6 vextractps $0x2, %xmm7, 0xc0(%r12,%rdx,4) vmovss %xmm6, 0x90(%r12,%rdx,4) vextractps $0x1, %xmm6, 0xb0(%r12,%rdx,4) vextractps $0x2, %xmm6, 0xd0(%r12,%rdx,4) incq %rdx addq $0x60, %rax cmpq %rdx, %rcx jne 0x1304b8e testb $0x1, 0xc0(%rsp) je 0x1304dd6 movq 0x8(%rsp), %rax addq $0x58, %rax leaq 0x5b0(%rsp), %rcx movq %rcx, 0x520(%rsp) movq %rax, 0x528(%rsp) movq 0x610(%rsp), %rbx movq 0x618(%rsp), %r15 vbroadcastss 0xbe6ced(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x330(%rsp) vbroadcastss 0xbe7e3f(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x340(%rsp) vmovaps %xmm0, 0x350(%rsp) vmovaps %xmm1, 0x360(%rsp) movq %r15, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x130582c leaq 0x230(%rsp), %rdx movq %rbx, (%rdx) movq %r15, 0x8(%rdx) leaq 0x180(%rsp), %rdi leaq 0x520(%rsp), %rsi vzeroupper callq 0x130d6d4 movq 0x70(%rsp), %rcx movq %r14, (%rcx) vmovaps 0x180(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rcx) vmovaps 0x190(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rcx) vmovaps 0x1a0(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rcx) vmovaps 0x1b0(%rsp), %xmm0 jmp 0x1304e08 movq 0x70(%rsp), %rcx movq %r14, (%rcx) vmovaps 0x30(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rcx) vmovaps 0x10(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rcx) vmovaps 0x60(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rcx) vmovaps 0x90(%rsp), %xmm0 vmovaps %xmm0, 0x40(%rcx) movq 0x638(%rsp), %rax movq %rax, 0x50(%rcx) cmpq $0x0, 0x1620(%rsp) je 0x1305783 xorl %r14d, %r14d movq 0x15a0(%rsp,%r14,8), %rax decq 0x8(%rax) jne 0x1304eb7 movq (%rax), %rbx testq %rbx, %rbx je 0x1304eb7 movq 0x18(%rbx), %r15 movq 0x20(%rbx), %rdi testq %rdi, %rdi je 0x1304e78 movq %r15, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1304e70 movzbl 0x8(%rbx), %edx vzeroupper callq 0x1ee67ca jmp 0x1304e78 vzeroupper callq 0x1ee612d testq %r15, %r15 je 0x1304e98 movq (%rbx), %rdi shlq $0x4, %r15 leaq (%r15,%r15,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %rbx, %rdi vzeroupper callq 0x6a4f0 incq %r14 cmpq 0x1620(%rsp), %r14 jb 0x1304e2b jmp 0x1305783 vmovups %ymm0, 0x98(%r15) vmovups %ymm0, 0x88(%r15) movq %rbx, %rax xchgq %rax, 0x8(%r15) movq %r15, 0x230(%rsp) leaq 0xe21983(%rip), %rdi # 0x2126878 movq %rdi, 0x6a0(%rsp) movb $0x1, 0x6a8(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%rbx), %rsi cmpq 0x140(%rbx), %rsi je 0x1304f32 movq 0x230(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%rbx) jmp 0x1304f46 leaq 0x130(%rbx), %rdi leaq 0x230(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x6a8(%rsp) jne 0x1304f5d movq 0x6a0(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x1838(%rsp) jne 0x1304f74 movq 0x1830(%rsp), %rdi callq 0x1ee7c24 movq 0x330(%rsp), %rsi addq %rsi, 0x28(%rbp) movq 0x10(%rbp), %rcx movl %ecx, %edx negl %edx andl $0xf, %edx leaq (%rcx,%rsi), %rax addq %rdx, %rax movq %rax, 0x10(%rbp) cmpq 0x18(%rbp), %rax ja 0x13058cd addq %rdx, 0x30(%rbp) subq %rsi, %rax addq 0x8(%rbp), %rax cmpq $0x7, %r12 movl $0x7, %ecx cmovbq %r12, %rcx orq %rax, %rcx orq $0x8, %rcx movq %rcx, 0x450(%rsp) cmpq $0x4, %r13 jae 0x1304ffd vbroadcastss 0xbe7bae(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xbe6a41(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, 0xe0(%rsp) vmovaps %xmm0, 0x50(%rsp) vmovaps %xmm0, 0x40(%rsp) vmovaps %xmm1, %xmm0 jmp 0x130573e movq %rax, 0x458(%rsp) cmpq $0x1, %r12 adcq $0x0, %r12 vbroadcastss 0xbe6a0a(%rip), %xmm0 # 0x1eeba20 vbroadcastss 0xbe7b65(%rip), %xmm1 # 0x1eecb84 movq $0x0, 0x28(%rsp) vxorps %xmm15, %xmm15, %xmm15 vxorps %xmm13, %xmm13, %xmm13 vmovss 0xbe76da(%rip), %xmm14 # 0x1eec714 vmovaps %xmm1, 0x40(%rsp) vmovaps %xmm0, 0xe0(%rsp) vmovaps %xmm1, 0x50(%rsp) movq %r12, 0x460(%rsp) movq 0x640(%rsp), %rax movq 0x20(%rax), %rcx movq 0x618(%rsp), %r9 movq 0x8(%rsp), %rax movq 0x70(%rax), %rax movq 0x68(%rax), %rdx vmovq 0x638(%rsp), %xmm1 vmovdqa %xmm1, 0xa0(%rsp) vpcmpeqd %xmm1, %xmm1, %xmm1 vmovdqa %xmm1, 0x6a0(%rsp) vmovdqa %xmm1, 0x1830(%rsp) leaq (%r14,%r14,4), %r10 shlq $0x4, %r10 addq %rcx, %r10 vmovaps %xmm15, 0x230(%rsp) vmovaps %xmm15, 0x330(%rsp) vmovaps %xmm15, 0x180(%rsp) vmovaps %xmm15, 0xf0(%rsp) xorl %r11d, %r11d cmpq %r9, %r14 jae 0x130514a movl 0xc(%r10), %eax movl %eax, 0x6a0(%rsp,%r11) movl 0x1c(%r10), %edi movl %edi, 0x1830(%rsp,%r11) movq 0x1e8(%rdx), %rbx movq (%rbx,%rax,8), %rax imulq 0x68(%rax), %rdi movq 0x58(%rax), %rbx movl 0xa0(%rax), %eax shrl $0x2, %eax movl (%rbx,%rdi), %ebp imull %eax, %ebp movl %ebp, 0x230(%rsp,%r11) movl 0x4(%rbx,%rdi), %ebp imull %eax, %ebp movl %ebp, 0x330(%rsp,%r11) movl 0x8(%rbx,%rdi), %ebp imull %eax, %ebp movl %ebp, 0x180(%rsp,%r11) imull 0xc(%rbx,%rdi), %eax movl %eax, 0xf0(%rsp,%r11) incq %r14 jmp 0x1305191 testq %r11, %r11 je 0x1305191 movl 0x6a0(%rsp), %eax movl %eax, 0x6a0(%rsp,%r11) movl $0xffffffff, 0x1830(%rsp,%r11) # imm = 0xFFFFFFFF movl 0x230(%rsp), %eax movl %eax, 0x230(%rsp,%r11) movl %eax, 0x330(%rsp,%r11) movl %eax, 0x180(%rsp,%r11) movl %eax, 0xf0(%rsp,%r11) leaq (%r14,%r14,4), %rax shlq $0x4, %rax addq %rcx, %rax cmpq %r9, %r14 cmovbq %rax, %r10 addq $0x4, %r11 cmpq $0x10, %r11 jne 0x13050d2 vmovaps %xmm0, 0x690(%rsp) movq 0x28(%rsp), %rax leaq (%rax,%rax,2), %rax shlq $0x5, %rax movq 0x458(%rsp), %rcx leaq (%rcx,%rax), %rsi vmovaps 0x230(%rsp), %xmm0 vmovaps %xmm0, (%rcx,%rax) vmovaps 0x330(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rcx,%rax) vmovaps 0x180(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rcx,%rax) vmovaps 0xf0(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rcx,%rax) vmovaps 0x6a0(%rsp), %xmm0 vmovaps %xmm0, 0x40(%rcx,%rax) vmovaps 0x1830(%rsp), %xmm0 vmovaps %xmm0, 0x50(%rcx,%rax) vmovshdup 0xa0(%rsp), %xmm0 # xmm0 = mem[1,1,3,3] vmovaps %xmm0, 0x320(%rsp) xorl %r10d, %r10d vbroadcastss 0xbe793a(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, %xmm2 vbroadcastss 0xbe67c9(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, %xmm3 vmovaps %xmm0, %xmm6 movq %rsi, 0x78(%rsp) movl 0x50(%rsi,%r10,4), %ecx movl $0xffffffff, %eax # imm = 0xFFFFFFFF cmpq %rax, %rcx je 0x13056dd vmovaps %xmm6, 0xc0(%rsp) vmovaps %xmm1, 0x30(%rsp) vmovaps %xmm3, 0x10(%rsp) vmovaps %xmm2, 0x90(%rsp) movl 0x40(%rsi,%r10,4), %eax movq %rdx, %r9 movq 0x1e8(%rdx), %rdi movq (%rdi,%rax,8), %rax vmovss 0x28(%rax), %xmm6 vmovss 0x2c(%rax), %xmm1 vmovaps 0xa0(%rsp), %xmm0 vsubss %xmm1, %xmm0, %xmm0 vmovss 0x30(%rax), %xmm2 vsubss %xmm1, %xmm2, %xmm2 vdivss %xmm2, %xmm0, %xmm3 vmovaps 0x320(%rsp), %xmm0 vsubss %xmm1, %xmm0, %xmm1 vdivss %xmm2, %xmm1, %xmm0 vmovss %xmm3, 0x60(%rsp) vmulss %xmm3, %xmm6, %xmm1 vmulss %xmm0, %xmm6, %xmm2 vroundss $0x9, %xmm1, %xmm1, %xmm4 vmovss %xmm2, 0xb0(%rsp) vroundss $0xa, %xmm2, %xmm2, %xmm8 vmaxss %xmm13, %xmm4, %xmm5 vcvttss2si %xmm5, %edi vcvttss2si %xmm4, %r11d vminss %xmm6, %xmm8, %xmm4 testl %r11d, %r11d movl $0xffffffff, %edx # imm = 0xFFFFFFFF cmovsl %edx, %r11d vcvttss2si %xmm8, %ebp vcvttss2si %xmm6, %ebx incl %ebx cmpl %ebx, %ebp cmovll %ebp, %ebx movslq %edi, %rdi movq 0x58(%rax), %rdx imulq 0x68(%rax), %rcx movl (%rdx,%rcx), %r15d movl 0x4(%rdx,%rcx), %r13d movq 0xe0(%rax), %r12 imulq $0x38, %rdi, %rdi movq (%r12,%rdi), %rax movq 0x10(%r12,%rdi), %r8 movq %r8, %rbp imulq %r15, %rbp vmovups (%rax,%rbp), %xmm8 movq %r8, %rbp imulq %r13, %rbp vmovups (%rax,%rbp), %xmm9 movl 0x8(%rdx,%rcx), %ebp movq %r8, %rsi imulq %rbp, %rsi vmovups (%rax,%rsi), %xmm15 vcvttss2si %xmm4, %esi movl 0xc(%rdx,%rcx), %ecx imulq %rcx, %r8 vmovups (%rax,%r8), %xmm2 vminps %xmm9, %xmm8, %xmm10 vminps %xmm2, %xmm15, %xmm7 vminps %xmm7, %xmm10, %xmm10 movslq %esi, %rax imulq $0x38, %rax, %rax movq (%r12,%rax), %rdx movq 0x10(%r12,%rax), %rsi movq %rsi, %r8 imulq %r15, %r8 vmovups (%rdx,%r8), %xmm7 movq %rsi, %r8 imulq %r13, %r8 vmovups (%rdx,%r8), %xmm3 movq %rsi, %r8 imulq %rbp, %r8 vmovups (%rdx,%r8), %xmm11 vmaxps %xmm9, %xmm8, %xmm8 imulq %rcx, %rsi vmovups (%rdx,%rsi), %xmm12 vmaxps %xmm2, %xmm15, %xmm2 vmaxps %xmm2, %xmm8, %xmm9 vminps %xmm3, %xmm7, %xmm2 vminps %xmm12, %xmm11, %xmm8 vminps %xmm8, %xmm2, %xmm15 vmaxps %xmm3, %xmm7, %xmm2 vmaxps %xmm12, %xmm11, %xmm3 vmaxps %xmm3, %xmm2, %xmm8 vsubss %xmm5, %xmm1, %xmm1 movl %ebx, %edx subl %r11d, %edx cmpl $0x1, %edx jne 0x1305477 vmaxss %xmm13, %xmm1, %xmm0 vsubss %xmm0, %xmm14, %xmm1 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps %xmm0, %xmm15, %xmm2 vshufps $0x0, %xmm1, %xmm1, %xmm3 # xmm3 = xmm1[0,0,0,0] vmulps %xmm3, %xmm10, %xmm1 vaddps %xmm2, %xmm1, %xmm1 vmulps %xmm0, %xmm8, %xmm0 vmulps %xmm3, %xmm9, %xmm2 vaddps %xmm0, %xmm2, %xmm5 vsubss 0xb0(%rsp), %xmm4, %xmm0 vmaxss %xmm13, %xmm0, %xmm0 vsubss %xmm0, %xmm14, %xmm2 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps %xmm0, %xmm10, %xmm3 vshufps $0x0, %xmm2, %xmm2, %xmm2 # xmm2 = xmm2[0,0,0,0] vmulps %xmm2, %xmm15, %xmm4 vaddps %xmm4, %xmm3, %xmm3 vmulps %xmm0, %xmm9, %xmm0 vmulps %xmm2, %xmm8, %xmm2 vaddps %xmm2, %xmm0, %xmm4 jmp 0x1305690 vmovss %xmm0, 0xd0(%rsp) movq 0x38(%r12,%rdi), %rdx movq 0x48(%r12,%rdi), %rsi movq %rsi, %rdi imulq %r15, %rdi vmovups (%rdx,%rdi), %xmm2 movq %rsi, %rdi imulq %r13, %rdi vmovups (%rdx,%rdi), %xmm3 movq %rsi, %rdi imulq %rbp, %rdi vmovups (%rdx,%rdi), %xmm5 imulq %rcx, %rsi vmovups (%rdx,%rsi), %xmm7 vminps %xmm3, %xmm2, %xmm11 vminps %xmm7, %xmm5, %xmm12 vminps %xmm12, %xmm11, %xmm11 vmaxps %xmm3, %xmm2, %xmm2 vmaxps %xmm7, %xmm5, %xmm3 vmaxps %xmm3, %xmm2, %xmm2 movq -0x38(%r12,%rax), %rdx movq -0x28(%r12,%rax), %rax movq %rax, %rsi imulq %r15, %rsi vmovups (%rdx,%rsi), %xmm3 movq %rax, %rsi imulq %r13, %rsi vmovups (%rdx,%rsi), %xmm5 movq %rax, %rsi imulq %rbp, %rsi vmovups (%rdx,%rsi), %xmm7 imulq %rcx, %rax vmovups (%rdx,%rax), %xmm12 vminps %xmm5, %xmm3, %xmm13 vmovaps %xmm14, %xmm0 vminps %xmm12, %xmm7, %xmm14 vminps %xmm14, %xmm13, %xmm13 vmovaps %xmm0, %xmm14 vmaxps %xmm5, %xmm3, %xmm3 vmaxps %xmm12, %xmm7, %xmm5 vmaxps %xmm5, %xmm3, %xmm7 vxorps %xmm12, %xmm12, %xmm12 vmaxss %xmm12, %xmm1, %xmm1 vsubss %xmm1, %xmm0, %xmm3 vshufps $0x0, %xmm1, %xmm1, %xmm5 # xmm5 = xmm1[0,0,0,0] vmulps %xmm5, %xmm11, %xmm1 vshufps $0x0, %xmm3, %xmm3, %xmm3 # xmm3 = xmm3[0,0,0,0] vmulps %xmm3, %xmm10, %xmm10 vaddps %xmm1, %xmm10, %xmm1 vmulps %xmm2, %xmm5, %xmm2 vmulps %xmm3, %xmm9, %xmm3 vaddps %xmm2, %xmm3, %xmm5 vsubss 0xb0(%rsp), %xmm4, %xmm2 vmaxss %xmm12, %xmm2, %xmm2 vsubss %xmm2, %xmm0, %xmm3 vshufps $0x0, %xmm2, %xmm2, %xmm2 # xmm2 = xmm2[0,0,0,0] vmulps %xmm2, %xmm13, %xmm4 vshufps $0x0, %xmm3, %xmm3, %xmm9 # xmm9 = xmm3[0,0,0,0] vmulps %xmm15, %xmm9, %xmm3 vaddps %xmm4, %xmm3, %xmm3 vmulps %xmm7, %xmm2, %xmm2 vmulps %xmm8, %xmm9, %xmm4 vaddps %xmm2, %xmm4, %xmm4 incl %r11d cmpl %ebx, %r11d jge 0x1305690 vmovss 0x60(%rsp), %xmm8 vmovss 0xd0(%rsp), %xmm2 vsubss %xmm8, %xmm2, %xmm2 movl %r11d, %eax imulq $0x38, %rax, %rax addq %rax, %r12 addq $0x10, %r12 vxorps %xmm15, %xmm15, %xmm15 vcvtsi2ss %r11d, %xmm6, %xmm7 vdivss %xmm6, %xmm7, %xmm7 vsubss %xmm8, %xmm7, %xmm7 vdivss %xmm2, %xmm7, %xmm7 vsubss %xmm7, %xmm14, %xmm8 vshufps $0x0, %xmm7, %xmm7, %xmm7 # xmm7 = xmm7[0,0,0,0] vmulps %xmm7, %xmm3, %xmm9 vshufps $0x0, %xmm8, %xmm8, %xmm8 # xmm8 = xmm8[0,0,0,0] vmulps %xmm1, %xmm8, %xmm10 vaddps %xmm10, %xmm9, %xmm9 vmulps %xmm7, %xmm4, %xmm7 vmulps %xmm5, %xmm8, %xmm8 vaddps %xmm7, %xmm8, %xmm7 movq -0x10(%r12), %rax movq (%r12), %rdx movq %rdx, %rsi imulq %r15, %rsi vmovups (%rax,%rsi), %xmm8 movq %rdx, %rsi imulq %r13, %rsi vmovups (%rax,%rsi), %xmm10 movq %rdx, %rsi imulq %rbp, %rsi vmovups (%rax,%rsi), %xmm11 imulq %rcx, %rdx vmovups (%rax,%rdx), %xmm12 vminps %xmm10, %xmm8, %xmm13 vminps %xmm12, %xmm11, %xmm14 vminps %xmm14, %xmm13, %xmm13 vmovaps %xmm0, %xmm14 vsubps %xmm9, %xmm13, %xmm9 vmaxps %xmm10, %xmm8, %xmm8 vmaxps %xmm12, %xmm11, %xmm10 vmaxps %xmm10, %xmm8, %xmm8 vsubps %xmm7, %xmm8, %xmm7 vminps %xmm15, %xmm9, %xmm8 vaddps %xmm1, %xmm8, %xmm1 vaddps %xmm3, %xmm8, %xmm3 vmovss 0x60(%rsp), %xmm8 vmaxps %xmm15, %xmm7, %xmm7 vaddps %xmm7, %xmm5, %xmm5 vaddps %xmm7, %xmm4, %xmm4 incl %r11d addq $0x38, %r12 cmpl %r11d, %ebx jne 0x13055c3 jmp 0x1305695 vxorps %xmm15, %xmm15, %xmm15 vmovaps 0xc0(%rsp), %xmm6 vminps %xmm1, %xmm6, %xmm6 vmovaps 0x30(%rsp), %xmm1 vmaxps %xmm5, %xmm1, %xmm1 vmovaps 0x10(%rsp), %xmm0 vminps %xmm3, %xmm0, %xmm3 vmovaps 0x90(%rsp), %xmm2 vmaxps %xmm4, %xmm2, %xmm2 incq %r10 cmpq $0x4, %r10 vxorps %xmm13, %xmm13, %xmm13 movq %r9, %rdx movq 0x78(%rsp), %rsi jne 0x1305264 vmovaps 0x40(%rsp), %xmm0 vmaxps %xmm2, %xmm0, %xmm0 vmovaps %xmm0, 0x40(%rsp) vmovaps 0xe0(%rsp), %xmm0 vminps %xmm3, %xmm0, %xmm0 vmovaps %xmm0, 0xe0(%rsp) vmovaps 0x50(%rsp), %xmm0 vmaxps %xmm1, %xmm0, %xmm0 vmovaps %xmm0, 0x50(%rsp) vmovaps 0x690(%rsp), %xmm0 vminps %xmm6, %xmm0, %xmm0 movq 0x28(%rsp), %rcx incq %rcx movq 0x460(%rsp), %r12 movq %rcx, 0x28(%rsp) cmpq %r12, %rcx jne 0x1305057 movq 0x70(%rsp), %rcx movq 0x450(%rsp), %rax movq %rax, (%rcx) vmovaps %xmm0, 0x10(%rcx) vmovaps 0x50(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rcx) vmovaps 0xe0(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rcx) vmovaps 0x40(%rsp), %xmm0 vmovaps %xmm0, 0x40(%rcx) movq 0x638(%rsp), %rax movq %rax, 0x50(%rcx) movq 0x70(%rsp), %rax addq $0x1e38, %rsp # imm = 0x1E38 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq xorl %edi, %edi testq %rcx, %rcx sete %dil leaq 0x1830(%rsp), %rdx movq %r12, %rsi vzeroupper callq 0x1f7b6a jmp 0x1304ceb movq %rcx, 0x10(%r14) shlq $0x2, %rax movq 0x20(%r14), %rcx cmpq %rcx, %rax jbe 0x1305901 leaq 0xf0(%rsp), %rsi movl $0x40, %edx movq %rbx, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 jmp 0x13048c3 movq %rcx, 0x10(%r14) shlq $0x2, %rax movq 0x20(%r14), %rcx cmpq %rcx, %rax jbe 0x1305972 leaq 0xf0(%rsp), %rsi movl $0x40, %edx movq %rbx, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 jmp 0x130496a leaq 0x230(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0xf0(%rsp), %rsi movq %r15, (%rsi) movq %rbx, 0x8(%rsi) movq $0x400, 0x10(%rsi) # imm = 0x400 leaq 0x490(%rsp), %rcx leaq 0x420(%rsp), %r8 movq %r8, (%rcx) leaq 0x520(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0x180(%rsp), %rdi leaq 0x330(%rsp), %rdx leaq 0x230(%rsp), %r9 callq 0x130dbf0 leaq 0x230(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1305c8f leaq 0x230(%rsp), %rdi callq 0x6aab0 jmp 0x1304d99 movq %rcx, 0x10(%rbp) shlq $0x2, %rsi movq 0x20(%rbp), %rcx cmpq %rcx, %rsi jbe 0x13059e3 leaq 0x330(%rsp), %rsi movl $0x40, %edx movq %rbx, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee jmp 0x1304fab leaq 0x230(%rsp), %rsi movq %rcx, (%rsi) movl $0x40, %edx movq %rbx, %rdi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x230(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0xf0(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx ja 0x1305ad0 movq %rax, 0x30(%r14) leaq 0x6a0(%rsp), %rbp jmp 0x13048c3 leaq 0x230(%rsp), %rsi movq %rcx, (%rsi) movl $0x40, %edx movq %rbx, %rdi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x230(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0xf0(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx ja 0x1305b46 movq %rax, 0x30(%r14) leaq 0x6a0(%rsp), %rbp jmp 0x130496a leaq 0x6a0(%rsp), %r15 movq %rcx, (%r15) movl $0x40, %edx movq %rbx, %rdi movq %r15, %rsi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %rsi movq %rax, 0x8(%rbp) movq 0x18(%rbp), %rax subq 0x10(%rbp), %rax addq 0x30(%rbp), %rax movq %rax, 0x30(%rbp) movq $0x0, 0x10(%rbp) movq (%r15), %rcx movq %rcx, 0x18(%rbp) movq 0x330(%rsp), %rdx movq %rdx, 0x10(%rbp) cmpq %rcx, %rdx ja 0x1305bbc movq %rax, 0x30(%rbp) movq %rsi, %rax jmp 0x1304fab movl $0x30, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0x6b0(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xbe71cb(%rip), %rsi # 0x1eecc38 leaq 0xbe71d7(%rip), %rdx # 0x1eecc4b leaq 0x6a0(%rsp), %rdi callq 0x8d7230 leaq 0xdfbe28(%rip), %rax # 0x21018b0 movq %rax, (%r14) movl $0x1, 0x8(%r14) leaq 0x10(%r14), %rdi movq %r14, %rax addq $0x20, %rax movq %rax, 0x10(%r14) movq 0x6a0(%rsp), %rsi movq 0x6a8(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xdfbd8f(%rip), %rsi # 0x2101850 leaq -0xa2ecac(%rip), %rdx # 0x8d6e1c movq %r14, %rdi callq 0x6a5d0 movq $0x0, 0x10(%r14) movq 0x20(%r14), %rax leaq 0x230(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %rbx, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x230(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0xf0(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx jbe 0x1305961 movq $0x0, 0x10(%r14) xorl %r12d, %r12d jmp 0x1305965 movq $0x0, 0x10(%r14) movq 0x20(%r14), %rax leaq 0x230(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %rbx, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x230(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0xf0(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx jbe 0x13059d2 movq $0x0, 0x10(%r14) xorl %r12d, %r12d jmp 0x13059d6 xorl %eax, %eax movq %rax, 0x10(%rbp) movq 0x20(%rbp), %rax leaq 0x6a0(%rsp), %r15 movq %rax, (%r15) movl $0x40, %edx movq %rbx, %rdi movq %r15, %rsi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %rsi movq %rax, 0x8(%rbp) movq 0x18(%rbp), %rax subq 0x10(%rbp), %rax addq 0x30(%rbp), %rax movq %rax, 0x30(%rbp) xorl %ecx, %ecx movq %rcx, 0x10(%rbp) movq (%r15), %rcx movq %rcx, 0x18(%rbp) movq 0x330(%rsp), %rdx movq %rdx, 0x10(%rbp) cmpq %rcx, %rdx jbe 0x1305a41 movq $0x0, 0x10(%rbp) xorl %esi, %esi jmp 0x1305a45 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbe5eb1(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe1efba(%rip), %rsi # 0x2124c08 movq 0xe1ed73(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbe5e7f(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe1ef88(%rip), %rsi # 0x2124c08 movq 0xe1ed41(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbe5e4d(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe1ef56(%rip), %rsi # 0x2124c08 movq 0xe1ed0f(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1305cf6 jmp 0x1305d14 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1305d17 jmp 0x1305d2d movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1305d30 jmp 0x1305f18 jmp 0x1305cf3 movq %rax, %rbx leaq 0x230(%rsp), %rdi callq 0x6aab0 jmp 0x1305e6c jmp 0x1305f18 jmp 0x1305f18 jmp 0x1305d14 movq %rax, %rbx leaq 0x520(%rsp), %rdi callq 0x6aab0 jmp 0x1305d52 jmp 0x1305f18 jmp 0x1305d2d movq %rax, %rbx leaq 0x490(%rsp), %rdi callq 0x6aab0 jmp 0x1305d52 jmp 0x1305f18 jmp 0x1305f18 jmp 0x1305d4f jmp 0x1305d4f jmp 0x1305d4f movq %rax, %rbx movq (%rsp), %rsi testq %rsi, %rsi je 0x1305e5f jmp 0x1305e57 movq %rax, %rbx movq %r12, %rdi callq 0x6a4f0 jmp 0x1305e6c movq %rax, %rbx leaq 0x6a0(%rsp), %rdi callq 0x8d6eda jmp 0x1305dcb jmp 0x1305f18 jmp 0x1305f18 movq %rax, %rbx leaq 0x230(%rsp), %rdi callq 0x8d6eda jmp 0x1305de5 movq %rax, %rbx leaq 0x230(%rsp), %rdi callq 0x8d6eda jmp 0x1305de5 jmp 0x1305f18 jmp 0x1305f18 jmp 0x1305f18 jmp 0x1305f18 movq %rax, %rbx leaq 0x1830(%rsp), %rdi callq 0x8d6eda jmp 0x1305f10 movq %rax, %rbx jmp 0x1305de5 movq %rax, %rbx leaq 0x330(%rsp), %rdi callq 0x8d6eda jmp 0x1305e6c movq %rax, %rbx xorl %ebp, %ebp jmp 0x1305e09 movq %rax, %rbx movq %r14, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x6a0(%rsp), %rdi cmpq %r15, %rdi je 0x1305e27 callq 0x6a4f0 jmp 0x1305e27 jmp 0x1305e69 jmp 0x1305e69 movq %rax, %rbx movb $0x1, %bpl testb %bpl, %bpl je 0x1305f10 movq %r14, %rdi callq 0x6a8a0 jmp 0x1305f10 jmp 0x1305f18 jmp 0x1305f18 jmp 0x1305e69 jmp 0x1305e4b movq %rax, %rbx movq (%rsp), %rsi testq %rsi, %rsi je 0x1305e5f movq %rsp, %rdi callq 0xbbb88e movq $0x0, (%rsp) jmp 0x1305e6c movq %rax, %rbx cmpq $0x0, 0x1620(%rsp) je 0x1305f10 xorl %r15d, %r15d movq 0x15a0(%rsp,%r15,8), %rax decq 0x8(%rax) jne 0x1305eff movq (%rax), %r14 testq %r14, %r14 je 0x1305eff movq 0x18(%r14), %r12 movq 0x20(%r14), %rdi testq %rdi, %rdi je 0x1305ec6 movq %r12, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1305ec1 movzbl 0x8(%r14), %edx callq 0x1ee67ca jmp 0x1305ec6 callq 0x1ee612d testq %r12, %r12 je 0x1305ee3 movq (%r14), %rdi shlq $0x4, %r12 leaq (%r12,%r12,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) leaq 0x10(%r14), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %r14, %rdi callq 0x6a4f0 incq %r15 cmpq 0x1620(%rsp), %r15 jb 0x1305e7e movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::QuadMesh>, 2ul>::TemporalBinInfo::merge2(embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::QuadMesh>, 2ul>::TemporalBinInfo const&, embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::QuadMesh>, 2ul>::TemporalBinInfo const&)
static __forceinline const TemporalBinInfo merge2(const TemporalBinInfo& a, const TemporalBinInfo& b) { TemporalBinInfo r = a; r.merge(b); return r; }
movq %rdi, %rax vmovaps (%rsi), %xmm0 vmovaps %xmm0, (%rdi) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, 0x10(%rdi) vmovaps 0x20(%rsi), %xmm1 vmovaps %xmm1, 0x20(%rdi) vmovaps 0x30(%rsi), %xmm2 vmovaps %xmm2, 0x30(%rdi) vmovaps 0x40(%rsi), %xmm2 vmovaps %xmm2, 0x40(%rdi) vmovaps 0x50(%rsi), %xmm2 vmovaps %xmm2, 0x50(%rdi) vmovaps 0x60(%rsi), %xmm3 vmovaps %xmm3, 0x60(%rdi) vmovaps 0x70(%rsi), %xmm3 vmovaps %xmm3, 0x70(%rdi) vmovaps 0x80(%rsi), %xmm3 vmovaps %xmm3, 0x80(%rdi) vmovdqa (%rdi), %xmm3 vpaddq (%rdx), %xmm3, %xmm3 vmovdqa %xmm3, (%rdi) vminps 0x10(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x10(%rdi) vmaxps 0x20(%rdx), %xmm1, %xmm0 vmovaps %xmm0, 0x20(%rdi) vmovaps 0x30(%rdi), %xmm0 vminps 0x30(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x30(%rdi) vmovaps 0x40(%rdi), %xmm0 vmaxps 0x40(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x40(%rdi) vminps 0x50(%rdx), %xmm2, %xmm0 vmovaps %xmm0, 0x50(%rdi) vmovaps 0x60(%rdi), %xmm0 vmaxps 0x60(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rdi) vmovaps 0x70(%rdi), %xmm0 vminps 0x70(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x70(%rdi) vmovaps 0x80(%rdi), %xmm0 vmaxps 0x80(%rdx), %xmm0, %xmm0 vmovaps %xmm0, 0x80(%rdi) retq
/embree[P]embree/kernels/bvh/../builders/heuristic_timesplit_array.h
embree::BVHNodeRecordMB4D<embree::NodeRefPtr<4>> const embree::avx::BVHBuilderMSMBlur::build<embree::NodeRefPtr<4>, embree::avx::RecalculatePrimRef<embree::UserGeometry>, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::UserGeometry, embree::Object>, embree::Scene::BuildProgressMonitorInterface>(embree::vector_t<embree::PrimRefMB, embree::aligned_monitored_allocator<embree::PrimRefMB, 16ul>>&, embree::PrimInfoMBT<embree::LBBox<embree::Vec3fa>> const&, embree::MemoryMonitorInterface*, embree::avx::RecalculatePrimRef<embree::UserGeometry>, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::UserGeometry, embree::Object>, embree::Scene::BuildProgressMonitorInterface, embree::avx::BVHBuilderMSMBlur::Settings const&)
static const BVHNodeRecordMB4D<NodeRef> build(mvector<PrimRefMB>& prims, const PrimInfoMB& pinfo, MemoryMonitorInterface* device, const RecalculatePrimRef recalculatePrimRef, const CreateAllocFunc createAlloc, const CreateNodeFunc createNode, const SetNodeFunc setNode, const CreateLeafFunc createLeaf, const ProgressMonitorFunc progressMonitor, const Settings& settings) { typedef BuilderT< NodeRef, RecalculatePrimRef, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, SetNodeFunc, CreateLeafFunc, ProgressMonitorFunc> Builder; Builder builder(device, recalculatePrimRef, createAlloc, createNode, setNode, createLeaf, progressMonitor, settings); return builder(prims,pinfo); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 movq %r9, %r10 movq %r8, %rax movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx movq 0x1b0(%rsp), %r8 movq 0x1c0(%rsp), %rdx movq 0x1b8(%rsp), %rsi leaq 0xdec58e(%rip), %rdi # 0x2102970 leaq 0x38(%rsp), %r9 movq %rdi, (%r9) movq 0x8(%rsi), %rsi movq %rsi, 0x8(%r9) movq %rdx, (%rsp) leaq 0x48(%rsp), %r12 movq %r12, %rdi movq %rcx, %rsi movq %rax, %rdx movq %r10, %rcx callq 0x13164ac vmovups (%r15), %ymm0 vmovups 0x20(%r15), %ymm1 vmovups 0x40(%r15), %ymm2 vmovups 0x70(%r15), %ymm3 leaq 0xd0(%rsp), %rdx vmovups %ymm3, 0x80(%rdx) movq $0x1, (%rdx) vmovups %ymm0, 0x10(%rdx) vmovups %ymm1, 0x30(%rdx) vmovups %ymm2, 0x50(%rdx) vmovaps 0x60(%r15), %xmm0 vmovaps %xmm0, 0x70(%rdx) movq %r14, 0xa0(%rdx) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) movq $0x0, 0x30(%rsp) movq 0x30(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x20(%rsp), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r12, %rsi movl $0x1, %ecx vzeroupper callq 0x13165c6 mfence movq %rbx, %rax addq $0x188, %rsp # imm = 0x188 popq %rbx popq %r12 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::UserGeometry>, 2ul>::TemporalBinInfo::bin(embree::PrimRefMB const*, unsigned long, unsigned long, embree::BBox<float>, embree::SetMB const&, embree::avx::RecalculatePrimRef<embree::UserGeometry> const&)
void bin(const PrimRefMB* prims, size_t begin, size_t end, BBox1f time_range, const SetMB& set, const RecalculatePrimRef& recalculatePrimRef) { for (int b=0; b<BINS-1; b++) { const float t = float(b+1)/float(BINS); const float ct = lerp(time_range.lower,time_range.upper,t); const float center_time = set.align_time(ct); if (center_time <= time_range.lower) continue; if (center_time >= time_range.upper) continue; const BBox1f dt0(time_range.lower,center_time); const BBox1f dt1(center_time,time_range.upper); /* find linear bounds for both time segments */ for (size_t i=begin; i<end; i++) { if (prims[i].time_range_overlap(dt0)) { const LBBox3fa bn0 = recalculatePrimRef.linearBounds(prims[i],dt0); #if MBLUR_BIN_LBBOX bounds0[b].extend(bn0); #else bounds0[b].extend(bn0.interpolate(0.5f)); #endif count0[b] += prims[i].timeSegmentRange(dt0).size(); } if (prims[i].time_range_overlap(dt1)) { const LBBox3fa bn1 = recalculatePrimRef.linearBounds(prims[i],dt1); #if MBLUR_BIN_LBBOX bounds1[b].extend(bn1); #else bounds1[b].extend(bn1.interpolate(0.5f)); #endif count1[b] += prims[i].timeSegmentRange(dt1).size(); } } } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x198, %rsp # imm = 0x198 movq %r9, 0x138(%rsp) vmovaps %xmm0, %xmm6 movq %rdx, %r15 movq %rdi, 0x130(%rsp) vmovshdup %xmm0, %xmm7 # xmm7 = xmm0[1,1,3,3] vmovss 0xbd0134(%rip), %xmm0 # 0x1eecb80 vmulss %xmm0, %xmm7, %xmm1 vmulss %xmm0, %xmm6, %xmm0 vaddss %xmm1, %xmm0, %xmm1 vmovss 0x80(%r8), %xmm0 vsubss %xmm0, %xmm1, %xmm2 vmovss 0x84(%r8), %xmm1 vsubss %xmm0, %xmm1, %xmm1 vdivss %xmm1, %xmm2, %xmm2 movq 0x78(%r8), %rax testq %rax, %rax js 0x131ca86 vcvtsi2ss %rax, %xmm3, %xmm3 jmp 0x131ca9b movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm3, %xmm3 vaddss %xmm3, %xmm3, %xmm3 vmulss %xmm3, %xmm2, %xmm2 vbroadcastss 0xc04418(%rip), %xmm4 # 0x1f20ec0 vandps %xmm4, %xmm2, %xmm4 vbroadcastss 0xc0746b(%rip), %xmm5 # 0x1f23f20 vorps %xmm5, %xmm4, %xmm4 vaddss %xmm4, %xmm2, %xmm2 vroundss $0xb, %xmm2, %xmm2, %xmm2 vdivss %xmm3, %xmm2, %xmm2 vmulss %xmm1, %xmm2, %xmm1 vaddss %xmm0, %xmm1, %xmm4 vcmpnleps %xmm4, %xmm7, %xmm0 vcmpnleps %xmm6, %xmm4, %xmm1 vandps %xmm0, %xmm1, %xmm0 vmovd %xmm0, %eax notb %al cmpq %rcx, %r15 setae %dl orb %al, %dl testb $0x1, %dl jne 0x131d5a0 vmovss 0xbe2568(%rip), %xmm1 # 0x1eff064 vmovss 0xbe2564(%rip), %xmm2 # 0x1eff068 vxorps %xmm5, %xmm5, %xmm5 vmovaps %xmm6, 0x180(%rsp) movq %rcx, 0x128(%rsp) movq %rsi, 0x120(%rsp) vmovaps %xmm7, 0x170(%rsp) vmovaps %xmm4, 0x160(%rsp) leaq (%r15,%r15,4), %rax shlq $0x4, %rax leaq (%rsi,%rax), %rdi vmulss 0x44(%rsi,%rax), %xmm1, %xmm0 vucomiss %xmm0, %xmm6 jae 0x131d077 vmulss 0x40(%rdi), %xmm2, %xmm0 vucomiss %xmm4, %xmm0 jae 0x131d077 movq %r15, 0xd0(%rsp) movq 0x138(%rsp), %rax movq (%rax), %rax movl 0xc(%rdi), %ecx movq 0x1e8(%rax), %rax movq (%rax,%rcx,8), %r12 vmovss 0x28(%r12), %xmm3 vmovss 0x2c(%r12), %xmm0 vsubss %xmm0, %xmm6, %xmm1 vmovss 0x30(%r12), %xmm2 vsubss %xmm0, %xmm2, %xmm2 vdivss %xmm2, %xmm1, %xmm1 vsubss %xmm0, %xmm4, %xmm0 vdivss %xmm2, %xmm0, %xmm0 vmovss %xmm1, 0x8(%rsp) vmulss %xmm1, %xmm3, %xmm2 vmovss %xmm2, 0x10(%rsp) vmovss %xmm0, 0x60(%rsp) vmulss %xmm0, %xmm3, %xmm1 vroundss $0x9, %xmm2, %xmm2, %xmm0 vmovss %xmm1, 0x30(%rsp) vroundss $0xa, %xmm1, %xmm1, %xmm1 vmaxss %xmm5, %xmm0, %xmm4 vmovss %xmm4, 0x80(%rsp) vminss %xmm3, %xmm1, %xmm2 vcvttss2si %xmm4, %r14d vmovss %xmm2, 0x20(%rsp) vcvttss2si %xmm2, %r15d vcvttss2si %xmm0, %ebp testl %ebp, %ebp movl $0xffffffff, %eax # imm = 0xFFFFFFFF cmovsl %eax, %ebp vcvttss2si %xmm1, %eax movq %rdi, 0xd8(%rsp) movl 0x1c(%rdi), %ebx vmovss %xmm3, 0xc(%rsp) vcvttss2si %xmm3, %r13d incl %r13d cmpl %r13d, %eax cmovll %eax, %r13d movq 0x18(%r12), %rax movq %rax, 0x40(%rsp) movl %ebx, 0x48(%rsp) movl %r14d, 0x4c(%rsp) leaq 0x140(%rsp), %rax movq %rax, 0x50(%rsp) leaq 0x40(%rsp), %rdi callq *0x58(%r12) movq 0x18(%r12), %rax movq %rax, 0xb0(%rsp) movl %ebx, 0xb8(%rsp) movl %r15d, 0x70(%rsp) movl %r15d, 0xbc(%rsp) leaq 0x40(%rsp), %rax movq %rax, 0xc0(%rsp) leaq 0xb0(%rsp), %rdi callq *0x58(%r12) movl %r13d, %eax subl %ebp, %eax vmovss 0x10(%rsp), %xmm0 vsubss 0x80(%rsp), %xmm0, %xmm0 cmpl $0x1, %eax jne 0x131cd44 vxorps %xmm10, %xmm10, %xmm10 vmaxss %xmm10, %xmm0, %xmm0 vmovss 0xbcfa5c(%rip), %xmm8 # 0x1eec714 vsubss %xmm0, %xmm8, %xmm1 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmovaps 0x40(%rsp), %xmm2 vmovaps 0x50(%rsp), %xmm3 vmulps %xmm2, %xmm0, %xmm4 vshufps $0x0, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[0,0,0,0] vmovaps 0x140(%rsp), %xmm5 vmovaps 0x150(%rsp), %xmm6 vmulps %xmm5, %xmm1, %xmm7 vaddps %xmm7, %xmm4, %xmm7 vmulps %xmm3, %xmm0, %xmm0 vmulps %xmm6, %xmm1, %xmm1 vaddps %xmm1, %xmm0, %xmm9 vmovss 0x20(%rsp), %xmm0 vsubss 0x30(%rsp), %xmm0, %xmm0 vmaxss %xmm10, %xmm0, %xmm0 vsubss %xmm0, %xmm8, %xmm1 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps %xmm5, %xmm0, %xmm4 vshufps $0x0, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[0,0,0,0] vmulps %xmm2, %xmm1, %xmm2 vaddps %xmm4, %xmm2, %xmm8 vmulps %xmm6, %xmm0, %xmm0 vmulps %xmm3, %xmm1, %xmm1 vaddps %xmm0, %xmm1, %xmm6 vxorps %xmm5, %xmm5, %xmm5 movq 0xd0(%rsp), %r15 jmp 0x131cf91 incl %r14d movq 0x18(%r12), %rax movq %rax, 0xe0(%rsp) movl %ebx, 0xe8(%rsp) movl %r14d, 0xec(%rsp) leaq 0xb0(%rsp), %rax movq %rax, 0xf0(%rsp) leaq 0xe0(%rsp), %r14 movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq *0x58(%r12) movl 0x70(%rsp), %ecx decl %ecx movq 0x18(%r12), %rax movq %rax, 0x90(%rsp) movl %ebx, 0x98(%rsp) movl %ecx, 0x9c(%rsp) movq %r14, 0xa0(%rsp) leaq 0x90(%rsp), %rdi callq *0x58(%r12) vxorps %xmm5, %xmm5, %xmm5 vmovss 0x10(%rsp), %xmm0 vmaxss %xmm5, %xmm0, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm1 # xmm1 = xmm0[0,0,0,0] vmulps 0xb0(%rsp), %xmm1, %xmm2 vmovss 0xbcf931(%rip), %xmm4 # 0x1eec714 vsubss %xmm0, %xmm4, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps 0x140(%rsp), %xmm0, %xmm3 vaddps %xmm3, %xmm2, %xmm7 vmulps 0xc0(%rsp), %xmm1, %xmm1 vmulps 0x150(%rsp), %xmm0, %xmm0 vaddps %xmm0, %xmm1, %xmm9 vmovss 0x20(%rsp), %xmm0 vsubss 0x30(%rsp), %xmm0, %xmm0 vmaxss %xmm5, %xmm0, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm1 # xmm1 = xmm0[0,0,0,0] vmulps 0xe0(%rsp), %xmm1, %xmm2 vsubss %xmm0, %xmm4, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps 0x40(%rsp), %xmm0, %xmm3 vmulps 0xf0(%rsp), %xmm1, %xmm1 vmulps 0x50(%rsp), %xmm0, %xmm0 vaddps %xmm3, %xmm2, %xmm8 vaddps %xmm0, %xmm1, %xmm6 incl %ebp cmpl %r13d, %ebp jge 0x131cf85 vmovss 0x8(%rsp), %xmm1 vmovss 0x60(%rsp), %xmm0 vsubss %xmm1, %xmm0, %xmm0 vmovss %xmm0, 0x4(%rsp) leaq 0x90(%rsp), %r14 movq 0xd0(%rsp), %r15 vmovaps %xmm6, 0x80(%rsp) vmovaps %xmm8, 0x10(%rsp) vmovaps %xmm9, 0x20(%rsp) vmovaps %xmm7, 0x30(%rsp) vcvtsi2ss %ebp, %xmm10, %xmm0 vdivss 0xc(%rsp), %xmm0, %xmm0 vsubss %xmm1, %xmm0, %xmm0 vdivss 0x4(%rsp), %xmm0, %xmm0 vsubss %xmm0, %xmm4, %xmm1 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps %xmm0, %xmm8, %xmm2 vshufps $0x0, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[0,0,0,0] vmulps %xmm1, %xmm7, %xmm3 vaddps %xmm3, %xmm2, %xmm2 vmovaps %xmm2, 0x70(%rsp) vmulps %xmm0, %xmm6, %xmm0 vmulps %xmm1, %xmm9, %xmm1 vaddps %xmm1, %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rsp) movq 0x18(%r12), %rax movq %rax, 0x108(%rsp) movl %ebx, 0x110(%rsp) movl %ebp, 0x114(%rsp) movq %r14, 0x118(%rsp) leaq 0x108(%rsp), %rdi callq *0x58(%r12) vmovaps 0x80(%rsp), %xmm6 vmovaps 0x10(%rsp), %xmm8 vmovaps 0x20(%rsp), %xmm9 vmovaps 0x30(%rsp), %xmm7 vmovss 0xbcf7dc(%rip), %xmm4 # 0x1eec714 vmovaps 0x90(%rsp), %xmm0 vsubps 0x70(%rsp), %xmm0, %xmm0 vmovaps 0xa0(%rsp), %xmm1 vsubps 0x60(%rsp), %xmm1, %xmm1 vxorps %xmm2, %xmm2, %xmm2 vminps %xmm2, %xmm0, %xmm0 vmaxps %xmm2, %xmm1, %xmm1 vaddps %xmm0, %xmm7, %xmm7 vaddps %xmm0, %xmm8, %xmm8 vaddps %xmm1, %xmm9, %xmm9 vaddps %xmm1, %xmm6, %xmm6 vmovss 0x8(%rsp), %xmm1 incl %ebp cmpl %ebp, %r13d jne 0x131ce84 jmp 0x131cf8d movq 0xd0(%rsp), %r15 vxorps %xmm5, %xmm5, %xmm5 movq 0x130(%rsp), %rdx vmovaps 0x10(%rdx), %xmm0 vminps %xmm7, %xmm0, %xmm0 vmovaps 0x20(%rdx), %xmm1 vmaxps %xmm9, %xmm1, %xmm1 vmovaps 0x30(%rdx), %xmm2 vminps %xmm8, %xmm2, %xmm2 vmovaps 0x40(%rdx), %xmm3 vmaxps %xmm6, %xmm3, %xmm3 vmovaps %xmm0, 0x10(%rdx) vmovaps %xmm1, 0x20(%rdx) vmovaps %xmm2, 0x30(%rdx) vmovaps %xmm3, 0x40(%rdx) movq 0xd8(%rsp), %rdi movl 0x3c(%rdi), %eax vcvtsi2ss %rax, %xmm11, %xmm0 vmovsd 0x40(%rdi), %xmm1 vmovshdup %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] vsubss %xmm1, %xmm2, %xmm2 vmovaps 0x180(%rsp), %xmm6 vsubss %xmm1, %xmm6, %xmm3 vdivss %xmm2, %xmm3, %xmm3 vmovaps 0x160(%rsp), %xmm4 vsubss %xmm1, %xmm4, %xmm1 vmulss 0xbd392a(%rip), %xmm3, %xmm3 # 0x1ef0940 vmulss %xmm0, %xmm3, %xmm3 vroundss $0x9, %xmm3, %xmm3, %xmm3 vmaxss %xmm3, %xmm5, %xmm3 vcvttss2si %xmm3, %eax vdivss %xmm2, %xmm1, %xmm1 vmulss 0xbd3910(%rip), %xmm1, %xmm1 # 0x1ef0944 vmulss %xmm0, %xmm1, %xmm1 vroundss $0xa, %xmm1, %xmm1, %xmm1 vminss %xmm0, %xmm1, %xmm0 vcvttss2si %xmm0, %ecx subl %eax, %ecx movslq %ecx, %rax addq %rax, (%rdx) movq 0x128(%rsp), %rcx movq 0x120(%rsp), %rsi vmovaps 0x170(%rsp), %xmm7 vmovss 0xbe1ff5(%rip), %xmm1 # 0x1eff064 vmovss 0xbe1ff1(%rip), %xmm2 # 0x1eff068 vmulss 0x44(%rdi), %xmm1, %xmm0 vucomiss %xmm0, %xmm4 jae 0x131d594 vmulss 0x40(%rdi), %xmm2, %xmm0 vucomiss %xmm7, %xmm0 jae 0x131d594 movq 0x138(%rsp), %rax movq (%rax), %rax movl 0xc(%rdi), %ecx movq 0x1e8(%rax), %rax movq (%rax,%rcx,8), %r12 vmovss 0x28(%r12), %xmm3 vmovss 0x2c(%r12), %xmm0 vsubss %xmm0, %xmm4, %xmm1 vmovss 0x30(%r12), %xmm2 vsubss %xmm0, %xmm2, %xmm2 vdivss %xmm2, %xmm1, %xmm1 vsubss %xmm0, %xmm7, %xmm0 vdivss %xmm2, %xmm0, %xmm0 vmovss %xmm1, 0x8(%rsp) vmulss %xmm1, %xmm3, %xmm2 vmovss %xmm2, 0x10(%rsp) vmovss %xmm0, 0x4(%rsp) vmulss %xmm0, %xmm3, %xmm1 vroundss $0x9, %xmm2, %xmm2, %xmm0 vmovss %xmm1, 0x30(%rsp) vroundss $0xa, %xmm1, %xmm1, %xmm1 vmaxss %xmm5, %xmm0, %xmm4 vmovss %xmm4, 0x80(%rsp) vminss %xmm3, %xmm1, %xmm2 vcvttss2si %xmm4, %ecx vmovss %xmm2, 0x20(%rsp) vcvttss2si %xmm2, %r14d vcvttss2si %xmm0, %ebp testl %ebp, %ebp movl $0xffffffff, %eax # imm = 0xFFFFFFFF cmovsl %eax, %ebp vcvttss2si %xmm1, %eax movq %rdi, 0xd8(%rsp) movl 0x1c(%rdi), %ebx vmovss %xmm3, 0xc(%rsp) vcvttss2si %xmm3, %r13d incl %r13d cmpl %r13d, %eax cmovll %eax, %r13d movq 0x18(%r12), %rax movq %rax, 0x40(%rsp) movl %ebx, 0x48(%rsp) movl %ecx, 0x60(%rsp) movl %ecx, 0x4c(%rsp) leaq 0x140(%rsp), %rax movq %rax, 0x50(%rsp) leaq 0x40(%rsp), %rdi callq *0x58(%r12) movq 0x18(%r12), %rax movq %rax, 0xb0(%rsp) movl %ebx, 0xb8(%rsp) movl %r14d, 0x70(%rsp) movl %r14d, 0xbc(%rsp) leaq 0x40(%rsp), %rax movq %rax, 0xc0(%rsp) leaq 0xb0(%rsp), %rdi callq *0x58(%r12) movl %r13d, %eax subl %ebp, %eax vmovss 0x10(%rsp), %xmm0 vsubss 0x80(%rsp), %xmm0, %xmm0 cmpl $0x1, %eax jne 0x131d26a vxorps %xmm10, %xmm10, %xmm10 vmaxss %xmm10, %xmm0, %xmm0 vmovss 0xbcf52a(%rip), %xmm8 # 0x1eec714 vsubss %xmm0, %xmm8, %xmm1 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmovaps 0x40(%rsp), %xmm2 vmovaps 0x50(%rsp), %xmm3 vmulps %xmm2, %xmm0, %xmm4 vshufps $0x0, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[0,0,0,0] vmovaps 0x140(%rsp), %xmm5 vmovaps 0x150(%rsp), %xmm6 vmulps %xmm5, %xmm1, %xmm7 vaddps %xmm7, %xmm4, %xmm7 vmulps %xmm3, %xmm0, %xmm0 vmulps %xmm6, %xmm1, %xmm1 vaddps %xmm1, %xmm0, %xmm9 vmovss 0x20(%rsp), %xmm0 vsubss 0x30(%rsp), %xmm0, %xmm0 vmaxss %xmm10, %xmm0, %xmm0 vsubss %xmm0, %xmm8, %xmm1 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps %xmm5, %xmm0, %xmm4 vshufps $0x0, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[0,0,0,0] vmulps %xmm2, %xmm1, %xmm2 vaddps %xmm4, %xmm2, %xmm8 vmulps %xmm6, %xmm0, %xmm0 vmulps %xmm3, %xmm1, %xmm1 vaddps %xmm0, %xmm1, %xmm6 jmp 0x131d4a3 movl 0x60(%rsp), %ecx incl %ecx movq 0x18(%r12), %rax movq %rax, 0xe0(%rsp) movl %ebx, 0xe8(%rsp) movl %ecx, 0xec(%rsp) leaq 0xb0(%rsp), %rax movq %rax, 0xf0(%rsp) leaq 0xe0(%rsp), %r14 movq %r14, %rdi vmovss %xmm0, 0x10(%rsp) callq *0x58(%r12) movl 0x70(%rsp), %ecx decl %ecx movq 0x18(%r12), %rax movq %rax, 0x90(%rsp) movl %ebx, 0x98(%rsp) movl %ecx, 0x9c(%rsp) movq %r14, 0xa0(%rsp) leaq 0x90(%rsp), %rdi callq *0x58(%r12) vxorps %xmm5, %xmm5, %xmm5 vmovss 0x10(%rsp), %xmm0 vmaxss %xmm5, %xmm0, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm1 # xmm1 = xmm0[0,0,0,0] vmulps 0xb0(%rsp), %xmm1, %xmm2 vmovss 0xbcf409(%rip), %xmm4 # 0x1eec714 vsubss %xmm0, %xmm4, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps 0x140(%rsp), %xmm0, %xmm3 vaddps %xmm3, %xmm2, %xmm7 vmulps 0xc0(%rsp), %xmm1, %xmm1 vmulps 0x150(%rsp), %xmm0, %xmm0 vaddps %xmm0, %xmm1, %xmm9 vmovss 0x20(%rsp), %xmm0 vsubss 0x30(%rsp), %xmm0, %xmm0 vmaxss %xmm5, %xmm0, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm1 # xmm1 = xmm0[0,0,0,0] vmulps 0xe0(%rsp), %xmm1, %xmm2 vsubss %xmm0, %xmm4, %xmm0 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps 0x40(%rsp), %xmm0, %xmm3 vmulps 0xf0(%rsp), %xmm1, %xmm1 vmulps 0x50(%rsp), %xmm0, %xmm0 vaddps %xmm3, %xmm2, %xmm8 vaddps %xmm0, %xmm1, %xmm6 incl %ebp cmpl %r13d, %ebp jge 0x131d4a3 vmovss 0x8(%rsp), %xmm1 vmovss 0x4(%rsp), %xmm0 vsubss %xmm1, %xmm0, %xmm0 vmovss %xmm0, 0x4(%rsp) leaq 0x90(%rsp), %r14 vmovaps %xmm6, 0x80(%rsp) vmovaps %xmm8, 0x10(%rsp) vmovaps %xmm9, 0x20(%rsp) vmovaps %xmm7, 0x30(%rsp) vcvtsi2ss %ebp, %xmm10, %xmm0 vdivss 0xc(%rsp), %xmm0, %xmm0 vsubss %xmm1, %xmm0, %xmm0 vdivss 0x4(%rsp), %xmm0, %xmm0 vsubss %xmm0, %xmm4, %xmm1 vshufps $0x0, %xmm0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmulps %xmm0, %xmm8, %xmm2 vshufps $0x0, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[0,0,0,0] vmulps %xmm1, %xmm7, %xmm3 vaddps %xmm3, %xmm2, %xmm2 vmovaps %xmm2, 0x70(%rsp) vmulps %xmm0, %xmm6, %xmm0 vmulps %xmm1, %xmm9, %xmm1 vaddps %xmm1, %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rsp) movq 0x18(%r12), %rax movq %rax, 0x108(%rsp) movl %ebx, 0x110(%rsp) movl %ebp, 0x114(%rsp) movq %r14, 0x118(%rsp) leaq 0x108(%rsp), %rdi callq *0x58(%r12) vmovaps 0x80(%rsp), %xmm6 vmovaps 0x10(%rsp), %xmm8 vmovaps 0x20(%rsp), %xmm9 vmovaps 0x30(%rsp), %xmm7 vmovss 0xbcf2bc(%rip), %xmm4 # 0x1eec714 vmovaps 0x90(%rsp), %xmm0 vsubps 0x70(%rsp), %xmm0, %xmm0 vmovaps 0xa0(%rsp), %xmm1 vsubps 0x60(%rsp), %xmm1, %xmm1 vxorps %xmm2, %xmm2, %xmm2 vminps %xmm2, %xmm0, %xmm0 vmaxps %xmm2, %xmm1, %xmm1 vaddps %xmm0, %xmm7, %xmm7 vaddps %xmm0, %xmm8, %xmm8 vaddps %xmm1, %xmm9, %xmm9 vaddps %xmm1, %xmm6, %xmm6 vmovss 0x8(%rsp), %xmm1 incl %ebp cmpl %ebp, %r13d jne 0x131d3a4 vxorps %xmm5, %xmm5, %xmm5 movq 0x130(%rsp), %rdx vmovaps 0x50(%rdx), %xmm0 vminps %xmm7, %xmm0, %xmm0 vmovaps 0x60(%rdx), %xmm1 vmaxps %xmm9, %xmm1, %xmm1 vmovaps 0x70(%rdx), %xmm2 vminps %xmm8, %xmm2, %xmm2 vmovaps 0x80(%rdx), %xmm3 vmaxps %xmm6, %xmm3, %xmm3 vmovaps %xmm0, 0x50(%rdx) vmovaps %xmm1, 0x60(%rdx) vmovaps %xmm2, 0x70(%rdx) vmovaps %xmm3, 0x80(%rdx) movq 0xd8(%rsp), %rcx movl 0x3c(%rcx), %eax vcvtsi2ss %rax, %xmm11, %xmm0 vmovsd 0x40(%rcx), %xmm1 vmovaps 0x160(%rsp), %xmm4 vsubss %xmm1, %xmm4, %xmm2 vmovshdup %xmm1, %xmm3 # xmm3 = xmm1[1,1,3,3] vsubss %xmm1, %xmm3, %xmm3 vdivss %xmm3, %xmm2, %xmm2 vmovaps 0x170(%rsp), %xmm7 vsubss %xmm1, %xmm7, %xmm1 vmulss 0xbd340e(%rip), %xmm2, %xmm2 # 0x1ef0940 vmulss %xmm0, %xmm2, %xmm2 vroundss $0x9, %xmm2, %xmm2, %xmm2 vmaxss %xmm2, %xmm5, %xmm2 vcvttss2si %xmm2, %eax vdivss %xmm3, %xmm1, %xmm1 vmulss 0xbd33f4(%rip), %xmm1, %xmm1 # 0x1ef0944 vmulss %xmm0, %xmm1, %xmm1 vroundss $0xa, %xmm1, %xmm1, %xmm1 vminss %xmm0, %xmm1, %xmm0 vcvttss2si %xmm0, %ecx subl %eax, %ecx movslq %ecx, %rax addq %rax, 0x8(%rdx) vmovaps 0x180(%rsp), %xmm6 movq 0x128(%rsp), %rcx movq 0x120(%rsp), %rsi vmovss 0xbe1ad8(%rip), %xmm1 # 0x1eff064 vmovss 0xbe1ad4(%rip), %xmm2 # 0x1eff068 incq %r15 cmpq %rcx, %r15 jne 0x131cb33 addq $0x198, %rsp # imm = 0x198 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/embree[P]embree/kernels/bvh/../builders/heuristic_timesplit_array.h
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<4>, embree::avx::RecalculatePrimRef<embree::UserGeometry>, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::UserGeometry, embree::Object>, embree::Scene::BuildProgressMonitorInterface>::findFallback(embree::SetMB const&)
__noinline Split findFallback(const SetMB& set) { /* split if primitives are not from same geometry */ if (!sameGeometry(set)) return Split(0.0f,Split::SPLIT_GEOMID); /* if a leaf can only hold a single time-segment, we might have to do additional temporal splits */ if (cfg.singleLeafTimeSegment) { /* test if one primitive has more than one time segment in time range, if so split time */ for (size_t i=set.begin(); i<set.end(); i++) { const PrimRefMB& prim = (*set.prims)[i]; const range<int> itime_range = prim.timeSegmentRange(set.time_range); const int localTimeSegments = itime_range.size(); assert(localTimeSegments > 0); if (localTimeSegments > 1) { const int icenter = (itime_range.begin() + itime_range.end())/2; const float splitTime = prim.timeStep(icenter); return Split(0.0f,(unsigned)Split::SPLIT_TEMPORAL,0,splitTime); } } } /* otherwise return fallback split */ return Split(0.0f,Split::SPLIT_FALLBACK); }
pushq %rbx movq %rdi, %rax movq 0x60(%rdx), %rdi movq 0x68(%rdx), %r8 cmpq %rdi, %r8 jne 0x131e2a8 movl $0x1, %ecx cmpb $0x1, 0x30(%rsi) jne 0x131e2e6 cmpq %r8, %rdi setb %sil jae 0x131e2e6 movq 0x90(%rdx), %r9 movq 0x20(%r9), %r10 vmovss 0x88(%rdx), %xmm0 vmovss 0x8c(%rdx), %xmm1 movq %r8, %r9 negq %r9 leaq (%rdi,%rdi,4), %rdx incq %rdi shlq $0x4, %rdx addq %r10, %rdx addq $0x44, %rdx vmovss 0xbd2720(%rip), %xmm2 # 0x1ef0940 vxorps %xmm3, %xmm3, %xmm3 vmovss 0xbd2718(%rip), %xmm4 # 0x1ef0944 movl -0x8(%rdx), %r10d vcvtsi2ss %r10, %xmm9, %xmm5 vmovsd -0x4(%rdx), %xmm6 vsubss %xmm6, %xmm0, %xmm7 vmovshdup %xmm6, %xmm8 # xmm8 = xmm6[1,1,3,3] vsubss %xmm6, %xmm8, %xmm8 vdivss %xmm8, %xmm7, %xmm7 vsubss %xmm6, %xmm1, %xmm6 vdivss %xmm8, %xmm6, %xmm6 vmulss %xmm2, %xmm7, %xmm7 vmulss %xmm5, %xmm7, %xmm7 vroundss $0x9, %xmm7, %xmm7, %xmm7 vmaxss %xmm7, %xmm3, %xmm7 vcvttss2si %xmm7, %r10d vmulss %xmm4, %xmm6, %xmm6 vmulss %xmm5, %xmm6, %xmm6 vroundss $0xa, %xmm6, %xmm6, %xmm6 vminss %xmm5, %xmm6, %xmm6 vcvttss2si %xmm6, %r11d movl %r11d, %ebx subl %r10d, %ebx cmpl $0x2, %ebx jge 0x131e2f9 cmpq %r8, %rdi setb %sil leaq (%r9,%rdi), %r10 incq %r10 incq %rdi addq $0x50, %rdx cmpq $0x1, %r10 jne 0x131e22c jmp 0x131e2e6 movq 0x90(%rdx), %rcx movq 0x20(%rcx), %rcx leaq (%rdi,%rdi,4), %r9 shlq $0x4, %r9 addq %rcx, %r9 addq $0x5c, %r9 movl -0x50(%r9), %ecx movq %rdi, %r10 incq %r10 cmpq %r8, %r10 jae 0x131e1c9 leaq 0x50(%r9), %r11 cmpl %ecx, (%r9) movq %r11, %r9 je 0x131e2c9 movl $0x3, %ecx movq $0x0, (%rax) movl $0x0, 0x8(%rax) movl %ecx, 0xc(%rax) popq %rbx retq addl %r10d, %r11d movl %r11d, %edi shrl $0x1f, %edi addl %r11d, %edi sarl %edi vmovss -0x4(%rdx), %xmm0 vmovss (%rdx), %xmm1 vcvtsi2ss %edi, %xmm9, %xmm2 vsubss %xmm0, %xmm1, %xmm1 vmulss %xmm2, %xmm1, %xmm1 vdivss %xmm5, %xmm1, %xmm1 vaddss %xmm1, %xmm0, %xmm0 movq $0x0, (%rax) vmovss %xmm0, 0x8(%rax) movl $0x2, 0xc(%rax) testb $0x1, %sil je 0x131e2e6 jmp 0x131e2f7 nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
unsigned long embree::parallel_filter<embree::PrimRefMB, unsigned long, embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::UserGeometry>, 2ul>::split(embree::avx::BinSplit<32ul> const&, embree::SetMB const&, embree::SetMB&, embree::SetMB&)::'lambda'(embree::PrimRefMB const&)>(embree::PrimRefMB*, unsigned long, unsigned long, unsigned long, embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::UserGeometry>, 2ul>::split(embree::avx::BinSplit<32ul> const&, embree::SetMB const&, embree::SetMB&, embree::SetMB&)::'lambda'(embree::PrimRefMB const&) const&)
inline Index parallel_filter( Ty* data, const Index begin, const Index end, const Index minStepSize, const Predicate& predicate) { /* sequential fallback */ if (end-begin <= minStepSize) return sequential_filter(data,begin,end,predicate); /* calculate number of tasks to use */ enum { MAX_TASKS = 64 }; const Index numThreads = TaskScheduler::threadCount(); const Index numBlocks = (end-begin+minStepSize-1)/minStepSize; const Index taskCount = min(numThreads,numBlocks,(Index)MAX_TASKS); /* filter blocks */ Index nused[MAX_TASKS]; Index nfree[MAX_TASKS]; parallel_for(taskCount, [&](const Index taskIndex) { const Index i0 = begin+(taskIndex+0)*(end-begin)/taskCount; const Index i1 = begin+(taskIndex+1)*(end-begin)/taskCount; const Index i2 = sequential_filter(data,i0,i1,predicate); nused[taskIndex] = i2-i0; nfree[taskIndex] = i1-i2; }); /* calculate offsets */ Index sused=0; Index sfree=0; Index pfree[MAX_TASKS]; for (Index i=0; i<taskCount; i++) { sused+=nused[i]; Index cfree = nfree[i]; pfree[i] = sfree; sfree+=cfree; } /* return if we did not filter out any element */ assert(sfree <= end-begin); assert(sused <= end-begin); if (sused == end-begin) return end; /* otherwise we have to copy misplaced elements around */ parallel_for(taskCount, [&](const Index taskIndex) { /* destination to write elements to */ Index dst = begin+(taskIndex+0)*(end-begin)/taskCount+nused[taskIndex]; Index dst_end = min(dst+nfree[taskIndex],begin+sused); if (dst_end <= dst) return; /* range of misplaced elements to copy to destination */ Index r0 = pfree[taskIndex]; Index r1 = r0+dst_end-dst; /* find range in misplaced elements in back to front order */ Index k0=0; for (Index i=taskCount-1; i>0; i--) { if (k0 > r1) break; Index k1 = k0+nused[i]; Index src = begin+(i+0)*(end-begin)/taskCount+nused[i]; for (Index i=max(r0,k0); i<min(r1,k1); i++) { Index isrc = src-i+k0-1; assert(dst >= begin && dst < end); assert(isrc >= begin && isrc < end); data[dst++] = data[isrc]; } k0 = k1; } }); return begin+sused; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x708, %rsp # imm = 0x708 movq %r8, %rbx movq %rdi, 0x28(%rsp) movq %rsi, 0x10(%rsp) movq %rdx, 0x20(%rsp) movq %rdx, %rax subq %rsi, %rax cmpq %rcx, %rax jbe 0x131e8e6 movq %rcx, %r14 xorl %edi, %edi callq 0x6ab80 movslq %eax, %rcx leaq 0x10(%rsp), %rsi movq (%rsi), %rax leaq 0x20(%rsp), %r13 movq (%r13), %rdx addq %r14, %rdx notq %rax addq %rdx, %rax xorl %edx, %edx divq %r14 cmpq %rax, %rcx cmovaeq %rax, %rcx cmpq $0x40, %rcx movl $0x40, %r14d cmovbq %rcx, %r14 leaq 0x30(%rsp), %rbp movq %r14, (%rbp) leaq 0x80(%rsp), %r12 movq %rsi, (%r12) movq %r13, 0x8(%r12) movq %rbp, 0x10(%r12) leaq 0x28(%rsp), %rax movq %rax, 0x18(%r12) movq %rbx, 0x20(%r12) leaq 0x500(%rsp), %rax movq %rax, 0x28(%r12) leaq 0x300(%rsp), %rax movq %rax, 0x30(%r12) leaq 0x100(%rsp), %r15 movw $0x401, 0xc(%r15) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r15) movq $0x8, 0x40(%r15) movq %r15, %rdi callq 0x6a660 leaq 0x40(%rsp), %rbx movq %r12, (%rbx) leaq 0x38(%rsp), %r8 movl $0x1, %edx xorl %edi, %edi movq %r14, %rsi movq %rbx, %rcx movq %r15, %r9 callq 0x13201c6 leaq 0x100(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x131ea6c leaq 0x100(%rsp), %r15 movq %r15, %rdi callq 0x6aab0 movq $0x0, 0x18(%rsp) movq 0x30(%rsp), %r14 testq %r14, %r14 je 0x131e983 xorl %ecx, %ecx xorl %edx, %edx xorl %eax, %eax addq 0x500(%rsp,%rcx,8), %rax movq %rdx, 0x100(%rsp,%rcx,8) addq 0x300(%rsp,%rcx,8), %rdx incq %rcx cmpq %rcx, %r14 jne 0x131e8c1 jmp 0x131e985 cmpq %rsi, %rdx jbe 0x131ea57 leaq (%rsi,%rsi,4), %rcx shlq $0x4, %rcx addq %rdi, %rcx addq $0x44, %rcx vmovss 0xbe075e(%rip), %xmm0 # 0x1eff064 vmovss 0xbe075a(%rip), %xmm1 # 0x1eff068 movq (%rbx), %rdx vmovss (%rcx), %xmm2 vmulss %xmm0, %xmm2, %xmm3 vmovss (%rdx), %xmm4 vucomiss %xmm3, %xmm4 jae 0x131e975 vmovss -0x4(%rcx), %xmm3 vmulss %xmm1, %xmm3, %xmm4 vucomiss 0x4(%rdx), %xmm4 jae 0x131e975 leaq (%rsi,%rsi,4), %rdx incq %rsi shlq $0x4, %rdx vmovaps -0x44(%rcx), %xmm4 vmovaps %xmm4, (%rdi,%rdx) vmovaps -0x34(%rcx), %xmm4 vmovaps %xmm4, 0x10(%rdi,%rdx) vmovaps -0x24(%rcx), %xmm4 vmovaps %xmm4, 0x20(%rdi,%rdx) vmovaps -0x14(%rcx), %xmm4 vmovaps %xmm4, 0x30(%rdi,%rdx) vmovss %xmm3, 0x40(%rdi,%rdx) vmovss %xmm2, 0x44(%rdi,%rdx) addq $0x50, %rcx decq %rax jne 0x131e90e jmp 0x131ea57 xorl %eax, %eax movq %rax, 0x18(%rsp) movq 0x20(%rsp), %rsi movq %rsi, %rcx subq 0x10(%rsp), %rcx cmpq %rcx, %rax je 0x131ea57 leaq 0x10(%rsp), %rax movq %rax, 0x40(%rsp) movq %r13, 0x48(%rsp) movq %rbp, 0x50(%rsp) leaq 0x500(%rsp), %rax movq %rax, 0x58(%rsp) leaq 0x300(%rsp), %rax movq %rax, 0x60(%rsp) leaq 0x18(%rsp), %rax movq %rax, 0x68(%rsp) movq %r15, 0x70(%rsp) leaq 0x28(%rsp), %rax movq %rax, 0x78(%rsp) leaq 0x80(%rsp), %r15 movw $0x401, 0xc(%r15) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r15) movq $0x8, 0x40(%r15) movq %r15, %rdi callq 0x6a660 leaq 0x38(%rsp), %rcx movq %rbx, (%rcx) leaq 0xf(%rsp), %r8 movl $0x1, %edx xorl %edi, %edi movq %r14, %rsi movq %r15, %r9 callq 0x1320a5a leaq 0x80(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x131ea9e leaq 0x80(%rsp), %rdi callq 0x6aab0 movq 0x18(%rsp), %rsi addq 0x10(%rsp), %rsi movq %rsi, %rax addq $0x708, %rsp # imm = 0x708 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbcd070(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe06179(%rip), %rsi # 0x2124c08 movq 0xe05f32(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbcd03e(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xe06147(%rip), %rsi # 0x2124c08 movq 0xe05f00(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x131eaf1 jmp 0x131eb1e jmp 0x131eaee movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x131eb09 movq %rax, %rbx leaq 0x80(%rsp), %rdi callq 0x6aab0 jmp 0x131eb16 jmp 0x131eb1e jmp 0x131eb1e jmp 0x131eb06 movq %rax, %rbx leaq 0x100(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/../../common/algorithms/parallel_filter.h
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<8>, embree::avx::RecalculatePrimRef<embree::UserGeometry>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeaf<8, embree::UserGeometry, embree::Object>, embree::Scene::BuildProgressMonitorInterface>::findFallback(embree::SetMB const&)
__noinline Split findFallback(const SetMB& set) { /* split if primitives are not from same geometry */ if (!sameGeometry(set)) return Split(0.0f,Split::SPLIT_GEOMID); /* if a leaf can only hold a single time-segment, we might have to do additional temporal splits */ if (cfg.singleLeafTimeSegment) { /* test if one primitive has more than one time segment in time range, if so split time */ for (size_t i=set.begin(); i<set.end(); i++) { const PrimRefMB& prim = (*set.prims)[i]; const range<int> itime_range = prim.timeSegmentRange(set.time_range); const int localTimeSegments = itime_range.size(); assert(localTimeSegments > 0); if (localTimeSegments > 1) { const int icenter = (itime_range.begin() + itime_range.end())/2; const float splitTime = prim.timeStep(icenter); return Split(0.0f,(unsigned)Split::SPLIT_TEMPORAL,0,splitTime); } } } /* otherwise return fallback split */ return Split(0.0f,Split::SPLIT_FALLBACK); }
pushq %rbx movq %rdi, %rax movq 0x60(%rdx), %rdi movq 0x68(%rdx), %r8 cmpq %rdi, %r8 jne 0x132b752 movl $0x1, %ecx cmpb $0x1, 0x30(%rsi) jne 0x132b790 cmpq %r8, %rdi setb %sil jae 0x132b790 movq 0x90(%rdx), %r9 movq 0x20(%r9), %r10 vmovss 0x88(%rdx), %xmm0 vmovss 0x8c(%rdx), %xmm1 movq %r8, %r9 negq %r9 leaq (%rdi,%rdi,4), %rdx incq %rdi shlq $0x4, %rdx addq %r10, %rdx addq $0x44, %rdx vmovss 0xbc5276(%rip), %xmm2 # 0x1ef0940 vxorps %xmm3, %xmm3, %xmm3 vmovss 0xbc526e(%rip), %xmm4 # 0x1ef0944 movl -0x8(%rdx), %r10d vcvtsi2ss %r10, %xmm9, %xmm5 vmovsd -0x4(%rdx), %xmm6 vsubss %xmm6, %xmm0, %xmm7 vmovshdup %xmm6, %xmm8 # xmm8 = xmm6[1,1,3,3] vsubss %xmm6, %xmm8, %xmm8 vdivss %xmm8, %xmm7, %xmm7 vsubss %xmm6, %xmm1, %xmm6 vdivss %xmm8, %xmm6, %xmm6 vmulss %xmm2, %xmm7, %xmm7 vmulss %xmm5, %xmm7, %xmm7 vroundss $0x9, %xmm7, %xmm7, %xmm7 vmaxss %xmm7, %xmm3, %xmm7 vcvttss2si %xmm7, %r10d vmulss %xmm4, %xmm6, %xmm6 vmulss %xmm5, %xmm6, %xmm6 vroundss $0xa, %xmm6, %xmm6, %xmm6 vminss %xmm5, %xmm6, %xmm6 vcvttss2si %xmm6, %r11d movl %r11d, %ebx subl %r10d, %ebx cmpl $0x2, %ebx jge 0x132b7a3 cmpq %r8, %rdi setb %sil leaq (%r9,%rdi), %r10 incq %r10 incq %rdi addq $0x50, %rdx cmpq $0x1, %r10 jne 0x132b6d6 jmp 0x132b790 movq 0x90(%rdx), %rcx movq 0x20(%rcx), %rcx leaq (%rdi,%rdi,4), %r9 shlq $0x4, %r9 addq %rcx, %r9 addq $0x5c, %r9 movl -0x50(%r9), %ecx movq %rdi, %r10 incq %r10 cmpq %r8, %r10 jae 0x132b673 leaq 0x50(%r9), %r11 cmpl %ecx, (%r9) movq %r11, %r9 je 0x132b773 movl $0x3, %ecx movq $0x0, (%rax) movl $0x0, 0x8(%rax) movl %ecx, 0xc(%rax) popq %rbx retq addl %r10d, %r11d movl %r11d, %edi shrl $0x1f, %edi addl %r11d, %edi sarl %edi vmovss -0x4(%rdx), %xmm0 vmovss (%rdx), %xmm1 vcvtsi2ss %edi, %xmm9, %xmm2 vsubss %xmm0, %xmm1, %xmm1 vmulss %xmm2, %xmm1, %xmm1 vdivss %xmm5, %xmm1, %xmm1 vaddss %xmm1, %xmm0, %xmm0 movq $0x0, (%rax) vmovss %xmm0, 0x8(%rax) movl $0x2, 0xc(%rax) testb $0x1, %sil je 0x132b790 jmp 0x132b7a1 nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::BVHNodeRecordMB4D<embree::NodeRefPtr<4>> const embree::avx::BVHBuilderMSMBlur::build<embree::NodeRefPtr<4>, embree::avx::RecalculatePrimRef<embree::Instance>, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::Instance, embree::InstancePrimitive>, embree::Scene::BuildProgressMonitorInterface>(embree::vector_t<embree::PrimRefMB, embree::aligned_monitored_allocator<embree::PrimRefMB, 16ul>>&, embree::PrimInfoMBT<embree::LBBox<embree::Vec3fa>> const&, embree::MemoryMonitorInterface*, embree::avx::RecalculatePrimRef<embree::Instance>, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeaf<4, embree::Instance, embree::InstancePrimitive>, embree::Scene::BuildProgressMonitorInterface, embree::avx::BVHBuilderMSMBlur::Settings const&)
static const BVHNodeRecordMB4D<NodeRef> build(mvector<PrimRefMB>& prims, const PrimInfoMB& pinfo, MemoryMonitorInterface* device, const RecalculatePrimRef recalculatePrimRef, const CreateAllocFunc createAlloc, const CreateNodeFunc createNode, const SetNodeFunc setNode, const CreateLeafFunc createLeaf, const ProgressMonitorFunc progressMonitor, const Settings& settings) { typedef BuilderT< NodeRef, RecalculatePrimRef, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, SetNodeFunc, CreateLeafFunc, ProgressMonitorFunc> Builder; Builder builder(device, recalculatePrimRef, createAlloc, createNode, setNode, createLeaf, progressMonitor, settings); return builder(prims,pinfo); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 movq %r9, %r10 movq %r8, %rax movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx movq 0x1b0(%rsp), %r8 movq 0x1c0(%rsp), %rdx movq 0x1b8(%rsp), %rsi leaq 0xdd5f7a(%rip), %rdi # 0x2102970 leaq 0x38(%rsp), %r9 movq %rdi, (%r9) movq 0x8(%rsi), %rsi movq %rsi, 0x8(%r9) movq %rdx, (%rsp) leaq 0x48(%rsp), %r12 movq %r12, %rdi movq %rcx, %rsi movq %rax, %rdx movq %r10, %rcx callq 0x132cac0 vmovups (%r15), %ymm0 vmovups 0x20(%r15), %ymm1 vmovups 0x40(%r15), %ymm2 vmovups 0x70(%r15), %ymm3 leaq 0xd0(%rsp), %rdx vmovups %ymm3, 0x80(%rdx) movq $0x1, (%rdx) vmovups %ymm0, 0x10(%rdx) vmovups %ymm1, 0x30(%rdx) vmovups %ymm2, 0x50(%rdx) vmovaps 0x60(%r15), %xmm0 vmovaps %xmm0, 0x70(%rdx) movq %r14, 0xa0(%rdx) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) movq $0x0, 0x30(%rsp) movq 0x30(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x20(%rsp), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r12, %rsi movl $0x1, %ecx vzeroupper callq 0x132cbda mfence movq %rbx, %rax addq $0x188, %rsp # imm = 0x188 popq %rbx popq %r12 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::Instance>, 2ul>::TemporalBinInfo::best(int, embree::BBox<float>, embree::SetMB const&)
Split best(int logBlockSize, BBox1f time_range, const SetMB& set) { float bestSAH = inf; float bestPos = 0.0f; for (int b=0; b<BINS-1; b++) { float t = float(b+1)/float(BINS); float ct = lerp(time_range.lower,time_range.upper,t); const float center_time = set.align_time(ct); if (center_time <= time_range.lower) continue; if (center_time >= time_range.upper) continue; const BBox1f dt0(time_range.lower,center_time); const BBox1f dt1(center_time,time_range.upper); /* calculate sah */ const size_t lCount = (count0[b]+(size_t(1) << logBlockSize)-1) >> int(logBlockSize); const size_t rCount = (count1[b]+(size_t(1) << logBlockSize)-1) >> int(logBlockSize); float sah0 = expectedApproxHalfArea(bounds0[b])*float(lCount)*dt0.size(); float sah1 = expectedApproxHalfArea(bounds1[b])*float(rCount)*dt1.size(); if (unlikely(lCount == 0)) sah0 = 0.0f; // happens for initial splits when objects not alive over entire shutter time if (unlikely(rCount == 0)) sah1 = 0.0f; const float sah = sah0+sah1; if (sah < bestSAH) { bestSAH = sah; bestPos = center_time; } } return Split(bestSAH*MBLUR_TIME_SPLIT_THRESHOLD,(unsigned)Split::SPLIT_TEMPORAL,0,bestPos); }
movq %rcx, %r8 movl %edx, %ecx movq %rdi, %rax vmovshdup %xmm0, %xmm1 # xmm1 = xmm0[1,1,3,3] vmovss 0xbba07e(%rip), %xmm2 # 0x1eecb80 vmulss %xmm2, %xmm1, %xmm3 vmulss %xmm2, %xmm0, %xmm4 vaddss %xmm3, %xmm4, %xmm4 vmovss 0x80(%r8), %xmm3 vsubss %xmm3, %xmm4, %xmm5 vmovss 0x84(%r8), %xmm4 vsubss %xmm3, %xmm4, %xmm4 vdivss %xmm4, %xmm5, %xmm5 movq 0x78(%r8), %rdx testq %rdx, %rdx js 0x1332b3c vcvtsi2ss %rdx, %xmm6, %xmm6 jmp 0x1332b51 movq %rdx, %rdi shrq %rdi andl $0x1, %edx orq %rdi, %rdx vcvtsi2ss %rdx, %xmm6, %xmm6 vaddss %xmm6, %xmm6, %xmm6 vmulss %xmm6, %xmm5, %xmm5 vbroadcastss 0xbee362(%rip), %xmm7 # 0x1f20ec0 vandps %xmm7, %xmm5, %xmm7 vbroadcastss 0xbf13b5(%rip), %xmm8 # 0x1f23f20 vorps %xmm7, %xmm8, %xmm7 vaddss %xmm7, %xmm5, %xmm5 vroundss $0xb, %xmm5, %xmm5, %xmm5 vdivss %xmm6, %xmm5, %xmm5 vmulss %xmm4, %xmm5, %xmm4 movl $0x1, %edi shlq %cl, %rdi vaddss %xmm3, %xmm4, %xmm3 movq 0x8(%rsi), %rdx addq %rdi, %rdx decq %rdx shrq %cl, %rdx testq %rdx, %rdx js 0x1332ba6 vcvtsi2ss %rdx, %xmm9, %xmm4 jmp 0x1332bbf movq %rdx, %r8 shrq %r8 movl %edx, %r9d andl $0x1, %r9d orq %r8, %r9 vcvtsi2ss %r9, %xmm9, %xmm4 vaddss %xmm4, %xmm4, %xmm4 vucomiss %xmm3, %xmm0 vmovss 0xbb8e55(%rip), %xmm5 # 0x1eeba20 vxorps %xmm6, %xmm6, %xmm6 jae 0x1332cf6 vucomiss %xmm1, %xmm3 jae 0x1332cf6 movl %ecx, %ecx movq (%rsi), %r8 addq %r8, %rdi decq %rdi shrq %cl, %rdi movq %rdi, %r8 shrq %r8 movl %edi, %ecx andl $0x1, %ecx orq %r8, %rcx testq %rdi, %rdi js 0x1332c07 vcvtsi2ss %rdi, %xmm9, %xmm5 jmp 0x1332c10 vcvtsi2ss %rcx, %xmm9, %xmm5 vaddss %xmm5, %xmm5, %xmm5 je 0x1332d0e vmovaps 0x20(%rsi), %xmm6 vsubps 0x10(%rsi), %xmm6, %xmm6 vmovaps 0x40(%rsi), %xmm7 vsubps 0x30(%rsi), %xmm7, %xmm7 vshufpd $0x1, %xmm6, %xmm6, %xmm8 # xmm8 = xmm6[1,0] vinsertps $0x4c, %xmm6, %xmm7, %xmm9 # xmm9 = xmm6[1],xmm7[1],zero,zero vinsertps $0x9c, %xmm7, %xmm8, %xmm8 # xmm8 = xmm8[0],xmm7[2],zero,zero vaddps %xmm8, %xmm9, %xmm10 vinsertps $0x1c, %xmm7, %xmm6, %xmm6 # xmm6 = xmm6[0],xmm7[0],zero,zero vmulps %xmm6, %xmm10, %xmm6 vmulps %xmm8, %xmm9, %xmm7 vaddps %xmm7, %xmm6, %xmm6 vhaddps %xmm6, %xmm6, %xmm6 vsubss %xmm0, %xmm3, %xmm0 vmulss 0xbb9f1d(%rip), %xmm6, %xmm6 # 0x1eecb80 vmulss %xmm5, %xmm6, %xmm5 vmulss %xmm5, %xmm0, %xmm0 testq %rdx, %rdx je 0x1332d17 vmovaps 0x60(%rsi), %xmm5 vsubps 0x50(%rsi), %xmm5, %xmm5 vmovaps 0x80(%rsi), %xmm6 vsubps 0x70(%rsi), %xmm6, %xmm6 vmovshdup %xmm5, %xmm7 # xmm7 = xmm5[1,1,3,3] vshufpd $0x1, %xmm5, %xmm5, %xmm8 # xmm8 = xmm5[1,0] vmovshdup %xmm6, %xmm9 # xmm9 = xmm6[1,1,3,3] vshufpd $0x1, %xmm6, %xmm6, %xmm10 # xmm10 = xmm6[1,0] vaddss %xmm7, %xmm8, %xmm11 vaddss %xmm10, %xmm9, %xmm12 vmulss %xmm7, %xmm8, %xmm7 vmulss %xmm5, %xmm11, %xmm5 vaddss %xmm7, %xmm5, %xmm5 vmulss %xmm10, %xmm9, %xmm7 vmulss %xmm6, %xmm12, %xmm6 vaddss %xmm7, %xmm6, %xmm6 vaddss %xmm6, %xmm5, %xmm5 vmulss %xmm2, %xmm5, %xmm2 vmulss %xmm4, %xmm2, %xmm2 vsubss %xmm3, %xmm1, %xmm1 vmulss %xmm2, %xmm1, %xmm1 vaddss %xmm1, %xmm0, %xmm0 vmovss 0xbb8d41(%rip), %xmm5 # 0x1eeba20 vucomiss %xmm5, %xmm0 vcmpneq_oqss %xmm5, %xmm0, %xmm1 vandps %xmm3, %xmm1, %xmm6 je 0x1332cf6 vmulss 0xbcc36a(%rip), %xmm0, %xmm5 # 0x1eff060 vmovss %xmm5, (%rax) movl $0x0, 0x4(%rax) vmovss %xmm6, 0x8(%rax) movl $0x2, 0xc(%rax) retq vxorps %xmm0, %xmm0, %xmm0 jmp 0x1332c6b vxorps %xmm1, %xmm1, %xmm1 jmp 0x1332cd3 nop
/embree[P]embree/kernels/bvh/../builders/heuristic_timesplit_array.h
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<8>, embree::avx::RecalculatePrimRef<embree::Instance>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeaf<8, embree::Instance, embree::InstancePrimitive>, embree::Scene::BuildProgressMonitorInterface>::recurse(embree::avx::BVHBuilderMSMBlur::BuildRecord const&, embree::FastAllocator::CachedAllocator, bool)
const NodeRecordMB4D recurse(const BuildRecord& current, Allocator alloc, bool toplevel) { /* get thread local allocator */ if (!alloc) alloc = createAlloc(); /* call memory monitor function to signal progress */ if (toplevel && current.size() <= cfg.singleThreadThreshold) progressMonitor(current.size()); /*! find best split */ const Split csplit = find(current.prims); /*! compute leaf and split cost */ const float leafSAH = cfg.intCost*current.prims.leafSAH(cfg.logBlockSize); const float splitSAH = cfg.travCost*current.prims.halfArea()+cfg.intCost*csplit.splitSAH(); assert((current.size() == 0) || ((leafSAH >= 0) && (splitSAH >= 0))); /*! create a leaf node when threshold reached or SAH tells us to stop */ if (current.size() <= cfg.minLeafSize || current.depth+MIN_LARGE_LEAF_LEVELS >= cfg.maxDepth || (current.size() <= cfg.maxLeafSize && leafSAH <= splitSAH)) { current.prims.deterministic_order(); return createLargeLeaf(current,alloc); } /*! perform initial split */ SetMB lprims,rprims; std::unique_ptr<mvector<PrimRefMB>> new_vector = split(csplit,current.prims,lprims,rprims); bool hasTimeSplits = new_vector != nullptr; NodeRecordMB4D values[MAX_BRANCHING_FACTOR]; LocalChildList children(current); { BuildRecord lrecord(lprims,current.depth+1); BuildRecord rrecord(rprims,current.depth+1); children.split(0,lrecord,rrecord,std::move(new_vector)); } /*! split until node is full or SAH tells us to stop */ while (children.size() < cfg.branchingFactor) { /*! find best child to split */ float bestArea = neg_inf; ssize_t bestChild = -1; for (size_t i=0; i<children.size(); i++) { if (children[i].size() <= cfg.minLeafSize) continue; if (expectedApproxHalfArea(children[i].prims.geomBounds) > bestArea) { bestChild = i; bestArea = expectedApproxHalfArea(children[i].prims.geomBounds); } } if (bestChild == -1) break; /* perform split */ BuildRecord& brecord = children[bestChild]; BuildRecord lrecord(current.depth+1); BuildRecord rrecord(current.depth+1); Split csplit = find(brecord.prims); std::unique_ptr<mvector<PrimRefMB>> new_vector = split(csplit,brecord.prims,lrecord.prims,rrecord.prims); hasTimeSplits |= new_vector != nullptr; children.split(bestChild,lrecord,rrecord,std::move(new_vector)); } /* detect time_ranges that have shrunken */ for (size_t i=0; i<children.size(); i++) { const BBox1f c = children[i].prims.time_range; const BBox1f p = current.prims.time_range; hasTimeSplits |= c.lower > p.lower || c.upper < p.upper; } /* sort buildrecords for simpler shadow ray traversal */ //std::sort(&children[0],&children[children.size()],std::greater<BuildRecord>()); // FIXME: reduces traversal performance of bvh8.triangle4 (need to verified) !! /*! create an inner node */ auto node = createNode(children.children.data(), children.numChildren, alloc, hasTimeSplits); LBBox3fa gbounds = empty; /* spawn tasks */ if (unlikely(current.size() > cfg.singleThreadThreshold)) { /*! parallel_for is faster than spawning sub-tasks */ parallel_for(size_t(0), children.size(), [&] (const range<size_t>& r) { for (size_t i=r.begin(); i<r.end(); i++) { values[i] = recurse(children[i],nullptr,true); _mm_mfence(); // to allow non-temporal stores during build } }); /*! merge bounding boxes */ for (size_t i=0; i<children.size(); i++) gbounds.extend(values[i].lbounds); } /* recurse into each child */ else { //for (size_t i=0; i<children.size(); i++) for (ssize_t i=children.size()-1; i>=0; i--) { values[i] = recurse(children[i],alloc,false); gbounds.extend(values[i].lbounds); } } setNode(current,children.children.data(),node,values,children.numChildren); /* calculate geometry bounds of this node */ if (unlikely(hasTimeSplits)) return NodeRecordMB4D(node,current.prims.linearBounds(recalculatePrimRef),current.prims.time_range); else return NodeRecordMB4D(node,gbounds,current.prims.time_range); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1ab8, %rsp # imm = 0x1AB8 movl %ecx, %ebp movq %rdx, (%rsp) movq %rsi, %r14 movq %rdi, %r15 leaq 0x1af0(%rsp), %rax cmpq $0x0, (%rax) movq %rsi, 0x8(%rsp) jne 0x133a95c movq 0x60(%r14), %r14 leaq 0xdea4e0(%rip), %rdi # 0x2124d38 callq 0x6a480 movq (%rax), %r12 testq %r12, %r12 jne 0x133a92f movq %rax, %rbx movl $0xc0, %edi movl $0x40, %esi callq 0x1ee60ac movq %rax, %r12 movq %rax, %rdi callq 0x1ee7b24 movq $0x0, 0x8(%r12) movq %r12, 0x40(%r12) vxorpd %xmm0, %xmm0, %xmm0 vmovupd %ymm0, 0x48(%r12) vmovupd %ymm0, 0x58(%r12) movq %r12, 0x80(%r12) vmovupd %ymm0, 0x88(%r12) vmovupd %ymm0, 0x98(%r12) movq %r12, (%rbx) leaq 0xdebfa9(%rip), %rdi # 0x2126878 movq %rdi, 0x720(%rsp) movb $0x1, 0x728(%rsp) vzeroupper callq 0x1ee7bb6 leaq 0x14b0(%rsp), %rsi movq %r12, (%rsi) leaq 0xdebf87(%rip), %rdi # 0x2126880 callq 0x90d91e movq 0x14b0(%rsp), %rsi testq %rsi, %rsi je 0x133a918 leaq 0x14b0(%rsp), %rdi callq 0x90da94 cmpb $0x1, 0x728(%rsp) jne 0x133a92f movq 0x720(%rsp), %rdi callq 0x1ee7c24 leaq 0x40(%r12), %rax subq $-0x80, %r12 cmpb $0x0, 0x10a(%r14) cmovneq %rax, %r12 leaq 0x1af0(%rsp), %rcx movq %r14, (%rcx) movq %rax, 0x8(%rcx) movq %r12, 0x10(%rcx) movq 0x8(%rsp), %r14 testb %bpl, %bpl je 0x133a998 movq (%rsp), %rcx movq 0x78(%rcx), %rax subq 0x70(%rcx), %rax cmpq 0x38(%r14), %rax ja 0x133a998 vmovq %rax, %xmm0 vpunpckldq 0xbb2140(%rip), %xmm0, %xmm0 # xmm0 = xmm0[0],mem[0],xmm0[1],mem[1] vsubpd 0xbb2148(%rip), %xmm0, %xmm0 # 0x1eecad0 movq 0x80(%r14), %rdi vhaddpd %xmm0, %xmm0, %xmm0 callq 0x91839e movq (%rsp), %rbx leaq 0x10(%rbx), %rdx leaq 0x6e0(%rsp), %r12 movq %r12, %rdi movq %r14, %rsi movq %rdx, 0x48(%rsp) callq 0x133dad6 movb 0x10(%r14), %cl movq $-0x1, %rax shlq %cl, %rax notq %rax addq 0x80(%rbx), %rax shrq %cl, %rax testq %rax, %rax movq %r15, 0x148(%rsp) js 0x133a9eb vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rax, %xmm1, %xmm0 jmp 0x133aa04 movq %rax, %rcx shrq %rcx andl $0x1, %eax orq %rcx, %rax vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rax, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 movq (%rsp), %rbx movq 0x78(%rbx), %r15 movq 0x70(%rbx), %rbp movq %r15, %r13 subq %rbp, %r13 cmpq 0x18(%r14), %r13 jbe 0x133aad4 movq (%rsp), %rax movq (%rax), %rax addq $0x8, %rax cmpq 0x8(%r14), %rax movq (%rsp), %rbx jae 0x133aad4 vmovss 0x98(%rbx), %xmm1 cmpq 0x20(%r14), %r13 ja 0x133ab2e vmovaps 0x20(%rbx), %xmm2 vmovaps 0x40(%rbx), %xmm3 vsubps 0x10(%rbx), %xmm2, %xmm2 vsubps 0x30(%rbx), %xmm3, %xmm3 vmovshdup %xmm2, %xmm4 # xmm4 = xmm2[1,1,3,3] vshufpd $0x1, %xmm2, %xmm2, %xmm5 # xmm5 = xmm2[1,0] vmovshdup %xmm3, %xmm6 # xmm6 = xmm3[1,1,3,3] vshufpd $0x1, %xmm3, %xmm3, %xmm7 # xmm7 = xmm3[1,0] vaddss %xmm5, %xmm4, %xmm8 vaddss %xmm7, %xmm6, %xmm9 vmulss %xmm5, %xmm4, %xmm4 vmulss %xmm2, %xmm8, %xmm2 vaddss %xmm4, %xmm2, %xmm2 vmulss %xmm7, %xmm6, %xmm4 vmulss %xmm3, %xmm9, %xmm3 vaddss %xmm4, %xmm3, %xmm3 vaddss %xmm3, %xmm2, %xmm2 vmovss 0x9c(%rbx), %xmm3 vmulss 0xbb20db(%rip), %xmm2, %xmm2 # 0x1eecb80 vsubss %xmm1, %xmm3, %xmm3 vmulss %xmm2, %xmm3, %xmm2 vmovss 0x2c(%r14), %xmm3 vmulss %xmm0, %xmm2, %xmm0 vmulss 0x6e0(%rsp), %xmm3, %xmm4 vmulss 0x28(%r14), %xmm2, %xmm2 vmulss %xmm0, %xmm3, %xmm0 vaddss %xmm4, %xmm2, %xmm2 vucomiss %xmm0, %xmm2 jb 0x133ab2e movq 0x48(%rsp), %rdi callq 0xba71d2 subq $0x20, %rsp leaq 0x1b10(%rsp), %rcx movq 0x10(%rcx), %rax movq %rax, 0x10(%rsp) vmovups (%rcx), %xmm0 vmovups %xmm0, (%rsp) movq 0x168(%rsp), %r12 movq %r12, %rdi movq %r14, %rsi movq %rbx, %rdx callq 0x133dc44 addq $0x20, %rsp movq %r12, %rax addq $0x1ab8, %rsp # imm = 0x1AB8 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movl 0x6ec(%rsp), %eax testl %eax, %eax jne 0x133caf4 vbroadcastss 0xbb0eda(%rip), %xmm0 # 0x1eeba20 leaq 0x720(%rsp), %r14 vmovaps %xmm0, (%r14) vbroadcastss 0xbb2028(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%r14) vmovaps %xmm0, 0x20(%r14) vmovaps %xmm1, 0x30(%r14) vmovaps %xmm0, 0x40(%r14) vmovaps %xmm1, 0x50(%r14) vxorps %xmm2, %xmm2, %xmm2 vmovups %ymm2, 0x60(%r14) xorl %eax, %eax movl %eax, 0x80(%r14) vbroadcastss 0xbb1b7e(%rip), %xmm3 # 0x1eec714 vmovlps %xmm3, 0x84(%r14) movl %eax, 0x8c(%r14) leaq 0x14b0(%rsp), %rbx vmovaps %xmm0, (%rbx) vmovaps %xmm1, 0x10(%rbx) vmovaps %xmm0, 0x20(%rbx) vmovaps %xmm1, 0x30(%rbx) vmovaps %xmm0, 0x40(%rbx) vmovaps %xmm1, 0x50(%rbx) vmovups %ymm2, 0x60(%rbx) movl %eax, 0x80(%rbx) vmovlps %xmm3, 0x84(%rbx) movl %eax, 0x8c(%rbx) vpermilps $0x0, 0x6e8(%rsp), %xmm0 # xmm0 = mem[0,0,0,0] leaq 0x2b0(%rsp), %rax vmovaps %xmm0, (%rax) movb 0x6e4(%rsp), %cl movl $0x1, %edx shll %cl, %edx movslq %edx, %rcx shlq $0x4, %rcx leaq 0xe15369(%rip), %rdx # 0x214ff80 vmovaps (%rdx,%rcx), %xmm0 leaq 0x220(%rsp), %rcx vmovaps %xmm0, (%rcx) leaq 0x90(%rsp), %r10 movq %r12, (%r10) movq %rax, 0x8(%r10) movq %rcx, 0x10(%r10) movq (%rsp), %r13 movq 0xa0(%r13), %rax movq 0x20(%rax), %rdi subq $0x8, %rsp movl $0x80, %eax leaq 0x3a8(%rsp), %r11 leaq 0x188(%rsp), %r12 leaq 0x5a8(%rsp), %rcx movq %rbp, %rsi movq %r15, %rdx movq %r14, %r8 movq %rbx, %r9 pushq $0xc00 # imm = 0xC00 pushq %rax pushq %r11 pushq %r12 pushq %r10 vzeroupper callq 0x125bcbe addq $0x30, %rsp movq 0xa0(%r13), %rcx vmovsd 0x98(%r13), %xmm0 vmovups (%r14), %ymm1 vmovups 0x20(%r14), %ymm2 vmovups 0x40(%r14), %ymm3 vmovups %ymm1, 0x500(%rsp) vmovups %ymm2, 0x520(%rsp) vmovups %ymm3, 0x540(%rsp) vmovaps 0x60(%r14), %xmm1 vmovaps %xmm1, 0x560(%rsp) vmovups 0x70(%r14), %ymm1 vmovups %ymm1, 0x570(%rsp) movq %rcx, 0x590(%rsp) movq %rbp, 0x560(%rsp) movq %rax, 0x568(%rsp) vmovsd 0x588(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm1, %xmm1 vmovlps %xmm1, 0x588(%rsp) vmovaps (%rbx), %xmm1 vmovaps %xmm1, 0x460(%rsp) vmovaps 0x10(%rbx), %xmm1 vmovaps %xmm1, 0x470(%rsp) vmovaps 0x20(%rbx), %xmm1 vmovaps %xmm1, 0x480(%rsp) vmovaps 0x30(%rbx), %xmm1 vmovaps %xmm1, 0x490(%rsp) vmovaps 0x40(%rbx), %xmm1 vmovaps %xmm1, 0x4a0(%rsp) vmovaps 0x50(%rbx), %xmm1 vmovaps %xmm1, 0x4b0(%rsp) vmovaps 0x60(%rbx), %xmm1 vmovaps %xmm1, 0x4c0(%rsp) vmovups 0x70(%rbx), %ymm1 vmovups %ymm1, 0x4d0(%rsp) movq %rcx, 0x4f0(%rsp) movq %rax, 0x4c0(%rsp) movq %r15, 0x4c8(%rsp) vmovsd 0x4e8(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovlps %xmm0, 0x4e8(%rsp) movq $0x0, 0x10(%rsp) cmpq $0x0, 0x10(%rsp) setne %r15b movl $0x1, %eax leaq 0x12a8(%rsp), %rdx movq %rax, -0x8(%rdx) movq %rax, 0x200(%rdx) movq (%rsp), %rax movq (%rax), %r14 movq %r14, -0xb88(%rdx) vmovaps 0x10(%rax), %xmm0 vmovaps %xmm0, -0xb78(%rdx) vmovaps 0x20(%rax), %xmm0 vmovaps %xmm0, -0xb68(%rdx) vmovaps 0x30(%rax), %xmm0 vmovaps %xmm0, -0xb58(%rdx) vmovaps 0x40(%rax), %xmm0 vmovaps %xmm0, -0xb48(%rdx) vmovaps 0x50(%rax), %xmm0 vmovaps %xmm0, -0xb38(%rdx) vmovaps 0x60(%rax), %xmm0 vmovaps %xmm0, -0xb28(%rdx) vmovups 0x70(%rax), %ymm0 vmovups %ymm0, -0xb18(%rdx) vmovaps 0x90(%rax), %xmm0 vmovaps %xmm0, -0xaf8(%rdx) movq 0xa0(%rax), %rax movq %rax, -0xae8(%rdx) movq %rax, (%rdx) movq $0x2, 0x8(%rdx) movq %rdx, -0x88(%rdx) vmovaps 0x500(%rsp), %xmm10 vmovaps 0x510(%rsp), %xmm11 vmovaps 0x520(%rsp), %xmm12 vmovaps 0x530(%rsp), %xmm13 vmovaps 0x540(%rsp), %xmm14 vmovaps 0x550(%rsp), %xmm15 vmovaps 0x560(%rsp), %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x570(%rsp), %xmm0 vmovaps %xmm0, 0x90(%rsp) vmovaps 0x580(%rsp), %xmm9 movq 0x590(%rsp), %r13 vmovaps 0x460(%rsp), %xmm2 vmovaps 0x470(%rsp), %xmm3 vmovaps 0x480(%rsp), %xmm4 vmovaps 0x490(%rsp), %xmm5 vmovaps 0x4a0(%rsp), %xmm6 vmovaps 0x4b0(%rsp), %xmm7 vmovaps 0x4c0(%rsp), %xmm8 vmovaps 0x4d0(%rsp), %xmm0 vmovaps %xmm0, 0x2b0(%rsp) vmovaps 0x4e0(%rsp), %xmm1 movq 0x4f0(%rsp), %r12 movq $0x0, 0x10(%rsp) cmpq %rax, %r13 je 0x133af9d leaq 0x12b8(%rsp), %rax movq $0x2, 0x1f0(%rax) movq %r13, (%rax) movq $0x1, 0x8(%rax) movq %rax, -0x98(%rax) jmp 0x133afb1 movq %rdx, 0x1220(%rsp) movq $0x3, 0x12b0(%rsp) cmpq 0x12a8(%rsp), %r12 je 0x133aff6 movq 0x14a8(%rsp), %rax leaq 0x1(%rax), %rcx movq %rcx, 0x14a8(%rsp) shlq $0x4, %rax leaq (%rdx,%rax), %rcx movq %r12, (%rdx,%rax) movq $0x1, 0x8(%rdx,%rax) movq 0x12a0(%rsp), %rax movq %rcx, 0x1220(%rsp,%rax,8) jmp 0x133b00e movq 0x12a0(%rsp), %rax movq %rdx, 0x1220(%rsp,%rax,8) incq 0x12b0(%rsp) decq 0x12b0(%rsp) jne 0x133b1a6 movq 0x12a8(%rsp), %rbx testq %rbx, %rbx je 0x133b1a6 vmovaps %xmm15, 0x660(%rsp) vmovaps %xmm14, 0x670(%rsp) vmovaps %xmm13, 0x680(%rsp) vmovaps %xmm12, 0x690(%rsp) vmovaps %xmm11, 0x6a0(%rsp) vmovaps %xmm10, 0x6b0(%rsp) vmovaps %xmm9, 0x6c0(%rsp) vmovaps %xmm8, 0x6d0(%rsp) vmovaps %xmm7, 0x390(%rsp) vmovaps %xmm6, 0x170(%rsp) vmovaps %xmm5, 0x210(%rsp) vmovaps %xmm4, 0x80(%rsp) vmovaps %xmm3, 0x30(%rsp) vmovaps %xmm2, 0x20(%rsp) vmovaps %xmm1, 0x50(%rsp) movq 0x18(%rbx), %rbp movq 0x20(%rbx), %rdi testq %rdi, %rdi je 0x133b0e2 movq %rbp, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x133b0da movzbl 0x8(%rbx), %edx vzeroupper callq 0x1ee67ca jmp 0x133b0e2 vzeroupper callq 0x1ee612d testq %rbp, %rbp je 0x133b109 movq (%rbx), %rdi shlq $0x4, %rbp leaq (,%rbp,4), %rsi addq %rbp, %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %rbx, %rdi vzeroupper callq 0x6a4f0 vmovaps 0x50(%rsp), %xmm1 vmovaps 0x20(%rsp), %xmm2 vmovaps 0x30(%rsp), %xmm3 vmovaps 0x80(%rsp), %xmm4 vmovaps 0x210(%rsp), %xmm5 vmovaps 0x170(%rsp), %xmm6 vmovaps 0x390(%rsp), %xmm7 vmovaps 0x6d0(%rsp), %xmm8 vmovaps 0x6c0(%rsp), %xmm9 vmovaps 0x6b0(%rsp), %xmm10 vmovaps 0x6a0(%rsp), %xmm11 vmovaps 0x690(%rsp), %xmm12 vmovaps 0x680(%rsp), %xmm13 vmovaps 0x670(%rsp), %xmm14 vmovaps 0x660(%rsp), %xmm15 incq %r14 leaq 0x798(%rsp), %rcx movq %r14, -0x78(%rcx) vmovaps %xmm10, -0x68(%rcx) vmovaps %xmm11, -0x58(%rcx) vmovaps %xmm12, -0x48(%rcx) vmovaps %xmm13, -0x38(%rcx) vmovaps %xmm14, -0x28(%rcx) vmovaps %xmm15, -0x18(%rcx) vmovaps 0x60(%rsp), %xmm0 vmovaps %xmm0, -0x8(%rcx) leaq 0xa0(%rsp), %rax vmovaps -0x10(%rax), %xmm0 vmovaps %xmm0, 0x8(%rcx) vmovaps %xmm9, 0x18(%rcx) movq %r13, 0x28(%rcx) imulq $0xb0, 0xb08(%rcx), %rax movq %r14, 0x720(%rsp,%rax) vmovaps %xmm2, 0x730(%rsp,%rax) vmovaps %xmm3, 0x740(%rsp,%rax) vmovaps %xmm4, 0x750(%rsp,%rax) vmovaps %xmm5, 0x760(%rsp,%rax) vmovaps %xmm6, 0x770(%rsp,%rax) vmovaps %xmm7, 0x780(%rsp,%rax) vmovaps %xmm8, 0x790(%rsp,%rax) leaq 0x2c0(%rsp), %rdx vmovaps -0x10(%rdx), %xmm0 vmovaps %xmm0, 0x7a0(%rsp,%rax) vmovaps %xmm1, 0x7b0(%rsp,%rax) movq %r12, 0x7c0(%rsp,%rax) incq 0xb08(%rcx) movq 0x8(%rsp), %rsi leaq 0x48(%rsi), %rax movq %rax, 0x210(%rsp) leaq 0x1e0(%rsp), %r12 movl $0x80, %eax leaq 0x90(%rsp), %rcx addq %rax, %rcx movq %rcx, 0x30(%rsp) leaq 0x2b0(%rsp), %rcx addq %rax, %rcx movq %rcx, 0x80(%rsp) vbroadcastss 0xbb075d(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x60(%rsp) vbroadcastss 0xbb18b2(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x50(%rsp) movq 0x12a0(%rsp), %rax cmpq (%rsi), %rax jae 0x133c11e testq %rax, %rax je 0x133b381 movq 0x18(%rsi), %rcx movq $-0x1, %r13 leaq 0x798(%rsp), %rdx vmovss 0xbb1877(%rip), %xmm0 # 0x1eecb84 xorl %r8d, %r8d movq (%rdx), %rdi subq -0x8(%rdx), %rdi cmpq %rcx, %rdi jbe 0x133b370 vmovaps -0x58(%rdx), %xmm1 vmovaps -0x38(%rdx), %xmm2 vsubps -0x68(%rdx), %xmm1, %xmm1 vsubps -0x48(%rdx), %xmm2, %xmm2 vinsertps $0x4c, %xmm1, %xmm2, %xmm3 # xmm3 = xmm1[1],xmm2[1],zero,zero vshufpd $0x1, %xmm1, %xmm1, %xmm4 # xmm4 = xmm1[1,0] vinsertps $0x9c, %xmm2, %xmm4, %xmm4 # xmm4 = xmm4[0],xmm2[2],zero,zero vaddps %xmm4, %xmm3, %xmm5 vmulps %xmm4, %xmm3, %xmm3 vinsertps $0x1c, %xmm2, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm2[0],zero,zero vmulps %xmm5, %xmm1, %xmm1 vaddps %xmm3, %xmm1, %xmm1 vhaddps %xmm1, %xmm1, %xmm1 vmulss 0xbb181d(%rip), %xmm1, %xmm1 # 0x1eecb80 vucomiss %xmm0, %xmm1 jbe 0x133b370 movq %r8, %r13 vmovaps %xmm1, %xmm0 incq %r8 addq $0xb0, %rdx cmpq %r8, %rax jne 0x133b310 jmp 0x133b388 movq $-0x1, %r13 cmpq $-0x1, %r13 je 0x133b9f0 imulq $0xb0, %r13, %rbx movq (%rsp), %rax movq (%rax), %rax incq %rax movq %rax, 0x90(%rsp) movq %rax, 0x2b0(%rsp) leaq (%rsp,%rbx), %rbp addq $0x730, %rbp # imm = 0x730 leaq 0x620(%rsp), %rdi movq %rbp, %rdx vzeroupper callq 0x133dad6 leaq (%rsp,%rbx), %r14 addq $0x720, %r14 # imm = 0x720 movl 0x62c(%rsp), %eax testl %eax, %eax movb %r15b, 0x20(%rsp) jne 0x133b9ff movq 0x70(%r14), %rbx movq %r12, %rbp movq 0x78(%r14), %r12 vbroadcastss 0xbb061c(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x220(%rsp) vbroadcastss 0xbb176e(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x230(%rsp) vmovaps %xmm0, 0x240(%rsp) vmovaps %xmm1, 0x250(%rsp) vmovaps %xmm0, 0x260(%rsp) vmovaps %xmm1, 0x270(%rsp) leaq 0x290(%rsp), %r15 movl $0x0, %eax movl %eax, 0x10(%r15) vxorps %xmm2, %xmm2, %xmm2 vmovups %ymm2, -0x10(%r15) vbroadcastss 0xbb12ad(%rip), %xmm3 # 0x1eec714 vmovlps %xmm3, 0x2a4(%rsp) movl %eax, 0x2ac(%rsp) vmovaps %xmm0, 0x180(%rsp) vmovaps %xmm1, 0x190(%rsp) vmovaps %xmm0, 0x1a0(%rsp) vmovaps %xmm1, 0x1b0(%rsp) vmovaps %xmm0, 0x1c0(%rsp) vmovaps %xmm1, 0x1d0(%rsp) movl %eax, 0x20(%rbp) vmovups %ymm2, (%rbp) vmovlps %xmm3, 0x204(%rsp) movl %eax, 0x20c(%rsp) vpermilps $0x0, 0x628(%rsp), %xmm0 # xmm0 = mem[0,0,0,0] vmovaps %xmm0, 0x5a0(%rsp) movb 0x624(%rsp), %cl movl $0x1, %eax shll %cl, %eax cltq shlq $0x4, %rax leaq 0xe14a8c(%rip), %rcx # 0x214ff80 vmovaps (%rcx,%rax), %xmm0 vmovaps %xmm0, 0x360(%rsp) leaq 0x620(%rsp), %rax movq %rax, 0x3a0(%rsp) leaq 0x5a0(%rsp), %rax movq %rax, 0x3a8(%rsp) leaq 0x360(%rsp), %rax movq %rax, 0x3b0(%rsp) movq 0xa0(%r14), %rax movq 0x20(%rax), %rdi subq $0x8, %rsp movq %rbx, %rsi movq %r12, %rdx leaq 0x458(%rsp), %rcx leaq 0x228(%rsp), %r8 leaq 0x188(%rsp), %r9 pushq $0xc00 # imm = 0xC00 pushq $0x80 leaq 0x450(%rsp), %rax pushq %rax leaq 0x178(%rsp), %rax pushq %rax leaq 0x3c8(%rsp), %rax pushq %rax vzeroupper callq 0x125bcbe addq $0x30, %rsp movq 0xa0(%r14), %rcx vmovsd 0x98(%r14), %xmm0 vmovups 0x220(%rsp), %ymm1 vmovups 0x240(%rsp), %ymm2 vmovups 0x260(%rsp), %ymm3 vmovups %ymm1, 0xa0(%rsp) vmovups %ymm2, 0xc0(%rsp) vmovups %ymm3, 0xe0(%rsp) vmovaps 0x280(%rsp), %xmm1 vmovaps %xmm1, 0x100(%rsp) vmovups (%r15), %ymm1 movq 0x30(%rsp), %rdx vmovups %ymm1, (%rdx) movq %rcx, 0x130(%rsp) movq %rbx, 0x100(%rsp) movq %rax, 0x108(%rsp) vmovsd 0x128(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm1, %xmm1 vmovlps %xmm1, 0x128(%rsp) vmovaps 0x180(%rsp), %xmm1 vmovaps %xmm1, 0x2c0(%rsp) vmovaps 0x190(%rsp), %xmm1 vmovaps %xmm1, 0x2d0(%rsp) vmovaps 0x1a0(%rsp), %xmm1 vmovaps %xmm1, 0x2e0(%rsp) vmovaps 0x1b0(%rsp), %xmm1 vmovaps %xmm1, 0x2f0(%rsp) vmovaps 0x1c0(%rsp), %xmm1 vmovaps %xmm1, 0x300(%rsp) vmovaps 0x1d0(%rsp), %xmm1 vmovaps %xmm1, 0x310(%rsp) vmovaps 0x1e0(%rsp), %xmm1 vmovaps %xmm1, 0x320(%rsp) vmovups 0x10(%rbp), %ymm1 movq 0x80(%rsp), %rdx vmovups %ymm1, (%rdx) movq %rcx, 0x350(%rsp) movq %rax, 0x320(%rsp) movq %r12, 0x328(%rsp) vmovsd 0x348(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovlps %xmm0, 0x348(%rsp) movq $0x0, 0x18(%rsp) leaq 0x12a8(%rsp), %rbx movb 0x20(%rsp), %r15b movq 0x18(%rsp), %r12 movq $0x0, 0x18(%rsp) movq 0x1220(%rsp,%r13,8), %rax movq 0x130(%rsp), %rcx cmpq (%rax), %rcx je 0x133b773 movq 0x14a8(%rsp), %rdx leaq 0x1(%rdx), %rsi movq %rsi, 0x14a8(%rsp) shlq $0x4, %rdx leaq (%rbx,%rdx), %rsi movq %rcx, (%rbx,%rdx) movq $0x1, 0x8(%rbx,%rdx) movq %rsi, 0x1220(%rsp,%r13,8) jmp 0x133b777 incq 0x8(%rax) movq 0x350(%rsp), %rcx cmpq (%rax), %rcx je 0x133b7bf movq 0x14a8(%rsp), %rdx leaq 0x1(%rdx), %rsi movq %rsi, 0x14a8(%rsp) shlq $0x4, %rdx leaq (%rbx,%rdx), %rsi movq %rcx, (%rbx,%rdx) movq $0x1, 0x8(%rbx,%rdx) movq 0x12a0(%rsp), %rcx movq %rsi, 0x1220(%rsp,%rcx,8) jmp 0x133b7d3 movq 0x12a0(%rsp), %rcx movq %rax, 0x1220(%rsp,%rcx,8) incq 0x8(%rax) decq 0x8(%rax) jne 0x133b862 movq (%rax), %rbx testq %rbx, %rbx je 0x133b862 movq 0x18(%rbx), %rbp movq 0x20(%rbx), %rdi testq %rdi, %rdi je 0x133b81c movq %rbp, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x133b814 movzbl 0x8(%rbx), %edx vzeroupper callq 0x1ee67ca jmp 0x133b81c vzeroupper callq 0x1ee612d testq %rbp, %rbp je 0x133b843 movq (%rbx), %rdi shlq $0x4, %rbp leaq (,%rbp,4), %rsi addq %rbp, %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %rbx, %rdi vzeroupper callq 0x6a4f0 testq %r12, %r12 setne %bl movq 0x90(%rsp), %rax movq %rax, (%r14) vmovaps 0xa0(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r14) vmovaps 0xb0(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r14) vmovaps 0xc0(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r14) vmovaps 0xd0(%rsp), %xmm0 vmovaps %xmm0, 0x40(%r14) vmovaps 0xe0(%rsp), %xmm0 vmovaps %xmm0, 0x50(%r14) vmovaps 0xf0(%rsp), %xmm0 vmovaps %xmm0, 0x60(%r14) leaq 0xa0(%rsp), %rax vmovups 0x60(%rax), %ymm0 vmovups %ymm0, 0x70(%r14) vmovaps 0x120(%rsp), %xmm0 vmovaps %xmm0, 0x90(%r14) movq 0x130(%rsp), %rax movq %rax, 0xa0(%r14) imulq $0xb0, 0x12a0(%rsp), %rax movq 0x2b0(%rsp), %rcx movq %rcx, 0x720(%rsp,%rax) vmovaps 0x2c0(%rsp), %xmm0 vmovaps %xmm0, 0x730(%rsp,%rax) vmovaps 0x2d0(%rsp), %xmm0 vmovaps %xmm0, 0x740(%rsp,%rax) vmovaps 0x2e0(%rsp), %xmm0 vmovaps %xmm0, 0x750(%rsp,%rax) vmovaps 0x2f0(%rsp), %xmm0 vmovaps %xmm0, 0x760(%rsp,%rax) vmovaps 0x300(%rsp), %xmm0 vmovaps %xmm0, 0x770(%rsp,%rax) vmovaps 0x310(%rsp), %xmm0 vmovaps %xmm0, 0x780(%rsp,%rax) leaq 0x2c0(%rsp), %rcx vmovups 0x60(%rcx), %ymm0 vmovups %ymm0, 0x790(%rsp,%rax) vmovaps 0x340(%rsp), %xmm0 vmovaps %xmm0, 0x7b0(%rsp,%rax) movq 0x350(%rsp), %rcx movq %rcx, 0x7c0(%rsp,%rax) incq 0x12a0(%rsp) movq 0x18(%rsp), %rsi testq %rsi, %rsi je 0x133b9e0 leaq 0x18(%rsp), %rdi vzeroupper callq 0xbbb88e orb %bl, %r15b movq 0x8(%rsp), %rsi leaq 0x1e0(%rsp), %r12 cmpq $-0x1, %r13 jne 0x133b2d8 jmp 0x133c11e cmpl $0x2, %eax jne 0x133c0bc vmovss 0x628(%rsp), %xmm0 vmovss 0x98(%r14), %xmm1 vmovss %xmm1, 0x78(%rsp) vmovss %xmm0, 0x7c(%rsp) vmovss %xmm0, 0x70(%rsp) vmovss 0x9c(%r14), %xmm0 vmovss %xmm0, 0x74(%rsp) movq 0xa0(%r14), %rax movq %rax, 0x170(%rsp) movl $0x28, %edi callq 0x6a170 movq %rax, %r12 movq 0x210(%rsp), %rax movq (%rax), %rdi movq 0x78(%r14), %rax movq %rdi, (%r12) movb $0x0, 0x8(%r12) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%r12) movq $0x0, 0x20(%r12) subq 0x70(%r14), %rax movb 0x20(%rsp), %r15b je 0x133bae7 movq %rax, 0x390(%rsp) shlq $0x4, %rax leaq (%rax,%rax,4), %rbx movq (%rdi), %rax movq %rbx, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %rbx # imm = 0x1C00000 jb 0x133bac3 movq %r12, %rsi addq $0x8, %rsi movq %rbx, %rdi callq 0x1ee665d jmp 0x133bad0 movl $0x10, %esi movq %rbx, %rdi callq 0x1ee60ac movq %rax, 0x20(%r12) movq 0x390(%rsp), %rax movq %rax, 0x10(%r12) movq %rax, 0x18(%r12) movq %r12, 0x18(%rsp) movq %r12, 0x150(%rsp) movq 0x170(%rsp), %rax movq %rax, 0x360(%rsp) leaq 0x78(%rsp), %rax movq %rax, 0x368(%rsp) movq 0x210(%rsp), %rax movq %rax, 0x370(%rsp) leaq 0x150(%rsp), %rax movq %rax, 0x378(%rsp) movq %rbp, 0x380(%rsp) movq 0x70(%r14), %rbx movq 0x78(%r14), %r12 vbroadcastss 0xbafed6(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x180(%rsp) vbroadcastss 0xbb1028(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x190(%rsp) vmovaps %xmm0, 0x1a0(%rsp) vmovaps %xmm1, 0x1b0(%rsp) vmovaps %xmm0, 0x1c0(%rsp) vmovaps %xmm1, 0x1d0(%rsp) leaq 0x1e0(%rsp), %rax xorl %ecx, %ecx movl %ecx, 0x20(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, (%rax) vbroadcastss 0xbb0b6d(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x204(%rsp) movl %ecx, 0x20c(%rsp) movq %r12, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x133bf14 movq %rbx, 0x3a0(%rsp) movq %r12, 0x3a8(%rsp) leaq 0x220(%rsp), %rdi leaq 0x360(%rsp), %rsi leaq 0x3a0(%rsp), %rdx vzeroupper callq 0x13349c8 movq 0x170(%rsp), %rbp movq 0x288(%rsp), %rax subq 0x280(%rsp), %rax movq 0x78(%r14), %rdx subq 0x70(%r14), %rdx cmpq %rdx, %rax leaq 0x290(%rsp), %rbx je 0x133bc5b movq 0x150(%rsp), %rax movq 0x20(%rax), %rdi leaq 0x78(%rsp), %rax movq %rax, 0x180(%rsp) movl $0x400, %ecx # imm = 0x400 xorl %esi, %esi leaq 0x180(%rsp), %r8 callq 0x133426d movq %rax, 0x288(%rsp) movq 0x150(%rsp), %rax vmovsd 0x78(%rsp), %xmm0 vmovups 0x220(%rsp), %ymm1 vmovups 0x240(%rsp), %ymm2 vmovups 0x260(%rsp), %ymm3 vmovaps (%rbx), %xmm4 movq 0x30(%rsp), %rcx vmovaps %xmm4, (%rcx) vmovups %ymm1, 0xa0(%rsp) vmovups %ymm2, 0xc0(%rsp) vmovups %ymm3, 0xe0(%rsp) movq 0x280(%rsp), %rcx movq %rcx, 0x100(%rsp) movq 0x288(%rsp), %rcx movq %rcx, 0x108(%rsp) vmovsd 0x2a0(%rsp), %xmm1 vmovsd 0x2a8(%rsp), %xmm2 vcmpltps %xmm0, %xmm2, %xmm3 vinsertps $0x50, %xmm2, %xmm0, %xmm4 # xmm4 = xmm0[0],xmm2[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm2, %xmm0 # xmm0 = xmm2[0],xmm0[1],xmm2[2,3] vblendvps %xmm3, %xmm4, %xmm0, %xmm0 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x120(%rsp) movq %rax, 0x130(%rsp) movq %rbp, 0x438(%rsp) leaq 0x70(%rsp), %rax movq %rax, 0x440(%rsp) movq 0x210(%rsp), %rax movq %rax, 0x448(%rsp) movq 0x70(%r14), %rbx movq 0x78(%r14), %r12 vbroadcastss 0xbafce0(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x3a0(%rsp) vbroadcastss 0xbb0e32(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x3b0(%rsp) vmovaps %xmm0, 0x3c0(%rsp) vmovaps %xmm1, 0x3d0(%rsp) vmovaps %xmm0, 0x3e0(%rsp) vmovaps %xmm1, 0x3f0(%rsp) leaq 0x3c0(%rsp), %rax xorl %ecx, %ecx movl %ecx, 0x60(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x40(%rax) vbroadcastss 0xbb0976(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x424(%rsp) movl %ecx, 0x42c(%rsp) movq %r12, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x133bfe4 movq %rbx, 0x5a0(%rsp) movq %r12, 0x5a8(%rsp) leaq 0x180(%rsp), %rdi leaq 0x438(%rsp), %rsi leaq 0x5a0(%rsp), %rdx vzeroupper callq 0x13369f0 movq 0x1e8(%rsp), %rax subq 0x1e0(%rsp), %rax movq 0x70(%r14), %rsi leaq (%rax,%rsi), %rcx movq %rsi, 0x1e0(%rsp) movq %rcx, 0x1e8(%rsp) movq 0x78(%r14), %rdx movq %rdx, %rcx subq %rsi, %rcx cmpq %rcx, %rax leaq 0x12a8(%rsp), %rbx leaq 0x1e0(%rsp), %r12 je 0x133be62 movq 0x20(%rbp), %rdi leaq 0x70(%rsp), %rax movq %rax, 0x3a0(%rsp) movl $0x400, %ecx # imm = 0x400 leaq 0x3a0(%rsp), %r8 callq 0x133461a movq %rax, 0x1e8(%rsp) vmovsd 0x70(%rsp), %xmm0 vmovups 0x180(%rsp), %ymm1 vmovups 0x1a0(%rsp), %ymm2 vmovups 0x1c0(%rsp), %ymm3 vmovaps 0x10(%r12), %xmm4 movq 0x80(%rsp), %rax vmovaps %xmm4, (%rax) vmovups %ymm1, 0x2c0(%rsp) vmovups %ymm2, 0x2e0(%rsp) vmovups %ymm3, 0x300(%rsp) movq 0x1e0(%rsp), %rax movq %rax, 0x320(%rsp) movq 0x1e8(%rsp), %rax movq %rax, 0x328(%rsp) vmovsd 0x208(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovsd 0x200(%rsp), %xmm1 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x340(%rsp) movq %rbp, 0x350(%rsp) jmp 0x133b71d movb $0x1, 0x3ac(%rsp) leaq 0x3c0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0x3e0(%rsp) movb $0x4, 0x3ad(%rsp) leaq 0x3a0(%rsp), %rdi vzeroupper callq 0x6a660 movq %r12, 0x5a0(%rsp) movq %rbx, 0x5a8(%rsp) movq $0x80, 0x5b0(%rsp) leaq -0x78be27(%rip), %r8 # 0xbb014c movq %r8, 0x158(%rsp) leaq 0x360(%rsp), %rax movq %rax, 0x160(%rsp) leaq 0x220(%rsp), %rdi leaq 0x5a0(%rsp), %rsi leaq 0x180(%rsp), %rdx leaq 0x158(%rsp), %rcx leaq 0x3a0(%rsp), %r9 callq 0x1334ddf leaq 0x3a0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x133d69f leaq 0x3a0(%rsp), %rdi callq 0x6aab0 movb 0x20(%rsp), %r15b jmp 0x133bbf9 movb $0x1, 0x5ac(%rsp) leaq 0x5c0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0x5e0(%rsp) movb $0x4, 0x5ad(%rsp) leaq 0x5a0(%rsp), %rdi vzeroupper callq 0x6a660 movq %r12, 0x158(%rsp) movq %rbx, 0x160(%rsp) movq $0x80, 0x168(%rsp) leaq -0x78bef7(%rip), %r8 # 0xbb014c movq %r8, 0x450(%rsp) leaq 0x438(%rsp), %rax movq %rax, 0x458(%rsp) leaq 0x180(%rsp), %rdi leaq 0x158(%rsp), %rsi leaq 0x3a0(%rsp), %rdx leaq 0x450(%rsp), %rcx leaq 0x5a0(%rsp), %r9 callq 0x1336d83 leaq 0x5a0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x133d6d1 leaq 0x5a0(%rsp), %rdi callq 0x6aab0 movb 0x20(%rsp), %r15b movq 0x170(%rsp), %rbp jmp 0x133bdf0 cmpl $0x1, %eax je 0x133c0f4 cmpl $0x3, %eax jne 0x133b707 movq %rbp, %rdi callq 0xba71d2 movq 0x8(%rsp), %rdi movq %rbp, %rsi leaq 0xa0(%rsp), %rdx leaq 0x2c0(%rsp), %rcx callq 0x1f490e jmp 0x133b707 movq %rbp, %rdi callq 0xba71d2 movq 0x8(%rsp), %rdi movq %rbp, %rsi leaq 0xa0(%rsp), %rdx leaq 0x2c0(%rsp), %rcx callq 0x134076e jmp 0x133b707 movq 0x12a0(%rsp), %rax testq %rax, %rax je 0x133c171 movq (%rsp), %rcx vmovss 0x98(%rcx), %xmm0 vmovss 0x9c(%rcx), %xmm1 leaq 0x7bc(%rsp), %rcx vmovss (%rcx), %xmm2 vmovss -0x4(%rcx), %xmm3 vcmpltps %xmm3, %xmm0, %xmm3 vcmpltps %xmm1, %xmm2, %xmm2 vorps %xmm2, %xmm3, %xmm2 vmovd %xmm2, %edx orb %dl, %r15b addq $0xb0, %rcx decq %rax jne 0x133c147 leaq 0x1af0(%rsp), %rax movq (%rax), %rbp movq 0x8(%rax), %r14 testb $0x1, %r15b je 0x133c25c movq $0x200, 0x180(%rsp) # imm = 0x200 movq (%r14), %rbx movq 0x8(%rbx), %rax cmpq %rax, %rbp je 0x133c43a movq %rbx, 0x2b0(%rsp) movb $0x1, 0x2b8(%rsp) movq %rbx, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%rbx), %rax testq %rax, %rax je 0x133c21d movq 0xa8(%rbx), %rax addq 0x68(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%rbx), %rax addq 0x98(%rbx), %rax movq 0x50(%rbx), %rcx addq 0x90(%rbx), %rcx subq %rcx, %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%rbx), %rax addq 0x70(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%rbx) vmovups %ymm0, 0x48(%rbx) testq %rbp, %rbp je 0x133c32b movq 0x10(%rbp), %rax movq %rax, 0x60(%rbx) vmovups %ymm0, 0x88(%rbx) vmovups %ymm0, 0x98(%rbx) movq 0x10(%rbp), %rax movq %rax, 0xa0(%rbx) jmp 0x133c33b movq $0x1c0, 0x180(%rsp) # imm = 0x1C0 movq (%r14), %rbx movq 0x8(%rbx), %rax cmpq %rax, %rbp je 0x133c56c movq %rbx, 0x2b0(%rsp) movb $0x1, 0x2b8(%rsp) movq %rbx, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%rbx), %rax testq %rax, %rax je 0x133c2ef movq 0xa8(%rbx), %rax addq 0x68(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%rbx), %rax addq 0x98(%rbx), %rax movq 0x50(%rbx), %rcx addq 0x90(%rbx), %rcx subq %rcx, %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%rbx), %rax addq 0x70(%rbx), %rax movq 0x8(%rbx), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%rbx) vmovups %ymm0, 0x48(%rbx) testq %rbp, %rbp je 0x133c38e movq 0x10(%rbp), %rax movq %rax, 0x60(%rbx) vmovups %ymm0, 0x88(%rbx) vmovups %ymm0, 0x98(%rbx) movq 0x10(%rbp), %rax movq %rax, 0xa0(%rbx) jmp 0x133c39e vmovups %ymm0, 0x98(%rbx) vmovups %ymm0, 0x88(%rbx) movq %rbp, %rax xchgq %rax, 0x8(%rbx) movq %rbx, 0x220(%rsp) leaq 0xdea527(%rip), %rdi # 0x2126878 movq %rdi, 0x90(%rsp) movb $0x1, 0x98(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%rbp), %rsi cmpq 0x140(%rbp), %rsi je 0x133c3f8 movq 0x220(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%rbp) jmp 0x133c40c vmovups %ymm0, 0x98(%rbx) vmovups %ymm0, 0x88(%rbx) movq %rbp, %rax xchgq %rax, 0x8(%rbx) movq %rbx, 0x220(%rsp) leaq 0xdea4c4(%rip), %rdi # 0x2126878 movq %rdi, 0x90(%rsp) movb $0x1, 0x98(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%rbp), %rsi cmpq 0x140(%rbp), %rsi je 0x133c52a movq 0x220(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%rbp) jmp 0x133c53e leaq 0x130(%rbp), %rdi leaq 0x220(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x98(%rsp) jne 0x133c423 movq 0x90(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x2b8(%rsp) jne 0x133c43a movq 0x2b0(%rsp), %rdi callq 0x1ee7c24 movq 0x180(%rsp), %rax addq %rax, 0x28(%r14) movq 0x10(%r14), %rcx movl %ecx, %edx negl %edx andl $0x1f, %edx leaq (%rcx,%rax), %r13 addq %rdx, %r13 movq %r13, 0x10(%r14) cmpq 0x18(%r14), %r13 ja 0x133ce0d addq %rdx, 0x30(%r14) subq %rax, %r13 addq 0x8(%r14), %r13 vbroadcastss 0xbaf5a6(%rip), %ymm0 # 0x1eeba20 vmovaps %ymm0, 0x1c0(%r13) vbroadcastss 0xbb06f8(%rip), %ymm1 # 0x1eecb84 vmovaps %ymm1, 0x1e0(%r13) vmovaps %ymm0, 0xc0(%r13) vmovaps %ymm0, 0x80(%r13) vmovaps %ymm0, 0x40(%r13) vmovaps %ymm1, 0xe0(%r13) vmovaps %ymm1, 0xa0(%r13) vmovaps %ymm1, 0x60(%r13) vxorps %xmm0, %xmm0, %xmm0 vmovaps %ymm0, 0x100(%r13) vmovaps %ymm0, 0x120(%r13) vmovaps %ymm0, 0x140(%r13) vmovaps %ymm0, 0x160(%r13) vmovaps %ymm0, 0x180(%r13) vmovaps %ymm0, 0x1a0(%r13) xorl %eax, %eax vbroadcastsd 0xbe596e(%rip), %ymm0 # 0x1f21e78 movq 0x8(%rsp), %rdx vmovups %ymm0, (%r13,%rax,8) addq $0x4, %rax cmpq $0x8, %rax jne 0x133c50f movl $0x6, %eax jmp 0x133c645 leaq 0x130(%rbp), %rdi leaq 0x220(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x98(%rsp) jne 0x133c555 movq 0x90(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x2b8(%rsp) jne 0x133c56c movq 0x2b0(%rsp), %rdi callq 0x1ee7c24 movq 0x180(%rsp), %rax addq %rax, 0x28(%r14) movq 0x10(%r14), %rcx movl %ecx, %edx negl %edx andl $0x1f, %edx leaq (%rcx,%rax), %r13 addq %rdx, %r13 movq %r13, 0x10(%r14) cmpq 0x18(%r14), %r13 ja 0x133ce44 addq %rdx, 0x30(%r14) subq %rax, %r13 addq 0x8(%r14), %r13 vbroadcastss 0xbaf474(%rip), %ymm0 # 0x1eeba20 vmovaps %ymm0, 0xc0(%r13) vmovaps %ymm0, 0x80(%r13) vmovaps %ymm0, 0x40(%r13) vbroadcastss 0xbb05b7(%rip), %ymm0 # 0x1eecb84 vmovaps %ymm0, 0xe0(%r13) vmovaps %ymm0, 0xa0(%r13) vmovaps %ymm0, 0x60(%r13) vxorps %xmm0, %xmm0, %xmm0 vmovaps %ymm0, 0x100(%r13) vmovaps %ymm0, 0x120(%r13) vmovaps %ymm0, 0x140(%r13) vmovaps %ymm0, 0x160(%r13) vmovaps %ymm0, 0x180(%r13) vmovaps %ymm0, 0x1a0(%r13) xorl %eax, %eax vbroadcastsd 0xbe584e(%rip), %ymm0 # 0x1f21e78 movq 0x8(%rsp), %rdx vmovups %ymm0, (%r13,%rax,8) addq $0x4, %rax cmpq $0x8, %rax jne 0x133c62f movl $0x1, %eax movq %rax, 0x80(%rsp) movq (%rsp), %rcx movq 0x78(%rcx), %rax subq 0x70(%rcx), %rax movq 0x12a0(%rsp), %rbp cmpq 0x38(%rdx), %rax ja 0x133cba3 movq %rbp, %rax decq %rax vbroadcastss 0xbb050a(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x20(%rsp) vbroadcastss 0xbaf397(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x30(%rsp) js 0x133c7f3 leaq (,%rbp,2), %rax addq %rbp, %rax shlq $0x5, %rax leaq 0x14b0(%rsp), %rcx leaq (%rax,%rcx), %r12 addq $-0x10, %r12 imulq $0xb0, %rbp, %rax leaq 0x720(%rsp), %rcx leaq (%rax,%rcx), %rbx addq $-0xb0, %rbx vbroadcastss 0xbb04ad(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xbaf340(%rip), %xmm1 # 0x1eeba20 leaq 0x90(%rsp), %r14 vmovaps %xmm1, 0x30(%rsp) vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x20(%rsp) vmovaps %xmm0, 0x50(%rsp) subq $0x20, %rsp leaq 0x1b10(%rsp), %rcx movq 0x10(%rcx), %rax movq %rax, 0x10(%rsp) vmovups (%rcx), %xmm0 vmovups %xmm0, (%rsp) movq %r14, %rdi movq 0x28(%rsp), %rsi movq %rbx, %rdx xorl %ecx, %ecx vzeroupper callq 0x133a818 addq $0x20, %rsp movq 0x90(%rsp), %rax movq %rax, -0x50(%r12) vmovaps 0xa0(%rsp), %xmm0 vmovaps %xmm0, -0x40(%r12) vmovaps 0xb0(%rsp), %xmm1 vmovaps %xmm1, -0x30(%r12) vmovaps 0xc0(%rsp), %xmm1 vmovaps %xmm1, -0x20(%r12) vmovaps 0xd0(%rsp), %xmm1 vmovaps %xmm1, -0x10(%r12) vmovsd 0xe0(%rsp), %xmm1 vmovsd %xmm1, (%r12) vmovaps 0x30(%rsp), %xmm1 vminps %xmm0, %xmm1, %xmm1 vmovaps %xmm1, 0x30(%rsp) vmovaps 0x20(%rsp), %xmm0 vmaxps -0x30(%r12), %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) vmovaps 0x60(%rsp), %xmm0 vminps -0x20(%r12), %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x50(%rsp), %xmm0 vmaxps -0x10(%r12), %xmm0, %xmm0 vmovaps %xmm0, 0x50(%rsp) decq %rbp addq $-0x60, %r12 addq $-0xb0, %rbx testq %rbp, %rbp jg 0x133c700 movq 0x80(%rsp), %rax orq %r13, %rax movq 0x12a0(%rsp), %rcx movq %rax, %rbp andl $0xf, %eax cmpl $0x1, %eax jne 0x133cd09 testq %rcx, %rcx je 0x133c9cb andq $-0x10, %r13 leaq 0x1504(%rsp), %rax xorl %edx, %edx vmovss 0xbafee0(%rip), %xmm0 # 0x1eec714 vbroadcastss 0xbe4683(%rip), %xmm1 # 0x1f20ec0 vbroadcastss 0xbe529e(%rip), %xmm2 # 0x1f21ae4 vbroadcastss 0xbe5291(%rip), %xmm3 # 0x1f21ae0 vbroadcastss 0xbe466c(%rip), %xmm4 # 0x1f20ec4 vbroadcastss 0xbb4783(%rip), %xmm5 # 0x1ef0fe4 movq -0x54(%rax), %rsi movq %rsi, (%r13,%rdx,8) vmovss (%rax), %xmm6 vmovss -0x4(%rax), %xmm7 vsubss %xmm7, %xmm6, %xmm6 vdivss %xmm6, %xmm0, %xmm6 vxorps %xmm1, %xmm7, %xmm8 vmulss %xmm6, %xmm8, %xmm8 vsubss %xmm8, %xmm0, %xmm9 vshufps $0x0, %xmm8, %xmm8, %xmm8 # xmm8 = xmm8[0,0,0,0] vmovaps -0x44(%rax), %xmm10 vmovaps -0x34(%rax), %xmm11 vmovaps -0x24(%rax), %xmm12 vmovaps -0x14(%rax), %xmm13 vmulps %xmm8, %xmm12, %xmm14 vshufps $0x0, %xmm9, %xmm9, %xmm9 # xmm9 = xmm9[0,0,0,0] vmulps %xmm9, %xmm10, %xmm15 vaddps %xmm15, %xmm14, %xmm14 vmulps %xmm8, %xmm13, %xmm8 vmulps %xmm9, %xmm11, %xmm9 vaddps %xmm9, %xmm8, %xmm8 vsubss %xmm7, %xmm0, %xmm7 vmulss %xmm6, %xmm7, %xmm6 vsubss %xmm6, %xmm0, %xmm7 vshufps $0x0, %xmm6, %xmm6, %xmm6 # xmm6 = xmm6[0,0,0,0] vmulps %xmm6, %xmm12, %xmm9 vshufps $0x0, %xmm7, %xmm7, %xmm7 # xmm7 = xmm7[0,0,0,0] vmulps %xmm7, %xmm10, %xmm10 vaddps %xmm10, %xmm9, %xmm9 vmulps %xmm6, %xmm13, %xmm6 vmulps %xmm7, %xmm11, %xmm7 vaddps %xmm7, %xmm6, %xmm6 vminps %xmm2, %xmm14, %xmm7 vmaxps %xmm3, %xmm8, %xmm8 vminps %xmm2, %xmm9, %xmm9 vmaxps %xmm3, %xmm6, %xmm6 vandps %xmm4, %xmm7, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vsubps %xmm10, %xmm7, %xmm7 vandps %xmm4, %xmm8, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vaddps %xmm10, %xmm8, %xmm8 vandps %xmm4, %xmm9, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vsubps %xmm10, %xmm9, %xmm9 vandps %xmm4, %xmm6, %xmm10 vmulps %xmm5, %xmm10, %xmm10 vmovss %xmm7, 0x40(%r13,%rdx,4) vextractps $0x1, %xmm7, 0x80(%r13,%rdx,4) vaddps %xmm6, %xmm10, %xmm6 vextractps $0x2, %xmm7, 0xc0(%r13,%rdx,4) vmovss %xmm8, 0x60(%r13,%rdx,4) vextractps $0x1, %xmm8, 0xa0(%r13,%rdx,4) vsubps %xmm7, %xmm9, %xmm7 vextractps $0x2, %xmm8, 0xe0(%r13,%rdx,4) vmovss %xmm7, 0x100(%r13,%rdx,4) vextractps $0x1, %xmm7, 0x140(%r13,%rdx,4) vsubps %xmm8, %xmm6, %xmm6 vextractps $0x2, %xmm7, 0x180(%r13,%rdx,4) vmovss %xmm6, 0x120(%r13,%rdx,4) vextractps $0x1, %xmm6, 0x160(%r13,%rdx,4) vextractps $0x2, %xmm6, 0x1a0(%r13,%rdx,4) incq %rdx addq $0x60, %rax cmpq %rdx, %rcx jne 0x133c861 testb $0x1, %r15b jne 0x133cd2a movq 0x148(%rsp), %r12 movq %rbp, (%r12) vmovaps 0x30(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r12) vmovaps 0x20(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r12) vmovaps 0x60(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r12) vmovaps 0x50(%rsp), %xmm0 vmovaps %xmm0, 0x40(%r12) movq (%rsp), %rax movq 0x98(%rax), %rax movq %rax, 0x50(%r12) cmpq $0x0, 0x12a0(%rsp) je 0x133cad4 xorl %r14d, %r14d movq 0x1220(%rsp,%r14,8), %rax decq 0x8(%rax) jne 0x133cac3 movq (%rax), %rbx testq %rbx, %rbx je 0x133cac3 movq 0x18(%rbx), %r15 movq 0x20(%rbx), %rdi testq %rdi, %rdi je 0x133ca84 movq %r15, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x133ca7c movzbl 0x8(%rbx), %edx vzeroupper callq 0x1ee67ca jmp 0x133ca84 vzeroupper callq 0x1ee612d testq %r15, %r15 je 0x133caa4 movq (%rbx), %rdi shlq $0x4, %r15 leaq (%r15,%r15,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx vzeroupper callq *(%rax) leaq 0x10(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %rbx, %rdi vzeroupper callq 0x6a4f0 incq %r14 cmpq 0x12a0(%rsp), %r14 jb 0x133ca37 movq 0x10(%rsp), %rsi testq %rsi, %rsi je 0x133ab16 leaq 0x10(%rsp), %rdi vzeroupper callq 0xbbb88e jmp 0x133ab16 cmpl $0x2, %eax jne 0x133d540 vmovss 0x6e8(%rsp), %xmm0 vmovss %xmm1, 0x620(%rsp) vmovss %xmm0, 0x624(%rsp) vmovss %xmm0, 0x360(%rsp) movq (%rsp), %rax vmovss 0x9c(%rax), %xmm0 vmovss %xmm0, 0x364(%rsp) movq 0xa0(%rax), %r15 movl $0x28, %edi callq 0x6a170 movq %rax, %r14 movq 0x8(%rsp), %rax movq 0x48(%rax), %rdi movq %rdi, (%r14) movb $0x0, 0x8(%r14) movq %r13, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rbx vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x10(%r14) movq $0x0, 0x20(%r14) movq (%rdi), %rax movq %rbx, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %rbx # imm = 0x1C00000 jb 0x133ce7b movq %r14, %rsi addq $0x8, %rsi movq %rbx, %rdi callq 0x1ee665d jmp 0x133ce88 leaq 0x14b0(%rsp), %rax movq %rax, 0x220(%rsp) movq %rdx, 0x228(%rsp) leaq 0x720(%rsp), %rax movq %rax, 0x230(%rsp) leaq 0x90(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x2b0(%rsp), %rdi movq %rbp, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x180(%rsp), %rsi leaq 0x220(%rsp), %rax movq %rax, (%rsi) leaq 0x3a0(%rsp), %rdx leaq 0x90(%rsp), %rcx callq 0x1340ba8 leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x133d66d leaq 0x90(%rsp), %rdi callq 0x6aab0 movq 0x12a0(%rsp), %rax vbroadcastss 0xbaff1c(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x20(%rsp) vbroadcastss 0xbaeda9(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x30(%rsp) testq %rax, %rax je 0x133c7f3 vbroadcastss 0xbafef5(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xbaed88(%rip), %xmm1 # 0x1eeba20 leaq 0x14f0(%rsp), %rcx vmovaps %xmm1, 0x30(%rsp) vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x20(%rsp) vmovaps %xmm0, 0x50(%rsp) vmovaps 0x30(%rsp), %xmm0 vminps -0x30(%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x30(%rsp) vmovaps 0x20(%rsp), %xmm0 vmaxps -0x20(%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) vmovaps 0x60(%rsp), %xmm0 vminps -0x10(%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x50(%rsp), %xmm0 vmaxps (%rcx), %xmm0, %xmm0 vmovaps %xmm0, 0x50(%rsp) addq $0x60, %rcx decq %rax jne 0x133ccb8 jmp 0x133c7f3 xorl %edi, %edi testq %rcx, %rcx sete %dil leaq 0x14b0(%rsp), %rdx movq %r13, %rsi vzeroupper callq 0x1f8c28 jmp 0x133c9cb movq 0x8(%rsp), %rax addq $0x58, %rax movq 0x48(%rsp), %rcx movq %rcx, 0x5a0(%rsp) movq %rax, 0x5a8(%rsp) movq (%rsp), %rax movq 0x70(%rax), %rbx movq 0x78(%rax), %r14 vbroadcastss 0xbaecc3(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x2b0(%rsp) vbroadcastss 0xbafe15(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x2c0(%rsp) vmovaps %xmm0, 0x2d0(%rsp) vmovaps %xmm1, 0x2e0(%rsp) movq %r14, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x133d35f leaq 0x90(%rsp), %rdx movq %rbx, (%rdx) movq %r14, 0x8(%rdx) leaq 0x220(%rsp), %rdi leaq 0x5a0(%rsp), %rsi vzeroupper callq 0x13394ae movq 0x148(%rsp), %r12 movq %rbp, (%r12) vmovaps 0x220(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r12) vmovaps 0x230(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r12) vmovaps 0x240(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r12) vmovaps 0x250(%rsp), %xmm0 jmp 0x133ca0e movq %rcx, 0x10(%r14) shlq $0x2, %rax movq 0x20(%r14), %rcx cmpq %rcx, %rax jbe 0x133d28d leaq 0x180(%rsp), %rsi movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 jmp 0x133c471 movq %rcx, 0x10(%r14) shlq $0x2, %rax movq 0x20(%r14), %rcx cmpq %rcx, %rax jbe 0x133d2f6 leaq 0x180(%rsp), %rsi movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 jmp 0x133c5a3 movl $0x10, %esi movq %rbx, %rdi callq 0x1ee60ac movq 0x8(%rsp), %rcx leaq 0x48(%rcx), %rbx movq %rax, 0x20(%r14) movq %r13, 0x10(%r14) movq %r13, 0x18(%r14) movq %r14, 0x10(%rsp) leaq 0x158(%rsp), %rax movq %r14, (%rax) movq %r15, 0x220(%rsp) leaq 0x620(%rsp), %r14 movq %r14, 0x228(%rsp) movq %rbx, 0x230(%rsp) movq %rax, 0x238(%rsp) movq 0x48(%rsp), %rax movq %rax, 0x240(%rsp) movq (%rsp), %rax movq 0x70(%rax), %r12 movq 0x78(%rax), %r13 vbroadcastss 0xbaeb29(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x14b0(%rsp) vbroadcastss 0xbafc7b(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x14c0(%rsp) vmovaps %xmm0, 0x14d0(%rsp) vmovaps %xmm1, 0x14e0(%rsp) vmovaps %xmm0, 0x14f0(%rsp) vmovaps %xmm1, 0x1500(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x1510(%rsp) xorl %eax, %eax movl %eax, 0x1530(%rsp) vbroadcastss 0xbaf7bf(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x1534(%rsp) movl %eax, 0x153c(%rsp) movq %r13, %rax subq %r12, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x133d400 leaq 0x90(%rsp), %rdx movq %r12, (%rdx) movq %r13, 0x8(%rdx) leaq 0x720(%rsp), %rdi leaq 0x220(%rsp), %rsi vzeroupper callq 0x13349c8 movq 0x788(%rsp), %rax subq 0x780(%rsp), %rax movq (%rsp), %rcx movq 0x78(%rcx), %rdx subq 0x70(%rcx), %rdx cmpq %rdx, %rax je 0x133cfea movq 0x158(%rsp), %rax movq 0x20(%rax), %rdi leaq 0x14b0(%rsp), %r8 movq %r14, (%r8) movl $0x400, %ecx # imm = 0x400 xorl %esi, %esi callq 0x133426d movq %rax, 0x788(%rsp) movq 0x158(%rsp), %rax vmovsd 0x620(%rsp), %xmm0 vmovups 0x720(%rsp), %ymm1 vmovups 0x740(%rsp), %ymm2 vmovups 0x760(%rsp), %ymm3 vmovaps 0x790(%rsp), %xmm4 vmovaps %xmm4, 0x570(%rsp) vmovups %ymm1, 0x500(%rsp) vmovups %ymm2, 0x520(%rsp) vmovups %ymm3, 0x540(%rsp) movq 0x780(%rsp), %rcx movq %rcx, 0x560(%rsp) movq 0x788(%rsp), %rcx movq %rcx, 0x568(%rsp) vmovsd 0x7a0(%rsp), %xmm1 vmovsd 0x7a8(%rsp), %xmm2 vcmpltps %xmm0, %xmm2, %xmm3 vinsertps $0x50, %xmm2, %xmm0, %xmm4 # xmm4 = xmm0[0],xmm2[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm2, %xmm0 # xmm0 = xmm2[0],xmm0[1],xmm2[2,3] vblendvps %xmm3, %xmm4, %xmm0, %xmm0 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x580(%rsp) movq %rax, 0x590(%rsp) movq %r15, 0x3a0(%rsp) leaq 0x360(%rsp), %r14 movq %r14, 0x3a8(%rsp) movq %rbx, 0x3b0(%rsp) movq (%rsp), %rax movq 0x70(%rax), %rbx movq 0x78(%rax), %r12 vbroadcastss 0xbae94a(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x90(%rsp) vbroadcastss 0xbafa9c(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0xa0(%rsp) vmovaps %xmm0, 0xb0(%rsp) vmovaps %xmm1, 0xc0(%rsp) vmovaps %xmm0, 0xd0(%rsp) vmovaps %xmm1, 0xe0(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0xf0(%rsp) xorl %eax, %eax movl %eax, 0x110(%rsp) vbroadcastss 0xbaf5e0(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x114(%rsp) movl %eax, 0x11c(%rsp) movq %r12, %rax subq %rbx, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x133d4a0 leaq 0x2b0(%rsp), %rdx movq %rbx, (%rdx) movq %r12, 0x8(%rdx) leaq 0x14b0(%rsp), %rdi leaq 0x3a0(%rsp), %rsi vzeroupper callq 0x13369f0 movq 0x1518(%rsp), %rax subq 0x1510(%rsp), %rax movq (%rsp), %rdx movq 0x70(%rdx), %rsi leaq (%rax,%rsi), %rcx movq %rsi, 0x1510(%rsp) movq %rcx, 0x1518(%rsp) movq 0x78(%rdx), %rdx movq %rdx, %rcx subq %rsi, %rcx cmpq %rcx, %rax je 0x133d1d9 movq 0x20(%r15), %rdi leaq 0x90(%rsp), %r8 movq %r14, (%r8) movl $0x400, %ecx # imm = 0x400 callq 0x133461a movq %rax, 0x1518(%rsp) vmovsd 0x360(%rsp), %xmm0 vmovups 0x14b0(%rsp), %ymm1 vmovups 0x14d0(%rsp), %ymm2 vmovups 0x14f0(%rsp), %ymm3 vmovaps 0x1520(%rsp), %xmm4 vmovaps %xmm4, 0x4d0(%rsp) vmovups %ymm1, 0x460(%rsp) vmovups %ymm2, 0x480(%rsp) vmovups %ymm3, 0x4a0(%rsp) movq 0x1510(%rsp), %rax movq %rax, 0x4c0(%rsp) movq 0x1518(%rsp), %rax movq %rax, 0x4c8(%rsp) vmovsd 0x1538(%rsp), %xmm1 vcmpltps %xmm0, %xmm1, %xmm2 vinsertps $0x50, %xmm1, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm1[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[1],xmm1[2,3] vblendvps %xmm2, %xmm3, %xmm0, %xmm0 vmovsd 0x1530(%rsp), %xmm1 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] vmovaps %xmm0, 0x4e0(%rsp) movq %r15, 0x4f0(%rsp) jmp 0x133ade3 leaq 0x90(%rsp), %rsi movq %rcx, (%rsi) movl $0x40, %edx movq %rbp, %rdi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx ja 0x133d581 movq %rax, 0x30(%r14) jmp 0x133c471 leaq 0x90(%rsp), %rsi movq %rcx, (%rsi) movl $0x40, %edx movq %rbp, %rdi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx ja 0x133d5f7 movq %rax, 0x30(%r14) jmp 0x133c5a3 leaq 0x90(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x180(%rsp), %rsi movq %r14, (%rsi) movq %rbx, 0x8(%rsi) movq $0x400, 0x10(%rsi) # imm = 0x400 leaq 0x3a0(%rsp), %rcx leaq 0x620(%rsp), %r8 movq %r8, (%rcx) leaq 0x5a0(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0x220(%rsp), %rdi leaq 0x2b0(%rsp), %rdx leaq 0x90(%rsp), %r9 callq 0x13395f3 leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x133d703 leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x133cdc3 leaq 0x90(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x2b0(%rsp), %rsi movq %r13, (%rsi) movq %r12, 0x8(%rsi) movq $0x80, 0x10(%rsi) leaq -0x78d2f9(%rip), %r8 # 0xbb014c leaq 0x180(%rsp), %rcx movq %r8, (%rcx) leaq 0x220(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0x720(%rsp), %rdi leaq 0x14b0(%rsp), %rdx leaq 0x90(%rsp), %r9 callq 0x1334ddf leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x133d735 leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x133cf9e leaq 0x2b0(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x180(%rsp), %rsi movq %r12, (%rsi) movq %rbx, 0x8(%rsi) movq $0x80, 0x10(%rsi) leaq -0x78d399(%rip), %r8 # 0xbb014c leaq 0x5a0(%rsp), %rcx movq %r8, (%rcx) leaq 0x3a0(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0x14b0(%rsp), %rdi leaq 0x90(%rsp), %rdx leaq 0x2b0(%rsp), %r9 callq 0x1336d83 leaq 0x2b0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x133d767 leaq 0x2b0(%rsp), %rdi callq 0x6aab0 jmp 0x133d17d cmpl $0x1, %eax je 0x133d799 cmpl $0x3, %eax jne 0x133adda movq 0x48(%rsp), %rbx movq %rbx, %rdi callq 0xba71d2 leaq 0x500(%rsp), %rdx leaq 0x460(%rsp), %rcx movq 0x8(%rsp), %rdi movq %rbx, %rsi callq 0x1f490e jmp 0x133adda movq $0x0, 0x10(%r14) movq 0x20(%r14), %rax leaq 0x90(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx jbe 0x133d2ed movq $0x0, 0x10(%r14) xorl %r13d, %r13d jmp 0x133c471 movq $0x0, 0x10(%r14) movq 0x20(%r14), %rax leaq 0x90(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %rbp, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r13 movq %rax, 0x8(%r14) movq 0x18(%r14), %rax subq 0x10(%r14), %rax addq 0x30(%r14), %rax movq %rax, 0x30(%r14) movq $0x0, 0x10(%r14) movq 0x90(%rsp), %rcx movq %rcx, 0x18(%r14) movq 0x180(%rsp), %rdx movq %rdx, 0x10(%r14) cmpq %rcx, %rdx jbe 0x133d356 movq $0x0, 0x10(%r14) xorl %r13d, %r13d jmp 0x133c5a3 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbae46f(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xde7578(%rip), %rsi # 0x2124c08 movq 0xde7331(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbae43d(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xde7546(%rip), %rsi # 0x2124c08 movq 0xde72ff(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbae40b(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xde7514(%rip), %rsi # 0x2124c08 movq 0xde72cd(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbae3d9(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xde74e2(%rip), %rsi # 0x2124c08 movq 0xde729b(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbae3a7(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xde74b0(%rip), %rsi # 0x2124c08 movq 0xde7269(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xbae375(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xde747e(%rip), %rsi # 0x2124c08 movq 0xde7237(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq 0x48(%rsp), %rbx movq %rbx, %rdi callq 0xba71d2 leaq 0x500(%rsp), %rdx leaq 0x460(%rsp), %rcx movq 0x8(%rsp), %rdi movq %rbx, %rsi callq 0x134076e jmp 0x133adda movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x133d807 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x133d821 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x133d83b jmp 0x133dacd jmp 0x133d804 jmp 0x133dacd jmp 0x133d81e jmp 0x133dacd jmp 0x133d838 movq %rax, %rbx leaq 0x2b0(%rsp), %rdi callq 0x6aab0 jmp 0x133d8bb jmp 0x133dacd movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x133d8bb jmp 0x133dacd movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x133da0d jmp 0x133dacd jmp 0x133d888 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x133d88b jmp 0x133d8a4 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x133d8a7 jmp 0x133d8b8 jmp 0x133d8b8 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x133d8fb jmp 0x133dacd jmp 0x133d888 movq %rax, %rbx leaq 0x5a0(%rsp), %rdi callq 0x6aab0 jmp 0x133d92b jmp 0x133dacd jmp 0x133d8a4 movq %rax, %rbx leaq 0x3a0(%rsp), %rdi callq 0x6aab0 jmp 0x133d92b jmp 0x133d8b8 movq %rax, %rbx movq 0x10(%rsp), %rsi testq %rsi, %rsi je 0x133d8cf leaq 0x10(%rsp), %rdi callq 0xbbb88e movq $0x0, 0x10(%rsp) jmp 0x133dac5 jmp 0x133da0a jmp 0x133dacd jmp 0x133d8f8 jmp 0x133da0a jmp 0x133dacd jmp 0x133dacd movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x133da0d jmp 0x133dacd movq %rax, %rbx movq %r14, %rdi callq 0x6a4f0 jmp 0x133dac5 jmp 0x133d928 jmp 0x133d928 jmp 0x133d928 movq %rax, %rbx movq 0x18(%rsp), %rsi testq %rsi, %rsi je 0x133d93f leaq 0x18(%rsp), %rdi callq 0xbbb88e movq $0x0, 0x18(%rsp) jmp 0x133da0d movq %rax, %rbx movq %r12, %rdi callq 0x6a4f0 jmp 0x133da0d movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x8d6eda jmp 0x133d9a2 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x8d6eda jmp 0x133d9a2 jmp 0x133dacd jmp 0x133dacd jmp 0x133dacd jmp 0x133dacd jmp 0x133dacd movq %rax, %rbx jmp 0x133d9a2 movq %rax, %rbx leaq 0x2b0(%rsp), %rdi callq 0x8d6eda jmp 0x133da0d jmp 0x133dacd jmp 0x133da0a movq %rax, %rbx movq 0x14b0(%rsp), %rsi testq %rsi, %rsi je 0x133d9d5 leaq 0x14b0(%rsp), %rdi callq 0x90da94 leaq 0x720(%rsp), %rdi callq 0x8d6eda jmp 0x133dac5 movq %rax, %rbx movq %r12, %rdi callq 0x1ee612d jmp 0x133dac5 jmp 0x133dacd jmp 0x133dacd jmp 0x133dacd jmp 0x133da0a jmp 0x133da0a movq %rax, %rbx cmpq $0x0, 0x12a0(%rsp) je 0x133dab1 xorl %r15d, %r15d movq 0x1220(%rsp,%r15,8), %rax decq 0x8(%rax) jne 0x133daa0 movq (%rax), %r14 testq %r14, %r14 je 0x133daa0 movq 0x18(%r14), %r12 movq 0x20(%r14), %rdi testq %rdi, %rdi je 0x133da67 movq %r12, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x133da62 movzbl 0x8(%r14), %edx callq 0x1ee67ca jmp 0x133da67 callq 0x1ee612d testq %r12, %r12 je 0x133da84 movq (%r14), %rdi shlq $0x4, %r12 leaq (%r12,%r12,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) leaq 0x10(%r14), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq %r14, %rdi callq 0x6a4f0 incq %r15 cmpq 0x12a0(%rsp), %r15 jb 0x133da1f movq 0x10(%rsp), %rsi testq %rsi, %rsi je 0x133dac5 leaq 0x10(%rsp), %rdi callq 0xbbb88e movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::RecalculatePrimRef<embree::InstanceArray>, 2ul>::find(embree::SetMB const&, unsigned long)
const Split find(const SetMB& set, const size_t logBlockSize) { assert(set.size() > 0); TemporalBinInfo binner(empty); binner.bin_parallel(set.prims->data(),set.begin(),set.end(),PARALLEL_FIND_BLOCK_SIZE,PARALLEL_THRESHOLD,set.time_range,set,recalculatePrimRef); Split tsplit = binner.best((int)logBlockSize,set.time_range,set); if (!tsplit.valid()) tsplit.data = Split::SPLIT_FALLBACK; // use fallback split return tsplit; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x298, %rsp # imm = 0x298 movq %rcx, %r15 movq %rdx, %r14 movq %rsi, %r9 movq %rdi, %rbx vxorps %xmm1, %xmm1, %xmm1 vmovaps %xmm1, 0x10(%rsp) vbroadcastss 0xba41d8(%rip), %xmm2 # 0x1eeba20 vmovaps %xmm2, 0x60(%rsp) vbroadcastss 0xba532d(%rip), %xmm3 # 0x1eecb84 vmovaps %xmm3, 0x70(%rsp) vmovaps %xmm2, 0x80(%rsp) vmovaps %xmm3, 0x90(%rsp) vmovaps %xmm2, 0x20(%rsp) vmovaps %xmm3, 0x30(%rsp) vmovaps %xmm2, 0x40(%rsp) vmovaps %xmm3, 0x50(%rsp) movq 0x90(%rdx), %rax movq 0x20(%rax), %rsi movq 0x60(%rdx), %r12 movq 0x68(%rdx), %r13 vmovsd 0x88(%rdx), %xmm0 addq $0x8, %r9 vmovsd %xmm0, 0x8(%rsp) movq %rsi, (%rsp) movq %r13, %rax subq %r12, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x1347912 leaq 0x10(%rsp), %rdi movq %r12, %rdx movq %r13, %rcx movq %r14, %r8 callq 0x1347cea vmovsd 0x88(%r14), %xmm0 leaq 0x10(%rsp), %rsi movq %rbx, %rdi movl %r15d, %edx movq %r14, %rcx vzeroupper callq 0x1347aba cmpl $-0x1, 0x4(%rbx) jne 0x13478fd movl $0x1, 0xc(%rbx) movq %rbx, %rax addq $0x298, %rsp # imm = 0x298 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rsp, %rax leaq 0xa0(%rsp), %rcx movq %rax, (%rcx) leaq 0x8(%rsp), %rax movq %rax, 0x8(%rcx) movq %r14, 0x10(%rcx) movq %r9, 0x18(%rcx) leaq 0x200(%rsp), %rbp vmovaps %xmm1, (%rbp) vmovaps %xmm2, 0x50(%rbp) vmovaps %xmm3, 0x60(%rbp) vmovaps %xmm2, 0x70(%rbp) vmovaps %xmm3, 0x80(%rbp) vmovaps %xmm2, 0x10(%rbp) vmovaps %xmm3, 0x20(%rbp) vmovaps %xmm2, 0x30(%rbp) vmovaps %xmm3, 0x40(%rbp) leaq 0x180(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vmovups %xmm1, 0x20(%rdi) movq $0x8, 0x40(%rdi) callq 0x6a660 leaq 0x168(%rsp), %rsi movq %r13, (%rsi) movq %r12, 0x8(%rsi) movq $0x400, 0x10(%rsi) # imm = 0x400 leaq 0x713(%rip), %r8 # 0x13480ba leaq 0xc0(%rsp), %rcx movq %r8, (%rcx) leaq 0xa0(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0xd0(%rsp), %rdi movq %rbp, %rdx leaq 0x180(%rsp), %r9 callq 0x134819a leaq 0x180(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1347a56 leaq 0x180(%rsp), %rdi callq 0x6aab0 vmovaps 0xd0(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rsp) vmovaps 0xe0(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rsp) vmovups 0xf0(%rsp), %ymm0 vmovups %ymm0, 0x30(%rsp) vmovups 0x110(%rsp), %ymm0 vmovups %ymm0, 0x50(%rsp) vmovups 0x130(%rsp), %ymm0 vmovups %ymm0, 0x70(%rsp) vmovaps 0x150(%rsp), %xmm0 vmovaps %xmm0, 0x90(%rsp) jmp 0x13478d1 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xba4086(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xddd18f(%rip), %rsi # 0x2124c08 movq 0xddcf48(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1347a9c jmp 0x1347ab1 jmp 0x1347a99 movq %rax, %rbx leaq 0x180(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/heuristic_timesplit_array.h
embree::BVHNodeRecordMB4D<embree::NodeRefPtr<8>> const embree::avx::BVHBuilderMSMBlur::build<embree::NodeRefPtr<8>, embree::avx::RecalculatePrimRef<embree::InstanceArray>, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeaf<8, embree::InstanceArray, embree::InstanceArrayPrimitive>, embree::Scene::BuildProgressMonitorInterface>(embree::vector_t<embree::PrimRefMB, embree::aligned_monitored_allocator<embree::PrimRefMB, 16ul>>&, embree::PrimInfoMBT<embree::LBBox<embree::Vec3fa>> const&, embree::MemoryMonitorInterface*, embree::avx::RecalculatePrimRef<embree::InstanceArray>, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeaf<8, embree::InstanceArray, embree::InstanceArrayPrimitive>, embree::Scene::BuildProgressMonitorInterface, embree::avx::BVHBuilderMSMBlur::Settings const&)
static const BVHNodeRecordMB4D<NodeRef> build(mvector<PrimRefMB>& prims, const PrimInfoMB& pinfo, MemoryMonitorInterface* device, const RecalculatePrimRef recalculatePrimRef, const CreateAllocFunc createAlloc, const CreateNodeFunc createNode, const SetNodeFunc setNode, const CreateLeafFunc createLeaf, const ProgressMonitorFunc progressMonitor, const Settings& settings) { typedef BuilderT< NodeRef, RecalculatePrimRef, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, SetNodeFunc, CreateLeafFunc, ProgressMonitorFunc> Builder; Builder builder(device, recalculatePrimRef, createAlloc, createNode, setNode, createLeaf, progressMonitor, settings); return builder(prims,pinfo); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x188, %rsp # imm = 0x188 movq %r9, %r10 movq %r8, %rax movq %rdx, %r15 movq %rsi, %r14 movq %rdi, %rbx movq 0x1b0(%rsp), %r8 movq 0x1c0(%rsp), %rdx movq 0x1b8(%rsp), %rsi leaq 0xdb31ec(%rip), %rdi # 0x2102970 leaq 0x38(%rsp), %r9 movq %rdi, (%r9) movq 0x8(%rsi), %rsi movq %rsi, 0x8(%r9) movq %rdx, (%rsp) leaq 0x48(%rsp), %r12 movq %r12, %rdi movq %rcx, %rsi movq %rax, %rdx movq %r10, %rcx callq 0x134f84e vmovups (%r15), %ymm0 vmovups 0x20(%r15), %ymm1 vmovups 0x40(%r15), %ymm2 vmovups 0x70(%r15), %ymm3 leaq 0xd0(%rsp), %rdx vmovups %ymm3, 0x80(%rdx) movq $0x1, (%rdx) vmovups %ymm0, 0x10(%rdx) vmovups %ymm1, 0x30(%rdx) vmovups %ymm2, 0x50(%rdx) vmovaps 0x60(%r15), %xmm0 vmovaps %xmm0, 0x70(%rdx) movq %r14, 0xa0(%rdx) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x20(%rsp) movq $0x0, 0x30(%rsp) movq 0x30(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x20(%rsp), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r12, %rsi movl $0x1, %ecx vzeroupper callq 0x134f968 mfence movq %rbx, %rax addq $0x188, %rsp # imm = 0x188 popq %rbx popq %r12 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<8>, embree::avx::RecalculatePrimRef<embree::InstanceArray>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeaf<8, embree::InstanceArray, embree::InstanceArrayPrimitive>, embree::Scene::BuildProgressMonitorInterface>::find(embree::SetMB const&)
const Split find(const SetMB& set) { /* first try standard object split */ const Split object_split = heuristicObjectSplit.find(set,cfg.logBlockSize); const float object_split_sah = object_split.splitSAH(); /* test temporal splits only when object split was bad */ const float leaf_sah = set.leafSAH(cfg.logBlockSize); if (object_split_sah < 0.50f*leaf_sah) return object_split; /* do temporal splits only if the time range is big enough */ if (set.time_range.size() > 1.01f/float(set.max_num_time_segments)) { const Split temporal_split = heuristicTemporalSplit.find(set,cfg.logBlockSize); const float temporal_split_sah = temporal_split.splitSAH(); /* take temporal split if it improved SAH */ if (temporal_split_sah < object_split_sah) return temporal_split; } return object_split; }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x58, %rsp movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %rbx addq $0x40, %rsi movq 0x10(%r15), %rcx leaq 0x10(%rsp), %r12 movq %r12, %rdi callq 0x12551d0 vmovss (%r12), %xmm6 vmovss 0x8c(%r14), %xmm0 vsubss 0x88(%r14), %xmm0, %xmm0 vmovaps 0x10(%r14), %xmm1 vmovaps 0x30(%r14), %xmm2 vsubps (%r14), %xmm1, %xmm1 vsubps 0x20(%r14), %xmm2, %xmm2 movq 0x10(%r15), %rcx vinsertps $0x4c, %xmm1, %xmm2, %xmm3 # xmm3 = xmm1[1],xmm2[1],zero,zero vshufpd $0x1, %xmm1, %xmm1, %xmm4 # xmm4 = xmm1[1,0] vinsertps $0x9c, %xmm2, %xmm4, %xmm4 # xmm4 = xmm4[0],xmm2[2],zero,zero vaddps %xmm4, %xmm3, %xmm5 vmulps %xmm4, %xmm3, %xmm3 vinsertps $0x1c, %xmm2, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm2[0],zero,zero vmulps %xmm5, %xmm1, %xmm1 vaddps %xmm3, %xmm1, %xmm1 vhaddps %xmm1, %xmm1, %xmm2 vmovss 0xb99ecb(%rip), %xmm1 # 0x1eecb80 vmulss %xmm1, %xmm2, %xmm2 vmulss %xmm2, %xmm0, %xmm2 movq $-0x1, %rax shlq %cl, %rax notq %rax addq 0x70(%r14), %rax shrq %cl, %rax testq %rax, %rax js 0x1352cdd vcvtsi2ss %rax, %xmm7, %xmm3 jmp 0x1352cf2 movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm7, %xmm3 vaddss %xmm3, %xmm3, %xmm3 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm1, %xmm2, %xmm1 vucomiss %xmm6, %xmm1 ja 0x1352d5b movq 0x78(%r14), %rax testq %rax, %rax js 0x1352d10 vcvtsi2ss %rax, %xmm7, %xmm1 jmp 0x1352d25 movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm7, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmovss 0xbac32f(%rip), %xmm2 # 0x1eff05c vdivss %xmm1, %xmm2, %xmm1 vucomiss %xmm1, %xmm0 jbe 0x1352d5b addq $0x48, %r15 movq %rbx, %rdi movq %r15, %rsi movq %r14, %rdx vmovss %xmm6, 0xc(%rsp) callq 0x1347818 vmovss 0xc(%rsp), %xmm0 vucomiss (%rbx), %xmm0 ja 0x1352d84 vmovaps 0x10(%rsp), %xmm0 vmovaps %xmm0, (%rbx) movq 0x20(%rsp), %rax movq %rax, 0x10(%rbx) vmovaps 0x30(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rbx) vmovaps 0x40(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rbx) movq %rbx, %rax addq $0x58, %rsp popq %rbx popq %r12 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::BVHNBuilderMBlurSAHGrid<4>::buildMultiSegment(unsigned long)
void buildMultiSegment(size_t numPrimitives) { /* create primref array */ mvector<PrimRefMB> prims(scene->device,numPrimitives); PrimInfoMB pinfo = createPrimRefArrayMSMBlurGrid(scene,prims,bvh->scene->progressInterface); /* early out if no valid primitives */ if (pinfo.size() == 0) { bvh->clear(); return; } GridRecalculatePrimRef recalculatePrimRef(scene,sgrids.data()); /* estimate acceleration structure size */ const size_t node_bytes = pinfo.num_time_segments*sizeof(AABBNodeMB)/(4*N); //FIXME: check leaf_bytes //const size_t leaf_bytes = size_t(1.2*Primitive::blocks(pinfo.num_time_segments)*sizeof(SubGridQBVHN<N>)); const size_t leaf_bytes = size_t(1.2*(float)numPrimitives/N * sizeof(SubGridQBVHN<N>)); bvh->alloc.init_estimate(node_bytes+leaf_bytes); /* settings for BVH build */ BVHBuilderMSMBlur::Settings settings; settings.branchingFactor = N; settings.maxDepth = BVH::maxDepth; settings.logBlockSize = bsr(sahBlockSize); settings.minLeafSize = min(minLeafSize,maxLeafSize); settings.maxLeafSize = maxLeafSize; settings.travCost = travCost; settings.intCost = intCost; settings.singleLeafTimeSegment = false; settings.singleThreadThreshold = bvh->alloc.fixSingleThreadThreshold(N,DEFAULT_SINGLE_THREAD_THRESHOLD,pinfo.size(),node_bytes+leaf_bytes); /* build hierarchy */ auto root = BVHBuilderMSMBlur::build<NodeRef>(prims,pinfo,scene->device, recalculatePrimRef, typename BVH::CreateAlloc(bvh), typename BVH::AABBNodeMB4D::Create(), typename BVH::AABBNodeMB4D::Set(), CreateMSMBlurLeafGrid<N>(scene,bvh,sgrids.data()), bvh->scene->progressInterface, settings); bvh->set(root.ref,root.lbounds,pinfo.num_time_segments); }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1b0, %rsp # imm = 0x1B0 movq %rsi, %r14 movq %rdi, %rbx movq 0x18(%rdi), %rax movq 0x188(%rax), %rax leaq 0x550(%rax), %rdi testq %rax, %rax cmoveq %rax, %rdi movq %rdi, 0x30(%rsp) movb $0x0, 0x38(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x40(%rsp) movq $0x0, 0x50(%rsp) testq %rsi, %rsi je 0x135684c movq %r14, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %r15 movq (%rdi), %rax movq %r15, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %r15 # imm = 0x1C00000 jb 0x1356830 leaq 0x38(%rsp), %rsi movq %r15, %rdi callq 0x1ee665d jmp 0x135683d movl $0x10, %esi movq %r15, %rdi callq 0x1ee60ac movq %rax, 0x50(%rsp) movq %r14, 0x40(%rsp) movq %r14, 0x48(%rsp) leaq 0x40(%rsp), %r13 movq 0x10(%rbx), %rax movq 0x18(%rbx), %rdx movl $0x260, %r8d # imm = 0x260 addq 0x68(%rax), %r8 vmovsd 0xb95e85(%rip), %xmm0 # 0x1eec6f0 leaq 0xc0(%rsp), %rdi leaq 0x30(%rsp), %rcx movq %rbx, %rsi callq 0x1356bb4 movq 0x128(%rsp), %rax cmpq 0x120(%rsp), %rax jne 0x13568a1 movq 0x10(%rbx), %rdi movq (%rdi), %rax callq *0x28(%rax) jmp 0x1356aba movq 0x10(%rbx), %rdi imulq $0xe0, 0x130(%rsp), %r12 shrq $0x4, %r12 testq %r14, %r14 js 0x13568c1 vcvtsi2ss %r14, %xmm1, %xmm0 jmp 0x13568d7 movq %r14, %rax shrq %rax andl $0x1, %r14d orq %rax, %r14 vcvtsi2ss %r14, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 movq 0x18(%rbx), %r14 movq 0x60(%rbx), %r15 vcvtss2sd %xmm0, %xmm0, %xmm0 vmulsd 0xbad86d(%rip), %xmm0, %xmm0 # 0x1f04158 vmulsd 0xbad89d(%rip), %xmm0, %xmm0 # 0x1f04190 vmulsd 0xbad89d(%rip), %xmm0, %xmm0 # 0x1f04198 vcvttsd2si %xmm0, %rax movq %rax, %rcx sarq $0x3f, %rcx vsubsd 0xb996d1(%rip), %xmm0, %xmm0 # 0x1eeffe0 vcvttsd2si %xmm0, %rdx andq %rcx, %rdx orq %rax, %rdx addq %rdx, %r12 addq $0x78, %rdi movq %r12, %rsi callq 0x90acb8 vmovaps 0xbcb4ef(%rip), %ymm0 # 0x1f21e20 vmovups %ymm0, 0x58(%rsp) movq $0x8, 0x78(%rsp) movl $0x3f800000, 0x80(%rsp) # imm = 0x3F800000 xorl %eax, %eax movb %al, 0x88(%rsp) movq $0x400, 0x90(%rsp) # imm = 0x400 movq $0x4, 0x58(%rsp) movq $0x50, 0x60(%rsp) movq 0x20(%rbx), %rcx bsrq %rcx, %rcx movq %rcx, 0x68(%rsp) movq 0x30(%rbx), %rcx movq 0x38(%rbx), %rdx cmpq %rdx, %rcx cmovaeq %rdx, %rcx movq %rcx, 0x70(%rsp) movq %rdx, 0x78(%rsp) vbroadcastss 0xb95d73(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x80(%rsp) vmovss 0x28(%rbx), %xmm0 vmovss %xmm0, 0x84(%rsp) movb %al, 0x88(%rsp) movq 0x10(%rbx), %rdi movq 0x128(%rsp), %rcx subq 0x120(%rsp), %rcx addq $0x78, %rdi movl $0x4, %esi movl $0x400, %edx # imm = 0x400 movq %r12, %r8 vzeroupper callq 0xbdde9e leaq 0x58(%rsp), %rdx movq %rax, 0x38(%rdx) movq 0x10(%rbx), %rax movq 0x18(%rbx), %rsi movq 0x188(%rsi), %rdi leaq 0x78(%rax), %r8 leaq 0x550(%rdi), %rcx testq %rdi, %rdi cmoveq %rdi, %rcx movq 0x60(%rbx), %rdi movq %rsi, 0x98(%rsp) movq %rax, 0xa0(%rsp) movq %rdi, 0xa8(%rsp) movq 0x68(%rax), %rax leaq 0xdabf33(%rip), %rsi # 0x2102970 leaq 0xb0(%rsp), %rdi movq %rsi, (%rdi) movq 0x268(%rax), %rax movq %rax, 0x8(%rdi) movq 0xa8(%rsp), %rax movq %rax, 0x18(%rsp) vmovupd 0x98(%rsp), %xmm0 vmovupd %xmm0, 0x8(%rsp) movq %rdx, 0x28(%rsp) movq %rdi, 0x20(%rsp) movq %r8, (%rsp) leaq 0x150(%rsp), %rdi leaq 0x30(%rsp), %rsi leaq 0xc0(%rsp), %rdx movq %r14, %r8 movq %r15, %r9 callq 0x1357680 movq 0x10(%rbx), %rdi leaq 0x160(%rsp), %rdx movq -0x10(%rdx), %rsi movq 0x130(%rsp), %rcx callq 0xb5ccb2 movq 0x48(%rsp), %rbx movq 0x50(%rsp), %rdi testq %rdi, %rdi je 0x1356aee movq %rbx, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1356ae9 movzbl 0x38(%rsp), %edx callq 0x1ee67ca jmp 0x1356aee callq 0x1ee612d testq %rbx, %rbx je 0x1356b0d movq 0x30(%rsp), %rdi shlq $0x4, %rbx leaq (%rbx,%rbx,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%r13) movq $0x0, 0x10(%r13) addq $0x1b0, %rsp # imm = 0x1B0 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq jmp 0x1356b38 jmp 0x1356b38 jmp 0x1356b38 jmp 0x1356b38 movq %rax, %rbx movq 0x48(%rsp), %r14 movq 0x50(%rsp), %rdi testq %rdi, %rdi jne 0x1356b83 testq %r14, %r14 je 0x1356b69 movq 0x30(%rsp), %rdi shlq $0x4, %r14 leaq (%r14,%r14,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, (%r13) movq $0x0, 0x10(%r13) movq %rbx, %rdi callq 0x6a600 movq %r14, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1356ba3 movzbl 0x38(%rsp), %edx callq 0x1ee67ca jmp 0x1356b4a callq 0x1ee612d jmp 0x1356b4a jmp 0x1356bac movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_sah_mb.cpp
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<4>, embree::avx::GridRecalculatePrimRef, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeafGrid<4>, embree::Scene::BuildProgressMonitorInterface>::BuilderT(embree::MemoryMonitorInterface*, embree::avx::GridRecalculatePrimRef, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeafGrid<4>, embree::Scene::BuildProgressMonitorInterface, embree::avx::BVHBuilderMSMBlur::Settings const&)
BuilderT (MemoryMonitorInterface* device, const RecalculatePrimRef recalculatePrimRef, const CreateAllocFunc createAlloc, const CreateNodeFunc createNode, const SetNodeFunc setNode, const CreateLeafFunc createLeaf, const ProgressMonitor progressMonitor, const Settings& settings) : cfg(settings), heuristicObjectSplit(), heuristicTemporalSplit(device, recalculatePrimRef), recalculatePrimRef(recalculatePrimRef), createAlloc(createAlloc), createNode(createNode), setNode(setNode), createLeaf(createLeaf), progressMonitor(progressMonitor) { if (cfg.branchingFactor > MAX_BRANCHING_FACTOR) throw_RTCError(RTC_ERROR_UNKNOWN,"bvh_builder: branching factor too large"); }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x28, %rsp movq 0x68(%rsp), %rax vmovups (%rax), %ymm0 vmovups 0x20(%rax), %ymm1 vmovups %ymm1, 0x20(%rdi) vmovups %ymm0, (%rdi) movq %rsi, 0x48(%rdi) movq %rdx, 0x50(%rdi) movq %rcx, 0x58(%rdi) movq %rdx, 0x60(%rdi) movq %rcx, 0x68(%rdi) movq %r8, 0x70(%rdi) vmovaps 0x50(%rsp), %xmm0 vmovups %xmm0, 0x80(%rdi) movq 0x60(%rsp), %rax movq %rax, 0x90(%rdi) leaq 0xda8cfc(%rip), %rax # 0x2102970 movq %rax, 0x98(%rdi) movq 0x8(%r9), %rax movq %rax, 0xa0(%rdi) cmpq $0x11, (%rdi) jae 0x1359c9a addq $0x28, %rsp popq %rbx popq %r14 popq %r15 popq %rbp vzeroupper retq movl $0x30, %edi vzeroupper callq 0x6a3b0 movq %rax, %rbx leaq 0x18(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb92f92(%rip), %rsi # 0x1eecc4c leaq 0xb92fb2(%rip), %rdx # 0x1eecc73 leaq 0x8(%rsp), %rdi callq 0x8d7230 leaq 0xda7bde(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x1, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x8(%rsp), %rsi movq 0x10(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xda7b4c(%rip), %rsi # 0x2101850 leaq -0xa82eef(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rax, %r14 xorl %ebp, %ebp jmp 0x1359d28 movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x8(%rsp), %rdi cmpq %r15, %rdi je 0x1359d3f callq 0x6a4f0 jmp 0x1359d3f movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x1359d4c movq %rbx, %rdi callq 0x6a8a0 movq %r14, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::HeuristicMBlurTemporalSplit<embree::PrimRefMB, embree::avx::GridRecalculatePrimRef, 2ul>::find(embree::SetMB const&, unsigned long)
const Split find(const SetMB& set, const size_t logBlockSize) { assert(set.size() > 0); TemporalBinInfo binner(empty); binner.bin_parallel(set.prims->data(),set.begin(),set.end(),PARALLEL_FIND_BLOCK_SIZE,PARALLEL_THRESHOLD,set.time_range,set,recalculatePrimRef); Split tsplit = binner.best((int)logBlockSize,set.time_range,set); if (!tsplit.valid()) tsplit.data = Split::SPLIT_FALLBACK; // use fallback split return tsplit; }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x298, %rsp # imm = 0x298 movq %rcx, %r15 movq %rdx, %r14 movq %rsi, %r9 movq %rdi, %rbx vxorps %xmm1, %xmm1, %xmm1 vmovaps %xmm1, 0x10(%rsp) vbroadcastss 0xb8ad5a(%rip), %xmm2 # 0x1eeba20 vmovaps %xmm2, 0x60(%rsp) vbroadcastss 0xb8beaf(%rip), %xmm3 # 0x1eecb84 vmovaps %xmm3, 0x70(%rsp) vmovaps %xmm2, 0x80(%rsp) vmovaps %xmm3, 0x90(%rsp) vmovaps %xmm2, 0x20(%rsp) vmovaps %xmm3, 0x30(%rsp) vmovaps %xmm2, 0x40(%rsp) vmovaps %xmm3, 0x50(%rsp) movq 0x90(%rdx), %rax movq 0x20(%rax), %rsi movq 0x60(%rdx), %r12 movq 0x68(%rdx), %r13 vmovsd 0x88(%rdx), %xmm0 addq $0x8, %r9 vmovsd %xmm0, 0x8(%rsp) movq %rsi, (%rsp) movq %r13, %rax subq %r12, %rax cmpq $0xbff, %rax # imm = 0xBFF ja 0x1360d90 leaq 0x10(%rsp), %rdi movq %r12, %rdx movq %r13, %rcx movq %r14, %r8 callq 0x1361168 vmovsd 0x88(%r14), %xmm0 leaq 0x10(%rsp), %rsi movq %rbx, %rdi movl %r15d, %edx movq %r14, %rcx vzeroupper callq 0x1360f38 cmpl $-0x1, 0x4(%rbx) jne 0x1360d7b movl $0x1, 0xc(%rbx) movq %rbx, %rax addq $0x298, %rsp # imm = 0x298 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %rsp, %rax leaq 0xa0(%rsp), %rcx movq %rax, (%rcx) leaq 0x8(%rsp), %rax movq %rax, 0x8(%rcx) movq %r14, 0x10(%rcx) movq %r9, 0x18(%rcx) leaq 0x200(%rsp), %rbp vmovaps %xmm1, (%rbp) vmovaps %xmm2, 0x50(%rbp) vmovaps %xmm3, 0x60(%rbp) vmovaps %xmm2, 0x70(%rbp) vmovaps %xmm3, 0x80(%rbp) vmovaps %xmm2, 0x10(%rbp) vmovaps %xmm3, 0x20(%rbp) vmovaps %xmm2, 0x30(%rbp) vmovaps %xmm3, 0x40(%rbp) leaq 0x180(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vmovups %xmm1, 0x20(%rdi) movq $0x8, 0x40(%rdi) callq 0x6a660 leaq 0x168(%rsp), %rsi movq %r13, (%rsi) movq %r12, 0x8(%rsi) movq $0x400, 0x10(%rsi) # imm = 0x400 leaq 0x14db(%rip), %r8 # 0x1362300 leaq 0xc0(%rsp), %rcx movq %r8, (%rcx) leaq 0xa0(%rsp), %rax movq %rax, 0x8(%rcx) leaq 0xd0(%rsp), %rdi movq %rbp, %rdx leaq 0x180(%rsp), %r9 callq 0x13623e0 leaq 0x180(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1360ed4 leaq 0x180(%rsp), %rdi callq 0x6aab0 vmovaps 0xd0(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rsp) vmovaps 0xe0(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rsp) vmovups 0xf0(%rsp), %ymm0 vmovups %ymm0, 0x30(%rsp) vmovups 0x110(%rsp), %ymm0 vmovups %ymm0, 0x50(%rsp) vmovups 0x130(%rsp), %ymm0 vmovups %ymm0, 0x70(%rsp) vmovaps 0x150(%rsp), %xmm0 vmovaps %xmm0, 0x90(%rsp) jmp 0x1360d4f movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb8ac08(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xdc3d11(%rip), %rsi # 0x2124c08 movq 0xdc3aca(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1360f1a jmp 0x1360f2f jmp 0x1360f17 movq %rax, %rbx leaq 0x180(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/heuristic_timesplit_array.h
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<4>, embree::avx::GridRecalculatePrimRef, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<4>, 4>::Set, embree::avx::CreateMSMBlurLeafGrid<4>, embree::Scene::BuildProgressMonitorInterface>::splitFallback(embree::SetMB const&, embree::SetMB&, embree::SetMB&)
void splitFallback(const SetMB& set, SetMB& lset, SetMB& rset) { mvector<PrimRefMB>& prims = *set.prims; const size_t begin = set.begin(); const size_t end = set.end(); const size_t center = (begin + end + 1) / 2; PrimInfoMB linfo = empty; for (size_t i=begin; i<center; i++) linfo.add_primref(prims[i]); PrimInfoMB rinfo = empty; for (size_t i=center; i<end; i++) rinfo.add_primref(prims[i]); new (&lset) SetMB(linfo,set.prims,range<size_t>(begin,center),set.time_range); new (&rset) SetMB(rinfo,set.prims,range<size_t>(center,end ),set.time_range); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x28, %rsp movq 0x90(%rsi), %r8 movq 0x60(%rsi), %r9 movq 0x68(%rsi), %rax leaq (%r9,%rax), %rdi incq %rdi shrq %rdi vmovss 0xb89657(%rip), %xmm0 # 0x1eec714 vmovaps %xmm0, -0x60(%rsp) vmovsd 0xb89625(%rip), %xmm0 # 0x1eec6f0 vmovaps %xmm0, -0x70(%rsp) vbroadcastss 0xb88946(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, -0x80(%rsp) vbroadcastss 0xb89a9b(%rip), %xmm4 # 0x1eecb84 movq %rdi, %rbx subq %r9, %rbx jbe 0x13631ff movq 0x20(%r8), %r10 leaq (%r9,%r9,4), %r11 shlq $0x4, %r11 leaq (%r11,%r10), %r14 addq $0x40, %r14 vmovss 0xb89603(%rip), %xmm9 # 0x1eec714 vmovsd 0xb895d7(%rip), %xmm6 # 0x1eec6f0 vbroadcastss 0xb89a62(%rip), %xmm8 # 0x1eecb84 vbroadcastss 0xb888f5(%rip), %xmm10 # 0x1eeba20 xorl %r10d, %r10d vmovaps %xmm10, %xmm7 vmovaps %xmm8, %xmm5 vmovaps %xmm10, %xmm13 vmovaps %xmm8, %xmm11 xorl %r11d, %r11d movq %r11, %r15 vmovaps -0x40(%r14), %xmm0 vmovaps %xmm5, %xmm3 vmovaps %xmm13, %xmm5 vmovaps -0x30(%r14), %xmm13 vmovaps -0x20(%r14), %xmm14 vmovaps -0x10(%r14), %xmm15 vbroadcastss 0xb89a11(%rip), %xmm2 # 0x1eecb80 vmulps %xmm2, %xmm14, %xmm12 vmulps %xmm2, %xmm0, %xmm1 vaddps %xmm1, %xmm12, %xmm1 vmulps %xmm2, %xmm15, %xmm12 vmulps %xmm2, %xmm13, %xmm2 vaddps %xmm2, %xmm12, %xmm2 vaddps %xmm2, %xmm1, %xmm1 vminps %xmm0, %xmm10, %xmm10 vmaxps %xmm13, %xmm8, %xmm8 vmovaps %xmm5, %xmm13 vminps %xmm14, %xmm7, %xmm7 vmaxps %xmm15, %xmm3, %xmm5 vminps %xmm1, %xmm13, %xmm13 vmovsd (%r14), %xmm0 vcmpltps %xmm0, %xmm9, %xmm2 vinsertps $0x50, %xmm0, %xmm9, %xmm12 # xmm12 = xmm9[0],xmm0[1],xmm9[2,3] vblendps $0x2, %xmm9, %xmm0, %xmm9 # xmm9 = xmm0[0],xmm9[1],xmm0[2,3] vblendvps %xmm2, %xmm12, %xmm9, %xmm9 vmaxps %xmm1, %xmm11, %xmm11 movl -0x14(%r14), %r11d addq %r11, %r10 movl -0x4(%r14), %r11d cmpq %r11, %r15 cmovaq %r15, %r11 setb %bpl vmovd %ebp, %xmm1 vpshufd $0x50, %xmm1, %xmm1 # xmm1 = xmm1[0,0,1,1] vpslld $0x1f, %xmm1, %xmm1 vblendvps %xmm1, %xmm0, %xmm6, %xmm6 addq $0x50, %r14 decq %rbx jne 0x1363143 jmp 0x1363233 xorl %r11d, %r11d xorl %r10d, %r10d vmovaps %xmm4, %xmm11 vbroadcastss 0xb8880e(%rip), %xmm10 # 0x1eeba20 vmovaps %xmm10, %xmm13 vmovaps %xmm4, %xmm5 vmovaps %xmm10, %xmm7 vmovaps %xmm4, %xmm8 vmovsd 0xb894c5(%rip), %xmm6 # 0x1eec6f0 vmovss 0xb894e1(%rip), %xmm9 # 0x1eec714 vmovaps %xmm13, 0x10(%rsp) movq %rax, %r15 subq %rdi, %r15 jbe 0x13633c0 vmovaps %xmm7, -0x50(%rsp) vmovaps %xmm5, -0x40(%rsp) vmovaps %xmm11, -0x30(%rsp) vmovaps %xmm10, -0x20(%rsp) vmovaps %xmm8, -0x10(%rsp) vmovaps %xmm6, (%rsp) movq 0x20(%r8), %rbx leaq (%rdi,%rdi,4), %r14 shlq $0x4, %r14 leaq (%r14,%rbx), %r12 addq $0x40, %r12 vmovss 0xb89490(%rip), %xmm0 # 0x1eec714 vmovaps %xmm0, -0x60(%rsp) vmovsd 0xb8945e(%rip), %xmm0 # 0x1eec6f0 vmovaps %xmm0, -0x70(%rsp) vbroadcastss 0xb898e3(%rip), %xmm12 # 0x1eecb84 vbroadcastss 0xb88776(%rip), %xmm13 # 0x1eeba20 xorl %ebx, %ebx vbroadcastss 0xb898cb(%rip), %xmm8 # 0x1eecb80 vmovaps %xmm13, %xmm15 vmovaps %xmm12, %xmm14 vmovaps %xmm13, -0x80(%rsp) vmovaps %xmm12, %xmm4 xorl %r14d, %r14d movq %r14, %r13 vmovaps -0x40(%r12), %xmm1 vmovaps -0x30(%r12), %xmm2 vmovaps -0x20(%r12), %xmm0 vmovaps -0x10(%r12), %xmm3 vmulps %xmm0, %xmm8, %xmm10 vmovaps -0x60(%rsp), %xmm6 vmulps %xmm1, %xmm8, %xmm5 vaddps %xmm5, %xmm10, %xmm11 vmulps %xmm3, %xmm8, %xmm5 vmulps %xmm2, %xmm8, %xmm7 vaddps %xmm5, %xmm7, %xmm5 vmovaps -0x80(%rsp), %xmm7 vaddps %xmm5, %xmm11, %xmm10 vminps %xmm1, %xmm13, %xmm13 vmaxps %xmm2, %xmm12, %xmm12 vminps %xmm0, %xmm15, %xmm15 vmaxps %xmm3, %xmm14, %xmm14 vminps %xmm10, %xmm7, %xmm7 vmovaps %xmm7, -0x80(%rsp) vmovsd (%r12), %xmm0 vcmpltps %xmm0, %xmm6, %xmm1 vinsertps $0x50, %xmm0, %xmm6, %xmm2 # xmm2 = xmm6[0],xmm0[1],xmm6[2,3] vblendps $0x2, %xmm6, %xmm0, %xmm3 # xmm3 = xmm0[0],xmm6[1],xmm0[2,3] vblendvps %xmm1, %xmm2, %xmm3, %xmm1 vmovaps %xmm1, -0x60(%rsp) vmovaps -0x70(%rsp), %xmm2 vmaxps %xmm10, %xmm4, %xmm4 movl -0x14(%r12), %r14d addq %r14, %rbx movl -0x4(%r12), %r14d cmpq %r14, %r13 setb %bpl vmovd %ebp, %xmm1 vpshufd $0x50, %xmm1, %xmm1 # xmm1 = xmm1[0,0,1,1] vpslld $0x1f, %xmm1, %xmm1 cmovaq %r13, %r14 vblendvps %xmm1, %xmm0, %xmm2, %xmm2 vmovaps %xmm2, -0x70(%rsp) addq $0x50, %r12 decq %r15 jne 0x13632cc vmovaps (%rsp), %xmm6 vmovaps -0x10(%rsp), %xmm8 vmovaps -0x20(%rsp), %xmm10 vmovaps -0x30(%rsp), %xmm11 vmovaps -0x40(%rsp), %xmm5 vmovaps -0x50(%rsp), %xmm7 jmp 0x13633db xorl %r14d, %r14d xorl %ebx, %ebx vmovaps %xmm4, %xmm14 vbroadcastss 0xb8864e(%rip), %xmm13 # 0x1eeba20 vmovaps %xmm13, %xmm15 vmovaps %xmm4, %xmm12 vmovsd 0x88(%rsi), %xmm0 vcmpltps %xmm0, %xmm9, %xmm1 vblendps $0x2, %xmm9, %xmm0, %xmm2 # xmm2 = xmm0[0],xmm9[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm9, %xmm0 # xmm0 = xmm9[0],xmm0[1],xmm9[2,3] vblendvps %xmm1, %xmm2, %xmm0, %xmm0 vmovaps %xmm10, (%rdx) vmovaps %xmm8, 0x10(%rdx) vmovaps %xmm7, 0x20(%rdx) vmovaps %xmm5, 0x30(%rdx) vmovaps 0x10(%rsp), %xmm1 vmovaps %xmm1, 0x40(%rdx) vmovaps %xmm11, 0x50(%rdx) movq %r10, 0x70(%rdx) movq %r11, 0x78(%rdx) vmovlhps %xmm0, %xmm6, %xmm0 # xmm0 = xmm6[0],xmm0[0] vmovaps %xmm0, 0x80(%rdx) movq %r8, 0x90(%rdx) movq %r9, 0x60(%rdx) movq %rdi, 0x68(%rdx) vmovsd 0x88(%rsi), %xmm0 vmovaps -0x60(%rsp), %xmm5 vcmpltps %xmm0, %xmm5, %xmm1 vblendps $0x2, %xmm5, %xmm0, %xmm2 # xmm2 = xmm0[0],xmm5[1],xmm0[2,3] vinsertps $0x50, %xmm0, %xmm5, %xmm0 # xmm0 = xmm5[0],xmm0[1],xmm5[2,3] vblendvps %xmm1, %xmm2, %xmm0, %xmm0 movq 0x90(%rsi), %rdx vmovaps %xmm13, (%rcx) vmovaps %xmm12, 0x10(%rcx) vmovaps %xmm15, 0x20(%rcx) vmovaps %xmm14, 0x30(%rcx) vmovaps -0x80(%rsp), %xmm1 vmovaps %xmm1, 0x40(%rcx) vmovaps %xmm4, 0x50(%rcx) vmovaps -0x70(%rsp), %xmm1 vmovlhps %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0] movq %rbx, 0x70(%rcx) movq %r14, 0x78(%rcx) vmovaps %xmm0, 0x80(%rcx) movq %rdx, 0x90(%rcx) movq %rdi, 0x60(%rcx) movq %rax, 0x68(%rcx) addq $0x28, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::BVHNBuilderMBlurSAHGrid<8>::buildMultiSegment(unsigned long)
void buildMultiSegment(size_t numPrimitives) { /* create primref array */ mvector<PrimRefMB> prims(scene->device,numPrimitives); PrimInfoMB pinfo = createPrimRefArrayMSMBlurGrid(scene,prims,bvh->scene->progressInterface); /* early out if no valid primitives */ if (pinfo.size() == 0) { bvh->clear(); return; } GridRecalculatePrimRef recalculatePrimRef(scene,sgrids.data()); /* estimate acceleration structure size */ const size_t node_bytes = pinfo.num_time_segments*sizeof(AABBNodeMB)/(4*N); //FIXME: check leaf_bytes //const size_t leaf_bytes = size_t(1.2*Primitive::blocks(pinfo.num_time_segments)*sizeof(SubGridQBVHN<N>)); const size_t leaf_bytes = size_t(1.2*(float)numPrimitives/N * sizeof(SubGridQBVHN<N>)); bvh->alloc.init_estimate(node_bytes+leaf_bytes); /* settings for BVH build */ BVHBuilderMSMBlur::Settings settings; settings.branchingFactor = N; settings.maxDepth = BVH::maxDepth; settings.logBlockSize = bsr(sahBlockSize); settings.minLeafSize = min(minLeafSize,maxLeafSize); settings.maxLeafSize = maxLeafSize; settings.travCost = travCost; settings.intCost = intCost; settings.singleLeafTimeSegment = false; settings.singleThreadThreshold = bvh->alloc.fixSingleThreadThreshold(N,DEFAULT_SINGLE_THREAD_THRESHOLD,pinfo.size(),node_bytes+leaf_bytes); /* build hierarchy */ auto root = BVHBuilderMSMBlur::build<NodeRef>(prims,pinfo,scene->device, recalculatePrimRef, typename BVH::CreateAlloc(bvh), typename BVH::AABBNodeMB4D::Create(), typename BVH::AABBNodeMB4D::Set(), CreateMSMBlurLeafGrid<N>(scene,bvh,sgrids.data()), bvh->scene->progressInterface, settings); bvh->set(root.ref,root.lbounds,pinfo.num_time_segments); }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x1b0, %rsp # imm = 0x1B0 movq %rsi, %r14 movq %rdi, %rbx movq 0x18(%rdi), %rax movq 0x188(%rax), %rax leaq 0x550(%rax), %rdi testq %rax, %rax cmoveq %rax, %rdi movq %rdi, 0x30(%rsp) movb $0x0, 0x38(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x40(%rsp) movq $0x0, 0x50(%rsp) testq %rsi, %rsi je 0x136a770 movq %r14, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %r15 movq (%rdi), %rax movq %r15, %rsi xorl %edx, %edx callq *(%rax) cmpq $0x1c00000, %r15 # imm = 0x1C00000 jb 0x136a754 leaq 0x38(%rsp), %rsi movq %r15, %rdi callq 0x1ee665d jmp 0x136a761 movl $0x10, %esi movq %r15, %rdi callq 0x1ee60ac movq %rax, 0x50(%rsp) movq %r14, 0x40(%rsp) movq %r14, 0x48(%rsp) leaq 0x40(%rsp), %r13 movq 0x10(%rbx), %rax movq 0x18(%rbx), %rdx movl $0x260, %r8d # imm = 0x260 addq 0x68(%rax), %r8 vmovsd 0xb81f61(%rip), %xmm0 # 0x1eec6f0 leaq 0xc0(%rsp), %rdi leaq 0x30(%rsp), %rcx movq %rbx, %rsi callq 0x136aad6 movq 0x128(%rsp), %rax cmpq 0x120(%rsp), %rax jne 0x136a7c5 movq 0x10(%rbx), %rdi movq (%rdi), %rax callq *0x28(%rax) jmp 0x136a9db movq 0x10(%rbx), %rdi imulq $0x1c0, 0x130(%rsp), %r12 # imm = 0x1C0 shrq $0x5, %r12 testq %r14, %r14 js 0x136a7e5 vcvtsi2ss %r14, %xmm1, %xmm0 jmp 0x136a7fb movq %r14, %rax shrq %rax andl $0x1, %r14d orq %rax, %r14 vcvtsi2ss %r14, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 movq 0x18(%rbx), %r14 movq 0x60(%rbx), %r15 vcvtss2sd %xmm0, %xmm0, %xmm0 vmulsd 0xb99949(%rip), %xmm0, %xmm0 # 0x1f04158 vmulsd 0xbbc739(%rip), %xmm0, %xmm0 # 0x1f26f50 vmulsd 0xbbc739(%rip), %xmm0, %xmm0 # 0x1f26f58 vcvttsd2si %xmm0, %rax movq %rax, %rcx sarq $0x3f, %rcx vsubsd 0xb857ad(%rip), %xmm0, %xmm0 # 0x1eeffe0 vcvttsd2si %xmm0, %rdx andq %rcx, %rdx orq %rax, %rdx addq %rdx, %r12 addq $0x78, %rdi movq %r12, %rsi callq 0x90acb8 vmovaps 0xbb75cb(%rip), %ymm0 # 0x1f21e20 vmovups %ymm0, 0x58(%rsp) movl $0x8, %eax movq %rax, 0x78(%rsp) movl $0x3f800000, 0x80(%rsp) # imm = 0x3F800000 xorl %ecx, %ecx movb %cl, 0x88(%rsp) movq $0x400, 0x90(%rsp) # imm = 0x400 movq %rax, 0x58(%rsp) movq $0x50, 0x60(%rsp) movq 0x20(%rbx), %rax bsrq %rax, %rax movq %rax, 0x68(%rsp) movq 0x30(%rbx), %rax movq 0x38(%rbx), %rdx cmpq %rdx, %rax cmovaeq %rdx, %rax movq %rax, 0x70(%rsp) movq %rdx, 0x78(%rsp) vbroadcastss 0xb81e52(%rip), %xmm0 # 0x1eec714 vmovlps %xmm0, 0x80(%rsp) vmovss 0x28(%rbx), %xmm0 vmovss %xmm0, 0x84(%rsp) movb %cl, 0x88(%rsp) movq 0x10(%rbx), %rdi movq 0x128(%rsp), %rcx subq 0x120(%rsp), %rcx addq $0x78, %rdi movl $0x8, %esi movl $0x400, %edx # imm = 0x400 movq %r12, %r8 vzeroupper callq 0xbdde9e leaq 0x58(%rsp), %rdx movq %rax, 0x38(%rdx) movq 0x18(%rbx), %rax movq 0x188(%rax), %rsi leaq 0x550(%rsi), %rcx testq %rsi, %rsi cmoveq %rsi, %rcx movq 0x10(%rbx), %rsi leaq 0x78(%rsi), %rdi movq 0x60(%rbx), %r8 movq %rax, 0x98(%rsp) movq %rsi, 0xa0(%rsp) movq %r8, 0xa8(%rsp) movq 0x68(%rsi), %rax leaq 0xd98012(%rip), %rsi # 0x2102970 leaq 0xb0(%rsp), %r8 movq %rsi, (%r8) movq 0x268(%rax), %rax movq %rax, 0x8(%r8) movq 0xa8(%rsp), %rax movq %rax, 0x18(%rsp) vmovupd 0x98(%rsp), %xmm0 vmovupd %xmm0, 0x8(%rsp) movq %rdx, 0x28(%rsp) movq %r8, 0x20(%rsp) movq %rdi, (%rsp) leaq 0x150(%rsp), %rdi leaq 0x30(%rsp), %rsi leaq 0xc0(%rsp), %rdx movq %r14, %r8 movq %r15, %r9 callq 0x136b5a2 movq 0x10(%rbx), %rdi leaq 0x160(%rsp), %rdx movq -0x10(%rdx), %rsi movq 0x130(%rsp), %rcx callq 0x14ce916 movq 0x48(%rsp), %rbx movq 0x50(%rsp), %rdi testq %rdi, %rdi je 0x136aa0f movq %rbx, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x136aa0a movzbl 0x38(%rsp), %edx callq 0x1ee67ca jmp 0x136aa0f callq 0x1ee612d testq %rbx, %rbx je 0x136aa2e movq 0x30(%rsp), %rdi shlq $0x4, %rbx leaq (%rbx,%rbx,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%r13) movq $0x0, 0x10(%r13) addq $0x1b0, %rsp # imm = 0x1B0 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq jmp 0x136aa59 jmp 0x136aa59 jmp 0x136aa59 jmp 0x136aa59 movq %rax, %rbx movq 0x48(%rsp), %r14 movq 0x50(%rsp), %rdi testq %rdi, %rdi jne 0x136aaa4 testq %r14, %r14 je 0x136aa8a movq 0x30(%rsp), %rdi shlq $0x4, %r14 leaq (%r14,%r14,4), %rsi negq %rsi movq (%rdi), %rax movl $0x1, %edx callq *(%rax) vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, (%r13) movq $0x0, 0x10(%r13) movq %rbx, %rdi callq 0x6a600 movq %r14, %rax shlq $0x4, %rax leaq (%rax,%rax,4), %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x136aac4 movzbl 0x38(%rsp), %edx callq 0x1ee67ca jmp 0x136aa6b callq 0x1ee612d jmp 0x136aa6b jmp 0x136aacd movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/bvh_builder_sah_mb.cpp
embree::avx::BVHBuilderMSMBlur::BuilderT<embree::NodeRefPtr<8>, embree::avx::GridRecalculatePrimRef, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeafGrid<8>, embree::Scene::BuildProgressMonitorInterface>::BuilderT(embree::MemoryMonitorInterface*, embree::avx::GridRecalculatePrimRef, embree::BVHN<8>::CreateAlloc, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Create, embree::AABBNodeMB4D_t<embree::NodeRefPtr<8>, 8>::Set, embree::avx::CreateMSMBlurLeafGrid<8>, embree::Scene::BuildProgressMonitorInterface, embree::avx::BVHBuilderMSMBlur::Settings const&)
BuilderT (MemoryMonitorInterface* device, const RecalculatePrimRef recalculatePrimRef, const CreateAllocFunc createAlloc, const CreateNodeFunc createNode, const SetNodeFunc setNode, const CreateLeafFunc createLeaf, const ProgressMonitor progressMonitor, const Settings& settings) : cfg(settings), heuristicObjectSplit(), heuristicTemporalSplit(device, recalculatePrimRef), recalculatePrimRef(recalculatePrimRef), createAlloc(createAlloc), createNode(createNode), setNode(setNode), createLeaf(createLeaf), progressMonitor(progressMonitor) { if (cfg.branchingFactor > MAX_BRANCHING_FACTOR) throw_RTCError(RTC_ERROR_UNKNOWN,"bvh_builder: branching factor too large"); }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x28, %rsp movq 0x68(%rsp), %rax vmovups (%rax), %ymm0 vmovups 0x20(%rax), %ymm1 vmovups %ymm1, 0x20(%rdi) vmovups %ymm0, (%rdi) movq %rsi, 0x48(%rdi) movq %rdx, 0x50(%rdi) movq %rcx, 0x58(%rdi) movq %rdx, 0x60(%rdi) movq %rcx, 0x68(%rdi) movq %r8, 0x70(%rdi) vmovaps 0x50(%rsp), %xmm0 vmovups %xmm0, 0x80(%rdi) movq 0x60(%rsp), %rax movq %rax, 0x90(%rdi) leaq 0xd94dda(%rip), %rax # 0x2102970 movq %rax, 0x98(%rdi) movq 0x8(%r9), %rax movq %rax, 0xa0(%rdi) cmpq $0x11, (%rdi) jae 0x136dbbc addq $0x28, %rsp popq %rbx popq %r14 popq %r15 popq %rbp vzeroupper retq movl $0x30, %edi vzeroupper callq 0x6a3b0 movq %rax, %rbx leaq 0x18(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb7f070(%rip), %rsi # 0x1eecc4c leaq 0xb7f090(%rip), %rdx # 0x1eecc73 leaq 0x8(%rsp), %rdi callq 0x8d7230 leaq 0xd93cbc(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x1, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x8(%rsp), %rsi movq 0x10(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd93c2a(%rip), %rsi # 0x2101850 leaq -0xa96e11(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rax, %r14 xorl %ebp, %ebp jmp 0x136dc4a movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x8(%rsp), %rdi cmpq %r15, %rdi je 0x136dc61 callq 0x6a4f0 jmp 0x136dc61 movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x136dc6e movq %rbx, %rdi callq 0x6a8a0 movq %r14, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/../builders/bvh_builder_msmblur.h
embree::avx::BVH4BuilderTwoLevelTriangle4vMeshSAH(void*, embree::Scene*, bool)
Builder* BVH4BuilderTwoLevelTriangle4vMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,TriangleMesh,Triangle4v>((BVH4*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); }
pushq %r15 pushq %r14 pushq %rbx movl %edx, %ebx movq %rsi, %r14 movq %rdi, %r15 movl $0xa8, %edi callq 0x6a170 xorl %ecx, %ecx movq %rcx, 0x8(%rax) leaq 0xda0caa(%rip), %rdx # 0x2116cf8 movq %rdx, (%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%rax) movq %rcx, 0x20(%rax) movq %r15, 0x28(%rax) movq %r14, 0x30(%rax) movq 0x188(%r14), %rdx leaq 0x550(%rdx), %rsi testq %rdx, %rdx cmoveq %rdx, %rsi movq %rsi, 0x38(%rax) movb %cl, 0x40(%rax) movq %rcx, 0x58(%rax) vmovups %xmm0, 0x48(%rax) movq %rsi, 0x60(%rax) movb %cl, 0x68(%rax) movq %rcx, 0x80(%rax) vmovups %xmm0, 0x70(%rax) movq $0x400, 0x90(%rax) # imm = 0x400 movq $0x100000, 0x98(%rax) # imm = 0x100000 movb %bl, 0xa0(%rax) popq %rbx popq %r14 popq %r15 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVH4BuilderTwoLevelQuadMeshSAH(void*, embree::Scene*, bool)
Builder* BVH4BuilderTwoLevelQuadMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,QuadMesh,Quad4v>((BVH4*)bvh,scene,QuadMesh::geom_type,useMortonBuilder); }
pushq %r15 pushq %r14 pushq %rbx movl %edx, %ebx movq %rsi, %r14 movq %rdi, %r15 movl $0xa8, %edi callq 0x6a170 xorl %ecx, %ecx movq %rcx, 0x8(%rax) leaq 0xda14ce(%rip), %rdx # 0x2117648 movq %rdx, (%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%rax) movq %rcx, 0x20(%rax) movq %r15, 0x28(%rax) movq %r14, 0x30(%rax) movq 0x188(%r14), %rdx leaq 0x550(%rdx), %rsi testq %rdx, %rdx cmoveq %rdx, %rsi movq %rsi, 0x38(%rax) movb %cl, 0x40(%rax) movq %rcx, 0x58(%rax) vmovups %xmm0, 0x48(%rax) movq %rsi, 0x60(%rax) movb %cl, 0x68(%rax) movq %rcx, 0x80(%rax) vmovups %xmm0, 0x70(%rax) movq $0x400, 0x90(%rax) # imm = 0x400 movq $0x200000, 0x98(%rax) # imm = 0x200000 movb %bl, 0xa0(%rax) popq %rbx popq %r14 popq %r15 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVH4BuilderTwoLevelInstanceSAH(void*, embree::Scene*, embree::Geometry::GTypeMask, bool)
Builder* BVH4BuilderTwoLevelInstanceSAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,Instance,InstancePrimitive>((BVH4*)bvh,scene,gtype,useMortonBuilder); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx pushq %rax movl %ecx, %ebx movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 movl $0xa8, %edi callq 0x6a170 xorl %ecx, %ecx movq %rcx, 0x8(%rax) leaq 0xda1cec(%rip), %rdx # 0x2117f98 movq %rdx, (%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%rax) movq %rcx, 0x20(%rax) movq %r12, 0x28(%rax) movq %r15, 0x30(%rax) movq 0x188(%r15), %rdx leaq 0x550(%rdx), %rsi testq %rdx, %rdx cmoveq %rdx, %rsi movq %rsi, 0x38(%rax) movb %cl, 0x40(%rax) movq %rcx, 0x58(%rax) vmovups %xmm0, 0x48(%rax) movq %rsi, 0x60(%rax) movb %cl, 0x68(%rax) movq %rcx, 0x80(%rax) vmovups %xmm0, 0x70(%rax) movq $0x400, 0x90(%rax) # imm = 0x400 movq %r14, 0x98(%rax) movb %bl, 0xa0(%rax) addq $0x8, %rsp popq %rbx popq %r12 popq %r14 popq %r15 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVH8BuilderTwoLevelTriangle4MeshSAH(void*, embree::Scene*, bool)
Builder* BVH8BuilderTwoLevelTriangle4MeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,TriangleMesh,Triangle4>((BVH8*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); }
pushq %r15 pushq %r14 pushq %rbx movl %edx, %ebx movq %rsi, %r14 movq %rdi, %r15 movl $0xa8, %edi callq 0x6a170 xorl %ecx, %ecx movq %rcx, 0x8(%rax) leaq 0xda2506(%rip), %rdx # 0x21188e8 movq %rdx, (%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%rax) movq %rcx, 0x20(%rax) movq %r15, 0x28(%rax) movq %r14, 0x30(%rax) movq 0x188(%r14), %rdx leaq 0x550(%rdx), %rsi testq %rdx, %rdx cmoveq %rdx, %rsi movq %rsi, 0x38(%rax) movb %cl, 0x40(%rax) movq %rcx, 0x58(%rax) vmovups %xmm0, 0x48(%rax) movq %rsi, 0x60(%rax) movb %cl, 0x68(%rax) movq %rcx, 0x80(%rax) vmovups %xmm0, 0x70(%rax) movq $0x400, 0x90(%rax) # imm = 0x400 movq $0x100000, 0x98(%rax) # imm = 0x100000 movb %bl, 0xa0(%rax) popq %rbx popq %r14 popq %r15 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVH8BuilderTwoLevelVirtualSAH(void*, embree::Scene*, bool)
Builder* BVH8BuilderTwoLevelVirtualSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,UserGeometry,Object>((BVH8*)bvh,scene,UserGeometry::geom_type,useMortonBuilder); }
pushq %r15 pushq %r14 pushq %rbx movl %edx, %ebx movq %rsi, %r14 movq %rdi, %r15 movl $0xa8, %edi callq 0x6a170 xorl %ecx, %ecx movq %rcx, 0x8(%rax) leaq 0xda354e(%rip), %rdx # 0x2119b88 movq %rdx, (%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%rax) movq %rcx, 0x20(%rax) movq %r15, 0x28(%rax) movq %r14, 0x30(%rax) movq 0x188(%r14), %rdx leaq 0x550(%rdx), %rsi testq %rdx, %rdx cmoveq %rdx, %rsi movq %rsi, 0x38(%rax) movb %cl, 0x40(%rax) movq %rcx, 0x58(%rax) vmovups %xmm0, 0x48(%rax) movq %rsi, 0x60(%rax) movb %cl, 0x68(%rax) movq %rcx, 0x80(%rax) vmovups %xmm0, 0x70(%rax) movq $0x400, 0x90(%rax) # imm = 0x400 movq $0x20000000, 0x98(%rax) # imm = 0x20000000 movb %bl, 0xa0(%rax) popq %rbx popq %r14 popq %r15 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVH8BuilderTwoLevelInstanceArraySAH(void*, embree::Scene*, embree::Geometry::GTypeMask, bool)
Builder* BVH8BuilderTwoLevelInstanceArraySAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,InstanceArray,InstanceArrayPrimitive>((BVH8*)bvh,scene,gtype,useMortonBuilder); }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx pushq %rax movl %ecx, %ebx movq %rdx, %r14 movq %rsi, %r15 movq %rdi, %r12 movl $0xa8, %edi callq 0x6a170 xorl %ecx, %ecx movq %rcx, 0x8(%rax) leaq 0xda3d64(%rip), %rdx # 0x211a4d8 movq %rdx, (%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x10(%rax) movq %rcx, 0x20(%rax) movq %r12, 0x28(%rax) movq %r15, 0x30(%rax) movq 0x188(%r15), %rdx leaq 0x550(%rdx), %rsi testq %rdx, %rdx cmoveq %rdx, %rsi movq %rsi, 0x38(%rax) movb %cl, 0x40(%rax) movq %rcx, 0x58(%rax) vmovups %xmm0, 0x48(%rax) movq %rsi, 0x60(%rax) movb %cl, 0x68(%rax) movq %rcx, 0x80(%rax) vmovups %xmm0, 0x70(%rax) movq $0x400, 0x90(%rax) # imm = 0x400 movq %r14, 0x98(%rax) movb %bl, 0xa0(%rax) addq $0x8, %rsp popq %rbx popq %r12 popq %r14 popq %r15 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::setupLargeBuildRefBuilder(unsigned long, embree::TriangleMesh const*)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupLargeBuildRefBuilder (size_t objectID, Mesh const * const mesh) { if (bvh->objects[objectID] == nullptr || // new mesh builders[objectID]->meshQualityChanged (mesh->quality) || // changed build quality dynamic_cast<RefBuilderLarge*>(builders[objectID].get()) == nullptr) // size change resulted in small->large change { Builder* builder = nullptr; delete bvh->objects[objectID]; createMeshAccel(objectID, builder); builders[objectID].reset (new RefBuilderLarge(objectID, builder, mesh->quality)); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rdx, %r14 movq %rsi, %rbx movq %rdi, %r15 movq 0x28(%rdi), %rax movq 0x200(%rax), %rax cmpq $0x0, (%rax,%rsi,8) je 0x13792ff movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi movzwl 0x3e(%r14), %esi andl $0x7, %esi movq (%rdi), %rax callq *0x18(%rax) testb %al, %al jne 0x13792ff movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13792ff leaq 0xd9d690(%rip), %rsi # 0x2116978 leaq 0xd9d6e1(%rip), %rdx # 0x21169d0 xorl %ecx, %ecx callq 0x6a9a0 testq %rax, %rax jne 0x13793a7 movq $0x0, (%rsp) movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x1379321 movq (%rdi), %rax callq *0x8(%rax) movq %rsp, %r12 movq %r15, %rdi movq %rbx, %rsi movq %r12, %rdx callq 0x1379c8e movq 0x10(%r15), %r13 movl $0x20, %edi callq 0x6a170 movq %rax, %r15 movq (%r12), %r12 testq %r12, %r12 je 0x1379356 movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) movzwl 0x3e(%r14), %ebp leaq 0xd9d696(%rip), %rax # 0x21169f8 movq %rax, (%r15) movq %rbx, 0x8(%r15) movq %r12, 0x10(%r15) testq %r12, %r12 je 0x137937c movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) andl $0x7, %ebp movl %ebp, 0x18(%r15) movq (%r13,%rbx,8), %rdi movq %r15, (%r13,%rbx,8) testq %rdi, %rdi je 0x1379398 movq (%rdi), %rax callq *0x8(%rax) testq %r12, %r12 je 0x13793a7 movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq jmp 0x13793c7 movq %rax, %rbx movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) jmp 0x13793d2 movq %rax, %rdi callq 0x8d6de8 movq %rax, %rbx movq %r15, %rdi callq 0x6a4f0 movq %rbx, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::NodeRefPtr<4> embree::avx::GeneralBVHBuilder::build<embree::NodeRefPtr<4>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(unsigned long)>(embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*, embree::avx::PrimInfoExtRange const&, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(unsigned long) const&, embree::avx::GeneralBVHBuilder::Settings const&)
__noinline static ReductionTy build(Heuristic& heuristic, PrimRef* prims, const Set& set, CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, const ProgressMonitor& progressMonitor, const Settings& settings) { typedef BuildRecordT<Set,typename Heuristic::Split> BuildRecord; typedef BuilderT< BuildRecord, Heuristic, Set, PrimRef, ReductionTy, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, UpdateNodeFunc, CreateLeafFunc, DefaultCanCreateLeafFunc<PrimRef, Set>, DefaultCanCreateLeafSplitFunc<PrimRef, Set>, ProgressMonitor> Builder; /* instantiate builder */ Builder builder(prims, heuristic, createAlloc, createNode, updateNode, createLeaf, DefaultCanCreateLeafFunc<PrimRef, Set>(), DefaultCanCreateLeafSplitFunc<PrimRef, Set>(), progressMonitor, settings); /* build hierarchy */ BuildRecord record(1,set); const ReductionTy root = builder.recurse(record,nullptr,true); _mm_mfence(); // to allow non-temporal stores during build return root; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 movq %r9, %rax movq %r8, %r10 movq %rdx, %r14 movq %rdi, %rdx movq 0x160(%rsp), %r12 leaq 0x28(%rsp), %r11 movq %rcx, (%r11) subq $0x8, %rsp leaq 0x15(%rsp), %r13 leaq 0x38(%rsp), %r15 leaq 0xb0(%rsp), %rbx leaq 0x17(%rsp), %r8 leaq 0x16(%rsp), %r9 movq %rbx, %rdi movq %r11, %rcx pushq %r12 pushq %rax pushq %r13 pushq %r15 pushq %r10 callq 0x137b2ba addq $0x30, %rsp movq $0x1, (%r15) movb $0x0, 0x8(%r15) vmovaps (%r14), %xmm0 vmovaps %xmm0, 0x10(%r15) vmovaps 0x10(%r14), %xmm0 vmovaps %xmm0, 0x20(%r15) vmovaps 0x20(%r14), %xmm0 vmovaps %xmm0, 0x30(%r15) vmovaps 0x30(%r14), %xmm0 vmovaps %xmm0, 0x40(%r15) vmovaps 0x40(%r14), %xmm0 vmovaps %xmm0, 0x50(%r15) movq 0x50(%r14), %rax movq %rax, 0x60(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x10(%rsp) movq $0x0, 0x20(%rsp) subq $0x20, %rsp movq 0x40(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x30(%rsp), %xmm0 vmovups %xmm0, (%rsp) leaq 0x50(%rsp), %rsi movq %rbx, %rdi movl $0x1, %edx callq 0x137b3c8 addq $0x20, %rsp mfence addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::GeneralBVHBuilder::BuilderT<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::NodeRefPtr<4>, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(unsigned long)>::recurse(embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>&, embree::FastAllocator::CachedAllocator, bool)
const ReductionTy recurse(BuildRecord& current, Allocator alloc, bool toplevel) { /* get thread local allocator */ if (!alloc) alloc = createAlloc(); /* call memory monitor function to signal progress */ if (toplevel && current.size() <= cfg.singleThreadThreshold) progressMonitor(current.size()); /*! find best split */ auto split = heuristic.find(current.prims,cfg.logBlockSize); /*! compute leaf and split cost */ const float leafSAH = cfg.intCost*current.prims.leafSAH(cfg.logBlockSize); const float splitSAH = cfg.travCost*halfArea(current.prims.geomBounds)+cfg.intCost*split.splitSAH(); assert((current.prims.size() == 0) || ((leafSAH >= 0) && (splitSAH >= 0))); /*! create a leaf node when threshold reached or SAH tells us to stop */ if (current.prims.size() <= cfg.minLeafSize || current.depth+MIN_LARGE_LEAF_LEVELS >= cfg.maxDepth || (current.prims.size() <= cfg.maxLeafSize && leafSAH <= splitSAH)) { heuristic.deterministic_order(current.prims); return createLargeLeaf(current,alloc); } /*! perform initial split */ Set lprims,rprims; heuristic.split(split,current.prims,lprims,rprims); /*! initialize child list with initial split */ ReductionTy values[MAX_BRANCHING_FACTOR]; BuildRecord children[MAX_BRANCHING_FACTOR]; children[0] = BuildRecord(current.depth+1,lprims); children[1] = BuildRecord(current.depth+1,rprims); size_t numChildren = 2; /*! split until node is full or SAH tells us to stop */ while (numChildren < cfg.branchingFactor) { /*! find best child to split */ float bestArea = neg_inf; ssize_t bestChild = -1; for (size_t i=0; i<numChildren; i++) { /* ignore leaves as they cannot get split */ if (children[i].prims.size() <= cfg.minLeafSize) continue; /* find child with largest surface area */ if (halfArea(children[i].prims.geomBounds) > bestArea) { bestChild = i; bestArea = halfArea(children[i].prims.geomBounds); } } if (bestChild == -1) break; /* perform best found split */ BuildRecord& brecord = children[bestChild]; BuildRecord lrecord(current.depth+1); BuildRecord rrecord(current.depth+1); auto split = heuristic.find(brecord.prims,cfg.logBlockSize); heuristic.split(split,brecord.prims,lrecord.prims,rrecord.prims); children[bestChild ] = lrecord; children[numChildren] = rrecord; numChildren++; } /* set barrier for primrefarrayalloc */ if (unlikely(current.size() > cfg.primrefarrayalloc)) for (size_t i=0; i<numChildren; i++) children[i].alloc_barrier = children[i].size() <= cfg.primrefarrayalloc; /* sort buildrecords for faster shadow ray traversal */ std::sort(&children[0],&children[numChildren],std::greater<BuildRecord>()); /*! create an inner node */ auto node = createNode(children,numChildren,alloc); /* spawn tasks */ if (current.size() > cfg.singleThreadThreshold) { /*! parallel_for is faster than spawning sub-tasks */ parallel_for(size_t(0), numChildren, [&] (const range<size_t>& r) { // FIXME: no range here for (size_t i=r.begin(); i<r.end(); i++) { values[i] = recurse(children[i],nullptr,true); _mm_mfence(); // to allow non-temporal stores during build } }); return updateNode(current,children,node,values,numChildren); } /* recurse into each child */ else { for (size_t i=0; i<numChildren; i++) values[i] = recurse(children[i],alloc,false); return updateNode(current,children,node,values,numChildren); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x998, %rsp # imm = 0x998 movl %edx, %ebp movq %rsi, %r14 movq %rdi, %rbx leaq 0x9d0(%rsp), %r13 cmpq $0x0, (%r13) jne 0x137b4f5 movq %r14, 0x20(%rsp) movq 0x50(%rbx), %rax movq (%rax), %r14 leaq 0xda9930(%rip), %rdi # 0x2124d38 callq 0x6a480 movq (%rax), %r15 testq %r15, %r15 jne 0x137b4d0 movq %rax, %r12 movl $0xc0, %edi movl $0x40, %esi callq 0x1ee60ac movq %rax, %r15 movq %rax, %rdi callq 0x1ee7b24 movq $0x0, 0x8(%r15) movq %r15, 0x40(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x48(%r15) vmovups %ymm0, 0x58(%r15) movq %r15, 0x80(%r15) vmovups %ymm0, 0x88(%r15) vmovups %ymm0, 0x98(%r15) movq %r15, (%r12) leaq 0xdab3ff(%rip), %rdi # 0x2126878 movq %rdi, 0x1d0(%rsp) movb $0x1, 0x1d8(%rsp) vzeroupper callq 0x1ee7bb6 leaq 0x40(%rsp), %rsi movq %r15, (%rsi) leaq 0xdab3e0(%rip), %rdi # 0x2126880 callq 0x90d91e movq 0x40(%rsp), %rsi testq %rsi, %rsi je 0x137b4b9 leaq 0x40(%rsp), %rdi callq 0x90da94 cmpb $0x1, 0x1d8(%rsp) jne 0x137b4d0 movq 0x1d0(%rsp), %rdi callq 0x1ee7c24 leaq 0x40(%r15), %rax subq $-0x80, %r15 cmpb $0x0, 0x10a(%r14) cmovneq %rax, %r15 movq %r14, (%r13) movq %rax, 0x8(%r13) movq %r15, 0x10(%r13) movq 0x20(%rsp), %r14 testb %bpl, %bpl je 0x137b523 movq 0x58(%r14), %rax subq 0x50(%r14), %rax cmpq 0x30(%rbx), %rax ja 0x137b523 movq 0x80(%rbx), %rax movq (%rax), %rax movq 0x28(%rax), %rax movq 0x68(%rax), %rdi vxorps %xmm0, %xmm0, %xmm0 callq 0x91839e movq 0x10(%rbx), %rcx movq 0x48(%rbx), %rsi leaq 0x10(%r14), %r15 leaq 0x190(%rsp), %rdi movq %r15, %rdx callq 0x137bf0a movb 0x10(%rbx), %cl movq 0x58(%r14), %rax movq 0x50(%r14), %rdx movq $-0x1, %rsi shlq %cl, %rsi notq %rsi addq %rax, %rsi subq %rdx, %rsi shrq %cl, %rsi testq %rsi, %rsi js 0x137b570 vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rsi, %xmm1, %xmm0 jmp 0x137b589 movq %rsi, %rcx shrq %rcx andl $0x1, %esi orq %rcx, %rsi vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rsi, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 subq %rdx, %rax cmpq 0x18(%rbx), %rax jbe 0x137b5f3 movq (%r14), %rcx addq $0x8, %rcx cmpq 0x8(%rbx), %rcx jae 0x137b5f3 cmpq 0x20(%rbx), %rax ja 0x137b63a vmovaps 0x20(%r14), %xmm1 vsubps 0x10(%r14), %xmm1, %xmm1 vmovshdup %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] vshufpd $0x1, %xmm1, %xmm1, %xmm3 # xmm3 = xmm1[1,0] vaddss %xmm3, %xmm2, %xmm4 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm4, %xmm1, %xmm1 vaddss %xmm2, %xmm1, %xmm1 vmovss 0x2c(%rbx), %xmm2 vmulss %xmm0, %xmm1, %xmm0 vmulss 0x190(%rsp), %xmm2, %xmm3 vmulss 0x28(%rbx), %xmm1, %xmm1 vmulss %xmm0, %xmm2, %xmm0 vaddss %xmm3, %xmm1, %xmm1 vucomiss %xmm0, %xmm1 jb 0x137b63a leaq 0x50(%r14), %rsi movq 0x48(%rbx), %rdi callq 0x137c06e movq 0x10(%r13), %rax movq %rax, 0x10(%rsp) vmovups (%r13), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r14, %rsi callq 0x137c0c2 movq %rax, %r12 movq %r12, %rax addq $0x998, %rsp # imm = 0x998 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movq 0x48(%rbx), %rdi leaq 0x190(%rsp), %rsi leaq 0x930(%rsp), %r13 leaq 0x8d0(%rsp), %r12 movq %r15, %rdx movq %r13, %rcx movq %r12, %r8 callq 0x137cd12 movq %r14, 0x20(%rsp) movq (%r14), %rax incq %rax vmovups (%r13), %ymm0 vmovups 0x20(%r13), %ymm1 movq 0x50(%r13), %rcx leaq 0x228(%rsp), %rdx movq %rax, -0x58(%rdx) xorl %esi, %esi movb %sil, -0x50(%rdx) vmovups %ymm0, -0x48(%rdx) vmovups %ymm1, -0x28(%rdx) vmovaps 0x40(%r13), %xmm0 vmovaps %xmm0, -0x8(%rdx) movq %rcx, 0x8(%rdx) vmovups (%r12), %ymm0 vmovups 0x20(%r12), %ymm1 movq 0x50(%r12), %rcx movq %rax, 0x18(%rdx) movb %sil, 0x20(%rdx) vmovups %ymm0, 0x28(%rdx) vmovups %ymm1, 0x48(%rdx) vmovaps 0x40(%r12), %xmm0 vmovaps %xmm0, 0x68(%rdx) movq %rcx, 0x78(%rdx) movl $0x2, %r15d cmpq (%rbx), %r15 jae 0x137b946 testq %r15, %r15 je 0x137b759 movq 0x18(%rbx), %rax movq $-0x1, %rbp leaq 0x228(%rsp), %rcx vmovss 0xb71477(%rip), %xmm0 # 0x1eecb84 xorl %edx, %edx movq (%rcx), %rsi subq -0x8(%rcx), %rsi cmpq %rax, %rsi jbe 0x137b74b vmovaps -0x38(%rcx), %xmm1 vsubps -0x48(%rcx), %xmm1, %xmm1 vmovshdup %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] vshufpd $0x1, %xmm1, %xmm1, %xmm3 # xmm3 = xmm1[1,0] vaddss %xmm3, %xmm2, %xmm4 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm4, %xmm1, %xmm1 vaddss %xmm2, %xmm1, %xmm1 vucomiss %xmm0, %xmm1 jbe 0x137b74b movq %rdx, %rbp vmovaps %xmm1, %xmm0 incq %rdx addq $0x70, %rcx cmpq %rdx, %r15 jne 0x137b70f jmp 0x137b760 movq $-0x1, %rbp cmpq $-0x1, %rbp je 0x137b93c imulq $0x70, %rbp, %rax movq 0x20(%rsp), %rcx movq (%rcx), %rcx incq %rcx movq %rcx, 0x40(%rsp) xorl %edx, %edx movb %dl, 0x48(%rsp) vbroadcastss 0xb70293(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, 0x50(%rsp) vbroadcastss 0xb713e8(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x60(%rsp) vmovaps %xmm1, 0x70(%rsp) vmovaps %xmm0, 0x80(%rsp) leaq 0x90(%rsp), %r14 movq %rdx, 0x10(%r14) vxorps %xmm2, %xmm2, %xmm2 vmovaps %xmm2, (%r14) movq %rcx, 0xc0(%rsp) movb %dl, 0xc8(%rsp) vmovaps %xmm1, 0xd0(%rsp) vmovaps %xmm0, 0xe0(%rsp) vmovaps %xmm1, 0xf0(%rsp) vmovaps %xmm0, 0x100(%rsp) movq %rbx, %r14 leaq 0x110(%rsp), %rbx movq %rdx, 0x10(%rbx) vmovaps %xmm2, (%rbx) movq 0x10(%r14), %rcx movq 0x48(%r14), %rsi leaq 0x1e0(%rsp,%rax), %r12 leaq 0x150(%rsp), %r13 movq %r13, %rdi movq %r12, %rdx vzeroupper callq 0x137bf0a movq 0x48(%r14), %rdi movq %r13, %rsi movq %r12, %rdx leaq 0x50(%rsp), %rcx leaq 0xd0(%rsp), %r8 callq 0x137cd12 movb 0x48(%rsp), %al movb %al, -0x8(%r12) movq 0x40(%rsp), %rax movq %rax, -0x10(%r12) vmovaps 0x50(%rsp), %xmm0 vmovaps %xmm0, (%r12) vmovaps 0x60(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r12) vmovaps 0x70(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r12) vmovaps 0x80(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r12) leaq 0x90(%rsp), %rcx movq 0x10(%rcx), %rax movq %rax, 0x50(%r12) vmovaps (%rcx), %xmm0 vmovaps %xmm0, 0x40(%r12) imulq $0x70, %r15, %rax movq 0xc0(%rsp), %rcx movq %rcx, 0x1d0(%rsp,%rax) movb 0xc8(%rsp), %cl movb %cl, 0x1d8(%rsp,%rax) vmovaps 0xd0(%rsp), %xmm0 vmovaps %xmm0, 0x1e0(%rsp,%rax) vmovaps 0xe0(%rsp), %xmm0 vmovaps %xmm0, 0x1f0(%rsp,%rax) vmovaps 0xf0(%rsp), %xmm0 vmovaps %xmm0, 0x200(%rsp,%rax) vmovaps 0x100(%rsp), %xmm0 vmovaps %xmm0, 0x210(%rsp,%rax) vmovaps (%rbx), %xmm0 vmovaps %xmm0, 0x220(%rsp,%rax) movq 0x10(%rbx), %rcx movq %r14, %rbx movq %rcx, 0x230(%rsp,%rax) incq %r15 cmpq $-0x1, %rbp jne 0x137b6e4 movq 0x20(%rsp), %r14 movq 0x58(%r14), %rax subq 0x50(%r14), %rax movq 0x38(%rbx), %rdx cmpq %rdx, %rax ja 0x137bd3a testq %r15, %r15 je 0x137b9a2 imulq $0x70, %r15, %rax leaq (%rsp,%rax), %r12 addq $0x1d0, %r12 # imm = 0x1D0 bsrq %r15, %rdx xorl $0x3f, %edx addl %edx, %edx xorq $0x7e, %rdx leaq 0x1d0(%rsp), %r13 movq %r13, %rdi movq %r12, %rsi vzeroupper callq 0x1383896 movq %r13, %rdi movq %r12, %rsi callq 0x1383954 leaq 0x9d0(%rsp), %rax movq (%rax), %r13 movq 0x8(%rax), %rbp movq $0x80, 0x28(%rsp) movq (%rbp), %r12 movq 0x8(%r12), %rax cmpq %rax, %r13 je 0x137bb2f movq %r12, 0xc0(%rsp) movb $0x1, 0xc8(%rsp) movq %r12, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%r12), %rax testq %rax, %rax je 0x137ba4f movq 0xa8(%r12), %rax addq 0x68(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%r12), %rax addq 0x98(%r12), %rax movq 0x50(%r12), %rcx addq 0x90(%r12), %rcx subq %rcx, %rax movq 0x8(%r12), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%r12), %rax addq 0x70(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%r12) vmovups %ymm0, 0x48(%r12) testq %r13, %r13 je 0x137ba91 movq 0x10(%r13), %rax movq %rax, 0x60(%r12) vmovups %ymm0, 0x88(%r12) vmovups %ymm0, 0x98(%r12) movq 0x10(%r13), %rax movq %rax, 0xa0(%r12) jmp 0x137baa5 vmovups %ymm0, 0x98(%r12) vmovups %ymm0, 0x88(%r12) movq %r13, %rax xchgq %rax, 0x8(%r12) movq %r12, 0x150(%rsp) leaq 0xdaadbc(%rip), %rdi # 0x2126878 movq %rdi, 0x40(%rsp) movb $0x1, 0x48(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%r13), %rsi cmpq 0x140(%r13), %rsi je 0x137baf3 movq 0x150(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%r13) jmp 0x137bb07 leaq 0x130(%r13), %rdi leaq 0x150(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x48(%rsp) jne 0x137bb18 movq 0x40(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0xc8(%rsp) jne 0x137bb2f movq 0xc0(%rsp), %rdi callq 0x1ee7c24 movq 0x28(%rsp), %rax addq %rax, 0x28(%rbp) movq 0x10(%rbp), %rcx movl %ecx, %edx negl %edx andl $0xf, %edx leaq (%rcx,%rax), %r12 addq %rdx, %r12 movq %r12, 0x10(%rbp) cmpq 0x18(%rbp), %r12 ja 0x137bd5a addq %rdx, 0x30(%rbp) subq %rax, %r12 addq 0x8(%rbp), %r12 vbroadcastss 0xb6feb4(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x60(%r12) vmovaps %xmm0, 0x40(%r12) vmovaps %xmm0, 0x20(%r12) vbroadcastss 0xb70ffa(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x70(%r12) vmovaps %xmm0, 0x50(%r12) vmovaps %xmm0, 0x30(%r12) vbroadcastsd 0xba62d0(%rip), %ymm0 # 0x1f21e78 vmovups %ymm0, (%r12) testq %r15, %r15 je 0x137bc00 leaq 0x1f0(%rsp), %rax xorl %ecx, %ecx vmovaps -0x10(%rax), %xmm0 vmovaps (%rax), %xmm1 vmovss %xmm0, 0x20(%r12,%rcx,4) vextractps $0x1, %xmm0, 0x40(%r12,%rcx,4) vextractps $0x2, %xmm0, 0x60(%r12,%rcx,4) vmovss %xmm1, 0x30(%r12,%rcx,4) vextractps $0x1, %xmm1, 0x50(%r12,%rcx,4) vextractps $0x2, %xmm1, 0x70(%r12,%rcx,4) incq %rcx addq $0x70, %rax cmpq %rcx, %r15 jne 0x137bbbd movq 0x58(%r14), %rax subq 0x50(%r14), %rax cmpq 0x30(%rbx), %rax jbe 0x137bccc leaq 0xc0(%rsp), %rax leaq 0x28(%rsp), %r14 movq %rax, (%r14) movq %rbx, 0x8(%r14) leaq 0x1d0(%rsp), %rax movq %rax, 0x10(%r14) leaq 0x40(%rsp), %rbx movw $0x401, 0xc(%rbx) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rbx) movq $0x8, 0x40(%rbx) movq %rbx, %rdi vzeroupper callq 0x6a660 leaq 0x150(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x148(%rsp), %rsi movq %r14, (%rsi) leaq 0x1f(%rsp), %rdx movq %rbx, %rcx callq 0x13844d4 leaq 0x40(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x137bde9 leaq 0x40(%rsp), %rdi callq 0x6aab0 testq %r15, %r15 je 0x137b622 xorl %eax, %eax movq 0xc0(%rsp,%rax,8), %rcx movq %rcx, (%r12,%rax,8) incq %rax cmpq %rax, %r15 jne 0x137bcb3 jmp 0x137b622 testq %r15, %r15 je 0x137b622 leaq 0x1d0(%rsp), %r14 xorl %r13d, %r13d leaq 0x9d0(%rsp), %rbp movq 0x10(%rbp), %rax movq %rax, 0x10(%rsp) vmovups (%rbp), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r14, %rsi xorl %edx, %edx vzeroupper callq 0x137b3c8 movq %rax, 0xc0(%rsp,%r13,8) incq %r13 addq $0x70, %r14 cmpq %r13, %r15 jne 0x137bce8 xorl %eax, %eax movq 0xc0(%rsp,%rax,8), %rcx movq %rcx, (%r12,%rax,8) incq %rax cmpq %rax, %r15 jne 0x137bd21 jmp 0x137b622 leaq 0x1d0(%rsp), %rsi movq %r15, %rdi vzeroupper callq 0x1ff2e1 testb $0x1, %al je 0x137b960 jmp 0x137b9a2 movq %rcx, 0x10(%rbp) shlq $0x2, %rax movq 0x20(%rbp), %rcx cmpq %rcx, %rax jbe 0x137bd87 leaq 0x28(%rsp), %rsi movl $0x40, %edx movq %r13, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 jmp 0x137bddf leaq 0x40(%rsp), %r14 movq %rcx, (%r14) movl $0x40, %edx movq %r13, %rdi movq %r14, %rsi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbp) movq 0x18(%rbp), %rax subq 0x10(%rbp), %rax addq 0x30(%rbp), %rax movq %rax, 0x30(%rbp) movq $0x0, 0x10(%rbp) movq (%r14), %rcx movq %rcx, 0x18(%rbp) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbp) cmpq %rcx, %rdx ja 0x137be1b movq %rax, 0x30(%rbp) movq 0x20(%rsp), %r14 jmp 0x137bb63 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb6fcf3(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xda8dfc(%rip), %rsi # 0x2124c08 movq 0xda8bb5(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 xorl %r14d, %r14d movq %r14, 0x10(%rbp) movq 0x20(%rbp), %rax leaq 0x40(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %r13, %rdi movq %rsi, %r13 xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbp) movq 0x18(%rbp), %rax subq 0x10(%rbp), %rax addq 0x30(%rbp), %rax movq %rax, 0x30(%rbp) movq %r14, 0x10(%rbp) movq (%r13), %rcx movq %rcx, 0x18(%rbp) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbp) cmpq %rcx, %rdx jbe 0x137bddb movq $0x0, 0x10(%rbp) xorl %r12d, %r12d jmp 0x137bddf movq %rax, %rbx leaq 0x40(%rsp), %rdi callq 0x8d6eda jmp 0x137beae movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x137bebb jmp 0x137bf02 jmp 0x137bf02 jmp 0x137bf02 jmp 0x137beb8 jmp 0x137bf02 movq %rax, %rbx leaq 0xc0(%rsp), %rdi jmp 0x137bee8 movq %rax, %rbx leaq 0x40(%rsp), %rdi callq 0x6aab0 jmp 0x137befa jmp 0x137bf02 movq %rax, %rbx movq 0x40(%rsp), %rsi testq %rsi, %rsi je 0x137bee0 leaq 0x40(%rsp), %rdi callq 0x90da94 leaq 0x1d0(%rsp), %rdi callq 0x8d6eda jmp 0x137befa movq %rax, %rbx movq %r15, %rdi callq 0x1ee612d movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
unsigned long embree::parallel_partitioning<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>, embree::EmptyTy, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&), embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&), embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&)>(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*, unsigned long, unsigned long, embree::EmptyTy const&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&) const&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&) const&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&) const&, unsigned long)
__noinline size_t parallel_partitioning(T* array, const size_t begin, const size_t end, const Vi &identity, V &leftReduction, V &rightReduction, const IsLeft& is_left, const Reduction_T& reduction_t, const Reduction_V& reduction_v, size_t BLOCK_SIZE = 128) { /* fall back to single threaded partitioning for small N */ if (unlikely(end-begin < BLOCK_SIZE)) return serial_partitioning(array,begin,end,leftReduction,rightReduction,is_left,reduction_t); /* otherwise use parallel code */ else { typedef parallel_partition_task<T,V,Vi,IsLeft,Reduction_T,Reduction_V> partition_task; std::unique_ptr<partition_task> p(new partition_task(&array[begin],end-begin,identity,is_left,reduction_t,reduction_v,BLOCK_SIZE)); return begin+p->partition(leftReduction,rightReduction); } }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x20, %rsp subq $0x100, %rsp # imm = 0x100 movq %r9, %r15 movq %r8, %r12 movq 0x18(%rbp), %r8 movq 0x10(%rbp), %r9 movq %rdx, %rbx subq %rsi, %rbx cmpq 0x28(%rbp), %rbx jb 0x13821aa movq %rdi, %r13 movl $0x34c0, %edi # imm = 0x34C0 movq %rsi, 0x8(%rsp) movl $0x40, %esi movq %rcx, 0x38(%rsp) callq 0x1ee60ac movq %rax, %r14 movq 0x8(%rsp), %rax shlq $0x6, %rax addq %rax, %r13 movq %r13, (%r14) movq %rbx, 0x8(%r14) movq 0x10(%rbp), %rax movq %rax, 0x10(%r14) movq 0x18(%rbp), %rax movq %rax, 0x18(%r14) movq 0x20(%rbp), %rax movq %rax, 0x20(%r14) movq 0x38(%rsp), %rax movq %rax, 0x28(%r14) xorl %edi, %edi callq 0x6ab80 movl %eax, %ecx movq 0x28(%rbp), %rsi leaq (%rbx,%rsi), %rax decq %rax xorl %edx, %edx divq %rsi cmpl $0x40, %ecx movl $0x40, %ebx cmovbl %ecx, %ebx cmpq %rbx, %rax cmovbq %rax, %rbx movq %rbx, 0x30(%r14) movq %r14, 0x28(%rsp) movq %r14, 0x30(%rsp) leaq 0x60(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) callq 0x6a660 leaq 0x30(%rsp), %rax leaq 0x40(%rsp), %rcx movq %rax, (%rcx) leaq 0x10(%rsp), %r8 leaq 0x60(%rsp), %r9 movl $0x1, %edx xorl %edi, %edi movq %rbx, %rsi callq 0x13824c8 leaq 0x60(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x138238c leaq 0x60(%rsp), %rdi callq 0x6aab0 movq 0x8(%r14), %rax movq 0x30(%r14), %rcx movq %rax, 0x40(%r14,%rcx,8) movq 0x30(%r14), %rax movq $0x0, 0x280(%r14,%rax,8) movq 0x30(%r14), %rdx testq %rdx, %rdx je 0x1381f25 leaq 0x2108(%r14), %rax xorl %ecx, %ecx vmovaps (%r12), %xmm0 vmovaps 0x10(%r12), %xmm1 vmovaps 0x20(%r12), %xmm2 vminps -0x1448(%rax), %xmm0, %xmm0 vmovaps 0x30(%r12), %xmm3 vmovaps %xmm0, (%r12) vmaxps -0x1438(%rax), %xmm1, %xmm0 vmovaps %xmm0, 0x10(%r12) vminps -0x1428(%rax), %xmm2, %xmm0 vmovaps %xmm0, 0x20(%r12) vmaxps -0x1418(%rax), %xmm3, %xmm0 vmovaps %xmm0, 0x30(%r12) movq -0x1408(%rax), %rdx addq %rdx, 0x40(%r12) movq -0x1400(%rax), %rdx addq %rdx, 0x48(%r12) vmovaps (%r15), %xmm0 vmovaps 0x10(%r15), %xmm1 vmovaps 0x20(%r15), %xmm2 vminps -0x48(%rax), %xmm0, %xmm0 vmovaps 0x30(%r15), %xmm3 vmovaps %xmm0, (%r15) vmaxps -0x38(%rax), %xmm1, %xmm0 vmovaps %xmm0, 0x10(%r15) vminps -0x28(%rax), %xmm2, %xmm0 vmovaps %xmm0, 0x20(%r15) vmaxps -0x18(%rax), %xmm3, %xmm0 vmovaps %xmm0, 0x30(%r15) movq -0x8(%rax), %rdx addq %rdx, 0x40(%r15) movq (%rax), %rdx addq %rdx, 0x48(%r15) incq %rcx movq 0x30(%r14), %rdx addq $0x50, %rax cmpq %rdx, %rcx jb 0x1381e52 movq 0x280(%r14), %r13 cmpq $0x2, %rdx jb 0x1382022 leaq 0x2(%rdx), %rax andq $-0x4, %rax addq $-0x2, %rdx vmovq %r13, %xmm0 vmovq %rdx, %xmm1 vpshufd $0x44, %xmm1, %xmm1 # xmm1 = xmm1[0,1,0,1] vinsertf128 $0x1, %xmm1, %ymm1, %ymm1 leaq 0x288(%r14), %rcx xorl %edx, %edx vmovaps 0xb9fed8(%rip), %ymm3 # 0x1f21e40 vextractf128 $0x1, %ymm1, %xmm4 vmovddup 0xb9ff12(%rip), %xmm2 # xmm2 = mem[0,0] vxorps %xmm2, %xmm4, %xmm4 vpcmpeqd %xmm5, %xmm5, %xmm5 vxorps %xmm2, %xmm1, %xmm6 movq 0x8(%rsp), %r15 vmovq %rdx, %xmm7 vpshufd $0x44, %xmm7, %xmm7 # xmm7 = xmm7[0,1,0,1] vinsertf128 $0x1, %xmm7, %ymm7, %ymm7 vorps %ymm3, %ymm7, %ymm7 vextractf128 $0x1, %ymm7, %xmm8 vxorps %xmm2, %xmm8, %xmm8 vpcmpgtq %xmm4, %xmm8, %xmm8 vpxor %xmm5, %xmm8, %xmm9 vxorps %xmm2, %xmm7, %xmm10 vpcmpgtq %xmm6, %xmm10, %xmm10 vpxor %xmm5, %xmm10, %xmm10 vinsertf128 $0x1, %xmm9, %ymm10, %ymm9 vmaskmovpd (%rcx,%rdx,8), %ymm9, %ymm11 vmovdqa %ymm0, %ymm9 vextractf128 $0x1, %ymm11, %xmm0 vextractf128 $0x1, %ymm9, %xmm10 vpaddq %xmm0, %xmm10, %xmm0 vpaddq %xmm9, %xmm11, %xmm11 vinsertf128 $0x1, %xmm0, %ymm11, %ymm0 addq $0x4, %rdx cmpq %rdx, %rax jne 0x1381f87 vxorps %xmm2, %xmm1, %xmm1 vxorps %xmm2, %xmm7, %xmm2 vpcmpgtq %xmm1, %xmm2, %xmm1 vextractf128 $0x1, %ymm0, %xmm2 vblendvpd %xmm8, %xmm10, %xmm2, %xmm2 vblendvpd %xmm1, %xmm9, %xmm0, %xmm0 vpaddq %xmm2, %xmm0, %xmm0 vpshufd $0xee, %xmm0, %xmm1 # xmm1 = xmm0[2,3,2,3] vpaddq %xmm1, %xmm0, %xmm0 vmovq %xmm0, %r13 jmp 0x1382027 movq 0x8(%rsp), %r15 movq 0x8(%r14), %rax xorl %ecx, %ecx movq %rcx, 0x10(%rsp) movq %rcx, 0x20(%rsp) movq %rcx, 0x18(%rsp) movq 0x30(%r14), %rbx testq %rbx, %rbx je 0x13820eb leaq 0x4c0(%r14), %rcx movq %r14, %rdx addq $0x8c0, %rdx # imm = 0x8C0 xorl %esi, %esi movq 0x40(%r14,%rsi,8), %rdi movq 0x280(%r14,%rsi,8), %r8 addq %rdi, %r8 testq %r8, %r8 movl $0x0, %r9d cmovgq %r8, %r9 movq 0x48(%r14,%rsi,8), %r10 cmpq %r10, %r13 cmovlq %r13, %r10 cmpq %rdi, %r13 cmovgq %r13, %rdi cmpq %r8, %rax cmovlq %rax, %r8 movq %r10, %r11 subq %r9, %r11 jle 0x13820bb addq %r11, 0x18(%rsp) movq 0x10(%rsp), %r11 leaq 0x1(%r11), %rbx movq %rbx, 0x10(%rsp) shlq $0x4, %r11 movq %r9, (%rcx,%r11) movq %r10, 0x8(%rcx,%r11) cmpq %rdi, %r8 jle 0x13820db movq 0x20(%rsp), %r9 leaq 0x1(%r9), %r10 movq %r10, 0x20(%rsp) shlq $0x4, %r9 movq %rdi, (%rdx,%r9) movq %r8, 0x8(%rdx,%r9) incq %rsi movq 0x30(%r14), %rbx cmpq %rbx, %rsi jb 0x138205c cmpq $0x0, 0x18(%rsp) je 0x1382185 leaq 0x18(%rsp), %rax movq %rax, 0x40(%rsp) movq %r14, 0x48(%rsp) leaq 0x10(%rsp), %rax movq %rax, 0x50(%rsp) leaq 0x20(%rsp), %rax movq %rax, 0x58(%rsp) leaq 0x60(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vpxor %xmm0, %xmm0, %xmm0 vmovdqu %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x30(%rsp), %rcx leaq 0x40(%rsp), %rax movq %rax, (%rcx) leaq 0x7(%rsp), %r8 leaq 0x60(%rsp), %r9 movl $0x1, %edx xorl %edi, %edi movq %rbx, %rsi callq 0x1382f38 leaq 0x60(%rsp), %rdi callq 0x6a770 testb %al, %al movq 0x8(%rsp), %r15 jne 0x13823be leaq 0x60(%rsp), %rdi callq 0x6aab0 addq %r15, %r13 leaq 0x28(%rsp), %rdi vzeroupper callq 0x138245e movq %r13, %rax leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movq %rdx, %r14 shlq $0x6, %rsi addq %rdi, %rsi shlq $0x6, %r14 addq %rdi, %r14 movq %rsi, %r13 addq $-0x40, %r14 cmpq %r14, %r13 ja 0x13822a7 movq (%r9), %rax movq 0x8(%r9), %rcx movq 0x10(%r9), %rdx vmovaps (%r13), %xmm2 vmovaps 0x10(%r13), %xmm1 vaddps %xmm1, %xmm2, %xmm0 vsubps 0x20(%rax), %xmm0, %xmm3 vmulps 0x30(%rax), %xmm3, %xmm3 vroundps $0x1, %xmm3, %xmm3 vcvtps2dq %xmm3, %xmm3 vmovdqa (%rcx), %xmm4 vpcmpgtd %xmm3, %xmm4, %xmm3 vtestps (%rdx), %xmm3 je 0x13822a7 vmovaps (%r12), %xmm3 vmovaps 0x10(%r12), %xmm4 vmovaps 0x20(%r12), %xmm5 vmovaps 0x30(%r12), %xmm6 vminps %xmm2, %xmm3, %xmm2 vmovaps %xmm2, (%r12) vmaxps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0x10(%r12) vminps %xmm0, %xmm5, %xmm1 vmovaps %xmm1, 0x20(%r12) vmaxps %xmm0, %xmm6, %xmm0 vmovaps %xmm0, 0x30(%r12) incq 0x48(%r12) addq $0x40, %r13 jmp 0x13821c2 vmovaps (%r15), %xmm3 vmovaps 0x10(%r15), %xmm4 vmovaps 0x20(%r15), %xmm5 vmovaps 0x30(%r15), %xmm6 vminps %xmm2, %xmm3, %xmm2 vmovaps %xmm2, (%r15) vmaxps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0x10(%r15) vminps %xmm0, %xmm5, %xmm1 vmovaps %xmm1, 0x20(%r15) vmaxps %xmm0, %xmm6, %xmm0 vmovaps %xmm0, 0x30(%r15) incq 0x48(%r15) addq $-0x40, %r14 cmpq %r14, %r13 seta %al ja 0x1382380 movq (%r9), %rcx movq 0x8(%r9), %rdx movq 0x10(%r9), %rsi vmovaps (%r14), %xmm2 vmovaps 0x10(%r14), %xmm1 vaddps %xmm1, %xmm2, %xmm0 vsubps 0x20(%rcx), %xmm0, %xmm3 vmulps 0x30(%rcx), %xmm3, %xmm3 vroundps $0x1, %xmm3, %xmm3 vcvtps2dq %xmm3, %xmm3 vmovdqa (%rdx), %xmm4 vpcmpgtd %xmm3, %xmm4, %xmm3 vtestps (%rsi), %xmm3 je 0x1382261 testb $0x1, %al jne 0x1382380 movq %rdi, 0x8(%rsp) movq %r8, %rdi movq %r12, %rsi movq %r14, %rdx movq %r8, %rbx vzeroupper callq 0x1382480 movq %rbx, %rdi movq %r15, %rsi movq %r13, %rdx callq 0x1382480 movq 0x10(%rbp), %r9 movq %rbx, %r8 movq 0x8(%rsp), %rdi vmovaps (%r13), %ymm0 movl 0x28(%r13), %eax movl %eax, 0x68(%rsp) movq 0x20(%r13), %rax movq %rax, 0x60(%rsp) vmovaps (%r14), %ymm1 vmovaps %ymm1, (%r13) movl 0x28(%r14), %eax movl %eax, 0x28(%r13) movq 0x20(%r14), %rax movq %rax, 0x20(%r13) vmovaps %ymm0, (%r14) movq 0x60(%rsp), %rax movq %rax, 0x20(%r14) movl 0x68(%rsp), %eax movl %eax, 0x28(%r14) addq $0x40, %r13 jmp 0x13821be subq %rdi, %r13 sarq $0x6, %r13 jmp 0x1382195 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb69750(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xda2859(%rip), %rsi # 0x2124c08 movq 0xda2612(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb6971e(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xda2827(%rip), %rsi # 0x2124c08 movq 0xda25e0(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1382411 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1382435 jmp 0x1382441 jmp 0x138240e movq %rax, %rbx leaq 0x60(%rsp), %rdi callq 0x6aab0 jmp 0x138244c jmp 0x1382441 jmp 0x1382441 jmp 0x1382432 movq %rax, %rbx movq %r14, %rdi callq 0x1ee612d jmp 0x1382456 jmp 0x1382441 movq %rax, %rbx leaq 0x60(%rsp), %rdi callq 0x6aab0 jmp 0x138244c movq %rax, %rdi callq 0x8d6de8 movq %rax, %rbx leaq 0x28(%rsp), %rdi callq 0x138245e movq %rbx, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/../builders/../../common/algorithms/parallel_partition.h
void std::__unguarded_linear_insert<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>*, __gnu_cxx::__ops::_Val_comp_iter<std::greater<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>>>>(embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>*, __gnu_cxx::__ops::_Val_comp_iter<std::greater<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>>>)
__forceinline BuildRecordT () {}
movb 0x8(%rdi), %al movb %al, -0x10(%rsp) movq (%rdi), %rax movq %rax, -0x18(%rsp) vmovaps 0x10(%rdi), %xmm0 vmovaps 0x20(%rdi), %xmm1 vmovaps 0x30(%rdi), %xmm2 vmovaps 0x40(%rdi), %xmm3 movq 0x50(%rdi), %rax vmovups 0x58(%rdi), %xmm4 movq 0x58(%rdi), %rdx subq %rax, %rdx movq -0x18(%rdi), %rcx subq -0x20(%rdi), %rcx cmpq %rcx, %rdx jbe 0x13844a1 leaq -0x70(%rdi), %rcx movb -0x68(%rdi), %sil movb %sil, 0x8(%rdi) movq -0x88(%rdi), %rsi movq -0x70(%rdi), %r8 movq %r8, (%rdi) vmovups -0x60(%rdi), %ymm5 vmovups %ymm5, 0x10(%rdi) vmovups -0x40(%rdi), %ymm5 vmovups %ymm5, 0x30(%rdi) vmovaps -0x20(%rdi), %xmm5 vmovaps %xmm5, 0x50(%rdi) movq -0x10(%rdi), %r8 movq %r8, 0x60(%rdi) subq -0x90(%rdi), %rsi movq %rcx, %rdi cmpq %rsi, %rdx ja 0x1384450 jmp 0x13844a4 movq %rdi, %rcx movb -0x10(%rsp), %dl movb %dl, 0x8(%rcx) movq -0x18(%rsp), %rdx movq %rdx, (%rcx) vmovaps %xmm0, 0x10(%rcx) vmovaps %xmm1, 0x20(%rcx) vmovaps %xmm2, 0x30(%rcx) vmovaps %xmm3, 0x40(%rcx) movq %rax, 0x50(%rcx) vmovups %xmm4, 0x58(%rcx) vzeroupper retq
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::build()
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::build() { /* delete some objects */ size_t num = scene->size(); if (num < bvh->objects.size()) { parallel_for(num, bvh->objects.size(), [&] (const range<size_t>& r) { for (size_t i=r.begin(); i<r.end(); i++) { builders[i].reset(); delete bvh->objects[i]; bvh->objects[i] = nullptr; } }); } #if PROFILE while(1) #endif { /* reset memory allocator */ bvh->alloc.reset(); /* skip build for empty scene */ const size_t numPrimitives = scene->getNumPrimitives(gtype,false); if (numPrimitives == 0) { prims.resize(0); bvh->set(BVH::emptyNode,empty,0); return; } /* calculate the size of the entire BVH */ const size_t numLeafBlocks = Primitive::blocks(numPrimitives); const size_t node_bytes = 2*numLeafBlocks*sizeof(typename BVH::AABBNode)/N; const size_t leaf_bytes = size_t(1.2*numLeafBlocks*sizeof(Primitive)); bvh->alloc.init_estimate(node_bytes+leaf_bytes); double t0 = bvh->preBuild(TOSTRING(isa) "::BVH" + toString(N) + "BuilderTwoLevel"); /* resize object array if scene got larger */ if (bvh->objects.size() < num) bvh->objects.resize(num); if (builders.size() < num) builders.resize(num); resizeRefsList (); nextRef.store(0); /* create acceleration structures */ parallel_for(size_t(0), num, [&] (const range<size_t>& r) { for (size_t objectID=r.begin(); objectID<r.end(); objectID++) { Mesh* mesh = scene->getSafe<Mesh>(objectID); /* ignore meshes we do not support */ if (mesh == nullptr || mesh->numTimeSteps != 1) continue; if (isSmallGeometry(mesh)) { setupSmallBuildRefBuilder (objectID, mesh); } else { setupLargeBuildRefBuilder (objectID, mesh); } } }); /* parallel build of acceleration structures */ parallel_for(size_t(0), num, [&] (const range<size_t>& r) { for (size_t objectID=r.begin(); objectID<r.end(); objectID++) { /* ignore if no triangle mesh or not enabled */ Mesh* mesh = scene->getSafe<Mesh>(objectID); if (mesh == nullptr || !mesh->isEnabled() || mesh->numTimeSteps != 1) continue; builders[objectID]->attachBuildRefs (this); } }); #if PROFILE double d0 = getSeconds(); #endif /* fast path for single geometry scenes */ if (nextRef == 1) { bvh->set(refs[0].node,LBBox3fa(refs[0].bounds()),numPrimitives); } else { /* open all large nodes */ refs.resize(nextRef); /* this probably needs some more tuning */ const size_t extSize = max(max((size_t)SPLIT_MIN_EXT_SPACE,refs.size()*SPLIT_MEMORY_RESERVE_SCALE),size_t((float)numPrimitives / SPLIT_MEMORY_RESERVE_FACTOR)); #if !ENABLE_DIRECT_SAH_MERGE_BUILDER #if ENABLE_OPEN_SEQUENTIAL open_sequential(extSize); #endif /* compute PrimRefs */ prims.resize(refs.size()); #endif { #if ENABLE_DIRECT_SAH_MERGE_BUILDER const PrimInfo pinfo = parallel_reduce(size_t(0), refs.size(), PrimInfo(empty), [&] (const range<size_t>& r) -> PrimInfo { PrimInfo pinfo(empty); for (size_t i=r.begin(); i<r.end(); i++) { pinfo.add_center2(refs[i]); } return pinfo; }, [] (const PrimInfo& a, const PrimInfo& b) { return PrimInfo::merge(a,b); }); #else const PrimInfo pinfo = parallel_reduce(size_t(0), refs.size(), PrimInfo(empty), [&] (const range<size_t>& r) -> PrimInfo { PrimInfo pinfo(empty); for (size_t i=r.begin(); i<r.end(); i++) { pinfo.add_center2(refs[i]); prims[i] = PrimRef(refs[i].bounds(),(size_t)refs[i].node); } return pinfo; }, [] (const PrimInfo& a, const PrimInfo& b) { return PrimInfo::merge(a,b); }); #endif /* skip if all objects where empty */ if (pinfo.size() == 0) bvh->set(BVH::emptyNode,empty,0); /* otherwise build toplevel hierarchy */ else { /* settings for BVH build */ GeneralBVHBuilder::Settings settings; settings.branchingFactor = N; settings.maxDepth = BVH::maxBuildDepthLeaf; settings.logBlockSize = bsr(N); settings.minLeafSize = 1; settings.maxLeafSize = 1; settings.travCost = 1.0f; settings.intCost = 1.0f; settings.singleThreadThreshold = singleThreadThreshold; #if ENABLE_DIRECT_SAH_MERGE_BUILDER refs.resize(extSize); NodeRef root = BVHBuilderBinnedOpenMergeSAH::build<NodeRef,BuildRef>( typename BVH::CreateAlloc(bvh), typename BVH::AABBNode::Create2(), typename BVH::AABBNode::Set2(), [&] (const BuildRef* refs, const range<size_t>& range, const FastAllocator::CachedAllocator& alloc) -> NodeRef { assert(range.size() == 1); return (NodeRef) refs[range.begin()].node; }, [&] (BuildRef &bref, BuildRef *refs) -> size_t { return openBuildRef(bref,refs); }, [&] (size_t dn) { bvh->scene->progressMonitor(0); }, refs.data(),extSize,pinfo,settings); #else NodeRef root = BVHBuilderBinnedSAH::build<NodeRef>( typename BVH::CreateAlloc(bvh), typename BVH::AABBNode::Create2(), typename BVH::AABBNode::Set2(), [&] (const PrimRef* prims, const range<size_t>& range, const FastAllocator::CachedAllocator& alloc) -> NodeRef { assert(range.size() == 1); return (NodeRef) prims[range.begin()].ID(); }, [&] (size_t dn) { bvh->scene->progressMonitor(0); }, prims.data(),pinfo,settings); #endif bvh->set(root,LBBox3fa(pinfo.geomBounds),numPrimitives); } } } bvh->alloc.cleanup(); bvh->postBuild(t0); #if PROFILE double d1 = getSeconds(); std::cout << "TOP_LEVEL OPENING/REBUILD TIME " << 1000.0*(d1-d0) << " ms" << std::endl; #endif } } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::deleteGeometry(size_t geomID) { if (geomID >= bvh->objects.size()) return; if (builders[geomID]) builders[geomID].reset(); delete bvh->objects [geomID]; bvh->objects [geomID] = nullptr; } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::clear() { for (size_t i=0; i<bvh->objects.size(); i++) if (bvh->objects[i]) bvh->objects[i]->clear(); for (size_t i=0; i<builders.size(); i++) if (builders[i]) builders[i].reset(); refs.clear(); } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::open_sequential(const size_t extSize) { if (refs.size() == 0) return; refs.reserve(extSize); #if 1 for (size_t i=0;i<refs.size();i++) { NodeRef ref = refs[i].node; if (ref.isAABBNode()) BVH::prefetch(ref); } #endif std::make_heap(refs.begin(),refs.end()); while (refs.size()+N-1 <= extSize) { std::pop_heap (refs.begin(),refs.end()); NodeRef ref = refs.back().node; if (ref.isLeaf()) break; refs.pop_back(); AABBNode* node = ref.getAABBNode(); for (size_t i=0; i<N; i++) { if (node->child(i) == BVH::emptyNode) continue; refs.push_back(BuildRef(node->bounds(i),node->child(i))); #if 1 NodeRef ref_pre = node->child(i); if (ref_pre.isAABBNode()) ref_pre.prefetch(); #endif std::push_heap (refs.begin(),refs.end()); } } } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupSmallBuildRefBuilder (size_t objectID, Mesh const * const /*mesh*/) { if (builders[objectID] == nullptr || // new mesh dynamic_cast<RefBuilderSmall*>(builders[objectID].get()) == nullptr) // size change resulted in large->small change { builders[objectID].reset (new RefBuilderSmall(objectID)); } } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupLargeBuildRefBuilder (size_t objectID, Mesh const * const mesh) { if (bvh->objects[objectID] == nullptr || // new mesh builders[objectID]->meshQualityChanged (mesh->quality) || // changed build quality dynamic_cast<RefBuilderLarge*>(builders[objectID].get()) == nullptr) // size change resulted in small->large change { Builder* builder = nullptr; delete bvh->objects[objectID]; createMeshAccel(objectID, builder); builders[objectID].reset (new RefBuilderLarge(objectID, builder, mesh->quality)); } } #if defined(EMBREE_GEOMETRY_TRIANGLE) Builder* BVH4BuilderTwoLevelTriangle4MeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,TriangleMesh,Triangle4>((BVH4*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH4BuilderTwoLevelTriangle4vMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,TriangleMesh,Triangle4v>((BVH4*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH4BuilderTwoLevelTriangle4iMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,TriangleMesh,Triangle4i>((BVH4*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_QUAD) Builder* BVH4BuilderTwoLevelQuadMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,QuadMesh,Quad4v>((BVH4*)bvh,scene,QuadMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_USER) Builder* BVH4BuilderTwoLevelVirtualSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,UserGeometry,Object>((BVH4*)bvh,scene,UserGeometry::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE) Builder* BVH4BuilderTwoLevelInstanceSAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,Instance,InstancePrimitive>((BVH4*)bvh,scene,gtype,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE_ARRAY) Builder* BVH4BuilderTwoLevelInstanceArraySAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,InstanceArray,InstanceArrayPrimitive>((BVH4*)bvh,scene,gtype,useMortonBuilder); } #endif #if defined(__AVX__) #if defined(EMBREE_GEOMETRY_TRIANGLE) Builder* BVH8BuilderTwoLevelTriangle4MeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,TriangleMesh,Triangle4>((BVH8*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH8BuilderTwoLevelTriangle4vMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,TriangleMesh,Triangle4v>((BVH8*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH8BuilderTwoLevelTriangle4iMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,TriangleMesh,Triangle4i>((BVH8*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_QUAD) Builder* BVH8BuilderTwoLevelQuadMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,QuadMesh,Quad4v>((BVH8*)bvh,scene,QuadMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_USER) Builder* BVH8BuilderTwoLevelVirtualSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,UserGeometry,Object>((BVH8*)bvh,scene,UserGeometry::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE) Builder* BVH8BuilderTwoLevelInstanceSAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,Instance,InstancePrimitive>((BVH8*)bvh,scene,gtype,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE_ARRAY) Builder* BVH8BuilderTwoLevelInstanceArraySAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,InstanceArray,InstanceArrayPrimitive>((BVH8*)bvh,scene,gtype,useMortonBuilder); } #endif #endif }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x208, %rsp # imm = 0x208 movq %rdi, %rbx movq 0x28(%rdi), %rax movq 0x30(%rdi), %rcx movq 0x1d8(%rcx), %r15 movq 0x208(%rax), %r12 subq 0x200(%rax), %r12 sarq $0x3, %r12 cmpq %r12, %r15 jae 0x1384dea leaq 0x110(%rsp), %r13 movq %rbx, (%r13) leaq 0x10(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x170(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0xa0(%rsp), %rsi movq %r13, (%rsi) leaq 0xf8(%rsp), %rdx movq %r14, %rcx callq 0x1385fd0 leaq 0x10(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1385977 leaq 0x10(%rsp), %rdi callq 0x6aab0 movq 0x28(%rbx), %rdi addq $0x78, %rdi callq 0x90ae4c movq 0x30(%rbx), %rax movq 0x98(%rbx), %rcx btl $0x14, %ecx jb 0x1384e0d xorl %r14d, %r14d jmp 0x1384e14 movq 0x290(%rax), %r14 btl $0x15, %ecx jae 0x1384e21 addq 0x2a0(%rax), %r14 testb $0xf, %cl je 0x1384e2d addq 0x2c0(%rax), %r14 testl $0x77770, %ecx # imm = 0x77770 je 0x1384e3c addq 0x2b0(%rax), %r14 testl $0xe000000, %ecx # imm = 0xE000000 je 0x1384e4b addq 0x340(%rax), %r14 btl $0x17, %ecx jae 0x1384e58 addq 0x2d0(%rax), %r14 btl $0x1d, %ecx jae 0x1384e65 addq 0x2e0(%rax), %r14 btl $0x1e, %ecx jae 0x1384e72 addq 0x2f0(%rax), %r14 testl %ecx, %ecx jns 0x1384e7d addq 0x300(%rax), %r14 btl $0x18, %ecx jae 0x1384e8a addq 0x310(%rax), %r14 btl $0x16, %ecx jae 0x1384e97 addq 0x320(%rax), %r14 testq %r14, %r14 je 0x1384f7a leaq 0x3(%r14), %rax movq %rax, %rcx shrq $0x2, %rcx shlq $0x4, %rax movabsq $0x3fffffffffffffc0, %rdx # imm = 0x3FFFFFFFFFFFFFC0 vcvtsi2sd %rcx, %xmm1, %xmm0 vmulsd 0xb7f292(%rip), %xmm0, %xmm0 # 0x1f04158 vmulsd 0xb7f292(%rip), %xmm0, %xmm0 # 0x1f04160 vcvttsd2si %xmm0, %rcx andq %rax, %rdx movq %rcx, %rax vsubsd 0xb6b0ff(%rip), %xmm0, %xmm0 # 0x1eeffe0 sarq $0x3f, %rax vcvttsd2si %xmm0, %rsi andq %rax, %rsi orq %rcx, %rsi addq %rdx, %rsi movq 0x28(%rbx), %rdi addq $0x78, %rdi callq 0x90acb8 movq 0x28(%rbx), %r12 leaq 0xb0(%rsp), %rax movq %rax, -0x10(%rax) leaq 0xa0(%rsp), %r13 movl $0x1, %esi movq %r13, %rdi movl $0x2d, %edx callq 0x6a580 movq (%r13), %rax movb $0x34, (%rax) leaq 0xb9cfb5(%rip), %rcx # 0x1f21eed movl $0x8, %r8d movq %r13, %rdi xorl %esi, %esi xorl %edx, %edx callq 0x6a1a0 leaq 0x180(%rsp), %r13 movq %r13, -0x10(%r13) movq (%rax), %rdx movq %rax, %rcx addq $0x10, %rcx cmpq %rcx, %rdx je 0x1384fa4 movq %rdx, 0x170(%rsp) movq (%rcx), %rdx movq %rdx, 0x180(%rsp) jmp 0x1384fae movq 0x78(%rbx), %r12 cmpq $0x0, 0x70(%rbx) je 0x1384f8d movq $0x0, 0x70(%rbx) cmpq %r12, 0x78(%rbx) jne 0x138525f movq $0x0, 0x70(%rbx) jmp 0x13853f5 vmovupd (%rcx), %xmm0 vmovupd %xmm0, (%r13) movq 0x8(%rax), %rdx leaq 0x170(%rsp), %rdi movq %rdx, 0x8(%rdi) movq %rcx, (%rax) movq $0x0, 0x8(%rax) movb $0x0, 0x10(%rax) leaq 0xb8d7bc(%rip), %rsi # 0x1f12790 callq 0x6a620 leaq 0x20(%rsp), %rbp movq %rbp, -0x10(%rbp) movq (%rax), %rdx movq %rax, %rcx addq $0x10, %rcx cmpq %rcx, %rdx je 0x1385000 movq %rdx, 0x10(%rsp) movq (%rcx), %rdx movq %rdx, 0x20(%rsp) jmp 0x1385009 vmovupd (%rcx), %xmm0 vmovupd %xmm0, (%rbp) movq 0x8(%rax), %rdx leaq 0x10(%rsp), %rsi movq %rdx, 0x8(%rsi) movq %rcx, (%rax) movq $0x0, 0x8(%rax) movb $0x0, 0x10(%rax) movq %r12, %rdi callq 0xb5d532 vmovsd %xmm0, 0x98(%rsp) movq 0x10(%rsp), %rdi cmpq %rbp, %rdi je 0x1385045 callq 0x6a4f0 movq 0x170(%rsp), %rdi cmpq %r13, %rdi je 0x1385057 callq 0x6a4f0 movq 0xa0(%rsp), %rdi leaq 0xb0(%rsp), %rax cmpq %rax, %rdi je 0x1385071 callq 0x6a4f0 movq 0x28(%rbx), %rdi movq 0x208(%rdi), %rax subq 0x200(%rdi), %rax sarq $0x3, %rax cmpq %r15, %rax jae 0x138509b addq $0x200, %rdi # imm = 0x200 movq %r15, %rsi callq 0xc882c0 movq 0x18(%rbx), %rax subq 0x10(%rbx), %rax sarq $0x3, %rax cmpq %r15, %rax jae 0x13850b8 leaq 0x10(%rbx), %rdi movq %r15, %rsi callq 0x1385c36 movq %rbx, %rdi callq 0x1385cae xorl %eax, %eax xchgl %eax, 0x88(%rbx) leaq 0x110(%rsp), %r13 movq %rbx, (%r13) leaq 0x10(%rsp), %r12 movw $0x401, 0xc(%r12) # imm = 0x401 vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x20(%r12) movq $0x8, 0x40(%r12) movq %r12, %rdi callq 0x6a660 leaq 0x170(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0xa0(%rsp), %rsi movq %r13, (%rsi) leaq 0xf8(%rsp), %rdx movq %r12, %rcx callq 0x1387032 leaq 0x10(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1385913 leaq 0x10(%rsp), %rdi callq 0x6aab0 movq %rbx, 0x110(%rsp) leaq 0x10(%rsp), %r12 movw $0x401, 0xc(%r12) # imm = 0x401 vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x20(%r12) movq $0x8, 0x40(%r12) movq %r12, %rdi callq 0x6a660 leaq 0x170(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0xa0(%rsp), %rsi movq %r13, (%rsi) leaq 0xf8(%rsp), %rdx movq %r12, %rcx callq 0x1388586 leaq 0x10(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1385945 leaq 0x10(%rsp), %rdi callq 0x6aab0 movl 0x88(%rbx), %eax cmpl $0x1, %eax jne 0x138520f movq 0x28(%rbx), %rdi movq 0x58(%rbx), %rax movq 0x20(%rax), %rsi vmovaps (%rax), %xmm0 vmovaps 0x10(%rax), %xmm1 leaq 0x10(%rsp), %rdx vmovaps %xmm0, (%rdx) vmovaps %xmm1, 0x10(%rdx) vmovaps %xmm0, 0x20(%rdx) vmovaps %xmm1, 0x30(%rdx) jmp 0x138589b movl 0x88(%rbx), %eax movslq %eax, %r13 movq 0x50(%rbx), %rax cmpq %r13, %rax jae 0x1385243 movq %r13, %rbp testq %rax, %rax je 0x1385246 cmpq %r13, %rax jae 0x1385243 movq %rax, %rbp addq %rbp, %rbp cmpq $0x1, %rbp adcq $0x0, %rbp cmpq %r13, %rbp jb 0x1385231 jmp 0x1385246 movq %rax, %rbp cmpq %r13, 0x48(%rbx) jbe 0x1385250 movq %r13, 0x48(%rbx) cmpq %rbp, 0x50(%rbx) jne 0x1385298 movq %r13, 0x48(%rbx) jmp 0x138546a movq 0x80(%rbx), %r14 testq %r12, %r12 je 0x138527d movq 0x60(%rbx), %rdi movq %r12, %rsi shlq $0x5, %rsi movq (%rdi), %rax xorl %edx, %edx callq *(%rax) movq %r12, %rdi shlq $0x5, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x13852ce leaq 0x68(%rbx), %rsi callq 0x1ee665d jmp 0x13852d8 movq 0x58(%rbx), %r15 testq %rbp, %rbp je 0x13852b3 movq 0x38(%rbx), %rdi movq %rbp, %rsi shlq $0x6, %rsi movq (%rdi), %rax xorl %edx, %edx callq *(%rax) movq %rbp, %rdi shlq $0x6, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x138533e leaq 0x40(%rbx), %rsi callq 0x1ee665d jmp 0x1385348 movl $0x20, %esi callq 0x1ee60ac movq %rax, 0x80(%rbx) cmpq $0x0, 0x70(%rbx) je 0x1385309 xorl %eax, %eax xorl %ecx, %ecx movq 0x80(%rbx), %rdx vmovapd (%r14,%rax), %ymm0 vmovapd %ymm0, (%rdx,%rax) incq %rcx addq $0x20, %rax cmpq 0x70(%rbx), %rcx jb 0x13852ea movq 0x78(%rbx), %r15 testq %r14, %r14 je 0x13853c9 movq %r15, %rsi shlq $0x5, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13853be movzbl 0x68(%rbx), %edx movq %r14, %rdi vzeroupper callq 0x1ee67ca jmp 0x13853c9 movl $0x20, %esi callq 0x1ee60ac movq %rax, 0x58(%rbx) cmpq $0x0, 0x48(%rbx) je 0x1385389 movl $0x20, %eax xorl %ecx, %ecx movq 0x58(%rbx), %rdx vmovapd -0x20(%r15,%rax), %ymm0 vmovapd %ymm0, -0x20(%rdx,%rax) movl 0x8(%r15,%rax), %esi movl %esi, 0x8(%rdx,%rax) movq (%r15,%rax), %rsi movq %rsi, (%rdx,%rax) incq %rcx addq $0x40, %rax cmpq 0x48(%rbx), %rcx jb 0x138535a movq 0x50(%rbx), %r12 testq %r15, %r15 je 0x1385442 movq %r12, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1385437 movzbl 0x40(%rbx), %edx movq %r15, %rdi vzeroupper callq 0x1ee67ca jmp 0x1385442 movq %r14, %rdi vzeroupper callq 0x1ee612d testq %r15, %r15 je 0x13853e9 movq 0x60(%rbx), %rdi shlq $0x5, %r15 negq %r15 movq (%rdi), %rax movq %r15, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq $0x0, 0x70(%rbx) movq %r12, 0x78(%rbx) movq 0x28(%rbx), %rdi vbroadcastss 0xb6661e(%rip), %xmm0 # 0x1eeba20 leaq 0x10(%rsp), %rdx vmovaps %xmm0, (%rdx) vbroadcastss 0xb67770(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%rdx) vmovaps %xmm0, 0x20(%rdx) vmovaps %xmm1, 0x30(%rdx) movl $0x8, %esi xorl %ecx, %ecx vzeroupper callq 0xb5ccb2 jmp 0x1385901 movq %r15, %rdi vzeroupper callq 0x1ee612d testq %r12, %r12 je 0x1385462 movq 0x38(%rbx), %rdi shlq $0x6, %r12 negq %r12 movq (%rdi), %rax movq %r12, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq %r13, 0x48(%rbx) movq %rbp, 0x50(%rbx) movq 0x48(%rbx), %r15 leaq (%r15,%r15), %rax testq %r14, %r14 js 0x138547e vcvtsi2ss %r14, %xmm1, %xmm0 jmp 0x1385496 movq %r14, %rcx shrq %rcx movl %r14d, %edx andl $0x1, %edx orq %rcx, %rdx vcvtsi2ss %rdx, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 vdivss 0xb8d262(%rip), %xmm0, %xmm0 # 0x1f12700 vcvttss2si %xmm0, %rcx movq %rcx, %rdx sarq $0x3f, %rdx vsubss 0xb66992(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rcx, %rsi cmpq %rsi, %rax cmovaq %rax, %rsi cmpq $0x3e9, %rsi # imm = 0x3E9 movl $0x3e8, %eax # imm = 0x3E8 cmovaeq %rsi, %rax movq %rax, 0x90(%rsp) vbroadcastss 0xb6653b(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x110(%rsp) vbroadcastss 0xb6768d(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x120(%rsp) vmovaps %xmm0, 0x130(%rsp) vmovaps %xmm1, 0x140(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x150(%rsp) leaq 0x168(%rsp), %rax movq %rbx, (%rax) leaq 0x1f0(%rsp), %rbp leaq 0x110(%rsp), %rcx movq %rcx, (%rbp) leaq 0xf(%rsp), %r13 movq %r13, 0x8(%rbp) movq %rax, 0x10(%rbp) leaq 0x10(%rsp), %r12 movw $0x401, 0xc(%r12) # imm = 0x401 vmovups %xmm0, 0x20(%r12) movq $0x8, 0x40(%r12) movq %r12, %rdi vzeroupper callq 0x6a660 leaq 0xf8(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x1e0(%rsp), %rax movq %r13, (%rax) movq %rbp, 0x8(%rax) leaq 0x170(%rsp), %rsi leaq 0x110(%rsp), %rcx movq %rcx, (%rsi) movq %rax, 0x8(%rsi) movq %r13, 0x10(%rsi) vmovups 0x110(%rsp), %ymm0 vmovups 0x130(%rsp), %ymm1 vmovups %ymm0, 0x20(%rsi) vmovups %ymm1, 0x40(%rsi) vmovapd 0x150(%rsp), %xmm0 vmovapd %xmm0, 0x60(%rsi) leaq 0xa0(%rsp), %rbp movq %rbp, %rdx movq %r12, %rcx vzeroupper callq 0x1388c78 leaq 0x1d0(%rsp), %rax vmovups -0x40(%rax), %ymm0 vmovups -0x20(%rax), %ymm1 vmovups %ymm0, 0xa0(%rsp) vmovups %ymm1, 0xc0(%rsp) vmovapd (%rax), %xmm0 vmovapd %xmm0, 0xe0(%rsp) leaq 0x10(%rsp), %rdi vzeroupper callq 0x6a770 testb %al, %al movq 0x90(%rsp), %r13 jne 0x13859a9 leaq 0x10(%rsp), %rdi callq 0x6aab0 movq 0xe8(%rsp), %rax cmpq 0xe0(%rsp), %rax jne 0x138569c movq 0x28(%rbx), %rdi leaq 0x10(%rsp), %rdx vbroadcastss 0xb663ac(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, (%rdx) vbroadcastss 0xb67503(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%rdx) vmovaps %xmm0, 0x20(%rdx) vmovaps %xmm1, 0x30(%rdx) movl $0x8, %esi xorl %ecx, %ecx jmp 0x138589e vmovaps 0xb9c77c(%rip), %ymm0 # 0x1f21e20 vmovups %ymm0, 0x10(%rsp) vmovaps 0xbb338e(%rip), %ymm0 # 0x1f38a40 vmovups %ymm0, 0x30(%rsp) movq $0x4, 0x10(%rsp) movq $0x28, 0x18(%rsp) movl $0x4, %eax bsrl %eax, %eax cltq movq %rax, 0x20(%rsp) movl $0x1, %eax movq %rax, 0x28(%rsp) movq %rax, 0x30(%rsp) movq 0x90(%rbx), %rax movq %rax, 0x40(%rsp) movq 0x50(%rbx), %rax cmpq %r13, %rax jae 0x1385722 movq %r13, %r15 testq %rax, %rax je 0x1385725 cmpq %r13, %rax jae 0x1385722 addq %rax, %rax cmpq $0x1, %rax movq %rax, %r15 adcq $0x0, %r15 cmpq %r13, %rax movq %r15, %rax jb 0x138570a jmp 0x1385725 movq %rax, %r15 cmpq %r13, 0x48(%rbx) jbe 0x138572f movq %r13, 0x48(%rbx) cmpq %r15, 0x50(%rbx) jne 0x138573e movq %r13, 0x48(%rbx) jmp 0x138582c movq 0x58(%rbx), %rbp testq %r15, %r15 je 0x138575c movq 0x38(%rbx), %rdi movq %r15, %rsi shlq $0x6, %rsi movq (%rdi), %rax xorl %edx, %edx vzeroupper callq *(%rax) movq %r15, %rdi shlq $0x6, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x138577a leaq 0x40(%rbx), %rsi vzeroupper callq 0x1ee665d jmp 0x1385787 movl $0x20, %esi vzeroupper callq 0x1ee60ac movq %rax, 0x58(%rbx) cmpq $0x0, 0x48(%rbx) je 0x13857c7 movl $0x20, %eax xorl %ecx, %ecx movq 0x58(%rbx), %rdx vmovaps -0x20(%rbp,%rax), %ymm0 vmovaps %ymm0, -0x20(%rdx,%rax) movl 0x8(%rbp,%rax), %esi movl %esi, 0x8(%rdx,%rax) movq (%rbp,%rax), %rsi movq %rsi, (%rdx,%rax) incq %rcx addq $0x40, %rax cmpq 0x48(%rbx), %rcx jb 0x1385799 movq 0x50(%rbx), %r12 testq %rbp, %rbp je 0x13857fc movq %r12, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13857f1 movzbl 0x40(%rbx), %edx movq %rbp, %rdi vzeroupper callq 0x1ee67ca jmp 0x13857fc movq %rbp, %rdi vzeroupper callq 0x1ee612d testq %r12, %r12 je 0x138581c movq 0x38(%rbx), %rdi shlq $0x6, %r12 negq %r12 movq (%rdi), %rax movq %r12, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq %r13, 0x48(%rbx) movq %r15, 0x50(%rbx) leaq 0xa0(%rsp), %rbp movq 0x28(%rbx), %rdi addq $0x78, %rdi movq 0x58(%rbx), %r8 leaq 0x170(%rsp), %r12 movq %r12, %rsi movq %rbx, %rdx movq %rbx, %rcx movq %r13, %r9 leaq 0x10(%rsp), %rax pushq %rax pushq %rbp vzeroupper callq 0x1385f2a addq $0x10, %rsp movq 0x28(%rbx), %rdi vmovaps 0xa0(%rsp), %xmm0 vmovaps 0xb0(%rsp), %xmm1 vmovaps %xmm0, (%r12) vmovaps %xmm1, 0x10(%r12) vmovaps %xmm0, 0x20(%r12) vmovaps %xmm1, 0x30(%r12) leaq 0x170(%rsp), %rdx movq %rax, %rsi movq %r14, %rcx callq 0xb5ccb2 movq 0x28(%rbx), %r15 leaq 0x78(%r15), %r14 movq %r14, %rdi callq 0x90af76 movq 0x1a8(%r15), %r12 movq 0x1b0(%r15), %r13 cmpq %r13, %r12 je 0x13858d8 movq (%r12), %rdi movq %r14, %rsi callq 0x90afd2 addq $0x8, %r12 jmp 0x13858c1 movq 0x1a8(%r15), %rax cmpq %rax, 0x1b0(%r15) je 0x13858ef movq %rax, 0x1b0(%r15) movq 0x28(%rbx), %rdi vmovsd 0x98(%rsp), %xmm0 callq 0xb5d720 addq $0x208, %rsp # imm = 0x208 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb661c9(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd9f2d2(%rip), %rsi # 0x2124c08 movq 0xd9f08b(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb66197(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd9f2a0(%rip), %rsi # 0x2124c08 movq 0xd9f059(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb66165(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd9f26e(%rip), %rsi # 0x2124c08 movq 0xd9f027(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb66133(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd9f23c(%rip), %rsi # 0x2124c08 movq 0xd9eff5(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1385a22 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1385a98 jmp 0x1385acc jmp 0x1385a1f movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1385aa9 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1385aba movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 jmp 0x1385ac4 jmp 0x1385acc jmp 0x1385acc jmp 0x1385a95 jmp 0x1385acc jmp 0x1385aa6 jmp 0x1385acc jmp 0x1385ab7 movq %rax, %rbx movq 0x10(%rsp), %rdi cmpq %rbp, %rdi je 0x1385a62 callq 0x6a4f0 jmp 0x1385a62 movq %rax, %rbx movq 0x170(%rsp), %rdi cmpq %r13, %rdi je 0x1385a79 callq 0x6a4f0 jmp 0x1385a79 movq %rax, %rbx movq 0xa0(%rsp), %rdi leaq 0xb0(%rsp), %rax cmpq %rax, %rdi je 0x1385ac4 callq 0x6a4f0 jmp 0x1385ac4 movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 jmp 0x1385ac4 jmp 0x1385acc movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 jmp 0x1385ac4 jmp 0x1385acc movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::resizeRefsList()
void resizeRefsList () { size_t num = parallel_reduce (size_t(0), scene->size(), size_t(0), [this](const range<size_t>& r)->size_t { size_t c = 0; for (auto i=r.begin(); i<r.end(); ++i) { Mesh* mesh = scene->getSafe<Mesh>(i); if (mesh == nullptr || mesh->numTimeSteps != 1) continue; size_t meshSize = mesh->size(); c += isSmallGeometry(mesh) ? Primitive::blocks(meshSize) : 1; } return c; }, std::plus<size_t>() ); if (refs.size() < num) { refs.resize(num); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xf8, %rsp movq %rdi, %rbx movq 0x30(%rdi), %rax movq 0x1d8(%rax), %r15 xorl %eax, %eax leaq 0x10(%rsp), %r13 movq %rax, (%r13) leaq 0x8(%rsp), %rax movq %rdi, (%rax) leaq 0x48(%rsp), %rbp movq %r13, (%rbp) leaq 0x6(%rsp), %r12 movq %r12, 0x8(%rbp) movq %rax, 0x10(%rbp) leaq 0x78(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x60(%rsp), %rdi movq %r15, (%rdi) xorl %eax, %eax movq %rax, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x18(%rsp), %rax movq %r12, (%rax) movq %rbp, 0x8(%rax) leaq 0x28(%rsp), %rsi movq %r13, (%rsi) movq %rax, 0x8(%rsi) movq %r12, 0x10(%rsi) movq (%r13), %rax movq %rax, 0x18(%rsi) leaq 0x7(%rsp), %rdx movq %r14, %rcx callq 0x1386816 movq 0x40(%rsp), %r12 leaq 0x78(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1385eca leaq 0x78(%rsp), %rdi callq 0x6aab0 movq 0x48(%rbx), %rax cmpq %r12, %rax jae 0x1385eb5 movq 0x50(%rbx), %rcx cmpq %r12, %rcx jae 0x1385dbc movq %r12, %r13 testq %rcx, %rcx je 0x1385dbf cmpq %r12, %rcx jae 0x1385dbc movq %rcx, %r13 addq %r13, %r13 cmpq $0x1, %r13 adcq $0x0, %r13 cmpq %r12, %r13 jb 0x1385daa jmp 0x1385dbf movq %rcx, %r13 cmpq %r12, %rax jbe 0x1385dc8 movq %r12, 0x48(%rbx) cmpq %r13, 0x50(%rbx) jne 0x1385dd7 movq %r12, 0x48(%rbx) jmp 0x1385eb5 movq 0x58(%rbx), %r14 testq %r13, %r13 je 0x1385df2 movq 0x38(%rbx), %rdi movq %r13, %rsi shlq $0x6, %rsi movq (%rdi), %rax xorl %edx, %edx callq *(%rax) movq %r13, %rdi shlq $0x6, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x1385e0d leaq 0x40(%rbx), %rsi callq 0x1ee665d jmp 0x1385e17 movl $0x20, %esi callq 0x1ee60ac movq %rax, 0x58(%rbx) cmpq $0x0, 0x48(%rbx) je 0x1385e58 movl $0x20, %eax xorl %ecx, %ecx movq 0x58(%rbx), %rdx vmovaps -0x20(%r14,%rax), %ymm0 vmovaps %ymm0, -0x20(%rdx,%rax) movl 0x8(%r14,%rax), %esi movl %esi, 0x8(%rdx,%rax) movq (%r14,%rax), %rsi movq %rsi, (%rdx,%rax) incq %rcx addq $0x40, %rax cmpq 0x48(%rbx), %rcx jb 0x1385e29 movq 0x50(%rbx), %r15 testq %r14, %r14 je 0x1385e8d movq %r15, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1385e82 movzbl 0x40(%rbx), %edx movq %r14, %rdi vzeroupper callq 0x1ee67ca jmp 0x1385e8d movq %r14, %rdi vzeroupper callq 0x1ee612d testq %r15, %r15 je 0x1385ead movq 0x38(%rbx), %rdi shlq $0x6, %r15 negq %r15 movq (%rdi), %rax movq %r15, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq %r12, 0x48(%rbx) movq %r13, 0x50(%rbx) addq $0xf8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb65c12(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd9ed1b(%rip), %rsi # 0x2124c08 movq 0xd9ead4(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1385f10 jmp 0x1385f22 jmp 0x1385f0d movq %rax, %rbx leaq 0x78(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.h
embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::RefBuilderLarge::attachBuildRefs(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>*)
void attachBuildRefs (BVHNBuilderTwoLevel* topBuilder) { BVH* object = topBuilder->getBVH(objectID_); assert(object); /* build object if it got modified */ if (topBuilder->isGeometryModified(objectID_)) builder_->build(); /* create build primitive */ if (!object->getBounds().empty()) { #if ENABLE_DIRECT_SAH_MERGE_BUILDER Mesh* mesh = topBuilder->getMesh(objectID_); topBuilder->refs[topBuilder->nextRef++] = BVHNBuilderTwoLevel::BuildRef(object->getBounds(),object->root,(unsigned int)objectID_,(unsigned int)mesh->size()); #else topBuilder->refs[topBuilder->nextRef++] = BVHNBuilderTwoLevel::BuildRef(object->getBounds(),object->root); #endif } }
pushq %r15 pushq %r14 pushq %rbx movq %rsi, %rbx movq %rdi, %r14 movq 0x8(%rdi), %rax movq 0x28(%rsi), %rdx movq 0x30(%rsi), %rcx movq 0x200(%rdx), %rdx movq (%rdx,%rax,8), %r15 movq 0x1e8(%rcx), %rdx movq (%rdx,%rax,8), %rdx testq %rdx, %rdx je 0x13883c3 movl 0x38(%rdx), %edx movq 0x208(%rcx), %rcx cmpl (%rcx,%rax,4), %edx jbe 0x13883c3 movq 0x10(%r14), %rdi movq (%rdi), %rax callq *0x20(%rax) vmovaps 0x10(%r15), %xmm0 vmovaps 0x20(%r15), %xmm1 vminps 0x30(%r15), %xmm0, %xmm0 vmaxps 0x40(%r15), %xmm1, %xmm1 vcmpleps %xmm1, %xmm0, %xmm2 vmovmskps %xmm2, %eax notb %al testb $0x7, %al jne 0x1388477 movq 0x8(%r14), %rcx movq 0x30(%rbx), %rax movq 0x1e8(%rax), %rax movq (%rax,%rcx,8), %rax xorl %edx, %edx cmpb $0x14, 0x3c(%rax) cmoveq %rax, %rdx movq 0x70(%r15), %rax vmovd %ecx, %xmm3 vinsertps $0x30, 0x20(%rdx), %xmm1, %xmm2 # xmm2 = xmm1[0,1,2],mem[0] vinsertps $0x30, %xmm3, %xmm0, %xmm3 # xmm3 = xmm0[0,1,2],xmm3[0] vinsertf128 $0x1, %xmm2, %ymm3, %ymm0 vxorps %xmm1, %xmm1, %xmm1 testb $0x8, %al jne 0x138844f vsubps %xmm3, %xmm2, %xmm1 vmovshdup %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] vshufpd $0x1, %xmm1, %xmm1, %xmm3 # xmm3 = xmm1[1,0] vaddss %xmm3, %xmm2, %xmm4 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm4, %xmm1, %xmm1 vaddss %xmm2, %xmm1, %xmm1 vaddss %xmm1, %xmm1, %xmm1 movl $0x1, %ecx lock xaddl %ecx, 0x88(%rbx) movslq %ecx, %rcx movq 0x58(%rbx), %rdx shlq $0x6, %rcx vmovaps %ymm0, (%rdx,%rcx) movq %rax, 0x20(%rdx,%rcx) vmovss %xmm1, 0x28(%rdx,%rcx) popq %rbx popq %r14 popq %r15 vzeroupper retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, 32ul>::split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)
__noinline void split(const Split& split, const PrimInfoExtRange& set_i, PrimInfoExtRange& lset, PrimInfoExtRange& rset) { PrimInfoExtRange set = set_i; /* valid split */ if (unlikely(!split.valid())) { deterministic_order(set); splitFallback(set,lset,rset); return; } std::pair<size_t,size_t> ext_weights(0,0); /* object split */ if (likely(set.size() < PARALLEL_THRESHOLD)) ext_weights = sequential_object_split(split,set,lset,rset); else ext_weights = parallel_object_split(split,set,lset,rset); /* if we have an extended range, set extended child ranges and move right split range */ if (unlikely(set.has_ext_range())) { setExtentedRanges(set,lset,rset,ext_weights.first,ext_weights.second); moveExtentedRange(set,lset,rset); } }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 movq %r8, %rbx movq %rcx, %r15 movq %rdi, %r14 vmovaps (%rdx), %xmm0 vmovaps %xmm0, 0x50(%rsp) vmovaps 0x10(%rdx), %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x20(%rdx), %xmm0 vmovaps %xmm0, 0x70(%rsp) vmovaps 0x30(%rdx), %xmm0 vmovaps %xmm0, 0x80(%rsp) vmovdqa 0x40(%rdx), %xmm0 vmovdqa %xmm0, 0x90(%rsp) movq 0x50(%rdx), %rax movq %rax, 0xa0(%rsp) cmpl $-0x1, 0x4(%rsi) je 0x138b1c5 movq 0x98(%rsp), %rax vmovq %xmm0, %rcx subq %rcx, %rax leaq 0x50(%rsp), %rdx movq %r14, %rdi movq %r15, %rcx movq %rbx, %r8 cmpq $0x3ff, %rax # imm = 0x3FF ja 0x138b25b callq 0x1fa83c movq 0xa0(%rsp), %rcx subq 0x98(%rsp), %rcx jne 0x138b265 addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq leaq 0x90(%rsp), %r12 movq %r14, %rdi movq %r12, %rsi callq 0x138a464 movq (%r12), %rdx movq 0x8(%r12), %rax leaq (%rax,%rdx), %rcx shrq %rcx vbroadcastss 0xb61993(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xb60826(%rip), %xmm1 # 0x1eeba20 movq %rcx, %rsi subq %rdx, %rsi jbe 0x138b278 movq (%r14), %rdi movq %rdx, %r8 shlq $0x6, %r8 addq %r8, %rdi addq $0x10, %rdi vbroadcastss 0xb60804(%rip), %xmm2 # 0x1eeba20 vbroadcastss 0xb6195f(%rip), %xmm3 # 0x1eecb84 vmovaps %xmm3, %xmm4 vmovaps %xmm2, %xmm5 vmovaps -0x10(%rdi), %xmm6 vmovaps (%rdi), %xmm7 vminps %xmm6, %xmm5, %xmm5 vmaxps %xmm7, %xmm4, %xmm4 vaddps %xmm7, %xmm6, %xmm6 vminps %xmm6, %xmm2, %xmm2 vmaxps %xmm6, %xmm3, %xmm3 addq $0x40, %rdi decq %rsi jne 0x138b22d movq %rcx, %rsi subq %rdx, %rsi jmp 0x138b28a callq 0x138ff60 jmp 0x138b19e testq %rax, %rax js 0x138b362 vcvtsi2ss %rax, %xmm1, %xmm0 jmp 0x138b379 xorl %esi, %esi vmovaps %xmm1, %xmm5 vmovaps %xmm0, %xmm4 vmovaps %xmm1, %xmm2 vmovaps %xmm0, %xmm3 movq %rax, %rdi subq %rcx, %rdi jbe 0x138b2f2 movq (%r14), %r8 movq %rcx, %r9 shlq $0x6, %r9 addq %r9, %r8 addq $0x10, %r8 vbroadcastss 0xb618d8(%rip), %xmm6 # 0x1eecb84 vbroadcastss 0xb6076b(%rip), %xmm7 # 0x1eeba20 vmovaps %xmm7, %xmm1 vmovaps %xmm6, %xmm0 vmovaps -0x10(%r8), %xmm8 vmovaps (%r8), %xmm9 vminps %xmm8, %xmm7, %xmm7 vmaxps %xmm9, %xmm6, %xmm6 vaddps %xmm9, %xmm8, %xmm8 vminps %xmm8, %xmm1, %xmm1 vmaxps %xmm8, %xmm0, %xmm0 addq $0x40, %r8 decq %rdi jne 0x138b2bd movq %rax, %rdi subq %rcx, %rdi jmp 0x138b2fc xorl %edi, %edi vmovaps %xmm0, %xmm6 vmovaps %xmm1, %xmm7 vmovaps %xmm5, (%r15) vmovaps %xmm4, 0x10(%r15) vmovaps %xmm2, 0x20(%r15) vmovaps %xmm3, 0x30(%r15) movq %rdx, 0x40(%r15) movq %rcx, 0x48(%r15) movq %rcx, 0x50(%r15) vmovaps %xmm7, (%rbx) vmovaps %xmm6, 0x10(%rbx) vmovaps %xmm1, 0x20(%rbx) vmovaps %xmm0, 0x30(%rbx) movq %rcx, 0x40(%rbx) movq %rax, 0x48(%rbx) movq %rax, 0x50(%rbx) movq 0xa0(%rsp), %rcx subq %rax, %rcx je 0x138b1b4 testq %rsi, %rsi js 0x138b4ca vcvtsi2ss %rsi, %xmm10, %xmm0 jmp 0x138b4e1 movq %rax, %rsi shrq %rsi movl %eax, %edi andl $0x1, %edi orq %rsi, %rdi vcvtsi2ss %rdi, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rdx, %rax js 0x138b385 vcvtsi2ss %rax, %xmm1, %xmm1 jmp 0x138b39a movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm1, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vdivss %xmm1, %xmm0, %xmm0 testq %rcx, %rcx js 0x138b3aa vcvtsi2ss %rcx, %xmm2, %xmm1 jmp 0x138b3c1 movq %rcx, %rax shrq %rax movl %ecx, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm2, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmulss %xmm1, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rax movq %rax, %rdx sarq $0x3f, %rdx vsubss 0xb60a65(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rax, %rsi cmpq %rcx, %rsi cmovaeq %rcx, %rsi subq %rsi, %rcx addq 0x48(%r15), %rsi movq %rsi, 0x50(%r15) movq 0x48(%rbx), %r12 addq %r12, %rcx movq %rcx, 0x50(%rbx) movq 0x50(%r15), %r13 movq 0x48(%r15), %rcx movq %r13, %rax subq %rcx, %rax movq %rax, 0x10(%rsp) movq 0x40(%rbx), %r15 movq %r12, %rax subq %r15, %rax movq %rax, 0x18(%rsp) subq %rcx, %r13 je 0x138b1b4 cmpq %rax, %r13 jae 0x138b632 addq %r15, %r13 leaq 0x28(%rsp), %r12 movq %r14, (%r12) leaq 0x18(%rsp), %rax movq %rax, 0x8(%r12) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r13, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r12, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x138f148 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x138b765 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x138b743 movq %rsi, %rax shrq %rax movl %esi, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm10, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rsi, %rdi js 0x138b4ed vcvtsi2ss %rdi, %xmm10, %xmm1 jmp 0x138b502 movq %rdi, %rax shrq %rax andl $0x1, %edi orq %rax, %rdi vcvtsi2ss %rdi, %xmm10, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vdivss %xmm1, %xmm0, %xmm0 testq %rcx, %rcx js 0x138b512 vcvtsi2ss %rcx, %xmm10, %xmm1 jmp 0x138b529 movq %rcx, %rax shrq %rax movl %ecx, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm10, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmulss %xmm1, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rax movq %rax, %rdx sarq $0x3f, %rdx vsubss 0xb608fd(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rax, %rsi cmpq %rcx, %rsi cmovaeq %rcx, %rsi subq %rsi, %rcx addq 0x48(%r15), %rsi movq %rsi, 0x50(%r15) movq 0x48(%rbx), %r12 addq %r12, %rcx movq %rcx, 0x50(%rbx) movq 0x50(%r15), %r13 movq 0x48(%r15), %rcx movq %r13, %rax subq %rcx, %rax movq %rax, 0x10(%rsp) movq 0x40(%rbx), %r15 movq %r12, %rax subq %r15, %rax movq %rax, 0x18(%rsp) subq %rcx, %r13 je 0x138b1b4 cmpq %rax, %r13 jae 0x138b6bd addq %r15, %r13 leaq 0x28(%rsp), %r12 movq %r14, (%r12) leaq 0x18(%rsp), %rax movq %rax, 0x8(%r12) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r13, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r12, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x138f148 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x138b7c9 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x138b743 leaq 0x28(%rsp), %r13 movq %r14, (%r13) leaq 0x10(%rsp), %rax movq %rax, 0x8(%r13) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r13, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x138f854 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x138b797 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x138b743 leaq 0x28(%rsp), %r13 movq %r14, (%r13) leaq 0x10(%rsp), %rax movq %rax, 0x8(%r13) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r13, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x138f854 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x138b7fb leaq 0xb0(%rsp), %rdi callq 0x6aab0 movq 0x10(%rsp), %rax vmovq %rax, %xmm0 vpshufd $0x44, %xmm0, %xmm0 # xmm0 = xmm0[0,1,0,1] vpaddq 0x40(%rbx), %xmm0, %xmm0 vmovdqa %xmm0, 0x40(%rbx) addq %rax, 0x50(%rbx) jmp 0x138b1b4 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb60377(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd99480(%rip), %rsi # 0x2124c08 movq 0xd99239(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb60345(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd9944e(%rip), %rsi # 0x2124c08 movq 0xd99207(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb60313(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd9941c(%rip), %rsi # 0x2124c08 movq 0xd991d5(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb602e1(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd993ea(%rip), %rsi # 0x2124c08 movq 0xd991a3(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x138b86c movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x138b880 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x138b89c movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x138b8ae jmp 0x138b8c5 jmp 0x138b8c5 jmp 0x138b869 jmp 0x138b87d movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x138b8bb jmp 0x138b8c5 movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x138b8bb jmp 0x138b8c5 jmp 0x138b8c5 jmp 0x138b8c5 jmp 0x138b899 jmp 0x138b8ab movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x138b8bb movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 jmp 0x138b8c5 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const&)::operator()(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const&) const
__forceinline operator Vec3fa () const { return Vec3fa(m128); }
vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vmovaps (%rsi), %xmm2 vmovaps 0x10(%rsi), %xmm3 vmovaps 0x20(%rsi), %xmm4 vmovaps 0x30(%rsi), %xmm5 vminps %xmm0, %xmm2, %xmm2 vmovaps %xmm2, (%rsi) vmaxps %xmm1, %xmm3, %xmm2 vmovaps %xmm2, 0x10(%rsi) vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm4, %xmm1 vmovaps %xmm1, 0x20(%rsi) vmaxps %xmm0, %xmm5, %xmm0 vmovaps %xmm0, 0x30(%rsi) incq 0x48(%rsi) retq
/embree[P]embree/kernels/bvh/../common/../../common/sys/../math/vec3fa.h
embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::clear()
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::clear() { for (size_t i=0; i<bvh->objects.size(); i++) if (bvh->objects[i]) bvh->objects[i]->clear(); for (size_t i=0; i<builders.size(); i++) if (builders[i]) builders[i].reset(); refs.clear(); }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx movq 0x28(%rdi), %rcx movq 0x200(%rcx), %rax cmpq %rax, 0x208(%rcx) je 0x1393350 xorl %r14d, %r14d movq (%rax,%r14,8), %rdi testq %rdi, %rdi je 0x139332f movq (%rdi), %rax callq *0x28(%rax) incq %r14 movq 0x28(%rbx), %rcx movq 0x200(%rcx), %rax movq 0x208(%rcx), %rcx subq %rax, %rcx sarq $0x3, %rcx cmpq %rcx, %r14 jb 0x1393320 movq 0x10(%rbx), %rax cmpq %rax, 0x18(%rbx) je 0x139338b xorl %r14d, %r14d movq (%rax,%r14,8), %rdi testq %rdi, %rdi je 0x1393374 movq $0x0, (%rax,%r14,8) movq (%rdi), %rax callq *0x8(%rax) incq %r14 movq 0x10(%rbx), %rax movq 0x18(%rbx), %rcx subq %rax, %rcx sarq $0x3, %rcx cmpq %rcx, %r14 jb 0x139335d movq 0x50(%rbx), %r14 movq 0x58(%rbx), %rdi testq %rdi, %rdi je 0x13933b8 movq %r14, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13933b3 movzbl 0x40(%rbx), %edx callq 0x1ee67ca jmp 0x13933b8 callq 0x1ee612d testq %r14, %r14 je 0x13933d5 movq 0x38(%rbx), %rdi shlq $0x6, %r14 negq %r14 movq (%rdi), %rax movq %r14, %rsi movl $0x1, %edx callq *(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x48(%rbx) movq $0x0, 0x58(%rbx) addq $0x8, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::__internal_two_level_builder__::MeshBuilder<4, embree::TriangleMesh, embree::TriangleMi<4>>::operator()(void*, embree::TriangleMesh*, unsigned long, embree::Geometry::GTypeMask, bool, embree::Builder*&)
void operator () (void* bvh, Mesh* mesh, size_t geomID, Geometry::GTypeMask gtype, bool useMortonBuilder, Builder*& builder) { if(useMortonBuilder) { builder = MortonBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); return; } switch (mesh->quality) { case RTC_BUILD_QUALITY_LOW: builder = MortonBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; case RTC_BUILD_QUALITY_MEDIUM: case RTC_BUILD_QUALITY_HIGH: builder = SAHBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; case RTC_BUILD_QUALITY_REFIT: builder = RefitBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; default: throw_RTCError(RTC_ERROR_UNKNOWN,"invalid build quality"); } }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x28, %rsp testl %r9d, %r9d je 0x13958d1 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x1447884 jmp 0x1395987 movzwl 0x3e(%rdx), %eax andl $0x7, %eax leal -0x1(%rax), %edi cmpl $0x2, %edi jb 0x1395967 cmpl $0x3, %eax je 0x1395978 testl %eax, %eax je 0x13958bd movl $0x30, %edi callq 0x6a3b0 movq %rax, %rbx leaq 0x18(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb56f18(%rip), %rsi # 0x1eec826 leaq 0xb56f26(%rip), %rdx # 0x1eec83b leaq 0x8(%rsp), %rdi callq 0x8d7230 leaq 0xd6bf8a(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x1, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x8(%rsp), %rsi movq 0x10(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd6bef8(%rip), %rsi # 0x2101850 leaq -0xabeb43(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x127a3c8 jmp 0x1395987 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x150253a movq 0x50(%rsp), %rcx movq %rax, (%rcx) addq $0x28, %rsp popq %rbx popq %r14 popq %r15 popq %rbp retq movq %rax, %r14 xorl %ebp, %ebp jmp 0x13959af movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x8(%rsp), %rdi cmpq %r15, %rdi je 0x13959c6 callq 0x6a4f0 jmp 0x13959c6 movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x13959d3 movq %rbx, %rdi callq 0x6a8a0 movq %r14, %rdi callq 0x6a600 nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel_internal.h
embree::NodeRefPtr<4> embree::avx::GeneralBVHBuilder::build<embree::NodeRefPtr<4>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(unsigned long)>(embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, 32ul>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*, embree::avx::PrimInfoExtRange const&, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(unsigned long) const&, embree::avx::GeneralBVHBuilder::Settings const&)
__noinline static ReductionTy build(Heuristic& heuristic, PrimRef* prims, const Set& set, CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, const ProgressMonitor& progressMonitor, const Settings& settings) { typedef BuildRecordT<Set,typename Heuristic::Split> BuildRecord; typedef BuilderT< BuildRecord, Heuristic, Set, PrimRef, ReductionTy, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, UpdateNodeFunc, CreateLeafFunc, DefaultCanCreateLeafFunc<PrimRef, Set>, DefaultCanCreateLeafSplitFunc<PrimRef, Set>, ProgressMonitor> Builder; /* instantiate builder */ Builder builder(prims, heuristic, createAlloc, createNode, updateNode, createLeaf, DefaultCanCreateLeafFunc<PrimRef, Set>(), DefaultCanCreateLeafSplitFunc<PrimRef, Set>(), progressMonitor, settings); /* build hierarchy */ BuildRecord record(1,set); const ReductionTy root = builder.recurse(record,nullptr,true); _mm_mfence(); // to allow non-temporal stores during build return root; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 movq %r9, %rax movq %r8, %r10 movq %rdx, %r14 movq %rdi, %rdx movq 0x160(%rsp), %r12 leaq 0x28(%rsp), %r11 movq %rcx, (%r11) subq $0x8, %rsp leaq 0x15(%rsp), %r13 leaq 0x38(%rsp), %r15 leaq 0xb0(%rsp), %rbx leaq 0x17(%rsp), %r8 leaq 0x16(%rsp), %r9 movq %rbx, %rdi movq %r11, %rcx pushq %r12 pushq %rax pushq %r13 pushq %r15 pushq %r10 callq 0x1396d66 addq $0x30, %rsp movq $0x1, (%r15) movb $0x0, 0x8(%r15) vmovaps (%r14), %xmm0 vmovaps %xmm0, 0x10(%r15) vmovaps 0x10(%r14), %xmm0 vmovaps %xmm0, 0x20(%r15) vmovaps 0x20(%r14), %xmm0 vmovaps %xmm0, 0x30(%r15) vmovaps 0x30(%r14), %xmm0 vmovaps %xmm0, 0x40(%r15) vmovaps 0x40(%r14), %xmm0 vmovaps %xmm0, 0x50(%r15) movq 0x50(%r14), %rax movq %rax, 0x60(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x10(%rsp) movq $0x0, 0x20(%rsp) subq $0x20, %rsp movq 0x40(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x30(%rsp), %xmm0 vmovups %xmm0, (%rsp) leaq 0x50(%rsp), %rsi movq %rbx, %rdi movl $0x1, %edx callq 0x1396e74 addq $0x20, %rsp mfence addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef const&)::operator()(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef const&) const
__forceinline operator Vec3fa () const { return Vec3fa(m128); }
vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vmovaps (%rsi), %xmm2 vmovaps 0x10(%rsi), %xmm3 vmovaps 0x20(%rsi), %xmm4 vmovaps 0x30(%rsi), %xmm5 vminps %xmm0, %xmm2, %xmm2 vmovaps %xmm2, (%rsi) vmaxps %xmm1, %xmm3, %xmm2 vmovaps %xmm2, 0x10(%rsi) vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm4, %xmm1 vmovaps %xmm1, 0x20(%rsi) vmaxps %xmm0, %xmm5, %xmm0 vmovaps %xmm0, 0x30(%rsi) incq 0x48(%rsi) retq
/embree[P]embree/kernels/bvh/../common/../../common/sys/../math/vec3fa.h
embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::build()
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::build() { /* delete some objects */ size_t num = scene->size(); if (num < bvh->objects.size()) { parallel_for(num, bvh->objects.size(), [&] (const range<size_t>& r) { for (size_t i=r.begin(); i<r.end(); i++) { builders[i].reset(); delete bvh->objects[i]; bvh->objects[i] = nullptr; } }); } #if PROFILE while(1) #endif { /* reset memory allocator */ bvh->alloc.reset(); /* skip build for empty scene */ const size_t numPrimitives = scene->getNumPrimitives(gtype,false); if (numPrimitives == 0) { prims.resize(0); bvh->set(BVH::emptyNode,empty,0); return; } /* calculate the size of the entire BVH */ const size_t numLeafBlocks = Primitive::blocks(numPrimitives); const size_t node_bytes = 2*numLeafBlocks*sizeof(typename BVH::AABBNode)/N; const size_t leaf_bytes = size_t(1.2*numLeafBlocks*sizeof(Primitive)); bvh->alloc.init_estimate(node_bytes+leaf_bytes); double t0 = bvh->preBuild(TOSTRING(isa) "::BVH" + toString(N) + "BuilderTwoLevel"); /* resize object array if scene got larger */ if (bvh->objects.size() < num) bvh->objects.resize(num); if (builders.size() < num) builders.resize(num); resizeRefsList (); nextRef.store(0); /* create acceleration structures */ parallel_for(size_t(0), num, [&] (const range<size_t>& r) { for (size_t objectID=r.begin(); objectID<r.end(); objectID++) { Mesh* mesh = scene->getSafe<Mesh>(objectID); /* ignore meshes we do not support */ if (mesh == nullptr || mesh->numTimeSteps != 1) continue; if (isSmallGeometry(mesh)) { setupSmallBuildRefBuilder (objectID, mesh); } else { setupLargeBuildRefBuilder (objectID, mesh); } } }); /* parallel build of acceleration structures */ parallel_for(size_t(0), num, [&] (const range<size_t>& r) { for (size_t objectID=r.begin(); objectID<r.end(); objectID++) { /* ignore if no triangle mesh or not enabled */ Mesh* mesh = scene->getSafe<Mesh>(objectID); if (mesh == nullptr || !mesh->isEnabled() || mesh->numTimeSteps != 1) continue; builders[objectID]->attachBuildRefs (this); } }); #if PROFILE double d0 = getSeconds(); #endif /* fast path for single geometry scenes */ if (nextRef == 1) { bvh->set(refs[0].node,LBBox3fa(refs[0].bounds()),numPrimitives); } else { /* open all large nodes */ refs.resize(nextRef); /* this probably needs some more tuning */ const size_t extSize = max(max((size_t)SPLIT_MIN_EXT_SPACE,refs.size()*SPLIT_MEMORY_RESERVE_SCALE),size_t((float)numPrimitives / SPLIT_MEMORY_RESERVE_FACTOR)); #if !ENABLE_DIRECT_SAH_MERGE_BUILDER #if ENABLE_OPEN_SEQUENTIAL open_sequential(extSize); #endif /* compute PrimRefs */ prims.resize(refs.size()); #endif { #if ENABLE_DIRECT_SAH_MERGE_BUILDER const PrimInfo pinfo = parallel_reduce(size_t(0), refs.size(), PrimInfo(empty), [&] (const range<size_t>& r) -> PrimInfo { PrimInfo pinfo(empty); for (size_t i=r.begin(); i<r.end(); i++) { pinfo.add_center2(refs[i]); } return pinfo; }, [] (const PrimInfo& a, const PrimInfo& b) { return PrimInfo::merge(a,b); }); #else const PrimInfo pinfo = parallel_reduce(size_t(0), refs.size(), PrimInfo(empty), [&] (const range<size_t>& r) -> PrimInfo { PrimInfo pinfo(empty); for (size_t i=r.begin(); i<r.end(); i++) { pinfo.add_center2(refs[i]); prims[i] = PrimRef(refs[i].bounds(),(size_t)refs[i].node); } return pinfo; }, [] (const PrimInfo& a, const PrimInfo& b) { return PrimInfo::merge(a,b); }); #endif /* skip if all objects where empty */ if (pinfo.size() == 0) bvh->set(BVH::emptyNode,empty,0); /* otherwise build toplevel hierarchy */ else { /* settings for BVH build */ GeneralBVHBuilder::Settings settings; settings.branchingFactor = N; settings.maxDepth = BVH::maxBuildDepthLeaf; settings.logBlockSize = bsr(N); settings.minLeafSize = 1; settings.maxLeafSize = 1; settings.travCost = 1.0f; settings.intCost = 1.0f; settings.singleThreadThreshold = singleThreadThreshold; #if ENABLE_DIRECT_SAH_MERGE_BUILDER refs.resize(extSize); NodeRef root = BVHBuilderBinnedOpenMergeSAH::build<NodeRef,BuildRef>( typename BVH::CreateAlloc(bvh), typename BVH::AABBNode::Create2(), typename BVH::AABBNode::Set2(), [&] (const BuildRef* refs, const range<size_t>& range, const FastAllocator::CachedAllocator& alloc) -> NodeRef { assert(range.size() == 1); return (NodeRef) refs[range.begin()].node; }, [&] (BuildRef &bref, BuildRef *refs) -> size_t { return openBuildRef(bref,refs); }, [&] (size_t dn) { bvh->scene->progressMonitor(0); }, refs.data(),extSize,pinfo,settings); #else NodeRef root = BVHBuilderBinnedSAH::build<NodeRef>( typename BVH::CreateAlloc(bvh), typename BVH::AABBNode::Create2(), typename BVH::AABBNode::Set2(), [&] (const PrimRef* prims, const range<size_t>& range, const FastAllocator::CachedAllocator& alloc) -> NodeRef { assert(range.size() == 1); return (NodeRef) prims[range.begin()].ID(); }, [&] (size_t dn) { bvh->scene->progressMonitor(0); }, prims.data(),pinfo,settings); #endif bvh->set(root,LBBox3fa(pinfo.geomBounds),numPrimitives); } } } bvh->alloc.cleanup(); bvh->postBuild(t0); #if PROFILE double d1 = getSeconds(); std::cout << "TOP_LEVEL OPENING/REBUILD TIME " << 1000.0*(d1-d0) << " ms" << std::endl; #endif } } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::deleteGeometry(size_t geomID) { if (geomID >= bvh->objects.size()) return; if (builders[geomID]) builders[geomID].reset(); delete bvh->objects [geomID]; bvh->objects [geomID] = nullptr; } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::clear() { for (size_t i=0; i<bvh->objects.size(); i++) if (bvh->objects[i]) bvh->objects[i]->clear(); for (size_t i=0; i<builders.size(); i++) if (builders[i]) builders[i].reset(); refs.clear(); } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::open_sequential(const size_t extSize) { if (refs.size() == 0) return; refs.reserve(extSize); #if 1 for (size_t i=0;i<refs.size();i++) { NodeRef ref = refs[i].node; if (ref.isAABBNode()) BVH::prefetch(ref); } #endif std::make_heap(refs.begin(),refs.end()); while (refs.size()+N-1 <= extSize) { std::pop_heap (refs.begin(),refs.end()); NodeRef ref = refs.back().node; if (ref.isLeaf()) break; refs.pop_back(); AABBNode* node = ref.getAABBNode(); for (size_t i=0; i<N; i++) { if (node->child(i) == BVH::emptyNode) continue; refs.push_back(BuildRef(node->bounds(i),node->child(i))); #if 1 NodeRef ref_pre = node->child(i); if (ref_pre.isAABBNode()) ref_pre.prefetch(); #endif std::push_heap (refs.begin(),refs.end()); } } } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupSmallBuildRefBuilder (size_t objectID, Mesh const * const /*mesh*/) { if (builders[objectID] == nullptr || // new mesh dynamic_cast<RefBuilderSmall*>(builders[objectID].get()) == nullptr) // size change resulted in large->small change { builders[objectID].reset (new RefBuilderSmall(objectID)); } } template<int N, typename Mesh, typename Primitive> void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupLargeBuildRefBuilder (size_t objectID, Mesh const * const mesh) { if (bvh->objects[objectID] == nullptr || // new mesh builders[objectID]->meshQualityChanged (mesh->quality) || // changed build quality dynamic_cast<RefBuilderLarge*>(builders[objectID].get()) == nullptr) // size change resulted in small->large change { Builder* builder = nullptr; delete bvh->objects[objectID]; createMeshAccel(objectID, builder); builders[objectID].reset (new RefBuilderLarge(objectID, builder, mesh->quality)); } } #if defined(EMBREE_GEOMETRY_TRIANGLE) Builder* BVH4BuilderTwoLevelTriangle4MeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,TriangleMesh,Triangle4>((BVH4*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH4BuilderTwoLevelTriangle4vMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,TriangleMesh,Triangle4v>((BVH4*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH4BuilderTwoLevelTriangle4iMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,TriangleMesh,Triangle4i>((BVH4*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_QUAD) Builder* BVH4BuilderTwoLevelQuadMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,QuadMesh,Quad4v>((BVH4*)bvh,scene,QuadMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_USER) Builder* BVH4BuilderTwoLevelVirtualSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,UserGeometry,Object>((BVH4*)bvh,scene,UserGeometry::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE) Builder* BVH4BuilderTwoLevelInstanceSAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,Instance,InstancePrimitive>((BVH4*)bvh,scene,gtype,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE_ARRAY) Builder* BVH4BuilderTwoLevelInstanceArraySAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<4,InstanceArray,InstanceArrayPrimitive>((BVH4*)bvh,scene,gtype,useMortonBuilder); } #endif #if defined(__AVX__) #if defined(EMBREE_GEOMETRY_TRIANGLE) Builder* BVH8BuilderTwoLevelTriangle4MeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,TriangleMesh,Triangle4>((BVH8*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH8BuilderTwoLevelTriangle4vMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,TriangleMesh,Triangle4v>((BVH8*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } Builder* BVH8BuilderTwoLevelTriangle4iMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,TriangleMesh,Triangle4i>((BVH8*)bvh,scene,TriangleMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_QUAD) Builder* BVH8BuilderTwoLevelQuadMeshSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,QuadMesh,Quad4v>((BVH8*)bvh,scene,QuadMesh::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_USER) Builder* BVH8BuilderTwoLevelVirtualSAH (void* bvh, Scene* scene, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,UserGeometry,Object>((BVH8*)bvh,scene,UserGeometry::geom_type,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE) Builder* BVH8BuilderTwoLevelInstanceSAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,Instance,InstancePrimitive>((BVH8*)bvh,scene,gtype,useMortonBuilder); } #endif #if defined(EMBREE_GEOMETRY_INSTANCE_ARRAY) Builder* BVH8BuilderTwoLevelInstanceArraySAH (void* bvh, Scene* scene, Geometry::GTypeMask gtype, bool useMortonBuilder) { return new BVHNBuilderTwoLevel<8,InstanceArray,InstanceArrayPrimitive>((BVH8*)bvh,scene,gtype,useMortonBuilder); } #endif #endif }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x208, %rsp # imm = 0x208 movq %rdi, %rbx movq 0x28(%rdi), %rax movq 0x30(%rdi), %rcx movq 0x1d8(%rcx), %r15 movq 0x208(%rax), %r12 subq 0x200(%rax), %r12 sarq $0x3, %r12 cmpq %r12, %r15 jae 0x139fc58 leaq 0x110(%rsp), %r13 movq %rbx, (%r13) leaq 0x10(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x170(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0xa0(%rsp), %rsi movq %r13, (%rsi) leaq 0xf8(%rsp), %rdx movq %r14, %rcx callq 0x13a0e3e leaq 0x10(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a07e5 leaq 0x10(%rsp), %rdi callq 0x6aab0 movq 0x28(%rbx), %rdi addq $0x78, %rdi callq 0x90ae4c movq 0x30(%rbx), %rax movq 0x98(%rbx), %rcx btl $0x14, %ecx jb 0x139fc7b xorl %r14d, %r14d jmp 0x139fc82 movq 0x290(%rax), %r14 btl $0x15, %ecx jae 0x139fc8f addq 0x2a0(%rax), %r14 testb $0xf, %cl je 0x139fc9b addq 0x2c0(%rax), %r14 testl $0x77770, %ecx # imm = 0x77770 je 0x139fcaa addq 0x2b0(%rax), %r14 testl $0xe000000, %ecx # imm = 0xE000000 je 0x139fcb9 addq 0x340(%rax), %r14 btl $0x17, %ecx jae 0x139fcc6 addq 0x2d0(%rax), %r14 btl $0x1d, %ecx jae 0x139fcd3 addq 0x2e0(%rax), %r14 btl $0x1e, %ecx jae 0x139fce0 addq 0x2f0(%rax), %r14 testl %ecx, %ecx jns 0x139fceb addq 0x300(%rax), %r14 btl $0x18, %ecx jae 0x139fcf8 addq 0x310(%rax), %r14 btl $0x16, %ecx jae 0x139fd05 addq 0x320(%rax), %r14 testq %r14, %r14 je 0x139fde8 leaq 0x3(%r14), %rax movq %rax, %rcx shrq $0x2, %rcx shlq $0x4, %rax movabsq $0x3fffffffffffffc0, %rdx # imm = 0x3FFFFFFFFFFFFFC0 vcvtsi2sd %rcx, %xmm1, %xmm0 vmulsd 0xb64424(%rip), %xmm0, %xmm0 # 0x1f04158 vmulsd 0xb64434(%rip), %xmm0, %xmm0 # 0x1f04170 vcvttsd2si %xmm0, %rcx andq %rax, %rdx movq %rcx, %rax vsubsd 0xb50291(%rip), %xmm0, %xmm0 # 0x1eeffe0 sarq $0x3f, %rax vcvttsd2si %xmm0, %rsi andq %rax, %rsi orq %rcx, %rsi addq %rdx, %rsi movq 0x28(%rbx), %rdi addq $0x78, %rdi callq 0x90acb8 movq 0x28(%rbx), %r12 leaq 0xb0(%rsp), %rax movq %rax, -0x10(%rax) leaq 0xa0(%rsp), %r13 movl $0x1, %esi movq %r13, %rdi movl $0x2d, %edx callq 0x6a580 movq (%r13), %rax movb $0x34, (%rax) leaq 0xb82147(%rip), %rcx # 0x1f21eed movl $0x8, %r8d movq %r13, %rdi xorl %esi, %esi xorl %edx, %edx callq 0x6a1a0 leaq 0x180(%rsp), %r13 movq %r13, -0x10(%r13) movq (%rax), %rdx movq %rax, %rcx addq $0x10, %rcx cmpq %rcx, %rdx je 0x139fe12 movq %rdx, 0x170(%rsp) movq (%rcx), %rdx movq %rdx, 0x180(%rsp) jmp 0x139fe1c movq 0x78(%rbx), %r12 cmpq $0x0, 0x70(%rbx) je 0x139fdfb movq $0x0, 0x70(%rbx) cmpq %r12, 0x78(%rbx) jne 0x13a00cd movq $0x0, 0x70(%rbx) jmp 0x13a0263 vmovupd (%rcx), %xmm0 vmovupd %xmm0, (%r13) movq 0x8(%rax), %rdx leaq 0x170(%rsp), %rdi movq %rdx, 0x8(%rdi) movq %rcx, (%rax) movq $0x0, 0x8(%rax) movb $0x0, 0x10(%rax) leaq 0xb7294e(%rip), %rsi # 0x1f12790 callq 0x6a620 leaq 0x20(%rsp), %rbp movq %rbp, -0x10(%rbp) movq (%rax), %rdx movq %rax, %rcx addq $0x10, %rcx cmpq %rcx, %rdx je 0x139fe6e movq %rdx, 0x10(%rsp) movq (%rcx), %rdx movq %rdx, 0x20(%rsp) jmp 0x139fe77 vmovupd (%rcx), %xmm0 vmovupd %xmm0, (%rbp) movq 0x8(%rax), %rdx leaq 0x10(%rsp), %rsi movq %rdx, 0x8(%rsi) movq %rcx, (%rax) movq $0x0, 0x8(%rax) movb $0x0, 0x10(%rax) movq %r12, %rdi callq 0xb5d532 vmovsd %xmm0, 0x98(%rsp) movq 0x10(%rsp), %rdi cmpq %rbp, %rdi je 0x139feb3 callq 0x6a4f0 movq 0x170(%rsp), %rdi cmpq %r13, %rdi je 0x139fec5 callq 0x6a4f0 movq 0xa0(%rsp), %rdi leaq 0xb0(%rsp), %rax cmpq %rax, %rdi je 0x139fedf callq 0x6a4f0 movq 0x28(%rbx), %rdi movq 0x208(%rdi), %rax subq 0x200(%rdi), %rax sarq $0x3, %rax cmpq %r15, %rax jae 0x139ff09 addq $0x200, %rdi # imm = 0x200 movq %r15, %rsi callq 0xc882c0 movq 0x18(%rbx), %rax subq 0x10(%rbx), %rax sarq $0x3, %rax cmpq %r15, %rax jae 0x139ff26 leaq 0x10(%rbx), %rdi movq %r15, %rsi callq 0x13a0aa4 movq %rbx, %rdi callq 0x13a0b1c xorl %eax, %eax xchgl %eax, 0x88(%rbx) leaq 0x110(%rsp), %r13 movq %rbx, (%r13) leaq 0x10(%rsp), %r12 movw $0x401, 0xc(%r12) # imm = 0x401 vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x20(%r12) movq $0x8, 0x40(%r12) movq %r12, %rdi callq 0x6a660 leaq 0x170(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0xa0(%rsp), %rsi movq %r13, (%rsi) leaq 0xf8(%rsp), %rdx movq %r12, %rcx callq 0x13a1ea0 leaq 0x10(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a0781 leaq 0x10(%rsp), %rdi callq 0x6aab0 movq %rbx, 0x110(%rsp) leaq 0x10(%rsp), %r12 movw $0x401, 0xc(%r12) # imm = 0x401 vxorpd %xmm0, %xmm0, %xmm0 vmovupd %xmm0, 0x20(%r12) movq $0x8, 0x40(%r12) movq %r12, %rdi callq 0x6a660 leaq 0x170(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0xa0(%rsp), %rsi movq %r13, (%rsi) leaq 0xf8(%rsp), %rdx movq %r12, %rcx callq 0x13a3464 leaq 0x10(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a07b3 leaq 0x10(%rsp), %rdi callq 0x6aab0 movl 0x88(%rbx), %eax cmpl $0x1, %eax jne 0x13a007d movq 0x28(%rbx), %rdi movq 0x58(%rbx), %rax movq 0x20(%rax), %rsi vmovaps (%rax), %xmm0 vmovaps 0x10(%rax), %xmm1 leaq 0x10(%rsp), %rdx vmovaps %xmm0, (%rdx) vmovaps %xmm1, 0x10(%rdx) vmovaps %xmm0, 0x20(%rdx) vmovaps %xmm1, 0x30(%rdx) jmp 0x13a0709 movl 0x88(%rbx), %eax movslq %eax, %r13 movq 0x50(%rbx), %rax cmpq %r13, %rax jae 0x13a00b1 movq %r13, %rbp testq %rax, %rax je 0x13a00b4 cmpq %r13, %rax jae 0x13a00b1 movq %rax, %rbp addq %rbp, %rbp cmpq $0x1, %rbp adcq $0x0, %rbp cmpq %r13, %rbp jb 0x13a009f jmp 0x13a00b4 movq %rax, %rbp cmpq %r13, 0x48(%rbx) jbe 0x13a00be movq %r13, 0x48(%rbx) cmpq %rbp, 0x50(%rbx) jne 0x13a0106 movq %r13, 0x48(%rbx) jmp 0x13a02d8 movq 0x80(%rbx), %r14 testq %r12, %r12 je 0x13a00eb movq 0x60(%rbx), %rdi movq %r12, %rsi shlq $0x5, %rsi movq (%rdi), %rax xorl %edx, %edx callq *(%rax) movq %r12, %rdi shlq $0x5, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x13a013c leaq 0x68(%rbx), %rsi callq 0x1ee665d jmp 0x13a0146 movq 0x58(%rbx), %r15 testq %rbp, %rbp je 0x13a0121 movq 0x38(%rbx), %rdi movq %rbp, %rsi shlq $0x6, %rsi movq (%rdi), %rax xorl %edx, %edx callq *(%rax) movq %rbp, %rdi shlq $0x6, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x13a01ac leaq 0x40(%rbx), %rsi callq 0x1ee665d jmp 0x13a01b6 movl $0x20, %esi callq 0x1ee60ac movq %rax, 0x80(%rbx) cmpq $0x0, 0x70(%rbx) je 0x13a0177 xorl %eax, %eax xorl %ecx, %ecx movq 0x80(%rbx), %rdx vmovapd (%r14,%rax), %ymm0 vmovapd %ymm0, (%rdx,%rax) incq %rcx addq $0x20, %rax cmpq 0x70(%rbx), %rcx jb 0x13a0158 movq 0x78(%rbx), %r15 testq %r14, %r14 je 0x13a0237 movq %r15, %rsi shlq $0x5, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13a022c movzbl 0x68(%rbx), %edx movq %r14, %rdi vzeroupper callq 0x1ee67ca jmp 0x13a0237 movl $0x20, %esi callq 0x1ee60ac movq %rax, 0x58(%rbx) cmpq $0x0, 0x48(%rbx) je 0x13a01f7 movl $0x20, %eax xorl %ecx, %ecx movq 0x58(%rbx), %rdx vmovapd -0x20(%r15,%rax), %ymm0 vmovapd %ymm0, -0x20(%rdx,%rax) movl 0x8(%r15,%rax), %esi movl %esi, 0x8(%rdx,%rax) movq (%r15,%rax), %rsi movq %rsi, (%rdx,%rax) incq %rcx addq $0x40, %rax cmpq 0x48(%rbx), %rcx jb 0x13a01c8 movq 0x50(%rbx), %r12 testq %r15, %r15 je 0x13a02b0 movq %r12, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13a02a5 movzbl 0x40(%rbx), %edx movq %r15, %rdi vzeroupper callq 0x1ee67ca jmp 0x13a02b0 movq %r14, %rdi vzeroupper callq 0x1ee612d testq %r15, %r15 je 0x13a0257 movq 0x60(%rbx), %rdi shlq $0x5, %r15 negq %r15 movq (%rdi), %rax movq %r15, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq $0x0, 0x70(%rbx) movq %r12, 0x78(%rbx) movq 0x28(%rbx), %rdi vbroadcastss 0xb4b7b0(%rip), %xmm0 # 0x1eeba20 leaq 0x10(%rsp), %rdx vmovaps %xmm0, (%rdx) vbroadcastss 0xb4c902(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%rdx) vmovaps %xmm0, 0x20(%rdx) vmovaps %xmm1, 0x30(%rdx) movl $0x8, %esi xorl %ecx, %ecx vzeroupper callq 0xb5ccb2 jmp 0x13a076f movq %r15, %rdi vzeroupper callq 0x1ee612d testq %r12, %r12 je 0x13a02d0 movq 0x38(%rbx), %rdi shlq $0x6, %r12 negq %r12 movq (%rdi), %rax movq %r12, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq %r13, 0x48(%rbx) movq %rbp, 0x50(%rbx) movq 0x48(%rbx), %r15 leaq (%r15,%r15), %rax testq %r14, %r14 js 0x13a02ec vcvtsi2ss %r14, %xmm1, %xmm0 jmp 0x13a0304 movq %r14, %rcx shrq %rcx movl %r14d, %edx andl $0x1, %edx orq %rcx, %rdx vcvtsi2ss %rdx, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 vdivss 0xb723f4(%rip), %xmm0, %xmm0 # 0x1f12700 vcvttss2si %xmm0, %rcx movq %rcx, %rdx sarq $0x3f, %rdx vsubss 0xb4bb24(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rcx, %rsi cmpq %rsi, %rax cmovaq %rax, %rsi cmpq $0x3e9, %rsi # imm = 0x3E9 movl $0x3e8, %eax # imm = 0x3E8 cmovaeq %rsi, %rax movq %rax, 0x90(%rsp) vbroadcastss 0xb4b6cd(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x110(%rsp) vbroadcastss 0xb4c81f(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x120(%rsp) vmovaps %xmm0, 0x130(%rsp) vmovaps %xmm1, 0x140(%rsp) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x150(%rsp) leaq 0x168(%rsp), %rax movq %rbx, (%rax) leaq 0x1f0(%rsp), %rbp leaq 0x110(%rsp), %rcx movq %rcx, (%rbp) leaq 0xf(%rsp), %r13 movq %r13, 0x8(%rbp) movq %rax, 0x10(%rbp) leaq 0x10(%rsp), %r12 movw $0x401, 0xc(%r12) # imm = 0x401 vmovups %xmm0, 0x20(%r12) movq $0x8, 0x40(%r12) movq %r12, %rdi vzeroupper callq 0x6a660 leaq 0xf8(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x1e0(%rsp), %rax movq %r13, (%rax) movq %rbp, 0x8(%rax) leaq 0x170(%rsp), %rsi leaq 0x110(%rsp), %rcx movq %rcx, (%rsi) movq %rax, 0x8(%rsi) movq %r13, 0x10(%rsi) vmovups 0x110(%rsp), %ymm0 vmovups 0x130(%rsp), %ymm1 vmovups %ymm0, 0x20(%rsi) vmovups %ymm1, 0x40(%rsi) vmovapd 0x150(%rsp), %xmm0 vmovapd %xmm0, 0x60(%rsi) leaq 0xa0(%rsp), %rbp movq %rbp, %rdx movq %r12, %rcx vzeroupper callq 0x13a3b56 leaq 0x1d0(%rsp), %rax vmovups -0x40(%rax), %ymm0 vmovups -0x20(%rax), %ymm1 vmovups %ymm0, 0xa0(%rsp) vmovups %ymm1, 0xc0(%rsp) vmovapd (%rax), %xmm0 vmovapd %xmm0, 0xe0(%rsp) leaq 0x10(%rsp), %rdi vzeroupper callq 0x6a770 testb %al, %al movq 0x90(%rsp), %r13 jne 0x13a0817 leaq 0x10(%rsp), %rdi callq 0x6aab0 movq 0xe8(%rsp), %rax cmpq 0xe0(%rsp), %rax jne 0x13a050a movq 0x28(%rbx), %rdi leaq 0x10(%rsp), %rdx vbroadcastss 0xb4b53e(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, (%rdx) vbroadcastss 0xb4c695(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%rdx) vmovaps %xmm0, 0x20(%rdx) vmovaps %xmm1, 0x30(%rdx) movl $0x8, %esi xorl %ecx, %ecx jmp 0x13a070c vmovaps 0xb8190e(%rip), %ymm0 # 0x1f21e20 vmovups %ymm0, 0x10(%rsp) vmovaps 0xb98520(%rip), %ymm0 # 0x1f38a40 vmovups %ymm0, 0x30(%rsp) movq $0x4, 0x10(%rsp) movq $0x28, 0x18(%rsp) movl $0x4, %eax bsrl %eax, %eax cltq movq %rax, 0x20(%rsp) movl $0x1, %eax movq %rax, 0x28(%rsp) movq %rax, 0x30(%rsp) movq 0x90(%rbx), %rax movq %rax, 0x40(%rsp) movq 0x50(%rbx), %rax cmpq %r13, %rax jae 0x13a0590 movq %r13, %r15 testq %rax, %rax je 0x13a0593 cmpq %r13, %rax jae 0x13a0590 addq %rax, %rax cmpq $0x1, %rax movq %rax, %r15 adcq $0x0, %r15 cmpq %r13, %rax movq %r15, %rax jb 0x13a0578 jmp 0x13a0593 movq %rax, %r15 cmpq %r13, 0x48(%rbx) jbe 0x13a059d movq %r13, 0x48(%rbx) cmpq %r15, 0x50(%rbx) jne 0x13a05ac movq %r13, 0x48(%rbx) jmp 0x13a069a movq 0x58(%rbx), %rbp testq %r15, %r15 je 0x13a05ca movq 0x38(%rbx), %rdi movq %r15, %rsi shlq $0x6, %rsi movq (%rdi), %rax xorl %edx, %edx vzeroupper callq *(%rax) movq %r15, %rdi shlq $0x6, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x13a05e8 leaq 0x40(%rbx), %rsi vzeroupper callq 0x1ee665d jmp 0x13a05f5 movl $0x20, %esi vzeroupper callq 0x1ee60ac movq %rax, 0x58(%rbx) cmpq $0x0, 0x48(%rbx) je 0x13a0635 movl $0x20, %eax xorl %ecx, %ecx movq 0x58(%rbx), %rdx vmovaps -0x20(%rbp,%rax), %ymm0 vmovaps %ymm0, -0x20(%rdx,%rax) movl 0x8(%rbp,%rax), %esi movl %esi, 0x8(%rdx,%rax) movq (%rbp,%rax), %rsi movq %rsi, (%rdx,%rax) incq %rcx addq $0x40, %rax cmpq 0x48(%rbx), %rcx jb 0x13a0607 movq 0x50(%rbx), %r12 testq %rbp, %rbp je 0x13a066a movq %r12, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13a065f movzbl 0x40(%rbx), %edx movq %rbp, %rdi vzeroupper callq 0x1ee67ca jmp 0x13a066a movq %rbp, %rdi vzeroupper callq 0x1ee612d testq %r12, %r12 je 0x13a068a movq 0x38(%rbx), %rdi shlq $0x6, %r12 negq %r12 movq (%rdi), %rax movq %r12, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq %r13, 0x48(%rbx) movq %r15, 0x50(%rbx) leaq 0xa0(%rsp), %rbp movq 0x28(%rbx), %rdi addq $0x78, %rdi movq 0x58(%rbx), %r8 leaq 0x170(%rsp), %r12 movq %r12, %rsi movq %rbx, %rdx movq %rbx, %rcx movq %r13, %r9 leaq 0x10(%rsp), %rax pushq %rax pushq %rbp vzeroupper callq 0x13a0d98 addq $0x10, %rsp movq 0x28(%rbx), %rdi vmovaps 0xa0(%rsp), %xmm0 vmovaps 0xb0(%rsp), %xmm1 vmovaps %xmm0, (%r12) vmovaps %xmm1, 0x10(%r12) vmovaps %xmm0, 0x20(%r12) vmovaps %xmm1, 0x30(%r12) leaq 0x170(%rsp), %rdx movq %rax, %rsi movq %r14, %rcx callq 0xb5ccb2 movq 0x28(%rbx), %r15 leaq 0x78(%r15), %r14 movq %r14, %rdi callq 0x90af76 movq 0x1a8(%r15), %r12 movq 0x1b0(%r15), %r13 cmpq %r13, %r12 je 0x13a0746 movq (%r12), %rdi movq %r14, %rsi callq 0x90afd2 addq $0x8, %r12 jmp 0x13a072f movq 0x1a8(%r15), %rax cmpq %rax, 0x1b0(%r15) je 0x13a075d movq %rax, 0x1b0(%r15) movq 0x28(%rbx), %rdi vmovsd 0x98(%rsp), %xmm0 callq 0xb5d720 addq $0x208, %rsp # imm = 0x208 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb4b35b(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd84464(%rip), %rsi # 0x2124c08 movq 0xd8421d(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb4b329(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd84432(%rip), %rsi # 0x2124c08 movq 0xd841eb(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb4b2f7(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd84400(%rip), %rsi # 0x2124c08 movq 0xd841b9(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb4b2c5(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd843ce(%rip), %rsi # 0x2124c08 movq 0xd84187(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a0890 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a0906 jmp 0x13a093a jmp 0x13a088d movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a0917 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a0928 movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 jmp 0x13a0932 jmp 0x13a093a jmp 0x13a093a jmp 0x13a0903 jmp 0x13a093a jmp 0x13a0914 jmp 0x13a093a jmp 0x13a0925 movq %rax, %rbx movq 0x10(%rsp), %rdi cmpq %rbp, %rdi je 0x13a08d0 callq 0x6a4f0 jmp 0x13a08d0 movq %rax, %rbx movq 0x170(%rsp), %rdi cmpq %r13, %rdi je 0x13a08e7 callq 0x6a4f0 jmp 0x13a08e7 movq %rax, %rbx movq 0xa0(%rsp), %rdi leaq 0xb0(%rsp), %rax cmpq %rax, %rdi je 0x13a0932 callq 0x6a4f0 jmp 0x13a0932 movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 jmp 0x13a0932 jmp 0x13a093a movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 jmp 0x13a0932 jmp 0x13a093a movq %rax, %rbx leaq 0x10(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::resizeRefsList()
void resizeRefsList () { size_t num = parallel_reduce (size_t(0), scene->size(), size_t(0), [this](const range<size_t>& r)->size_t { size_t c = 0; for (auto i=r.begin(); i<r.end(); ++i) { Mesh* mesh = scene->getSafe<Mesh>(i); if (mesh == nullptr || mesh->numTimeSteps != 1) continue; size_t meshSize = mesh->size(); c += isSmallGeometry(mesh) ? Primitive::blocks(meshSize) : 1; } return c; }, std::plus<size_t>() ); if (refs.size() < num) { refs.resize(num); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xf8, %rsp movq %rdi, %rbx movq 0x30(%rdi), %rax movq 0x1d8(%rax), %r15 xorl %eax, %eax leaq 0x10(%rsp), %r13 movq %rax, (%r13) leaq 0x8(%rsp), %rax movq %rdi, (%rax) leaq 0x48(%rsp), %rbp movq %r13, (%rbp) leaq 0x6(%rsp), %r12 movq %r12, 0x8(%rbp) movq %rax, 0x10(%rbp) leaq 0x78(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x60(%rsp), %rdi movq %r15, (%rdi) xorl %eax, %eax movq %rax, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x18(%rsp), %rax movq %r12, (%rax) movq %rbp, 0x8(%rax) leaq 0x28(%rsp), %rsi movq %r13, (%rsi) movq %rax, 0x8(%rsi) movq %r12, 0x10(%rsi) movq (%r13), %rax movq %rax, 0x18(%rsi) leaq 0x7(%rsp), %rdx movq %r14, %rcx callq 0x13a1684 movq 0x40(%rsp), %r12 leaq 0x78(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a0d38 leaq 0x78(%rsp), %rdi callq 0x6aab0 movq 0x48(%rbx), %rax cmpq %r12, %rax jae 0x13a0d23 movq 0x50(%rbx), %rcx cmpq %r12, %rcx jae 0x13a0c2a movq %r12, %r13 testq %rcx, %rcx je 0x13a0c2d cmpq %r12, %rcx jae 0x13a0c2a movq %rcx, %r13 addq %r13, %r13 cmpq $0x1, %r13 adcq $0x0, %r13 cmpq %r12, %r13 jb 0x13a0c18 jmp 0x13a0c2d movq %rcx, %r13 cmpq %r12, %rax jbe 0x13a0c36 movq %r12, 0x48(%rbx) cmpq %r13, 0x50(%rbx) jne 0x13a0c45 movq %r12, 0x48(%rbx) jmp 0x13a0d23 movq 0x58(%rbx), %r14 testq %r13, %r13 je 0x13a0c60 movq 0x38(%rbx), %rdi movq %r13, %rsi shlq $0x6, %rsi movq (%rdi), %rax xorl %edx, %edx callq *(%rax) movq %r13, %rdi shlq $0x6, %rdi cmpq $0x1c00000, %rdi # imm = 0x1C00000 jb 0x13a0c7b leaq 0x40(%rbx), %rsi callq 0x1ee665d jmp 0x13a0c85 movl $0x20, %esi callq 0x1ee60ac movq %rax, 0x58(%rbx) cmpq $0x0, 0x48(%rbx) je 0x13a0cc6 movl $0x20, %eax xorl %ecx, %ecx movq 0x58(%rbx), %rdx vmovaps -0x20(%r14,%rax), %ymm0 vmovaps %ymm0, -0x20(%rdx,%rax) movl 0x8(%r14,%rax), %esi movl %esi, 0x8(%rdx,%rax) movq (%r14,%rax), %rsi movq %rsi, (%rdx,%rax) incq %rcx addq $0x40, %rax cmpq 0x48(%rbx), %rcx jb 0x13a0c97 movq 0x50(%rbx), %r15 testq %r14, %r14 je 0x13a0cfb movq %r15, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13a0cf0 movzbl 0x40(%rbx), %edx movq %r14, %rdi vzeroupper callq 0x1ee67ca jmp 0x13a0cfb movq %r14, %rdi vzeroupper callq 0x1ee612d testq %r15, %r15 je 0x13a0d1b movq 0x38(%rbx), %rdi shlq $0x6, %r15 negq %r15 movq (%rdi), %rax movq %r15, %rsi movl $0x1, %edx vzeroupper callq *(%rax) movq %r12, 0x48(%rbx) movq %r13, 0x50(%rbx) addq $0xf8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb4ada4(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd83ead(%rip), %rsi # 0x2124c08 movq 0xd83c66(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a0d7e jmp 0x13a0d90 jmp 0x13a0d7b movq %rax, %rbx leaq 0x78(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.h
embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::setupLargeBuildRefBuilder(unsigned long, embree::QuadMesh const*)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupLargeBuildRefBuilder (size_t objectID, Mesh const * const mesh) { if (bvh->objects[objectID] == nullptr || // new mesh builders[objectID]->meshQualityChanged (mesh->quality) || // changed build quality dynamic_cast<RefBuilderLarge*>(builders[objectID].get()) == nullptr) // size change resulted in small->large change { Builder* builder = nullptr; delete bvh->objects[objectID]; createMeshAccel(objectID, builder); builders[objectID].reset (new RefBuilderLarge(objectID, builder, mesh->quality)); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rdx, %r14 movq %rsi, %rbx movq %rdi, %r15 movq 0x28(%rdi), %rax movq 0x200(%rax), %rax cmpq $0x0, (%rax,%rsi,8) je 0x13a256f movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi movzwl 0x3e(%r14), %esi andl $0x7, %esi movq (%rdi), %rax callq *0x18(%rax) testb %al, %al jne 0x13a256f movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13a256f leaq 0xd75218(%rip), %rsi # 0x2117770 leaq 0xd75269(%rip), %rdx # 0x21177c8 xorl %ecx, %ecx callq 0x6a9a0 testq %rax, %rax jne 0x13a2617 movq $0x0, (%rsp) movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13a2591 movq (%rdi), %rax callq *0x8(%rax) movq %rsp, %r12 movq %r15, %rdi movq %rbx, %rsi movq %r12, %rdx callq 0x13a2f62 movq 0x10(%r15), %r13 movl $0x20, %edi callq 0x6a170 movq %rax, %r15 movq (%r12), %r12 testq %r12, %r12 je 0x13a25c6 movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) movzwl 0x3e(%r14), %ebp leaq 0xd7521e(%rip), %rax # 0x21177f0 movq %rax, (%r15) movq %rbx, 0x8(%r15) movq %r12, 0x10(%r15) testq %r12, %r12 je 0x13a25ec movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) andl $0x7, %ebp movl %ebp, 0x18(%r15) movq (%r13,%rbx,8), %rdi movq %r15, (%r13,%rbx,8) testq %rdi, %rdi je 0x13a2608 movq (%rdi), %rax callq *0x8(%rax) testq %r12, %r12 je 0x13a2617 movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq jmp 0x13a2637 movq %rax, %rbx movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) jmp 0x13a2642 movq %rax, %rdi callq 0x8d6de8 movq %rax, %rbx movq %r15, %rdi callq 0x6a4f0 movq %rbx, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::createMeshAccel(unsigned long, embree::Builder*&)
void createMeshAccel (size_t geomID, Builder*& builder) { bvh->objects[geomID] = new BVH(Primitive::type,scene); BVH* accel = bvh->objects[geomID]; auto mesh = scene->getSafe<Mesh>(geomID); if (nullptr == mesh) { throw_RTCError(RTC_ERROR_INVALID_ARGUMENT,"geomID does not return correct type"); return; } __internal_two_level_builder__::MeshBuilder<N,Mesh,Primitive>()(accel, mesh, geomID, this->gtype, this->useMortonBuilder_, builder); }
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x30, %rsp movq %rdx, %r14 movq %rsi, %rbx movq %rdi, %r15 movl $0x240, %edi # imm = 0x240 movl $0x10, %esi callq 0x1ee60ac movq %rax, %r12 movq 0x30(%r15), %rdx leaq 0xd8223c(%rip), %rsi # 0x21251d0 movq %rax, %rdi callq 0xb5c9a0 movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq %r12, (%rax,%rbx,8) movq 0x30(%r15), %rax movq 0x1e8(%rax), %rax movq (%rax,%rbx,8), %rax xorl %edx, %edx testq %rax, %rax je 0x13a2fc9 cmpb $0x15, 0x3c(%rax) cmoveq %rax, %rdx testq %rdx, %rdx je 0x13a300a movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rsi movq 0x98(%r15), %r8 movzbl 0xa0(%r15), %r9d movq %r14, (%rsp) leaq 0x10(%rsp), %rdi movq %rbx, %rcx callq 0x13a30d6 addq $0x30, %rsp popq %rbx popq %r12 popq %r14 popq %r15 popq %rbp retq movl $0x30, %edi callq 0x6a3b0 movq %rax, %rbx leaq 0x20(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb6f779(%rip), %rsi # 0x1f127a0 leaq 0xb6f795(%rip), %rdx # 0x1f127c3 leaq 0x10(%rsp), %rdi callq 0x8d7230 leaq 0xd5e871(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x2, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x10(%rsp), %rsi movq 0x18(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd5e7df(%rip), %rsi # 0x2101850 leaq -0xacc25c(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rax, %r14 xorl %ebp, %ebp jmp 0x13a3095 movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x10(%rsp), %rdi cmpq %r15, %rdi je 0x13a30ac callq 0x6a4f0 jmp 0x13a30ac movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x13a30c6 movq %rbx, %rdi callq 0x6a8a0 jmp 0x13a30c6 movq %rax, %r14 movq %r12, %rdi callq 0x1ee612d movq %r14, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef, 32ul>::split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)
__noinline void split(const Split& split, const PrimInfoExtRange& set_i, PrimInfoExtRange& lset, PrimInfoExtRange& rset) { PrimInfoExtRange set = set_i; /* valid split */ if (unlikely(!split.valid())) { deterministic_order(set); splitFallback(set,lset,rset); return; } std::pair<size_t,size_t> ext_weights(0,0); /* object split */ if (likely(set.size() < PARALLEL_THRESHOLD)) ext_weights = sequential_object_split(split,set,lset,rset); else ext_weights = parallel_object_split(split,set,lset,rset); /* if we have an extended range, set extended child ranges and move right split range */ if (unlikely(set.has_ext_range())) { setExtentedRanges(set,lset,rset,ext_weights.first,ext_weights.second); moveExtentedRange(set,lset,rset); } }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 movq %r8, %rbx movq %rcx, %r15 movq %rdi, %r14 vmovaps (%rdx), %xmm0 vmovaps %xmm0, 0x50(%rsp) vmovaps 0x10(%rdx), %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x20(%rdx), %xmm0 vmovaps %xmm0, 0x70(%rsp) vmovaps 0x30(%rdx), %xmm0 vmovaps %xmm0, 0x80(%rsp) vmovdqa 0x40(%rdx), %xmm0 vmovdqa %xmm0, 0x90(%rsp) movq 0x50(%rdx), %rax movq %rax, 0xa0(%rsp) cmpl $-0x1, 0x4(%rsi) je 0x13a60a3 movq 0x98(%rsp), %rax vmovq %xmm0, %rcx subq %rcx, %rax leaq 0x50(%rsp), %rdx movq %r14, %rdi movq %r15, %rcx movq %rbx, %r8 cmpq $0x3ff, %rax # imm = 0x3FF ja 0x13a6139 callq 0x1fb434 movq 0xa0(%rsp), %rcx subq 0x98(%rsp), %rcx jne 0x13a6143 addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq leaq 0x90(%rsp), %r12 movq %r14, %rdi movq %r12, %rsi callq 0x13a5342 movq (%r12), %rdx movq 0x8(%r12), %rax leaq (%rax,%rdx), %rcx shrq %rcx vbroadcastss 0xb46ab5(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xb45948(%rip), %xmm1 # 0x1eeba20 movq %rcx, %rsi subq %rdx, %rsi jbe 0x13a6156 movq (%r14), %rdi movq %rdx, %r8 shlq $0x6, %r8 addq %r8, %rdi addq $0x10, %rdi vbroadcastss 0xb45926(%rip), %xmm2 # 0x1eeba20 vbroadcastss 0xb46a81(%rip), %xmm3 # 0x1eecb84 vmovaps %xmm3, %xmm4 vmovaps %xmm2, %xmm5 vmovaps -0x10(%rdi), %xmm6 vmovaps (%rdi), %xmm7 vminps %xmm6, %xmm5, %xmm5 vmaxps %xmm7, %xmm4, %xmm4 vaddps %xmm7, %xmm6, %xmm6 vminps %xmm6, %xmm2, %xmm2 vmaxps %xmm6, %xmm3, %xmm3 addq $0x40, %rdi decq %rsi jne 0x13a610b movq %rcx, %rsi subq %rdx, %rsi jmp 0x13a6168 callq 0x13aae3e jmp 0x13a607c testq %rax, %rax js 0x13a6240 vcvtsi2ss %rax, %xmm1, %xmm0 jmp 0x13a6257 xorl %esi, %esi vmovaps %xmm1, %xmm5 vmovaps %xmm0, %xmm4 vmovaps %xmm1, %xmm2 vmovaps %xmm0, %xmm3 movq %rax, %rdi subq %rcx, %rdi jbe 0x13a61d0 movq (%r14), %r8 movq %rcx, %r9 shlq $0x6, %r9 addq %r9, %r8 addq $0x10, %r8 vbroadcastss 0xb469fa(%rip), %xmm6 # 0x1eecb84 vbroadcastss 0xb4588d(%rip), %xmm7 # 0x1eeba20 vmovaps %xmm7, %xmm1 vmovaps %xmm6, %xmm0 vmovaps -0x10(%r8), %xmm8 vmovaps (%r8), %xmm9 vminps %xmm8, %xmm7, %xmm7 vmaxps %xmm9, %xmm6, %xmm6 vaddps %xmm9, %xmm8, %xmm8 vminps %xmm8, %xmm1, %xmm1 vmaxps %xmm8, %xmm0, %xmm0 addq $0x40, %r8 decq %rdi jne 0x13a619b movq %rax, %rdi subq %rcx, %rdi jmp 0x13a61da xorl %edi, %edi vmovaps %xmm0, %xmm6 vmovaps %xmm1, %xmm7 vmovaps %xmm5, (%r15) vmovaps %xmm4, 0x10(%r15) vmovaps %xmm2, 0x20(%r15) vmovaps %xmm3, 0x30(%r15) movq %rdx, 0x40(%r15) movq %rcx, 0x48(%r15) movq %rcx, 0x50(%r15) vmovaps %xmm7, (%rbx) vmovaps %xmm6, 0x10(%rbx) vmovaps %xmm1, 0x20(%rbx) vmovaps %xmm0, 0x30(%rbx) movq %rcx, 0x40(%rbx) movq %rax, 0x48(%rbx) movq %rax, 0x50(%rbx) movq 0xa0(%rsp), %rcx subq %rax, %rcx je 0x13a6092 testq %rsi, %rsi js 0x13a63a8 vcvtsi2ss %rsi, %xmm10, %xmm0 jmp 0x13a63bf movq %rax, %rsi shrq %rsi movl %eax, %edi andl $0x1, %edi orq %rsi, %rdi vcvtsi2ss %rdi, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rdx, %rax js 0x13a6263 vcvtsi2ss %rax, %xmm1, %xmm1 jmp 0x13a6278 movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm1, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vdivss %xmm1, %xmm0, %xmm0 testq %rcx, %rcx js 0x13a6288 vcvtsi2ss %rcx, %xmm2, %xmm1 jmp 0x13a629f movq %rcx, %rax shrq %rax movl %ecx, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm2, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmulss %xmm1, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rax movq %rax, %rdx sarq $0x3f, %rdx vsubss 0xb45b87(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rax, %rsi cmpq %rcx, %rsi cmovaeq %rcx, %rsi subq %rsi, %rcx addq 0x48(%r15), %rsi movq %rsi, 0x50(%r15) movq 0x48(%rbx), %r12 addq %r12, %rcx movq %rcx, 0x50(%rbx) movq 0x50(%r15), %r13 movq 0x48(%r15), %rcx movq %r13, %rax subq %rcx, %rax movq %rax, 0x10(%rsp) movq 0x40(%rbx), %r15 movq %r12, %rax subq %r15, %rax movq %rax, 0x18(%rsp) subq %rcx, %r13 je 0x13a6092 cmpq %rax, %r13 jae 0x13a6510 addq %r15, %r13 leaq 0x28(%rsp), %r12 movq %r14, (%r12) leaq 0x18(%rsp), %rax movq %rax, 0x8(%r12) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r13, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r12, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x13aa026 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a6643 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x13a6621 movq %rsi, %rax shrq %rax movl %esi, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm10, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rsi, %rdi js 0x13a63cb vcvtsi2ss %rdi, %xmm10, %xmm1 jmp 0x13a63e0 movq %rdi, %rax shrq %rax andl $0x1, %edi orq %rax, %rdi vcvtsi2ss %rdi, %xmm10, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vdivss %xmm1, %xmm0, %xmm0 testq %rcx, %rcx js 0x13a63f0 vcvtsi2ss %rcx, %xmm10, %xmm1 jmp 0x13a6407 movq %rcx, %rax shrq %rax movl %ecx, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm10, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmulss %xmm1, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rax movq %rax, %rdx sarq $0x3f, %rdx vsubss 0xb45a1f(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rax, %rsi cmpq %rcx, %rsi cmovaeq %rcx, %rsi subq %rsi, %rcx addq 0x48(%r15), %rsi movq %rsi, 0x50(%r15) movq 0x48(%rbx), %r12 addq %r12, %rcx movq %rcx, 0x50(%rbx) movq 0x50(%r15), %r13 movq 0x48(%r15), %rcx movq %r13, %rax subq %rcx, %rax movq %rax, 0x10(%rsp) movq 0x40(%rbx), %r15 movq %r12, %rax subq %r15, %rax movq %rax, 0x18(%rsp) subq %rcx, %r13 je 0x13a6092 cmpq %rax, %r13 jae 0x13a659b addq %r15, %r13 leaq 0x28(%rsp), %r12 movq %r14, (%r12) leaq 0x18(%rsp), %rax movq %rax, 0x8(%r12) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r13, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r12, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x13aa026 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a66a7 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x13a6621 leaq 0x28(%rsp), %r13 movq %r14, (%r13) leaq 0x10(%rsp), %rax movq %rax, 0x8(%r13) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r13, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x13aa732 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a6675 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x13a6621 leaq 0x28(%rsp), %r13 movq %r14, (%r13) leaq 0x10(%rsp), %rax movq %rax, 0x8(%r13) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r13, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x13aa732 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13a66d9 leaq 0xb0(%rsp), %rdi callq 0x6aab0 movq 0x10(%rsp), %rax vmovq %rax, %xmm0 vpshufd $0x44, %xmm0, %xmm0 # xmm0 = xmm0[0,1,0,1] vpaddq 0x40(%rbx), %xmm0, %xmm0 vmovdqa %xmm0, 0x40(%rbx) addq %rax, 0x50(%rbx) jmp 0x13a6092 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb45499(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd7e5a2(%rip), %rsi # 0x2124c08 movq 0xd7e35b(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb45467(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd7e570(%rip), %rsi # 0x2124c08 movq 0xd7e329(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb45435(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd7e53e(%rip), %rsi # 0x2124c08 movq 0xd7e2f7(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb45403(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd7e50c(%rip), %rsi # 0x2124c08 movq 0xd7e2c5(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a674a movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a675e movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a677a movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13a678c jmp 0x13a67a3 jmp 0x13a67a3 jmp 0x13a6747 jmp 0x13a675b movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x13a6799 jmp 0x13a67a3 movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x13a6799 jmp 0x13a67a3 jmp 0x13a67a3 jmp 0x13a67a3 jmp 0x13a6777 jmp 0x13a6789 movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x13a6799 movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 jmp 0x13a67a3 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef const&)::operator()(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::QuadMesh, embree::QuadMv<4>>::BuildRef const&) const
__forceinline operator Vec3fa () const { return Vec3fa(m128); }
vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vmovaps (%rsi), %xmm2 vmovaps 0x10(%rsi), %xmm3 vmovaps 0x20(%rsi), %xmm4 vmovaps 0x30(%rsi), %xmm5 vminps %xmm0, %xmm2, %xmm2 vmovaps %xmm2, (%rsi) vmaxps %xmm1, %xmm3, %xmm2 vmovaps %xmm2, 0x10(%rsi) vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm4, %xmm1 vmovaps %xmm1, 0x20(%rsi) vmaxps %xmm0, %xmm5, %xmm0 vmovaps %xmm0, 0x30(%rsi) incq 0x48(%rsi) retq
/embree[P]embree/kernels/bvh/../common/../../common/sys/../math/vec3fa.h
embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::deleteGeometry(unsigned long)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::deleteGeometry(size_t geomID) { if (geomID >= bvh->objects.size()) return; if (builders[geomID]) builders[geomID].reset(); delete bvh->objects [geomID]; bvh->objects [geomID] = nullptr; }
movq 0x28(%rdi), %rax movq 0x208(%rax), %rcx subq 0x200(%rax), %rcx sarq $0x3, %rcx cmpq %rsi, %rcx jbe 0x13ae1ea pushq %r14 pushq %rbx pushq %rax movq %rsi, %rbx movq %rdi, %r14 movq 0x10(%rdi), %rax movq (%rax,%rsi,8), %rdi testq %rdi, %rdi je 0x13ae1b6 movq $0x0, (%rax,%rbx,8) movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13ae1d0 movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq $0x0, (%rax,%rbx,8) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::setupLargeBuildRefBuilder(unsigned long, embree::UserGeometry const*)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupLargeBuildRefBuilder (size_t objectID, Mesh const * const mesh) { if (bvh->objects[objectID] == nullptr || // new mesh builders[objectID]->meshQualityChanged (mesh->quality) || // changed build quality dynamic_cast<RefBuilderLarge*>(builders[objectID].get()) == nullptr) // size change resulted in small->large change { Builder* builder = nullptr; delete bvh->objects[objectID]; createMeshAccel(objectID, builder); builders[objectID].reset (new RefBuilderLarge(objectID, builder, mesh->quality)); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rdx, %r14 movq %rsi, %rbx movq %rdi, %r15 movq 0x28(%rdi), %rax movq 0x200(%rax), %rax cmpq $0x0, (%rax,%rsi,8) je 0x13afd99 movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi movzwl 0x3e(%r14), %esi andl $0x7, %esi movq (%rdi), %rax callq *0x18(%rax) testb %al, %al jne 0x13afd99 movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13afd99 leaq 0xd67e96(%rip), %rsi # 0x2117c18 leaq 0xd67ee7(%rip), %rdx # 0x2117c70 xorl %ecx, %ecx callq 0x6a9a0 testq %rax, %rax jne 0x13afe41 movq $0x0, (%rsp) movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13afdbb movq (%rdi), %rax callq *0x8(%rax) movq %rsp, %r12 movq %r15, %rdi movq %rbx, %rsi movq %r12, %rdx callq 0x13b0510 movq 0x10(%r15), %r13 movl $0x20, %edi callq 0x6a170 movq %rax, %r15 movq (%r12), %r12 testq %r12, %r12 je 0x13afdf0 movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) movzwl 0x3e(%r14), %ebp leaq 0xd67e9c(%rip), %rax # 0x2117c98 movq %rax, (%r15) movq %rbx, 0x8(%r15) movq %r12, 0x10(%r15) testq %r12, %r12 je 0x13afe16 movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) andl $0x7, %ebp movl %ebp, 0x18(%r15) movq (%r13,%rbx,8), %rdi movq %r15, (%r13,%rbx,8) testq %rdi, %rdi je 0x13afe32 movq (%rdi), %rax callq *0x8(%rax) testq %r12, %r12 je 0x13afe41 movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq jmp 0x13afe61 movq %rax, %rbx movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) jmp 0x13afe6c movq %rax, %rdi callq 0x8d6de8 movq %rax, %rbx movq %r15, %rdi callq 0x6a4f0 movq %rbx, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::GeneralBVHBuilder::BuilderT<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef, embree::NodeRefPtr<4>, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::build()::'lambda'(unsigned long)>::createLargeLeaf(embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>> const&, embree::FastAllocator::CachedAllocator)
const ReductionTy createLargeLeaf(const BuildRecord& current, Allocator alloc) { /* this should never occur but is a fatal error */ if (current.depth > cfg.maxDepth) throw_RTCError(RTC_ERROR_UNKNOWN,"depth limit reached"); /* create leaf for few primitives */ if (current.prims.size() <= cfg.maxLeafSize && canCreateLeaf(prims,current.prims)) return createLeaf(prims,current.prims,alloc); /* fill all children by always splitting the largest one */ ReductionTy values[MAX_BRANCHING_FACTOR]; BuildRecord children[MAX_BRANCHING_FACTOR]; size_t numChildren = 1; children[0] = current; do { /* find best child with largest bounding box area */ size_t bestChild = -1; size_t bestSize = 0; for (size_t i=0; i<numChildren; i++) { /* ignore leaves as they cannot get split */ if (children[i].prims.size() <= cfg.maxLeafSize && canCreateLeaf(prims,children[i].prims)) continue; /* remember child with largest size */ if (children[i].prims.size() > bestSize) { bestSize = children[i].prims.size(); bestChild = i; } } if (bestChild == (size_t)-1) break; /*! split best child into left and right child */ BuildRecord left(current.depth+1); BuildRecord right(current.depth+1); if (!canCreateLeaf(prims,children[bestChild].prims)) { canCreateLeafSplit(prims,children[bestChild].prims,left.prims,right.prims); } else { heuristic.splitFallback(children[bestChild].prims,left.prims,right.prims); } /* add new children left and right */ children[bestChild] = children[numChildren-1]; children[numChildren-1] = left; children[numChildren+0] = right; numChildren++; } while (numChildren < cfg.branchingFactor); /* set barrier for primrefarrayalloc */ if (unlikely(current.size() > cfg.primrefarrayalloc)) for (size_t i=0; i<numChildren; i++) children[i].alloc_barrier = children[i].size() <= cfg.primrefarrayalloc; /* create node */ auto node = createNode(children,numChildren,alloc); /* recurse into each child and perform reduction */ for (size_t i=0; i<numChildren; i++) values[i] = createLargeLeaf(children[i],alloc); /* perform reduction */ return updateNode(current,children,node,values,numChildren); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x898, %rsp # imm = 0x898 movq (%rsi), %rax cmpq 0x8(%rdi), %rax ja 0x13b33d9 movq %rdi, %rbp movq 0x50(%rsi), %rax movq 0x58(%rsi), %rcx subq %rax, %rcx cmpq 0x20(%rdi), %rcx jbe 0x13b30d7 leaq 0x8d0(%rsp), %r13 leaq 0x50(%rsi), %rax movb 0x8(%rsi), %cl leaq 0x1e8(%rsp), %rdx movb %cl, -0x50(%rdx) movq (%rsi), %rcx movq %rcx, -0x58(%rdx) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, -0x48(%rdx) vmovaps 0x20(%rsi), %xmm0 vmovaps %xmm0, -0x38(%rdx) vmovaps 0x30(%rsi), %xmm0 vmovaps %xmm0, -0x28(%rdx) movq %rsi, 0x60(%rsp) vmovaps 0x40(%rsi), %xmm0 vmovaps %xmm0, -0x18(%rdx) vmovaps (%rax), %xmm0 vmovaps %xmm0, -0x8(%rdx) movq 0x10(%rax), %rax movq %rax, 0x8(%rdx) movl $0x1, %r14d leaq 0x90(%rsp), %rdi movq %rbp, 0x58(%rsp) testq %r14, %r14 je 0x13b2a2d movq 0x20(%rbp), %rax movq $-0x1, %r12 leaq 0x1e8(%rsp), %rcx xorl %edx, %edx xorl %esi, %esi movq (%rcx), %r8 subq -0x8(%rcx), %r8 cmpq %rax, %r8 jbe 0x13b2a1f cmpq %rsi, %r8 cmovaq %rdx, %r12 cmovaq %r8, %rsi incq %rdx addq $0x70, %rcx cmpq %rdx, %r14 jne 0x13b2a08 jmp 0x13b2a34 movq $-0x1, %r12 cmpq $-0x1, %r12 je 0x13b2fcf imulq $0x70, %r12, %rdx movq 0x48(%rbp), %rax movq 0x1e0(%rsp,%rdx), %r10 movq 0x1e8(%rsp,%rdx), %rbx leaq (%rbx,%r10), %r15 shrq %r15 movq %r15, %rcx subq %r10, %rcx jbe 0x13b2ae9 movq (%rax), %rsi movq %r10, %r8 shlq $0x6, %r8 addq %r8, %rsi addq $0x10, %rsi vbroadcastss 0xb3a101(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, %xmm2 vbroadcastss 0xb38f90(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, %xmm3 vmovaps %xmm0, %xmm4 vmovaps %xmm1, %xmm5 vmovaps -0x10(%rsi), %xmm0 vmovaps (%rsi), %xmm1 vminps %xmm0, %xmm5, %xmm5 vmaxps %xmm1, %xmm4, %xmm4 vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm3, %xmm3 vmaxps %xmm0, %xmm2, %xmm2 addq $0x40, %rsi decq %rcx jne 0x13b2a9c vmovaps %xmm5, 0x140(%rsp) vmovaps %xmm4, 0x150(%rsp) vmovaps %xmm3, 0x160(%rsp) movq %r15, %rsi subq %r10, %rsi vmovaps %xmm2, %xmm0 jmp 0x13b2b18 vbroadcastss 0xb38f2e(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, 0x140(%rsp) vbroadcastss 0xb3a080(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x150(%rsp) vmovaps %xmm1, 0x160(%rsp) xorl %esi, %esi vmovaps %xmm0, 0x180(%rsp) movq %rbx, %rcx subq %r15, %rcx movq %r10, 0x80(%rsp) jbe 0x13b2bba movq (%rax), %r9 movq %r15, %r8 shlq $0x6, %r8 addq %r8, %r9 addq $0x10, %r9 movq %rcx, %r8 vbroadcastss 0xb3a032(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, %xmm2 vbroadcastss 0xb38ec1(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, %xmm3 vmovaps %xmm0, %xmm4 vmovaps %xmm1, %xmm5 vmovaps -0x10(%r9), %xmm0 vmovaps (%r9), %xmm1 vminps %xmm0, %xmm5, %xmm5 vmaxps %xmm1, %xmm4, %xmm4 vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm3, %xmm3 vmaxps %xmm0, %xmm2, %xmm2 addq $0x40, %r9 decq %r8 jne 0x13b2b6b vmovaps %xmm5, 0x110(%rsp) vmovaps %xmm4, 0x120(%rsp) vmovaps %xmm3, 0x130(%rsp) movq %rbx, %r9 subq %r15, %r9 vmovaps %xmm2, %xmm0 jmp 0x13b2bea vbroadcastss 0xb38e5d(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, 0x110(%rsp) vbroadcastss 0xb39faf(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x120(%rsp) vmovaps %xmm1, 0x130(%rsp) xorl %r9d, %r9d movq 0x60(%rsp), %r8 movq (%r8), %r8 movq %r8, 0x78(%rsp) leaq (%rsp,%rdx), %rbp addq $0x190, %rbp # imm = 0x190 movq 0x60(%rbp), %r13 subq %rbx, %r13 vmovaps %xmm0, 0x170(%rsp) jne 0x13b2c22 movq %r15, %r10 movq %rbx, %rcx movq %r15, %rax jmp 0x13b2e5c testq %rsi, %rsi js 0x13b2c2e vcvtsi2ss %rsi, %xmm6, %xmm0 jmp 0x13b2c47 movq %rsi, %rdx shrq %rdx movl %esi, %r8d andl $0x1, %r8d orq %rdx, %r8 vcvtsi2ss %r8, %xmm6, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rsi, %r9 js 0x13b2c53 vcvtsi2ss %r9, %xmm6, %xmm1 jmp 0x13b2c69 movq %r9, %rdx shrq %rdx andl $0x1, %r9d orq %rdx, %r9 vcvtsi2ss %r9, %xmm6, %xmm1 vaddss %xmm1, %xmm1, %xmm1 testq %r13, %r13 js 0x13b2c75 vcvtsi2ss %r13, %xmm6, %xmm2 jmp 0x13b2c8d movq %r13, %rdx shrq %rdx movl %r13d, %esi andl $0x1, %esi orq %rdx, %rsi vcvtsi2ss %rsi, %xmm6, %xmm2 vaddss %xmm2, %xmm2, %xmm2 vdivss %xmm1, %xmm0, %xmm0 vmulss %xmm2, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rsi movq %rsi, %r8 vsubss 0xb39199(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rdx sarq $0x3f, %r8 andq %r8, %rdx orq %rsi, %rdx cmpq %r13, %rdx cmovaeq %r13, %rdx subq %rdx, %r13 leaq (%rdx,%r15), %rsi movq %rsi, 0x68(%rsp) addq %rbx, %r13 movq %rdx, 0x70(%rsp) movq %rcx, 0x88(%rsp) testq %rdx, %rdx je 0x13b2d97 cmpq %rcx, %rdx jae 0x13b2d9f movq %rax, 0x30(%rsp) leaq 0x88(%rsp), %rax movq %rax, 0x38(%rsp) movb $0x1, 0x9c(%rsp) leaq 0xb0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0xd0(%rsp) movb $0x4, 0x9d(%rsp) callq 0x6a660 movq 0x68(%rsp), %rax movq %rax, 0x40(%rsp) movq %r15, 0x48(%rsp) movq $0x40, 0x50(%rsp) leaq 0x30(%rsp), %rax movq %rax, 0x28(%rsp) leaq 0x40(%rsp), %rdi leaq 0x28(%rsp), %rsi leaq 0x27(%rsp), %rdx leaq 0x90(%rsp), %rcx callq 0x13b75d4 leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13b33a7 leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x13b2e3a movq %r15, %rax jmp 0x13b2e51 movq %rax, 0x30(%rsp) leaq 0x70(%rsp), %rax movq %rax, 0x38(%rsp) movb $0x1, 0x9c(%rsp) leaq 0xb0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0xd0(%rsp) movb $0x4, 0x9d(%rsp) callq 0x6a660 movq %rbx, 0x40(%rsp) movq %r15, 0x48(%rsp) movq $0x40, 0x50(%rsp) leaq 0x30(%rsp), %rax movq %rax, 0x28(%rsp) leaq 0x40(%rsp), %rdi leaq 0x28(%rsp), %rsi leaq 0x27(%rsp), %rdx leaq 0x90(%rsp), %rcx callq 0x13b7ce0 leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13b3375 leaq 0x90(%rsp), %rdi callq 0x6aab0 leaq 0x90(%rsp), %rdi movq 0x70(%rsp), %rcx leaq (%rcx,%r15), %rax addq %rcx, %rbx addq %rcx, %r13 movq %rbx, %rcx movq %r13, %rbx movq 0x68(%rsp), %r10 movq 0x78(%rsp), %r9 incq %r9 imulq $0x70, %r14, %rdx movb 0x128(%rsp,%rdx), %sil movb %sil, 0x8(%rbp) movq 0x120(%rsp,%rdx), %rsi movq %rsi, (%rbp) vmovaps 0x130(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x10(%rbp) vmovaps 0x140(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x20(%rbp) vmovaps 0x150(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x30(%rbp) vmovaps 0x160(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x40(%rbp) movq 0x180(%rsp,%rdx), %rsi movq %rsi, 0x60(%rbp) vmovaps 0x170(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x50(%rbp) movq %r9, 0x120(%rsp,%rdx) movl $0x0, %esi movb %sil, 0x128(%rsp,%rdx) vmovaps 0x140(%rsp), %xmm0 vmovaps %xmm0, 0x130(%rsp,%rdx) vmovaps 0x150(%rsp), %xmm0 vmovaps %xmm0, 0x140(%rsp,%rdx) vmovaps 0x160(%rsp), %xmm0 vmovaps %xmm0, 0x150(%rsp,%rdx) vmovaps 0x180(%rsp), %xmm0 vmovaps %xmm0, 0x160(%rsp,%rdx) movq 0x80(%rsp), %r8 movq %r8, 0x170(%rsp,%rdx) movq %r15, 0x178(%rsp,%rdx) movq %r10, 0x180(%rsp,%rdx) movq %r9, 0x190(%rsp,%rdx) movb %sil, 0x198(%rsp,%rdx) vmovaps 0x110(%rsp), %xmm0 vmovaps %xmm0, 0x1a0(%rsp,%rdx) vmovaps 0x120(%rsp), %xmm0 vmovaps %xmm0, 0x1b0(%rsp,%rdx) vmovaps 0x130(%rsp), %xmm0 vmovaps %xmm0, 0x1c0(%rsp,%rdx) vmovaps 0x170(%rsp), %xmm0 vmovaps %xmm0, 0x1d0(%rsp,%rdx) movq %rax, 0x1e0(%rsp,%rdx) movq %rcx, 0x1e8(%rsp,%rdx) movq %rbx, 0x1f0(%rsp,%rdx) incq %r14 movq 0x58(%rsp), %rbp leaq 0x8d0(%rsp), %r13 cmpq $-0x1, %r12 je 0x13b2fdf cmpq (%rbp), %r14 jb 0x13b29ec movq 0x60(%rsp), %rcx movq 0x58(%rcx), %rax subq 0x50(%rcx), %rax movq 0x38(%rbp), %rdx cmpq %rdx, %rax ja 0x13b32c8 movq (%r13), %r15 movq 0x8(%r13), %rbx movq $0x80, 0x28(%rsp) movq (%rbx), %r12 movq 0x8(%r12), %rax cmpq %rax, %r15 je 0x13b3184 movq %r12, 0x40(%rsp) movb $0x1, 0x48(%rsp) movq %r12, %rdi callq 0x1ee7bb6 movq 0x8(%r12), %rax testq %rax, %rax je 0x13b3095 movq 0xa8(%r12), %rax addq 0x68(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%r12), %rax addq 0x98(%r12), %rax movq 0x50(%r12), %rcx addq 0x90(%r12), %rcx subq %rcx, %rax movq 0x8(%r12), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%r12), %rax addq 0x70(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%r12) vmovups %ymm0, 0x48(%r12) testq %r15, %r15 je 0x13b30e9 movq 0x10(%r15), %rax movq %rax, 0x60(%r12) vmovups %ymm0, 0x88(%r12) vmovups %ymm0, 0x98(%r12) movq 0x10(%r15), %rax movq %rax, 0xa0(%r12) jmp 0x13b30fd movq 0x40(%rbp), %rcx shlq $0x6, %rax movq 0x20(%rcx,%rax), %r12 jmp 0x13b32b0 vmovups %ymm0, 0x98(%r12) vmovups %ymm0, 0x88(%r12) movq %r15, %rax xchgq %rax, 0x8(%r12) movq %r12, 0x30(%rsp) leaq 0xd73767(%rip), %rdi # 0x2126878 movq %rdi, 0x90(%rsp) movb $0x1, 0x98(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%r15), %rsi cmpq 0x140(%r15), %rsi je 0x13b314b movq 0x30(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%r15) jmp 0x13b315c leaq 0x130(%r15), %rdi leaq 0x30(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x98(%rsp) jne 0x13b3173 movq 0x90(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x48(%rsp) jne 0x13b3184 movq 0x40(%rsp), %rdi callq 0x1ee7c24 movq 0x28(%rsp), %rax addq %rax, 0x28(%rbx) movq 0x10(%rbx), %rcx movl %ecx, %edx negl %edx andl $0xf, %edx leaq (%rcx,%rax), %r12 addq %rdx, %r12 movq %r12, 0x10(%rbx) cmpq 0x18(%rbx), %r12 ja 0x13b32dd addq %rdx, 0x30(%rbx) subq %rax, %r12 addq 0x8(%rbx), %r12 vbroadcastss 0xb3885f(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x60(%r12) vmovaps %xmm0, 0x40(%r12) vmovaps %xmm0, 0x20(%r12) vbroadcastss 0xb399a5(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x70(%r12) vmovaps %xmm0, 0x50(%r12) vmovaps %xmm0, 0x30(%r12) vbroadcastsd 0xb6ec7b(%rip), %ymm0 # 0x1f21e78 vmovups %ymm0, (%r12) testq %r14, %r14 je 0x13b3255 leaq 0x1b0(%rsp), %rax xorl %ecx, %ecx vmovaps -0x10(%rax), %xmm0 vmovaps (%rax), %xmm1 vmovss %xmm0, 0x20(%r12,%rcx,4) vextractps $0x1, %xmm0, 0x40(%r12,%rcx,4) vextractps $0x2, %xmm0, 0x60(%r12,%rcx,4) vmovss %xmm1, 0x30(%r12,%rcx,4) vextractps $0x1, %xmm1, 0x50(%r12,%rcx,4) vextractps $0x2, %xmm1, 0x70(%r12,%rcx,4) incq %rcx addq $0x70, %rax cmpq %rcx, %r14 jne 0x13b3212 testq %r14, %r14 je 0x13b32b0 leaq 0x190(%rsp), %r15 xorl %ebx, %ebx movq 0x10(%r13), %rax movq %rax, 0x10(%rsp) vmovups (%r13), %xmm0 vmovups %xmm0, (%rsp) movq %rbp, %rdi movq %r15, %rsi vzeroupper callq 0x13b2944 movq %rax, 0x90(%rsp,%rbx,8) incq %rbx addq $0x70, %r15 cmpq %rbx, %r14 jne 0x13b3264 xorl %eax, %eax movq 0x90(%rsp,%rax,8), %rcx movq %rcx, (%r12,%rax,8) incq %rax cmpq %rax, %r14 jne 0x13b329c movq %r12, %rax addq $0x898, %rsp # imm = 0x898 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq leaq 0x190(%rsp), %rsi movq %r14, %rdi callq 0x1ff467 jmp 0x13b2ff9 movq %rcx, 0x10(%rbx) shlq $0x2, %rax movq 0x20(%rbx), %rcx cmpq %rcx, %rax jbe 0x13b330a leaq 0x28(%rsp), %rsi movl $0x40, %edx movq %r15, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 jmp 0x13b31b8 leaq 0x90(%rsp), %r13 movq %rcx, (%r13) movl $0x40, %edx movq %r15, %rdi movq %r13, %rsi movl $0x1, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbx) movq 0x18(%rbx), %rax subq 0x10(%rbx), %rax addq 0x30(%rbx), %rax movq %rax, 0x30(%rbx) movq $0x0, 0x10(%rbx) movq (%r13), %rcx movq %rcx, 0x18(%rbx) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbx) cmpq %rcx, %rdx ja 0x13b345c movq %rax, 0x30(%rbx) leaq 0x8d0(%rsp), %r13 jmp 0x13b31b8 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb38767(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd71870(%rip), %rsi # 0x2124c08 movq 0xd71629(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb38735(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd7183e(%rip), %rsi # 0x2124c08 movq 0xd715f7(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x30, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0x1a0(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb3983f(%rip), %rsi # 0x1eecc38 leaq 0xb3984b(%rip), %rdx # 0x1eecc4b leaq 0x190(%rsp), %rdi callq 0x8d7230 leaq 0xd4e49c(%rip), %rax # 0x21018b0 movq %rax, (%r14) movl $0x1, 0x8(%r14) leaq 0x10(%r14), %rdi movq %r14, %rax addq $0x20, %rax movq %rax, 0x10(%r14) movq 0x190(%rsp), %rsi movq 0x198(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd4e403(%rip), %rsi # 0x2101850 leaq -0xadc638(%rip), %rdx # 0x8d6e1c movq %r14, %rdi callq 0x6a5d0 xorl %ebp, %ebp movq %rbp, 0x10(%rbx) movq 0x20(%rbx), %rax leaq 0x90(%rsp), %r13 movq %rax, (%r13) movl $0x40, %edx movq %r15, %rdi movq %r13, %rsi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbx) movq 0x18(%rbx), %rax subq 0x10(%rbx), %rax addq 0x30(%rbx), %rax movq %rax, 0x30(%rbx) movq %rbp, 0x10(%rbx) movq (%r13), %rcx movq %rcx, 0x18(%rbx) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbx) cmpq %rcx, %rdx ja 0x13b34c3 movq %rax, 0x30(%rbx) movq 0x58(%rsp), %rbp jmp 0x13b3368 movq $0x0, 0x10(%rbx) xorl %r12d, %r12d jmp 0x13b34b9 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x8d6eda jmp 0x13b3518 jmp 0x13b358b jmp 0x13b358b movq %rax, %rbx xorl %ebp, %ebp jmp 0x13b3501 movq %rax, %rbx movq %r14, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x190(%rsp), %rdi cmpq %r15, %rdi je 0x13b352a callq 0x6a4f0 jmp 0x13b352a movq %rax, %rbx leaq 0x40(%rsp), %rdi callq 0x8d6eda jmp 0x13b3581 movq %rax, %rbx movb $0x1, %bpl testb %bpl, %bpl je 0x13b3581 movq %r14, %rdi callq 0x6a8a0 jmp 0x13b3581 jmp 0x13b355b movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13b355e jmp 0x13b3571 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13b3574 jmp 0x13b358b jmp 0x13b355b movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x13b3581 jmp 0x13b358b jmp 0x13b3571 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 jmp 0x13b358b movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef, 32ul>::parallel_object_find(embree::avx::PrimInfoExtRange const&, unsigned long)
__noinline const Split parallel_object_find(const PrimInfoExtRange& set, const size_t logBlockSize) { Binner binner(empty); const BinMapping<OBJECT_BINS> mapping(set.centBounds); const BinMapping<OBJECT_BINS>& _mapping = mapping; // CLANG 3.4 parser bug workaround auto body = [&] (const range<size_t>& r) -> Binner { Binner binner(empty); binner.bin(prims0+r.begin(),r.size(),_mapping); return binner; }; auto reduction = [&] (const Binner& b0, const Binner& b1) -> Binner { Binner r = b0; r.merge(b1,_mapping.size()); return r; }; binner = parallel_reduce(set.begin(),set.end(),PARALLEL_FIND_BLOCK_SIZE,binner,body,reduction); return binner.best(mapping,logBlockSize); }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x40, %rsp subq $0x1f00, %rsp # imm = 0x1F00 movq %rcx, (%rsp) movq %rdi, %rbx movl $0xc00, %eax # imm = 0xC00 leaq 0x310(%rsp), %rcx vbroadcastss 0xb35d4f(%rip), %xmm1 # 0x1eeba20 vbroadcastss 0xb36eaa(%rip), %xmm2 # 0x1eecb84 vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm1, -0x10(%rcx) vmovaps %xmm2, (%rcx) vmovaps %xmm1, -0x30(%rcx) vmovaps %xmm2, -0x20(%rcx) vmovaps %xmm1, -0x50(%rcx) vmovaps %xmm2, -0x40(%rcx) vmovaps %xmm0, 0x2c0(%rsp,%rax) addq $0x10, %rax addq $0x60, %rcx cmpq $0xe00, %rax # imm = 0xE00 jne 0x13b5cde leaq 0x80(%rsp), %rax movq $0x20, (%rax) vmovaps 0x20(%rdx), %xmm0 vmovaps 0x30(%rdx), %xmm1 vsubps %xmm0, %xmm1, %xmm1 vbroadcastss 0xb6c136(%rip), %xmm2 # 0x1f21e70 vmaxps %xmm1, %xmm2, %xmm1 vcmpnleps %xmm2, %xmm1, %xmm2 vbroadcastss 0xb82cdc(%rip), %xmm3 # 0x1f38a28 vdivps %xmm1, %xmm3, %xmm1 vandps %xmm1, %xmm2, %xmm1 vmovaps %xmm1, 0x20(%rax) vmovaps %xmm0, 0x10(%rax) leaq 0xb0(%rsp), %r12 movq %rsi, (%r12) movq %rax, 0x8(%r12) leaq 0x78(%rsp), %r15 movq %rax, (%r15) leaq 0xc0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) vpermilps $0x4e, 0x40(%rdx), %xmm0 # xmm0 = mem[2,3,0,1] vmovaps %xmm0, 0x10(%rsp) movq %r14, %rdi callq 0x6a660 leaq 0x30(%rsp), %rsi vmovdqa 0x10(%rsp), %xmm0 vmovdqa %xmm0, (%rsi) movq $0x200, 0x10(%rsi) # imm = 0x200 leaq 0x20(%rsp), %rcx movq %r15, (%rcx) movq %r12, 0x8(%rcx) leaq 0x10c0(%rsp), %r12 leaq 0x2c0(%rsp), %r13 movq %r12, %rdi movq %r13, %rdx movq %r15, %r8 movq %r14, %r9 callq 0x13b624f leaq 0xc0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13b61ec leaq 0xc0(%rsp), %r15 movq %r15, %rdi callq 0x6aab0 xorl %eax, %eax movl $0x10, %ecx vmovaps -0x10(%r12,%rcx), %xmm0 vmovaps %xmm0, -0x10(%r13,%rcx) vmovdqa (%r12,%rcx), %xmm0 vmovdqa %xmm0, (%r13,%rcx) addq $0x20, %rcx cmpq $0x70, %rcx jne 0x13b5e1d incq %rax addq $0x60, %r12 addq $0x60, %r13 cmpq $0x20, %rax jne 0x13b5e18 leaq 0x1cc0(%rsp), %rsi movl $0x200, %edx # imm = 0x200 leaq 0xec0(%rsp), %rdi callq 0x6a0f0 movq 0x80(%rsp), %rax movq %rax, %rdx decq %rdx vbroadcastss 0xb35b9c(%rip), %xmm10 # 0x1eeba20 vbroadcastss 0xb36cf7(%rip), %xmm11 # 0x1eecb84 je 0x13b5f8b movq %rax, %rdi shlq $0x4, %rdi leaq (%rdi,%r15), %rcx addq $-0x10, %rcx leaq 0x10c0(%rsp), %rsi leaq -0x4(%rdi,%rsi), %rsi leaq 0x2c0(%rsp), %r8 addq %r8, %rdi addq $0xbf0, %rdi # imm = 0xBF0 leaq (%rax,%rax,2), %r9 shlq $0x5, %r9 addq %r9, %r8 addq $-0x10, %r8 vpxor %xmm0, %xmm0, %xmm0 xorl %r9d, %r9d movq %rdx, %r10 vmovaps %xmm11, %xmm1 vmovaps %xmm10, %xmm2 vmovaps %xmm11, %xmm3 vmovaps %xmm10, %xmm4 vpaddd (%rdi,%r9), %xmm0, %xmm0 vminps -0x50(%r8), %xmm10, %xmm10 vmaxps -0x40(%r8), %xmm11, %xmm11 vmovdqa %xmm0, (%rcx,%r9) vsubps %xmm10, %xmm11, %xmm5 vminps -0x30(%r8), %xmm2, %xmm2 vmaxps -0x20(%r8), %xmm1, %xmm1 vsubps %xmm2, %xmm1, %xmm6 vinsertps $0x4c, %xmm5, %xmm6, %xmm7 # xmm7 = xmm5[1],xmm6[1],zero,zero vshufpd $0x1, %xmm5, %xmm5, %xmm8 # xmm8 = xmm5[1,0] vinsertps $0x9c, %xmm6, %xmm8, %xmm8 # xmm8 = xmm8[0],xmm6[2],zero,zero vaddps %xmm7, %xmm8, %xmm9 vmulps %xmm7, %xmm8, %xmm7 vunpcklps %xmm6, %xmm5, %xmm5 # xmm5 = xmm5[0],xmm6[0],xmm5[1],xmm6[1] vmulps %xmm5, %xmm9, %xmm5 vaddps %xmm7, %xmm5, %xmm5 vmovlps %xmm5, -0xc(%rsi,%r9) vminps -0x10(%r8), %xmm4, %xmm4 vmaxps (%r8), %xmm3, %xmm3 vsubps %xmm4, %xmm3, %xmm5 vmovshdup %xmm5, %xmm6 # xmm6 = xmm5[1,1,3,3] vshufpd $0x1, %xmm5, %xmm5, %xmm7 # xmm7 = xmm5[1,0] vaddss %xmm7, %xmm6, %xmm8 vmulss %xmm7, %xmm6, %xmm6 vmulss %xmm5, %xmm8, %xmm5 vaddss %xmm6, %xmm5, %xmm5 vmovss %xmm5, -0x4(%rsi,%r9) movl $0x0, (%rsi,%r9) addq $-0x10, %r9 addq $-0x60, %r8 decq %r10 jne 0x13b5eea movl $0xffffffff, %esi # imm = 0xFFFFFFFF movl $0xffffffff, %edi # imm = 0xFFFFFFFF movq (%rsp), %rcx shll %cl, %edi vbroadcastss 0xb35a7c(%rip), %xmm10 # 0x1eeba20 vmovaps %xmm10, 0x30(%rsp) vpxor %xmm0, %xmm0, %xmm0 vmovdqa %xmm0, 0x20(%rsp) cmpq $0x2, %rax jb 0x13b6157 notl %edi vmovd %edi, %xmm0 vpshufd $0x0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmovdqa %xmm0, 0x60(%rsp) vpxor %xmm0, %xmm0, %xmm0 vmovddup 0xb6af0d(%rip), %xmm2 # xmm2 = mem[0,0] vbroadcastss 0xb35a3c(%rip), %xmm3 # 0x1eeba20 vbroadcastss 0xb36b97(%rip), %xmm4 # 0x1eecb84 vxorps %xmm5, %xmm5, %xmm5 leaq 0x310(%rsp), %r8 xorl %edi, %edi vmovd %ecx, %xmm1 vmovdqa %xmm1, (%rsp) vmovddup 0xb6be74(%rip), %xmm1 # xmm1 = mem[0,0] vmovaps %xmm1, 0x50(%rsp) vmovaps %xmm4, %xmm11 vmovaps %xmm3, %xmm12 vmovaps %xmm4, %xmm13 vmovaps %xmm3, %xmm14 vmovaps %xmm3, %xmm10 vminps -0x50(%r8), %xmm3, %xmm3 vmaxps -0x40(%r8), %xmm4, %xmm4 vmovaps %xmm4, 0x10(%rsp) vminps -0x30(%r8), %xmm12, %xmm12 vmaxps -0x20(%r8), %xmm11, %xmm11 vsubps %xmm3, %xmm4, %xmm15 vsubps %xmm12, %xmm11, %xmm9 vminps -0x10(%r8), %xmm14, %xmm14 vmaxps (%r8), %xmm13, %xmm13 vsubps %xmm14, %xmm13, %xmm8 vshufps $0x65, %xmm8, %xmm15, %xmm1 # xmm1 = xmm15[1,1],xmm8[2,1] vblendps $0x2, %xmm9, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm9[1],xmm1[2,3] vshufps $0x96, %xmm8, %xmm15, %xmm6 # xmm6 = xmm15[2,1],xmm8[1,2] vinsertps $0x90, %xmm9, %xmm6, %xmm6 # xmm6 = xmm6[0],xmm9[2],xmm6[2,3] vaddps %xmm6, %xmm1, %xmm7 vmulps %xmm6, %xmm1, %xmm1 vshufps $0x0, %xmm8, %xmm15, %xmm6 # xmm6 = xmm15[0,0],xmm8[0,0] vpaddd 0xec0(%rsp,%rdi), %xmm5, %xmm5 vinsertps $0x10, %xmm9, %xmm6, %xmm6 # xmm6 = xmm6[0],xmm9[0],xmm6[2,3] vmulps %xmm7, %xmm6, %xmm6 vaddps %xmm1, %xmm6, %xmm1 vmovdqa 0x60(%rsp), %xmm7 vpaddd %xmm7, %xmm5, %xmm6 vmovdqa (%rsp), %xmm9 vpsrld %xmm9, %xmm6, %xmm6 vpaddd 0xd0(%rsp,%rdi), %xmm7, %xmm7 vpsrad $0x1f, %xmm6, %xmm8 vmovdqa 0x50(%rsp), %xmm15 vpand %xmm15, %xmm8, %xmm8 vmovaps %xmm3, %xmm4 vmovdqa %xmm5, %xmm3 vmovaps %xmm2, %xmm5 vmovdqa %xmm0, %xmm2 vbroadcastss 0xb6ade4(%rip), %xmm0 # 0x1f20ec4 vpand %xmm0, %xmm6, %xmm6 vcvtdq2ps %xmm6, %xmm6 vaddps %xmm6, %xmm8, %xmm6 vpsrld %xmm9, %xmm7, %xmm7 vmulps %xmm6, %xmm1, %xmm1 vpsrad $0x1f, %xmm7, %xmm6 vpand %xmm6, %xmm15, %xmm6 vpand %xmm0, %xmm7, %xmm7 vmovdqa %xmm2, %xmm0 vmovaps %xmm5, %xmm2 vmovdqa %xmm3, %xmm5 vmovaps %xmm4, %xmm3 vmovaps 0x10(%rsp), %xmm4 vcvtdq2ps %xmm7, %xmm7 vaddps %xmm6, %xmm7, %xmm6 vmulps 0x10d0(%rsp,%rdi), %xmm6, %xmm6 vaddps %xmm6, %xmm1, %xmm1 vcmpltps %xmm10, %xmm1, %xmm6 vblendvps %xmm6, %xmm2, %xmm0, %xmm0 vminps %xmm10, %xmm1, %xmm10 vpsubd 0xb35cda(%rip), %xmm2, %xmm2 # 0x1eebe20 addq $0x60, %r8 addq $0x10, %rdi decq %rdx jne 0x13b6026 vmovaps %xmm10, 0x30(%rsp) vmovdqa %xmm0, 0x20(%rsp) vmovss 0xb358b5(%rip), %xmm0 # 0x1eeba20 xorl %ecx, %ecx vxorps %xmm1, %xmm1, %xmm1 xorl %edx, %edx vmovss 0xa0(%rsp,%rcx,4), %xmm2 vucomiss %xmm1, %xmm2 jne 0x13b6184 jnp 0x13b61a0 vmovss 0x30(%rsp,%rcx,4), %xmm2 vucomiss %xmm2, %xmm0 jbe 0x13b61a0 movl 0x20(%rsp,%rcx,4), %edi testl %edi, %edi je 0x13b61a0 vmovaps %xmm2, %xmm0 movl %ecx, %esi movl %edi, %edx incq %rcx cmpq $0x3, %rcx jne 0x13b6173 vmovss %xmm0, (%rbx) movl %esi, 0x4(%rbx) movl %edx, 0x8(%rbx) movl $0x0, 0xc(%rbx) movq %rax, 0x10(%rbx) vmovaps 0x90(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rbx) vmovaps 0xa0(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rbx) movq %rbx, %rax leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb358f0(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd6e9f9(%rip), %rsi # 0x2124c08 movq 0xd6e7b2(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13b6232 jmp 0x13b6247 jmp 0x13b622f movq %rax, %rbx leaq 0xc0(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef const&)::operator()(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::UserGeometry, embree::Object>::BuildRef const&) const
__forceinline operator Vec3fa () const { return Vec3fa(m128); }
vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vmovaps (%rsi), %xmm2 vmovaps 0x10(%rsi), %xmm3 vmovaps 0x20(%rsi), %xmm4 vmovaps 0x30(%rsi), %xmm5 vminps %xmm0, %xmm2, %xmm2 vmovaps %xmm2, (%rsi) vmaxps %xmm1, %xmm3, %xmm2 vmovaps %xmm2, 0x10(%rsi) vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm4, %xmm1 vmovaps %xmm1, 0x20(%rsi) vmaxps %xmm0, %xmm5, %xmm0 vmovaps %xmm0, 0x30(%rsi) incq 0x48(%rsi) retq
/embree[P]embree/kernels/bvh/../common/../../common/sys/../math/vec3fa.h
embree::NodeRefPtr<4> embree::avx::BVHBuilderBinnedOpenMergeSAH::build<embree::NodeRefPtr<4>, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(unsigned long)>(embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(unsigned long), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*, unsigned long, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&, embree::avx::GeneralBVHBuilder::Settings const&)
static ReductionTy build(CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, NodeOpenerFunc nodeOpenerFunc, ProgressMonitor progressMonitor, BuildRef* prims, const size_t extSize, const PrimInfo& pinfo, const Settings& settings) { typedef HeuristicArrayOpenMergeSAH<NodeOpenerFunc,BuildRef,NUM_OBJECT_BINS_HQ> Heuristic; Heuristic heuristic(nodeOpenerFunc,prims,settings.branchingFactor); return GeneralBVHBuilder::build<ReductionTy,Heuristic,Set,BuildRef>( heuristic, prims, PrimInfoExtRange(0,pinfo.size(),extSize,pinfo), createAlloc, createNode, updateNode, createLeaf, progressMonitor, settings); }
pushq %r14 pushq %rbx subq $0x98, %rsp movq %r8, %rax movq %rcx, %r10 movq %rsi, %r8 movq %rdi, %rcx movq 0xb0(%rsp), %rsi movq 0xb8(%rsp), %rbx leaq 0x10(%rsp), %r14 movq %rdx, (%r14) leaq 0x8(%rsp), %r11 movq %r10, (%r11) movq (%rbx), %rdx leaq 0x18(%rsp), %rdi movq %rax, (%rdi) movq %r14, 0x8(%rdi) movq %rdx, 0x10(%rdi) movq 0x48(%rsi), %r10 subq 0x40(%rsi), %r10 vmovaps (%rsi), %xmm0 leaq 0x30(%rsp), %rdx vmovaps %xmm0, (%rdx) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, 0x10(%rdx) vmovaps 0x20(%rsi), %xmm0 vmovaps %xmm0, 0x20(%rdx) vmovaps 0x30(%rsi), %xmm0 vmovaps %xmm0, 0x30(%rdx) movq $0x0, 0x40(%rdx) movq %r10, 0x48(%rdx) movq %r9, 0x50(%rdx) movq %rbx, (%rsp) movq %rax, %rsi movq %r11, %r9 callq 0x13bf040 addq $0x98, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::NodeRefPtr<4> embree::avx::GeneralBVHBuilder::build<embree::NodeRefPtr<4>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(unsigned long)>(embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, 32ul>&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*, embree::avx::PrimInfoExtRange const&, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(unsigned long) const&, embree::avx::GeneralBVHBuilder::Settings const&)
__noinline static ReductionTy build(Heuristic& heuristic, PrimRef* prims, const Set& set, CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, const ProgressMonitor& progressMonitor, const Settings& settings) { typedef BuildRecordT<Set,typename Heuristic::Split> BuildRecord; typedef BuilderT< BuildRecord, Heuristic, Set, PrimRef, ReductionTy, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, UpdateNodeFunc, CreateLeafFunc, DefaultCanCreateLeafFunc<PrimRef, Set>, DefaultCanCreateLeafSplitFunc<PrimRef, Set>, ProgressMonitor> Builder; /* instantiate builder */ Builder builder(prims, heuristic, createAlloc, createNode, updateNode, createLeaf, DefaultCanCreateLeafFunc<PrimRef, Set>(), DefaultCanCreateLeafSplitFunc<PrimRef, Set>(), progressMonitor, settings); /* build hierarchy */ BuildRecord record(1,set); const ReductionTy root = builder.recurse(record,nullptr,true); _mm_mfence(); // to allow non-temporal stores during build return root; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 movq %r9, %rax movq %r8, %r10 movq %rdx, %r14 movq %rdi, %rdx movq 0x160(%rsp), %r12 leaq 0x28(%rsp), %r11 movq %rcx, (%r11) subq $0x8, %rsp leaq 0x15(%rsp), %r13 leaq 0x38(%rsp), %r15 leaq 0xb0(%rsp), %rbx leaq 0x17(%rsp), %r8 leaq 0x16(%rsp), %r9 movq %rbx, %rdi movq %r11, %rcx pushq %r12 pushq %rax pushq %r13 pushq %r15 pushq %r10 callq 0x13bf14a addq $0x30, %rsp movq $0x1, (%r15) movb $0x0, 0x8(%r15) vmovaps (%r14), %xmm0 vmovaps %xmm0, 0x10(%r15) vmovaps 0x10(%r14), %xmm0 vmovaps %xmm0, 0x20(%r15) vmovaps 0x20(%r14), %xmm0 vmovaps %xmm0, 0x30(%r15) vmovaps 0x30(%r14), %xmm0 vmovaps %xmm0, 0x40(%r15) vmovaps 0x40(%r14), %xmm0 vmovaps %xmm0, 0x50(%r15) movq 0x50(%r14), %rax movq %rax, 0x60(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x10(%rsp) movq $0x0, 0x20(%rsp) subq $0x20, %rsp movq 0x40(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x30(%rsp), %xmm0 vmovups %xmm0, (%rsp) leaq 0x50(%rsp), %rsi movq %rbx, %rdi movl $0x1, %edx callq 0x13bf258 addq $0x20, %rsp mfence addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::GeneralBVHBuilder::BuilderT<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, embree::NodeRefPtr<4>, embree::FastAllocator::CachedAllocator, embree::BVHN<4>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Create2, embree::AABBNode_t<embree::NodeRefPtr<4>, 4>::Set2, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(unsigned long)>::createLargeLeaf(embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>> const&, embree::FastAllocator::CachedAllocator)
const ReductionTy createLargeLeaf(const BuildRecord& current, Allocator alloc) { /* this should never occur but is a fatal error */ if (current.depth > cfg.maxDepth) throw_RTCError(RTC_ERROR_UNKNOWN,"depth limit reached"); /* create leaf for few primitives */ if (current.prims.size() <= cfg.maxLeafSize && canCreateLeaf(prims,current.prims)) return createLeaf(prims,current.prims,alloc); /* fill all children by always splitting the largest one */ ReductionTy values[MAX_BRANCHING_FACTOR]; BuildRecord children[MAX_BRANCHING_FACTOR]; size_t numChildren = 1; children[0] = current; do { /* find best child with largest bounding box area */ size_t bestChild = -1; size_t bestSize = 0; for (size_t i=0; i<numChildren; i++) { /* ignore leaves as they cannot get split */ if (children[i].prims.size() <= cfg.maxLeafSize && canCreateLeaf(prims,children[i].prims)) continue; /* remember child with largest size */ if (children[i].prims.size() > bestSize) { bestSize = children[i].prims.size(); bestChild = i; } } if (bestChild == (size_t)-1) break; /*! split best child into left and right child */ BuildRecord left(current.depth+1); BuildRecord right(current.depth+1); if (!canCreateLeaf(prims,children[bestChild].prims)) { canCreateLeafSplit(prims,children[bestChild].prims,left.prims,right.prims); } else { heuristic.splitFallback(children[bestChild].prims,left.prims,right.prims); } /* add new children left and right */ children[bestChild] = children[numChildren-1]; children[numChildren-1] = left; children[numChildren+0] = right; numChildren++; } while (numChildren < cfg.branchingFactor); /* set barrier for primrefarrayalloc */ if (unlikely(current.size() > cfg.primrefarrayalloc)) for (size_t i=0; i<numChildren; i++) children[i].alloc_barrier = children[i].size() <= cfg.primrefarrayalloc; /* create node */ auto node = createNode(children,numChildren,alloc); /* recurse into each child and perform reduction */ for (size_t i=0; i<numChildren; i++) values[i] = createLargeLeaf(children[i],alloc); /* perform reduction */ return updateNode(current,children,node,values,numChildren); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x898, %rsp # imm = 0x898 movq (%rsi), %rax cmpq 0x8(%rdi), %rax ja 0x13c09e7 movq %rdi, %rbp movq 0x50(%rsi), %rax movq 0x58(%rsi), %rcx subq %rax, %rcx cmpq 0x20(%rdi), %rcx jbe 0x13c06e5 leaq 0x8d0(%rsp), %r13 leaq 0x50(%rsi), %rax movb 0x8(%rsi), %cl leaq 0x1e8(%rsp), %rdx movb %cl, -0x50(%rdx) movq (%rsi), %rcx movq %rcx, -0x58(%rdx) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, -0x48(%rdx) vmovaps 0x20(%rsi), %xmm0 vmovaps %xmm0, -0x38(%rdx) vmovaps 0x30(%rsi), %xmm0 vmovaps %xmm0, -0x28(%rdx) movq %rsi, 0x60(%rsp) vmovaps 0x40(%rsi), %xmm0 vmovaps %xmm0, -0x18(%rdx) vmovaps (%rax), %xmm0 vmovaps %xmm0, -0x8(%rdx) movq 0x10(%rax), %rax movq %rax, 0x8(%rdx) movl $0x1, %r14d leaq 0x90(%rsp), %rdi movq %rbp, 0x58(%rsp) testq %r14, %r14 je 0x13c003b movq 0x20(%rbp), %rax movq $-0x1, %r12 leaq 0x1e8(%rsp), %rcx xorl %edx, %edx xorl %esi, %esi movq (%rcx), %r8 subq -0x8(%rcx), %r8 cmpq %rax, %r8 jbe 0x13c002d cmpq %rsi, %r8 cmovaq %rdx, %r12 cmovaq %r8, %rsi incq %rdx addq $0x70, %rcx cmpq %rdx, %r14 jne 0x13c0016 jmp 0x13c0042 movq $-0x1, %r12 cmpq $-0x1, %r12 je 0x13c05dd imulq $0x70, %r12, %rdx movq 0x48(%rbp), %rax movq 0x1e0(%rsp,%rdx), %r10 movq 0x1e8(%rsp,%rdx), %rbx leaq (%rbx,%r10), %r15 shrq %r15 movq %r15, %rcx subq %r10, %rcx jbe 0x13c00f7 movq (%rax), %rsi movq %r10, %r8 shlq $0x6, %r8 addq %r8, %rsi addq $0x10, %rsi vbroadcastss 0xb2caf3(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, %xmm2 vbroadcastss 0xb2b982(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, %xmm3 vmovaps %xmm0, %xmm4 vmovaps %xmm1, %xmm5 vmovaps -0x10(%rsi), %xmm0 vmovaps (%rsi), %xmm1 vminps %xmm0, %xmm5, %xmm5 vmaxps %xmm1, %xmm4, %xmm4 vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm3, %xmm3 vmaxps %xmm0, %xmm2, %xmm2 addq $0x40, %rsi decq %rcx jne 0x13c00aa vmovaps %xmm5, 0x140(%rsp) vmovaps %xmm4, 0x150(%rsp) vmovaps %xmm3, 0x160(%rsp) movq %r15, %rsi subq %r10, %rsi vmovaps %xmm2, %xmm0 jmp 0x13c0126 vbroadcastss 0xb2b920(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, 0x140(%rsp) vbroadcastss 0xb2ca72(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x150(%rsp) vmovaps %xmm1, 0x160(%rsp) xorl %esi, %esi vmovaps %xmm0, 0x180(%rsp) movq %rbx, %rcx subq %r15, %rcx movq %r10, 0x80(%rsp) jbe 0x13c01c8 movq (%rax), %r9 movq %r15, %r8 shlq $0x6, %r8 addq %r8, %r9 addq $0x10, %r9 movq %rcx, %r8 vbroadcastss 0xb2ca24(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, %xmm2 vbroadcastss 0xb2b8b3(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, %xmm3 vmovaps %xmm0, %xmm4 vmovaps %xmm1, %xmm5 vmovaps -0x10(%r9), %xmm0 vmovaps (%r9), %xmm1 vminps %xmm0, %xmm5, %xmm5 vmaxps %xmm1, %xmm4, %xmm4 vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm3, %xmm3 vmaxps %xmm0, %xmm2, %xmm2 addq $0x40, %r9 decq %r8 jne 0x13c0179 vmovaps %xmm5, 0x110(%rsp) vmovaps %xmm4, 0x120(%rsp) vmovaps %xmm3, 0x130(%rsp) movq %rbx, %r9 subq %r15, %r9 vmovaps %xmm2, %xmm0 jmp 0x13c01f8 vbroadcastss 0xb2b84f(%rip), %xmm1 # 0x1eeba20 vmovaps %xmm1, 0x110(%rsp) vbroadcastss 0xb2c9a1(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x120(%rsp) vmovaps %xmm1, 0x130(%rsp) xorl %r9d, %r9d movq 0x60(%rsp), %r8 movq (%r8), %r8 movq %r8, 0x78(%rsp) leaq (%rsp,%rdx), %rbp addq $0x190, %rbp # imm = 0x190 movq 0x60(%rbp), %r13 subq %rbx, %r13 vmovaps %xmm0, 0x170(%rsp) jne 0x13c0230 movq %r15, %r10 movq %rbx, %rcx movq %r15, %rax jmp 0x13c046a testq %rsi, %rsi js 0x13c023c vcvtsi2ss %rsi, %xmm6, %xmm0 jmp 0x13c0255 movq %rsi, %rdx shrq %rdx movl %esi, %r8d andl $0x1, %r8d orq %rdx, %r8 vcvtsi2ss %r8, %xmm6, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rsi, %r9 js 0x13c0261 vcvtsi2ss %r9, %xmm6, %xmm1 jmp 0x13c0277 movq %r9, %rdx shrq %rdx andl $0x1, %r9d orq %rdx, %r9 vcvtsi2ss %r9, %xmm6, %xmm1 vaddss %xmm1, %xmm1, %xmm1 testq %r13, %r13 js 0x13c0283 vcvtsi2ss %r13, %xmm6, %xmm2 jmp 0x13c029b movq %r13, %rdx shrq %rdx movl %r13d, %esi andl $0x1, %esi orq %rdx, %rsi vcvtsi2ss %rsi, %xmm6, %xmm2 vaddss %xmm2, %xmm2, %xmm2 vdivss %xmm1, %xmm0, %xmm0 vmulss %xmm2, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rsi movq %rsi, %r8 vsubss 0xb2bb8b(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rdx sarq $0x3f, %r8 andq %r8, %rdx orq %rsi, %rdx cmpq %r13, %rdx cmovaeq %r13, %rdx subq %rdx, %r13 leaq (%rdx,%r15), %rsi movq %rsi, 0x68(%rsp) addq %rbx, %r13 movq %rdx, 0x70(%rsp) movq %rcx, 0x88(%rsp) testq %rdx, %rdx je 0x13c03a5 cmpq %rcx, %rdx jae 0x13c03ad movq %rax, 0x30(%rsp) leaq 0x88(%rsp), %rax movq %rax, 0x38(%rsp) movb $0x1, 0x9c(%rsp) leaq 0xb0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0xd0(%rsp) movb $0x4, 0x9d(%rsp) callq 0x6a660 movq 0x68(%rsp), %rax movq %rax, 0x40(%rsp) movq %r15, 0x48(%rsp) movq $0x40, 0x50(%rsp) leaq 0x30(%rsp), %rax movq %rax, 0x28(%rsp) leaq 0x40(%rsp), %rdi leaq 0x28(%rsp), %rsi leaq 0x27(%rsp), %rdx leaq 0x90(%rsp), %rcx callq 0x13c4be2 leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13c09b5 leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x13c0448 movq %r15, %rax jmp 0x13c045f movq %rax, 0x30(%rsp) leaq 0x70(%rsp), %rax movq %rax, 0x38(%rsp) movb $0x1, 0x9c(%rsp) leaq 0xb0(%rsp), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x8, 0xd0(%rsp) movb $0x4, 0x9d(%rsp) callq 0x6a660 movq %rbx, 0x40(%rsp) movq %r15, 0x48(%rsp) movq $0x40, 0x50(%rsp) leaq 0x30(%rsp), %rax movq %rax, 0x28(%rsp) leaq 0x40(%rsp), %rdi leaq 0x28(%rsp), %rsi leaq 0x27(%rsp), %rdx leaq 0x90(%rsp), %rcx callq 0x13c52ee leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13c0983 leaq 0x90(%rsp), %rdi callq 0x6aab0 leaq 0x90(%rsp), %rdi movq 0x70(%rsp), %rcx leaq (%rcx,%r15), %rax addq %rcx, %rbx addq %rcx, %r13 movq %rbx, %rcx movq %r13, %rbx movq 0x68(%rsp), %r10 movq 0x78(%rsp), %r9 incq %r9 imulq $0x70, %r14, %rdx movb 0x128(%rsp,%rdx), %sil movb %sil, 0x8(%rbp) movq 0x120(%rsp,%rdx), %rsi movq %rsi, (%rbp) vmovaps 0x130(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x10(%rbp) vmovaps 0x140(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x20(%rbp) vmovaps 0x150(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x30(%rbp) vmovaps 0x160(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x40(%rbp) movq 0x180(%rsp,%rdx), %rsi movq %rsi, 0x60(%rbp) vmovaps 0x170(%rsp,%rdx), %xmm0 vmovaps %xmm0, 0x50(%rbp) movq %r9, 0x120(%rsp,%rdx) movl $0x0, %esi movb %sil, 0x128(%rsp,%rdx) vmovaps 0x140(%rsp), %xmm0 vmovaps %xmm0, 0x130(%rsp,%rdx) vmovaps 0x150(%rsp), %xmm0 vmovaps %xmm0, 0x140(%rsp,%rdx) vmovaps 0x160(%rsp), %xmm0 vmovaps %xmm0, 0x150(%rsp,%rdx) vmovaps 0x180(%rsp), %xmm0 vmovaps %xmm0, 0x160(%rsp,%rdx) movq 0x80(%rsp), %r8 movq %r8, 0x170(%rsp,%rdx) movq %r15, 0x178(%rsp,%rdx) movq %r10, 0x180(%rsp,%rdx) movq %r9, 0x190(%rsp,%rdx) movb %sil, 0x198(%rsp,%rdx) vmovaps 0x110(%rsp), %xmm0 vmovaps %xmm0, 0x1a0(%rsp,%rdx) vmovaps 0x120(%rsp), %xmm0 vmovaps %xmm0, 0x1b0(%rsp,%rdx) vmovaps 0x130(%rsp), %xmm0 vmovaps %xmm0, 0x1c0(%rsp,%rdx) vmovaps 0x170(%rsp), %xmm0 vmovaps %xmm0, 0x1d0(%rsp,%rdx) movq %rax, 0x1e0(%rsp,%rdx) movq %rcx, 0x1e8(%rsp,%rdx) movq %rbx, 0x1f0(%rsp,%rdx) incq %r14 movq 0x58(%rsp), %rbp leaq 0x8d0(%rsp), %r13 cmpq $-0x1, %r12 je 0x13c05ed cmpq (%rbp), %r14 jb 0x13bfffa movq 0x60(%rsp), %rcx movq 0x58(%rcx), %rax subq 0x50(%rcx), %rax movq 0x38(%rbp), %rdx cmpq %rdx, %rax ja 0x13c08d6 movq (%r13), %r15 movq 0x8(%r13), %rbx movq $0x80, 0x28(%rsp) movq (%rbx), %r12 movq 0x8(%r12), %rax cmpq %rax, %r15 je 0x13c0792 movq %r12, 0x40(%rsp) movb $0x1, 0x48(%rsp) movq %r12, %rdi callq 0x1ee7bb6 movq 0x8(%r12), %rax testq %rax, %rax je 0x13c06a3 movq 0xa8(%r12), %rax addq 0x68(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%r12), %rax addq 0x98(%r12), %rax movq 0x50(%r12), %rcx addq 0x90(%r12), %rcx subq %rcx, %rax movq 0x8(%r12), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%r12), %rax addq 0x70(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%r12) vmovups %ymm0, 0x48(%r12) testq %r15, %r15 je 0x13c06f7 movq 0x10(%r15), %rax movq %rax, 0x60(%r12) vmovups %ymm0, 0x88(%r12) vmovups %ymm0, 0x98(%r12) movq 0x10(%r15), %rax movq %rax, 0xa0(%r12) jmp 0x13c070b movq 0x40(%rbp), %rcx shlq $0x6, %rax movq 0x20(%rcx,%rax), %r12 jmp 0x13c08be vmovups %ymm0, 0x98(%r12) vmovups %ymm0, 0x88(%r12) movq %r15, %rax xchgq %rax, 0x8(%r12) movq %r12, 0x30(%rsp) leaq 0xd66159(%rip), %rdi # 0x2126878 movq %rdi, 0x90(%rsp) movb $0x1, 0x98(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%r15), %rsi cmpq 0x140(%r15), %rsi je 0x13c0759 movq 0x30(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%r15) jmp 0x13c076a leaq 0x130(%r15), %rdi leaq 0x30(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x98(%rsp) jne 0x13c0781 movq 0x90(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0x48(%rsp) jne 0x13c0792 movq 0x40(%rsp), %rdi callq 0x1ee7c24 movq 0x28(%rsp), %rax addq %rax, 0x28(%rbx) movq 0x10(%rbx), %rcx movl %ecx, %edx negl %edx andl $0xf, %edx leaq (%rcx,%rax), %r12 addq %rdx, %r12 movq %r12, 0x10(%rbx) cmpq 0x18(%rbx), %r12 ja 0x13c08eb addq %rdx, 0x30(%rbx) subq %rax, %r12 addq 0x8(%rbx), %r12 vbroadcastss 0xb2b251(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x60(%r12) vmovaps %xmm0, 0x40(%r12) vmovaps %xmm0, 0x20(%r12) vbroadcastss 0xb2c397(%rip), %xmm0 # 0x1eecb84 vmovaps %xmm0, 0x70(%r12) vmovaps %xmm0, 0x50(%r12) vmovaps %xmm0, 0x30(%r12) vbroadcastsd 0xb6166d(%rip), %ymm0 # 0x1f21e78 vmovups %ymm0, (%r12) testq %r14, %r14 je 0x13c0863 leaq 0x1b0(%rsp), %rax xorl %ecx, %ecx vmovaps -0x10(%rax), %xmm0 vmovaps (%rax), %xmm1 vmovss %xmm0, 0x20(%r12,%rcx,4) vextractps $0x1, %xmm0, 0x40(%r12,%rcx,4) vextractps $0x2, %xmm0, 0x60(%r12,%rcx,4) vmovss %xmm1, 0x30(%r12,%rcx,4) vextractps $0x1, %xmm1, 0x50(%r12,%rcx,4) vextractps $0x2, %xmm1, 0x70(%r12,%rcx,4) incq %rcx addq $0x70, %rax cmpq %rcx, %r14 jne 0x13c0820 testq %r14, %r14 je 0x13c08be leaq 0x190(%rsp), %r15 xorl %ebx, %ebx movq 0x10(%r13), %rax movq %rax, 0x10(%rsp) vmovups (%r13), %xmm0 vmovups %xmm0, (%rsp) movq %rbp, %rdi movq %r15, %rsi vzeroupper callq 0x13bff52 movq %rax, 0x90(%rsp,%rbx,8) incq %rbx addq $0x70, %r15 cmpq %rbx, %r14 jne 0x13c0872 xorl %eax, %eax movq 0x90(%rsp,%rax,8), %rcx movq %rcx, (%r12,%rax,8) incq %rax cmpq %rax, %r14 jne 0x13c08aa movq %r12, %rax addq $0x898, %rsp # imm = 0x898 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq leaq 0x190(%rsp), %rsi movq %r14, %rdi callq 0x1ff4bf jmp 0x13c0607 movq %rcx, 0x10(%rbx) shlq $0x2, %rax movq 0x20(%rbx), %rcx cmpq %rcx, %rax jbe 0x13c0918 leaq 0x28(%rsp), %rsi movl $0x40, %edx movq %r15, %rdi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 jmp 0x13c07c6 leaq 0x90(%rsp), %r13 movq %rcx, (%r13) movl $0x40, %edx movq %r15, %rdi movq %r13, %rsi movl $0x1, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbx) movq 0x18(%rbx), %rax subq 0x10(%rbx), %rax addq 0x30(%rbx), %rax movq %rax, 0x30(%rbx) movq $0x0, 0x10(%rbx) movq (%r13), %rcx movq %rcx, 0x18(%rbx) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbx) cmpq %rcx, %rdx ja 0x13c0a6a movq %rax, 0x30(%rbx) leaq 0x8d0(%rsp), %r13 jmp 0x13c07c6 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb2b159(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd64262(%rip), %rsi # 0x2124c08 movq 0xd6401b(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb2b127(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd64230(%rip), %rsi # 0x2124c08 movq 0xd63fe9(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x30, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0x1a0(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb2c231(%rip), %rsi # 0x1eecc38 leaq 0xb2c23d(%rip), %rdx # 0x1eecc4b leaq 0x190(%rsp), %rdi callq 0x8d7230 leaq 0xd40e8e(%rip), %rax # 0x21018b0 movq %rax, (%r14) movl $0x1, 0x8(%r14) leaq 0x10(%r14), %rdi movq %r14, %rax addq $0x20, %rax movq %rax, 0x10(%r14) movq 0x190(%rsp), %rsi movq 0x198(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd40df5(%rip), %rsi # 0x2101850 leaq -0xae9c46(%rip), %rdx # 0x8d6e1c movq %r14, %rdi callq 0x6a5d0 xorl %ebp, %ebp movq %rbp, 0x10(%rbx) movq 0x20(%rbx), %rax leaq 0x90(%rsp), %r13 movq %rax, (%r13) movl $0x40, %edx movq %r15, %rdi movq %r13, %rsi xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbx) movq 0x18(%rbx), %rax subq 0x10(%rbx), %rax addq 0x30(%rbx), %rax movq %rax, 0x30(%rbx) movq %rbp, 0x10(%rbx) movq (%r13), %rcx movq %rcx, 0x18(%rbx) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbx) cmpq %rcx, %rdx ja 0x13c0ad1 movq %rax, 0x30(%rbx) movq 0x58(%rsp), %rbp jmp 0x13c0976 movq $0x0, 0x10(%rbx) xorl %r12d, %r12d jmp 0x13c0ac7 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x8d6eda jmp 0x13c0b26 jmp 0x13c0b99 jmp 0x13c0b99 movq %rax, %rbx xorl %ebp, %ebp jmp 0x13c0b0f movq %rax, %rbx movq %r14, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x190(%rsp), %rdi cmpq %r15, %rdi je 0x13c0b38 callq 0x6a4f0 jmp 0x13c0b38 movq %rax, %rbx leaq 0x40(%rsp), %rdi callq 0x8d6eda jmp 0x13c0b8f movq %rax, %rbx movb $0x1, %bpl testb %bpl, %bpl je 0x13c0b8f movq %r14, %rdi callq 0x6a8a0 jmp 0x13c0b8f jmp 0x13c0b69 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13c0b6c jmp 0x13c0b7f movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13c0b82 jmp 0x13c0b99 jmp 0x13c0b69 movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 jmp 0x13c0b8f jmp 0x13c0b99 jmp 0x13c0b7f movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 jmp 0x13c0b99 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, 32ul>::sequential_object_find(embree::avx::PrimInfoExtRange const&, unsigned long)
__noinline const Split sequential_object_find(const PrimInfoExtRange& set, const size_t logBlockSize) { Binner binner(empty); const BinMapping<OBJECT_BINS> mapping(set.centBounds); binner.bin(prims0,set.begin(),set.end(),mapping); return binner.best(mapping,logBlockSize); }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x40, %rsp subq $0x1300, %rsp # imm = 0x1300 movq %rdi, %rax movl $0xc00, %edi # imm = 0xC00 leaq 0x110(%rsp), %r8 vbroadcastss 0xb28ea3(%rip), %xmm2 # 0x1eeba20 vbroadcastss 0xb29ffe(%rip), %xmm3 # 0x1eecb84 vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm2, -0x10(%r8) vmovaps %xmm3, (%r8) vmovaps %xmm2, -0x30(%r8) vmovaps %xmm3, -0x20(%r8) vmovaps %xmm2, -0x50(%r8) vmovaps %xmm3, -0x40(%r8) vmovaps %xmm0, 0xc0(%rsp,%rdi) addq $0x10, %rdi addq $0x60, %r8 cmpq $0xe00, %rdi # imm = 0xE00 jne 0x13c2b8a movq $0x20, 0x90(%rsp) vmovaps 0x20(%rdx), %xmm0 vmovaps 0x30(%rdx), %xmm1 vsubps %xmm0, %xmm1, %xmm1 vbroadcastss 0xb5f286(%rip), %xmm4 # 0x1f21e70 vmaxps %xmm1, %xmm4, %xmm1 vcmpnleps %xmm4, %xmm1, %xmm4 vbroadcastss 0xb75e2c(%rip), %xmm5 # 0x1f38a28 vdivps %xmm1, %xmm5, %xmm1 vandps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0xb0(%rsp) vmovaps %xmm0, 0xa0(%rsp) movq 0x40(%rdx), %rdi movq 0x48(%rdx), %r8 subq %rdi, %r8 je 0x13c2fa5 movq (%rsi), %rdx shlq $0x6, %rdi movq %r8, %r9 decq %r9 movq %rdx, 0x30(%rsp) je 0x13c2e83 leaq 0xcc0(%rsp), %r10 leaq 0xcc4(%rsp), %r11 leaq 0xcc8(%rsp), %rbx leaq (%rdi,%rdx), %r14 addq $0x5c, %r14 xorl %esi, %esi vbroadcastss 0xb75dc2(%rip), %xmm4 # 0x1f38a2c vxorps %xmm5, %xmm5, %xmm5 vmovaps -0x5c(%r14), %xmm9 vmovaps -0x4c(%r14), %xmm8 vmovaps -0x1c(%r14), %xmm7 vmovaps -0xc(%r14), %xmm6 vaddps %xmm8, %xmm9, %xmm10 vsubps %xmm0, %xmm10, %xmm10 vmulps %xmm1, %xmm10, %xmm10 vroundps $0x1, %xmm10, %xmm10 vcvtps2dq %xmm10, %xmm10 vpminsd %xmm4, %xmm10, %xmm10 vpmaxsd %xmm5, %xmm10, %xmm11 vaddps %xmm6, %xmm7, %xmm10 vsubps %xmm0, %xmm10, %xmm10 vmulps %xmm1, %xmm10, %xmm10 vroundps $0x1, %xmm10, %xmm10 vcvtps2dq %xmm10, %xmm10 vmovd %xmm11, %r15d leaq (%r15,%r15,2), %r13 shll $0x5, %r13d vmovaps 0xc0(%rsp,%r13), %xmm12 vmovaps 0xd0(%rsp,%r13), %xmm13 vminps %xmm9, %xmm12, %xmm12 vmovaps %xmm12, 0xc0(%rsp,%r13) vmaxps %xmm8, %xmm13, %xmm12 vpextrd $0x1, %xmm11, %r12d vmovaps %xmm12, 0xd0(%rsp,%r13) leaq (%r12,%r12,2), %r13 shll $0x5, %r13d vmovaps 0xe0(%rsp,%r13), %xmm12 vmovaps 0xf0(%rsp,%r13), %xmm13 vminps %xmm9, %xmm12, %xmm12 vmovaps %xmm12, 0xe0(%rsp,%r13) vmaxps %xmm8, %xmm13, %xmm12 vmovaps %xmm12, 0xf0(%rsp,%r13) vpextrd $0x2, %xmm11, %r13d leaq (,%r13,2), %rdx addq %r13, %rdx shll $0x5, %edx vmovaps 0x100(%rsp,%rdx), %xmm11 vmovaps 0x110(%rsp,%rdx), %xmm12 vminps %xmm9, %xmm11, %xmm9 vmovaps %xmm9, 0x100(%rsp,%rdx) vmaxps %xmm8, %xmm12, %xmm8 vmovaps %xmm8, 0x110(%rsp,%rdx) movl -0x40(%r14), %edx shlq $0x4, %r15 addl %edx, (%r10,%r15) shlq $0x4, %r12 addl %edx, (%r11,%r12) shlq $0x4, %r13 addl %edx, (%rbx,%r13) vpminsd %xmm4, %xmm10, %xmm8 vpmaxsd %xmm5, %xmm8, %xmm8 vmovd %xmm8, %r15d leaq (%r15,%r15,2), %rdx shll $0x5, %edx vmovaps 0xc0(%rsp,%rdx), %xmm9 vmovaps 0xd0(%rsp,%rdx), %xmm10 vminps %xmm7, %xmm9, %xmm9 vmovaps %xmm9, 0xc0(%rsp,%rdx) vmaxps %xmm6, %xmm10, %xmm9 vpextrd $0x1, %xmm8, %r12d vmovaps %xmm9, 0xd0(%rsp,%rdx) leaq (%r12,%r12,2), %rdx shll $0x5, %edx vmovaps 0xe0(%rsp,%rdx), %xmm9 vmovaps 0xf0(%rsp,%rdx), %xmm10 vminps %xmm7, %xmm9, %xmm9 vmovaps %xmm9, 0xe0(%rsp,%rdx) vmaxps %xmm6, %xmm10, %xmm9 vmovaps %xmm9, 0xf0(%rsp,%rdx) vpextrd $0x2, %xmm8, %edx leaq (%rdx,%rdx,2), %r13 shll $0x5, %r13d vmovaps 0x100(%rsp,%r13), %xmm8 vmovaps 0x110(%rsp,%r13), %xmm9 vminps %xmm7, %xmm8, %xmm7 vmovaps %xmm7, 0x100(%rsp,%r13) vmaxps %xmm6, %xmm9, %xmm6 vmovaps %xmm6, 0x110(%rsp,%r13) movl (%r14), %r13d shlq $0x4, %r15 addl %r13d, (%r10,%r15) shlq $0x4, %r12 addl %r13d, (%r11,%r12) shlq $0x4, %rdx addl %r13d, (%rbx,%rdx) addq $0x2, %rsi subq $-0x80, %r14 cmpq %r9, %rsi jb 0x13c2c6e jmp 0x13c2e85 xorl %esi, %esi cmpq %r8, %rsi jae 0x13c2fa5 movq 0x30(%rsp), %rdx addq %rdi, %rdx shlq $0x6, %rsi vmovaps (%rdx,%rsi), %xmm5 vmovaps 0x10(%rdx,%rsi), %xmm4 vaddps %xmm4, %xmm5, %xmm6 vsubps %xmm0, %xmm6, %xmm6 vmulps %xmm6, %xmm1, %xmm6 vroundps $0x1, %xmm6, %xmm6 vcvtps2dq %xmm6, %xmm6 vbroadcastss 0xb75b68(%rip), %xmm7 # 0x1f38a2c vpminsd %xmm7, %xmm6, %xmm6 vpxor %xmm7, %xmm7, %xmm7 vpmaxsd %xmm7, %xmm6, %xmm6 movl 0x1c(%rdx,%rsi), %edx vmovd %xmm6, %edi leaq (%rdi,%rdi,2), %rsi shlq $0x4, %rdi addl %edx, 0xcc0(%rsp,%rdi) shlq $0x5, %rsi vmovaps 0xc0(%rsp,%rsi), %xmm7 vmovaps 0xd0(%rsp,%rsi), %xmm8 vminps %xmm5, %xmm7, %xmm7 vmovaps %xmm7, 0xc0(%rsp,%rsi) vmaxps %xmm4, %xmm8, %xmm7 vmovaps %xmm7, 0xd0(%rsp,%rsi) vpextrd $0x1, %xmm6, %edi movl %edi, %esi shll $0x4, %esi addl %edx, 0xcc4(%rsp,%rsi) leaq (%rdi,%rdi,2), %rsi shlq $0x5, %rsi vmovaps 0xe0(%rsp,%rsi), %xmm7 vmovaps 0xf0(%rsp,%rsi), %xmm8 vminps %xmm5, %xmm7, %xmm7 vmovaps %xmm7, 0xe0(%rsp,%rsi) vmaxps %xmm4, %xmm8, %xmm7 vmovaps %xmm7, 0xf0(%rsp,%rsi) vpextrd $0x2, %xmm6, %edi movl %edi, %esi shll $0x4, %esi addl %edx, 0xcc8(%rsp,%rsi) leaq (%rdi,%rdi,2), %rdx shlq $0x5, %rdx vmovaps 0x100(%rsp,%rdx), %xmm6 vmovaps 0x110(%rsp,%rdx), %xmm7 vminps %xmm5, %xmm6, %xmm5 vmovaps %xmm5, 0x100(%rsp,%rdx) vmaxps %xmm4, %xmm7, %xmm4 vmovaps %xmm4, 0x110(%rsp,%rdx) vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x70(%rsp) vxorps %xmm4, %xmm4, %xmm4 movl $0xbf0, %edx # imm = 0xBF0 movl $0xdf0, %esi # imm = 0xDF0 vmovaps %xmm3, %xmm5 vmovaps %xmm2, %xmm6 vmovaps %xmm3, %xmm7 vmovaps %xmm2, %xmm8 vpaddd 0xc0(%rsp,%rsi), %xmm4, %xmm4 vminps 0x70(%rsp,%rdx), %xmm2, %xmm2 vmaxps 0x80(%rsp,%rdx), %xmm3, %xmm3 vmovdqa %xmm4, 0x4c0(%rsp,%rsi) vsubps %xmm2, %xmm3, %xmm9 vminps 0x90(%rsp,%rdx), %xmm6, %xmm6 vmaxps 0xa0(%rsp,%rdx), %xmm5, %xmm5 vsubps %xmm6, %xmm5, %xmm10 vinsertps $0x4c, %xmm9, %xmm10, %xmm11 # xmm11 = xmm9[1],xmm10[1],zero,zero vshufpd $0x1, %xmm9, %xmm9, %xmm12 # xmm12 = xmm9[1,0] vinsertps $0x9c, %xmm10, %xmm12, %xmm12 # xmm12 = xmm12[0],xmm10[2],zero,zero vaddps %xmm12, %xmm11, %xmm13 vmulps %xmm12, %xmm11, %xmm11 vunpcklps %xmm10, %xmm9, %xmm9 # xmm9 = xmm9[0],xmm10[0],xmm9[1],xmm10[1] vmulps %xmm13, %xmm9, %xmm9 vaddps %xmm11, %xmm9, %xmm9 vmovlps %xmm9, 0x2c0(%rsp,%rsi) vminps 0xb0(%rsp,%rdx), %xmm8, %xmm8 vmaxps 0xc0(%rsp,%rdx), %xmm7, %xmm7 vsubps %xmm8, %xmm7, %xmm9 vmovshdup %xmm9, %xmm10 # xmm10 = xmm9[1,1,3,3] vshufpd $0x1, %xmm9, %xmm9, %xmm11 # xmm11 = xmm9[1,0] vaddss %xmm11, %xmm10, %xmm12 vmulss %xmm11, %xmm10, %xmm10 vmulss %xmm12, %xmm9, %xmm9 vaddss %xmm10, %xmm9, %xmm9 vmovss %xmm9, 0x2c8(%rsp,%rsi) movl $0x0, 0x2cc(%rsp,%rsi) addq $-0x60, %rdx addq $-0x10, %rsi cmpq $0x50, %rdx jne 0x13c2fcf movl $0xffffffff, %edx # imm = 0xFFFFFFFF shll %cl, %edx notl %edx vmovd %edx, %xmm2 vpshufd $0x0, %xmm2, %xmm0 # xmm0 = xmm2[0,0,0,0] vmovdqa %xmm0, 0x30(%rsp) vbroadcastss 0xb28965(%rip), %xmm3 # 0x1eeba20 vmovaps %xmm3, 0x50(%rsp) vpxor %xmm4, %xmm4, %xmm4 vmovdqa %xmm4, 0x40(%rsp) vmovddup 0xb5de15(%rip), %xmm5 # xmm5 = mem[0,0] vbroadcastss 0xb29aa8(%rip), %xmm6 # 0x1eecb84 movl $0xc00, %edx # imm = 0xC00 leaq 0x110(%rsp), %rsi vmovddup 0xb5ed8f(%rip), %xmm0 # xmm0 = mem[0,0] vmovaps %xmm0, 0x80(%rsp) vmovaps %xmm6, %xmm10 vmovaps %xmm3, %xmm11 vmovaps %xmm6, %xmm12 vmovaps %xmm3, %xmm13 vmovaps %xmm3, %xmm14 vxorps %xmm15, %xmm15, %xmm15 vminps -0x50(%rsi), %xmm3, %xmm3 vmaxps -0x40(%rsi), %xmm6, %xmm6 vsubps %xmm3, %xmm6, %xmm1 vminps -0x30(%rsi), %xmm11, %xmm11 vmaxps -0x20(%rsi), %xmm10, %xmm10 vminps -0x10(%rsi), %xmm13, %xmm13 vmaxps (%rsi), %xmm12, %xmm12 vsubps %xmm11, %xmm10, %xmm0 vsubps %xmm13, %xmm12, %xmm9 vshufps $0x65, %xmm9, %xmm1, %xmm8 # xmm8 = xmm1[1,1],xmm9[2,1] vblendps $0x2, %xmm0, %xmm8, %xmm8 # xmm8 = xmm8[0],xmm0[1],xmm8[2,3] vshufps $0x96, %xmm9, %xmm1, %xmm2 # xmm2 = xmm1[2,1],xmm9[1,2] vinsertps $0x90, %xmm0, %xmm2, %xmm2 # xmm2 = xmm2[0],xmm0[2],xmm2[2,3] vaddps %xmm2, %xmm8, %xmm7 vmulps %xmm2, %xmm8, %xmm2 vshufps $0x0, %xmm9, %xmm1, %xmm1 # xmm1 = xmm1[0,0],xmm9[0,0] vpaddd 0xc0(%rsp,%rdx), %xmm4, %xmm4 vinsertps $0x10, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0],xmm1[2,3] vmulps %xmm7, %xmm0, %xmm0 vaddps %xmm2, %xmm0, %xmm0 vmovdqa 0x30(%rsp), %xmm7 vpaddd %xmm7, %xmm4, %xmm1 vmovd %ecx, %xmm2 vpsrld %xmm2, %xmm1, %xmm1 vpaddd 0x4d0(%rsp,%rdx), %xmm7, %xmm7 vpsrld %xmm2, %xmm7, %xmm2 vpsrad $0x1f, %xmm1, %xmm7 vmovdqa 0x80(%rsp), %xmm8 vpand %xmm7, %xmm8, %xmm7 vbroadcastss 0xb5dd0f(%rip), %xmm9 # 0x1f20ec4 vpand %xmm1, %xmm9, %xmm1 vcvtdq2ps %xmm1, %xmm1 vaddps %xmm7, %xmm1, %xmm1 vmulps %xmm1, %xmm0, %xmm0 vpsrad $0x1f, %xmm2, %xmm1 vpand %xmm1, %xmm8, %xmm1 vpand %xmm2, %xmm9, %xmm2 vcvtdq2ps %xmm2, %xmm2 vaddps %xmm1, %xmm2, %xmm1 vmulps 0x2d0(%rsp,%rdx), %xmm1, %xmm1 vaddps %xmm1, %xmm0, %xmm0 vcmpltps %xmm14, %xmm0, %xmm1 vblendvps %xmm1, %xmm5, %xmm15, %xmm15 vminps %xmm14, %xmm0, %xmm14 vpsubd 0xb28c20(%rip), %xmm5, %xmm5 # 0x1eebe20 addq $0x60, %rsi addq $0x10, %rdx cmpq $0xdf0, %rdx # imm = 0xDF0 jne 0x13c3113 vmovaps %xmm14, 0x50(%rsp) vmovaps %xmm15, 0x40(%rsp) vmovss 0xb287f7(%rip), %xmm2 # 0x1eeba20 movl $0xffffffff, %edx # imm = 0xFFFFFFFF xorl %ecx, %ecx vxorps %xmm3, %xmm3, %xmm3 xorl %esi, %esi vmovaps 0x60(%rsp), %xmm1 vmovss 0xb0(%rsp,%rcx,4), %xmm0 vucomiss %xmm3, %xmm0 jne 0x13c324d jnp 0x13c3269 vmovss 0x50(%rsp,%rcx,4), %xmm4 vucomiss %xmm4, %xmm2 jbe 0x13c3269 movl 0x40(%rsp,%rcx,4), %edi testl %edi, %edi je 0x13c3269 vmovaps %xmm4, %xmm2 movl %ecx, %edx movl %edi, %esi incq %rcx cmpq $0x3, %rcx jne 0x13c323c vmovss %xmm2, (%rax) movl %edx, 0x4(%rax) movl %esi, 0x8(%rax) movl $0x0, 0xc(%rax) movq $0x20, 0x10(%rax) vmovaps 0x70(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rax) vmovaps %xmm1, 0x30(%rax) leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef const&)::operator()(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::Instance, embree::InstancePrimitive>::BuildRef const&) const
__forceinline operator Vec3fa () const { return Vec3fa(m128); }
vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vmovaps (%rsi), %xmm2 vmovaps 0x10(%rsi), %xmm3 vmovaps 0x20(%rsi), %xmm4 vmovaps 0x30(%rsi), %xmm5 vminps %xmm0, %xmm2, %xmm2 vmovaps %xmm2, (%rsi) vmaxps %xmm1, %xmm3, %xmm2 vmovaps %xmm2, 0x10(%rsi) vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm4, %xmm1 vmovaps %xmm1, 0x20(%rsi) vmaxps %xmm0, %xmm5, %xmm0 vmovaps %xmm0, 0x30(%rsi) incq 0x48(%rsi) retq
/embree[P]embree/kernels/bvh/../common/../../common/sys/../math/vec3fa.h
embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::setupLargeBuildRefBuilder(unsigned long, embree::InstanceArray const*)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupLargeBuildRefBuilder (size_t objectID, Mesh const * const mesh) { if (bvh->objects[objectID] == nullptr || // new mesh builders[objectID]->meshQualityChanged (mesh->quality) || // changed build quality dynamic_cast<RefBuilderLarge*>(builders[objectID].get()) == nullptr) // size change resulted in small->large change { Builder* builder = nullptr; delete bvh->objects[objectID]; createMeshAccel(objectID, builder); builders[objectID].reset (new RefBuilderLarge(objectID, builder, mesh->quality)); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rdx, %r14 movq %rsi, %rbx movq %rdi, %r15 movq 0x28(%rdi), %rax movq 0x200(%rax), %rax cmpq $0x0, (%rax,%rsi,8) je 0x13ca955 movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi movzwl 0x3e(%r14), %esi andl $0x7, %esi movq (%rdi), %rax callq *0x18(%rax) testb %al, %al jne 0x13ca955 movq 0x10(%r15), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13ca955 leaq 0xd4dc2a(%rip), %rsi # 0x2118568 leaq 0xd4dc7b(%rip), %rdx # 0x21185c0 xorl %ecx, %ecx callq 0x6a9a0 testq %rax, %rax jne 0x13ca9fd movq $0x0, (%rsp) movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13ca977 movq (%rdi), %rax callq *0x8(%rax) movq %rsp, %r12 movq %r15, %rdi movq %rbx, %rsi movq %r12, %rdx callq 0x13cb0cc movq 0x10(%r15), %r13 movl $0x20, %edi callq 0x6a170 movq %rax, %r15 movq (%r12), %r12 testq %r12, %r12 je 0x13ca9ac movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) movzwl 0x3e(%r14), %ebp leaq 0xd4dc30(%rip), %rax # 0x21185e8 movq %rax, (%r15) movq %rbx, 0x8(%r15) movq %r12, 0x10(%r15) testq %r12, %r12 je 0x13ca9d2 movq (%r12), %rax movq %r12, %rdi callq *0x10(%rax) andl $0x7, %ebp movl %ebp, 0x18(%r15) movq (%r13,%rbx,8), %rdi movq %r15, (%r13,%rbx,8) testq %rdi, %rdi je 0x13ca9ee movq (%rdi), %rax callq *0x8(%rax) testq %r12, %r12 je 0x13ca9fd movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq jmp 0x13caa1d movq %rax, %rbx movq (%r12), %rax movq %r12, %rdi callq *0x18(%rax) jmp 0x13caa28 movq %rax, %rdi callq 0x8d6de8 movq %rax, %rbx movq %r15, %rdi callq 0x6a4f0 movq %rbx, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::RefBuilderLarge::attachBuildRefs(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>*)
void attachBuildRefs (BVHNBuilderTwoLevel* topBuilder) { BVH* object = topBuilder->getBVH(objectID_); assert(object); /* build object if it got modified */ if (topBuilder->isGeometryModified(objectID_)) builder_->build(); /* create build primitive */ if (!object->getBounds().empty()) { #if ENABLE_DIRECT_SAH_MERGE_BUILDER Mesh* mesh = topBuilder->getMesh(objectID_); topBuilder->refs[topBuilder->nextRef++] = BVHNBuilderTwoLevel::BuildRef(object->getBounds(),object->root,(unsigned int)objectID_,(unsigned int)mesh->size()); #else topBuilder->refs[topBuilder->nextRef++] = BVHNBuilderTwoLevel::BuildRef(object->getBounds(),object->root); #endif } }
pushq %r15 pushq %r14 pushq %rbx movq %rsi, %rbx movq %rdi, %r14 movq 0x8(%rdi), %rax movq 0x28(%rsi), %rdx movq 0x30(%rsi), %rcx movq 0x200(%rdx), %rdx movq (%rdx,%rax,8), %r15 movq 0x1e8(%rcx), %rdx movq (%rdx,%rax,8), %rdx testq %rdx, %rdx je 0x13cb411 movl 0x38(%rdx), %edx movq 0x208(%rcx), %rcx cmpl (%rcx,%rax,4), %edx jbe 0x13cb411 movq 0x10(%r14), %rdi movq (%rdi), %rax callq *0x20(%rax) vmovaps 0x10(%r15), %xmm0 vmovaps 0x20(%r15), %xmm1 vminps 0x30(%r15), %xmm0, %xmm0 vmaxps 0x40(%r15), %xmm1, %xmm1 vcmpleps %xmm1, %xmm0, %xmm2 vmovmskps %xmm2, %eax notb %al testb $0x7, %al jne 0x13cb4c5 movq 0x8(%r14), %rcx movq 0x30(%rbx), %rax movq 0x1e8(%rax), %rax movq (%rax,%rcx,8), %rax xorl %edx, %edx cmpb $0x18, 0x3c(%rax) cmoveq %rax, %rdx movq 0x70(%r15), %rax vmovd %ecx, %xmm3 vinsertps $0x30, 0x20(%rdx), %xmm1, %xmm2 # xmm2 = xmm1[0,1,2],mem[0] vinsertps $0x30, %xmm3, %xmm0, %xmm3 # xmm3 = xmm0[0,1,2],xmm3[0] vinsertf128 $0x1, %xmm2, %ymm3, %ymm0 vxorps %xmm1, %xmm1, %xmm1 testb $0x8, %al jne 0x13cb49d vsubps %xmm3, %xmm2, %xmm1 vmovshdup %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] vshufpd $0x1, %xmm1, %xmm1, %xmm3 # xmm3 = xmm1[1,0] vaddss %xmm3, %xmm2, %xmm4 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm4, %xmm1, %xmm1 vaddss %xmm2, %xmm1, %xmm1 vaddss %xmm1, %xmm1, %xmm1 movl $0x1, %ecx lock xaddl %ecx, 0x88(%rbx) movslq %ecx, %rcx movq 0x58(%rbx), %rdx shlq $0x6, %rcx vmovaps %ymm0, (%rdx,%rcx) movq %rax, 0x20(%rdx,%rcx) vmovss %xmm1, 0x28(%rdx,%rcx) popq %rbx popq %r14 popq %r15 vzeroupper retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::find(embree::avx::PrimInfoExtRange&, unsigned long)
__noinline const Split find(PrimInfoExtRange& set, const size_t logBlockSize) { /* single element */ if (set.size() <= 1) return Split(); /* disable opening if there is no overlap */ const size_t D = 4; if (unlikely(set.has_ext_range() && set.size() <= D)) { bool disjoint = true; for (size_t j=set.begin(); j<set.end()-1; j++) { for (size_t i=set.begin()+1; i<set.end(); i++) { if (conjoint(prims0[j].bounds(),prims0[i].bounds())) { disjoint = false; break; } } } if (disjoint) set.set_ext_range(set.end()); /* disables opening */ } std::pair<size_t,bool> p(0,false); /* disable opening when all primitives are from same geometry */ if (unlikely(set.has_ext_range())) { p = getProperties(set); #if EQUAL_GEOMID_STOP_CRITERIA == 1 if (p.second) set.set_ext_range(set.end()); /* disable opening */ #endif } /* open nodes when we have sufficient space available */ if (unlikely(set.has_ext_range())) { #if USE_LOOP_OPENING == 1 openNodesBasedOnExtendLoop(set,p.first); #else if (p.first <= set.ext_range_size()) openNodesBasedOnExtend(set); #endif /* disable opening when insufficient space for opening a node available */ if (set.ext_range_size() < max_open_size-1) set.set_ext_range(set.end()); /* disable opening */ } /* find best split */ return object_find(set,logBlockSize); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rcx, %r14 movq %rdx, %r15 movq %rdi, %rbx movq 0x40(%rdx), %rcx movq 0x48(%rdx), %rax movq %rax, %rdx subq %rcx, %rdx cmpq $0x1, %rdx ja 0x13cd384 vmovsd 0xb30b22(%rip), %xmm0 # 0x1efdea0 vmovaps %xmm0, (%rbx) jmp 0x13cd3d1 movq %rsi, %r12 cmpq $0x4, %rdx ja 0x13cd397 cmpq %rax, 0x50(%r15) jne 0x13cd42e cmpq %rax, 0x50(%r15) jne 0x13cd3e3 xorl %eax, %eax movq 0x50(%r15), %rcx subq 0x48(%r15), %rcx jne 0x13cd3fd movq 0x48(%r15), %rax subq 0x40(%r15), %rax movq %rbx, %rdi movq %r12, %rsi movq %r15, %rdx movq %r14, %rcx cmpq $0x3ff, %rax # imm = 0x3FF ja 0x13cd3cc callq 0x13d0100 jmp 0x13cd3d1 callq 0x13d085e movq %rbx, %rax addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %r12, %rdi movq %r15, %rsi callq 0x13ce91c testb $0x1, %dl je 0x13cd39f movq 0x48(%r15), %rcx movq %rcx, 0x50(%r15) jmp 0x13cd39f cmpq %rcx, %rax ja 0x13cd40d movq %r12, %rdi movq %r15, %rsi callq 0x13ceb2e movq 0x48(%r15), %rax movq 0x50(%r15), %rcx subq %rax, %rcx movq 0x10(%r12), %rdx decq %rdx cmpq %rdx, %rcx jae 0x13cd3a9 movq %rax, 0x50(%r15) jmp 0x13cd3a9 leaq -0x1(%rax), %rdx cmpq %rdx, %rcx jae 0x13cd4a9 movq (%r12), %rsi movq %rcx, %rdi shlq $0x6, %rdi addq %rsi, %rdi addq $0x50, %rdi movb $0x1, %r8b movq %rcx, %r9 movq %r9, %r10 shlq $0x6, %r10 addq %rsi, %r10 movq %rdi, %r13 movq %rcx, %r11 incq %r11 cmpq %rax, %r11 jae 0x13cd497 vmovaps (%r10), %xmm0 vmovaps 0x10(%r10), %xmm1 vmaxps -0x10(%r13), %xmm0, %xmm0 vminps (%r13), %xmm1, %xmm1 addq $0x40, %r13 vcmpleps %xmm1, %xmm0, %xmm0 vmovmskps %xmm0, %ebp notb %bpl testb $0x7, %bpl jne 0x13cd45f xorl %r8d, %r8d incq %r9 cmpq %rdx, %r9 jne 0x13cd44f testb $0x1, %r8b je 0x13cd397 movq %rax, 0x50(%r15) jmp 0x13cd397
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::openNodesBasedOnExtend(embree::avx::PrimInfoExtRange&)
__noinline void openNodesBasedOnExtend(PrimInfoExtRange& set) { const OpenHeuristic heuristic(set); const size_t ext_range_start = set.end(); if (false && set.size() < PARALLEL_THRESHOLD) { size_t extra_elements = 0; for (size_t i=set.begin(); i<set.end(); i++) { if (heuristic(prims0[i])) { PrimRef tmp[MAX_OPENED_CHILD_NODES]; const size_t n = nodeOpenerFunc(prims0[i],tmp); assert(extra_elements + n-1 <= set.ext_range_size()); for (size_t j=0; j<n; j++) set.extend_center2(tmp[j]); prims0[i] = tmp[0]; for (size_t j=1; j<n; j++) prims0[ext_range_start+extra_elements+j-1] = tmp[j]; extra_elements += n-1; } } set._end += extra_elements; } else { std::atomic<size_t> ext_elements; ext_elements.store(0); PrimInfo info = parallel_reduce( set.begin(), set.end(), CREATE_SPLITS_STEP_SIZE, PrimInfo(empty), [&](const range<size_t>& r) -> PrimInfo { PrimInfo info(empty); for (size_t i=r.begin(); i<r.end(); i++) if (heuristic(prims0[i])) { PrimRef tmp[MAX_OPENED_CHILD_NODES]; const size_t n = nodeOpenerFunc(prims0[i],tmp); const size_t ID = ext_elements.fetch_add(n-1); assert(ID + n-1 <= set.ext_range_size()); for (size_t j=0; j<n; j++) info.extend_center2(tmp[j]); prims0[i] = tmp[0]; for (size_t j=1; j<n; j++) prims0[ext_range_start+ID+j-1] = tmp[j]; } return info; }, [] (const PrimInfo& a, const PrimInfo& b) { return PrimInfo::merge(a,b); }); set.centBounds.extend(info.centBounds); assert(ext_elements.load() <= set.ext_range_size()); set._end += ext_elements.load(); } }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x1f8, %rsp # imm = 0x1F8 movq %rsi, %rbx vmovaps 0x10(%rsi), %xmm0 vsubps (%rsi), %xmm0, %xmm0 vbroadcastss 0xb52373(%rip), %xmm1 # 0x1f20ec4 vandps %xmm1, %xmm0, %xmm1 vshufps $0xf5, %xmm1, %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] xorl %eax, %eax vucomiss %xmm2, %xmm1 setbe %al vmaxss %xmm2, %xmm1, %xmm2 vshufps $0x4e, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[2,3,0,1] vucomiss %xmm1, %xmm2 movl $0x2, %ecx cmovaq %rax, %rcx leaq 0x90(%rsp), %r14 vmovaps %xmm0, (%r14) leaq 0x50(%rsp), %rax movq %rcx, (%rax) vmovss 0xb1db7e(%rip), %xmm0 # 0x1eec714 vdivss 0x90(%rsp,%rcx,4), %xmm0, %xmm0 vmovss %xmm0, 0x8(%rax) movq 0x48(%rsi), %rcx leaq 0x48(%rsp), %rdx movq %rcx, (%rdx) xorl %ecx, %ecx leaq 0x28(%rsp), %rsi xchgq %rcx, (%rsi) vbroadcastss 0xb1ce5d(%rip), %xmm0 # 0x1eeba20 leaq 0x1a0(%rsp), %r15 vmovaps %xmm0, (%r15) vbroadcastss 0xb1dfab(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%r15) vmovaps %xmm0, 0x20(%r15) vmovaps %xmm1, 0x30(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x40(%r15) leaq 0x110(%rsp), %r12 movq %rax, (%r12) movq %rdi, 0x8(%r12) movq %rsi, 0x10(%r12) movq %rdx, 0x18(%r12) movw $0x401, 0xc(%r14) # imm = 0x401 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) vpermilps $0x4e, 0x40(%rbx), %xmm0 # xmm0 = mem[2,3,0,1] vmovaps %xmm0, 0x10(%rsp) movq %r14, %rdi callq 0x6a660 leaq 0x70(%rsp), %rdi vmovaps 0x10(%rsp), %xmm0 vmovaps %xmm0, (%rdi) movq $0x80, 0x10(%rdi) leaq 0xe(%rsp), %rax leaq 0x60(%rsp), %rcx movq %rax, (%rcx) movq %r12, 0x8(%rcx) leaq 0x130(%rsp), %rsi movq %r15, (%rsi) movq %rcx, 0x8(%rsi) movq %rax, 0x10(%rsi) vmovups (%r15), %ymm0 vmovups 0x20(%r15), %ymm1 vmovups %ymm0, 0x20(%rsi) vmovups %ymm1, 0x40(%rsi) vmovaps 0x40(%r15), %xmm0 vmovaps %xmm0, 0x60(%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx vzeroupper callq 0x13cf62a vmovaps 0x170(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rsp) vmovaps 0x180(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rsp) leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13ced19 leaq 0x90(%rsp), %rdi callq 0x6aab0 vmovaps 0x20(%rbx), %xmm0 vminps 0x30(%rsp), %xmm0, %xmm0 vmovaps 0x30(%rbx), %xmm1 vmaxps 0x10(%rsp), %xmm1, %xmm1 vmovaps %xmm0, 0x20(%rbx) vmovaps %xmm1, 0x30(%rbx) movq 0x28(%rsp), %rax addq %rax, 0x48(%rbx) addq $0x1f8, %rsp # imm = 0x1F8 popq %rbx popq %r12 popq %r14 popq %r15 retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb1cdc3(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd55ecc(%rip), %rsi # 0x2124c08 movq 0xd55c85(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13ced5f jmp 0x13ced74 jmp 0x13ced5c movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
unsigned long embree::parallel_partitioning<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>, embree::EmptyTy, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef const&), embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef const&), embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&)>(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*, unsigned long, unsigned long, embree::EmptyTy const&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef const&) const&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef const&) const&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<4, embree::InstanceArray, embree::InstanceArrayPrimitive>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&) const&, unsigned long)
__noinline size_t parallel_partitioning(T* array, const size_t begin, const size_t end, const Vi &identity, V &leftReduction, V &rightReduction, const IsLeft& is_left, const Reduction_T& reduction_t, const Reduction_V& reduction_v, size_t BLOCK_SIZE = 128) { /* fall back to single threaded partitioning for small N */ if (unlikely(end-begin < BLOCK_SIZE)) return serial_partitioning(array,begin,end,leftReduction,rightReduction,is_left,reduction_t); /* otherwise use parallel code */ else { typedef parallel_partition_task<T,V,Vi,IsLeft,Reduction_T,Reduction_V> partition_task; std::unique_ptr<partition_task> p(new partition_task(&array[begin],end-begin,identity,is_left,reduction_t,reduction_v,BLOCK_SIZE)); return begin+p->partition(leftReduction,rightReduction); } }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x20, %rsp subq $0x100, %rsp # imm = 0x100 movq %r9, %r15 movq %r8, %r12 movq 0x18(%rbp), %r8 movq 0x10(%rbp), %r9 movq %rdx, %rbx subq %rsi, %rbx cmpq 0x28(%rbp), %rbx jb 0x13d35ee movq %rdi, %r13 movl $0x34c0, %edi # imm = 0x34C0 movq %rsi, 0x8(%rsp) movl $0x40, %esi movq %rcx, 0x38(%rsp) callq 0x1ee60ac movq %rax, %r14 movq 0x8(%rsp), %rax shlq $0x6, %rax addq %rax, %r13 movq %r13, (%r14) movq %rbx, 0x8(%r14) movq 0x10(%rbp), %rax movq %rax, 0x10(%r14) movq 0x18(%rbp), %rax movq %rax, 0x18(%r14) movq 0x20(%rbp), %rax movq %rax, 0x20(%r14) movq 0x38(%rsp), %rax movq %rax, 0x28(%r14) xorl %edi, %edi callq 0x6ab80 movl %eax, %ecx movq 0x28(%rbp), %rsi leaq (%rbx,%rsi), %rax decq %rax xorl %edx, %edx divq %rsi cmpl $0x40, %ecx movl $0x40, %ebx cmovbl %ecx, %ebx cmpq %rbx, %rax cmovbq %rax, %rbx movq %rbx, 0x30(%r14) movq %r14, 0x28(%rsp) movq %r14, 0x30(%rsp) leaq 0x60(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) callq 0x6a660 leaq 0x30(%rsp), %rax leaq 0x40(%rsp), %rcx movq %rax, (%rcx) leaq 0x10(%rsp), %r8 leaq 0x60(%rsp), %r9 movl $0x1, %edx xorl %edi, %edi movq %rbx, %rsi callq 0x13d390c leaq 0x60(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13d37d0 leaq 0x60(%rsp), %rdi callq 0x6aab0 movq 0x8(%r14), %rax movq 0x30(%r14), %rcx movq %rax, 0x40(%r14,%rcx,8) movq 0x30(%r14), %rax movq $0x0, 0x280(%r14,%rax,8) movq 0x30(%r14), %rdx testq %rdx, %rdx je 0x13d3369 leaq 0x2108(%r14), %rax xorl %ecx, %ecx vmovaps (%r12), %xmm0 vmovaps 0x10(%r12), %xmm1 vmovaps 0x20(%r12), %xmm2 vminps -0x1448(%rax), %xmm0, %xmm0 vmovaps 0x30(%r12), %xmm3 vmovaps %xmm0, (%r12) vmaxps -0x1438(%rax), %xmm1, %xmm0 vmovaps %xmm0, 0x10(%r12) vminps -0x1428(%rax), %xmm2, %xmm0 vmovaps %xmm0, 0x20(%r12) vmaxps -0x1418(%rax), %xmm3, %xmm0 vmovaps %xmm0, 0x30(%r12) movq -0x1408(%rax), %rdx addq %rdx, 0x40(%r12) movq -0x1400(%rax), %rdx addq %rdx, 0x48(%r12) vmovaps (%r15), %xmm0 vmovaps 0x10(%r15), %xmm1 vmovaps 0x20(%r15), %xmm2 vminps -0x48(%rax), %xmm0, %xmm0 vmovaps 0x30(%r15), %xmm3 vmovaps %xmm0, (%r15) vmaxps -0x38(%rax), %xmm1, %xmm0 vmovaps %xmm0, 0x10(%r15) vminps -0x28(%rax), %xmm2, %xmm0 vmovaps %xmm0, 0x20(%r15) vmaxps -0x18(%rax), %xmm3, %xmm0 vmovaps %xmm0, 0x30(%r15) movq -0x8(%rax), %rdx addq %rdx, 0x40(%r15) movq (%rax), %rdx addq %rdx, 0x48(%r15) incq %rcx movq 0x30(%r14), %rdx addq $0x50, %rax cmpq %rdx, %rcx jb 0x13d3296 movq 0x280(%r14), %r13 cmpq $0x2, %rdx jb 0x13d3466 leaq 0x2(%rdx), %rax andq $-0x4, %rax addq $-0x2, %rdx vmovq %r13, %xmm0 vmovq %rdx, %xmm1 vpshufd $0x44, %xmm1, %xmm1 # xmm1 = xmm1[0,1,0,1] vinsertf128 $0x1, %xmm1, %ymm1, %ymm1 leaq 0x288(%r14), %rcx xorl %edx, %edx vmovaps 0xb4ea94(%rip), %ymm3 # 0x1f21e40 vextractf128 $0x1, %ymm1, %xmm4 vmovddup 0xb4eace(%rip), %xmm2 # xmm2 = mem[0,0] vxorps %xmm2, %xmm4, %xmm4 vpcmpeqd %xmm5, %xmm5, %xmm5 vxorps %xmm2, %xmm1, %xmm6 movq 0x8(%rsp), %r15 vmovq %rdx, %xmm7 vpshufd $0x44, %xmm7, %xmm7 # xmm7 = xmm7[0,1,0,1] vinsertf128 $0x1, %xmm7, %ymm7, %ymm7 vorps %ymm3, %ymm7, %ymm7 vextractf128 $0x1, %ymm7, %xmm8 vxorps %xmm2, %xmm8, %xmm8 vpcmpgtq %xmm4, %xmm8, %xmm8 vpxor %xmm5, %xmm8, %xmm9 vxorps %xmm2, %xmm7, %xmm10 vpcmpgtq %xmm6, %xmm10, %xmm10 vpxor %xmm5, %xmm10, %xmm10 vinsertf128 $0x1, %xmm9, %ymm10, %ymm9 vmaskmovpd (%rcx,%rdx,8), %ymm9, %ymm11 vmovdqa %ymm0, %ymm9 vextractf128 $0x1, %ymm11, %xmm0 vextractf128 $0x1, %ymm9, %xmm10 vpaddq %xmm0, %xmm10, %xmm0 vpaddq %xmm9, %xmm11, %xmm11 vinsertf128 $0x1, %xmm0, %ymm11, %ymm0 addq $0x4, %rdx cmpq %rdx, %rax jne 0x13d33cb vxorps %xmm2, %xmm1, %xmm1 vxorps %xmm2, %xmm7, %xmm2 vpcmpgtq %xmm1, %xmm2, %xmm1 vextractf128 $0x1, %ymm0, %xmm2 vblendvpd %xmm8, %xmm10, %xmm2, %xmm2 vblendvpd %xmm1, %xmm9, %xmm0, %xmm0 vpaddq %xmm2, %xmm0, %xmm0 vpshufd $0xee, %xmm0, %xmm1 # xmm1 = xmm0[2,3,2,3] vpaddq %xmm1, %xmm0, %xmm0 vmovq %xmm0, %r13 jmp 0x13d346b movq 0x8(%rsp), %r15 movq 0x8(%r14), %rax xorl %ecx, %ecx movq %rcx, 0x10(%rsp) movq %rcx, 0x20(%rsp) movq %rcx, 0x18(%rsp) movq 0x30(%r14), %rbx testq %rbx, %rbx je 0x13d352f leaq 0x4c0(%r14), %rcx movq %r14, %rdx addq $0x8c0, %rdx # imm = 0x8C0 xorl %esi, %esi movq 0x40(%r14,%rsi,8), %rdi movq 0x280(%r14,%rsi,8), %r8 addq %rdi, %r8 testq %r8, %r8 movl $0x0, %r9d cmovgq %r8, %r9 movq 0x48(%r14,%rsi,8), %r10 cmpq %r10, %r13 cmovlq %r13, %r10 cmpq %rdi, %r13 cmovgq %r13, %rdi cmpq %r8, %rax cmovlq %rax, %r8 movq %r10, %r11 subq %r9, %r11 jle 0x13d34ff addq %r11, 0x18(%rsp) movq 0x10(%rsp), %r11 leaq 0x1(%r11), %rbx movq %rbx, 0x10(%rsp) shlq $0x4, %r11 movq %r9, (%rcx,%r11) movq %r10, 0x8(%rcx,%r11) cmpq %rdi, %r8 jle 0x13d351f movq 0x20(%rsp), %r9 leaq 0x1(%r9), %r10 movq %r10, 0x20(%rsp) shlq $0x4, %r9 movq %rdi, (%rdx,%r9) movq %r8, 0x8(%rdx,%r9) incq %rsi movq 0x30(%r14), %rbx cmpq %rbx, %rsi jb 0x13d34a0 cmpq $0x0, 0x18(%rsp) je 0x13d35c9 leaq 0x18(%rsp), %rax movq %rax, 0x40(%rsp) movq %r14, 0x48(%rsp) leaq 0x10(%rsp), %rax movq %rax, 0x50(%rsp) leaq 0x20(%rsp), %rax movq %rax, 0x58(%rsp) leaq 0x60(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vpxor %xmm0, %xmm0, %xmm0 vmovdqu %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x30(%rsp), %rcx leaq 0x40(%rsp), %rax movq %rax, (%rcx) leaq 0x7(%rsp), %r8 leaq 0x60(%rsp), %r9 movl $0x1, %edx xorl %edi, %edi movq %rbx, %rsi callq 0x13d437c leaq 0x60(%rsp), %rdi callq 0x6a770 testb %al, %al movq 0x8(%rsp), %r15 jne 0x13d3802 leaq 0x60(%rsp), %rdi callq 0x6aab0 addq %r15, %r13 leaq 0x28(%rsp), %rdi vzeroupper callq 0x13d38a2 movq %r13, %rax leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movq %rdx, %r14 shlq $0x6, %rsi addq %rdi, %rsi shlq $0x6, %r14 addq %rdi, %r14 movq %rsi, %r13 addq $-0x40, %r14 cmpq %r14, %r13 ja 0x13d36eb movq (%r9), %rax movq 0x8(%r9), %rcx movq 0x10(%r9), %rdx vmovaps (%r13), %xmm2 vmovaps 0x10(%r13), %xmm1 vaddps %xmm1, %xmm2, %xmm0 vsubps 0x20(%rax), %xmm0, %xmm3 vmulps 0x30(%rax), %xmm3, %xmm3 vroundps $0x1, %xmm3, %xmm3 vcvtps2dq %xmm3, %xmm3 vmovdqa (%rcx), %xmm4 vpcmpgtd %xmm3, %xmm4, %xmm3 vtestps (%rdx), %xmm3 je 0x13d36eb vmovaps (%r12), %xmm3 vmovaps 0x10(%r12), %xmm4 vmovaps 0x20(%r12), %xmm5 vmovaps 0x30(%r12), %xmm6 vminps %xmm2, %xmm3, %xmm2 vmovaps %xmm2, (%r12) vmaxps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0x10(%r12) vminps %xmm0, %xmm5, %xmm1 vmovaps %xmm1, 0x20(%r12) vmaxps %xmm0, %xmm6, %xmm0 vmovaps %xmm0, 0x30(%r12) incq 0x48(%r12) addq $0x40, %r13 jmp 0x13d3606 vmovaps (%r15), %xmm3 vmovaps 0x10(%r15), %xmm4 vmovaps 0x20(%r15), %xmm5 vmovaps 0x30(%r15), %xmm6 vminps %xmm2, %xmm3, %xmm2 vmovaps %xmm2, (%r15) vmaxps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0x10(%r15) vminps %xmm0, %xmm5, %xmm1 vmovaps %xmm1, 0x20(%r15) vmaxps %xmm0, %xmm6, %xmm0 vmovaps %xmm0, 0x30(%r15) incq 0x48(%r15) addq $-0x40, %r14 cmpq %r14, %r13 seta %al ja 0x13d37c4 movq (%r9), %rcx movq 0x8(%r9), %rdx movq 0x10(%r9), %rsi vmovaps (%r14), %xmm2 vmovaps 0x10(%r14), %xmm1 vaddps %xmm1, %xmm2, %xmm0 vsubps 0x20(%rcx), %xmm0, %xmm3 vmulps 0x30(%rcx), %xmm3, %xmm3 vroundps $0x1, %xmm3, %xmm3 vcvtps2dq %xmm3, %xmm3 vmovdqa (%rdx), %xmm4 vpcmpgtd %xmm3, %xmm4, %xmm3 vtestps (%rsi), %xmm3 je 0x13d36a5 testb $0x1, %al jne 0x13d37c4 movq %rdi, 0x8(%rsp) movq %r8, %rdi movq %r12, %rsi movq %r14, %rdx movq %r8, %rbx vzeroupper callq 0x13d38c4 movq %rbx, %rdi movq %r15, %rsi movq %r13, %rdx callq 0x13d38c4 movq 0x10(%rbp), %r9 movq %rbx, %r8 movq 0x8(%rsp), %rdi vmovaps (%r13), %ymm0 movl 0x28(%r13), %eax movl %eax, 0x68(%rsp) movq 0x20(%r13), %rax movq %rax, 0x60(%rsp) vmovaps (%r14), %ymm1 vmovaps %ymm1, (%r13) movl 0x28(%r14), %eax movl %eax, 0x28(%r13) movq 0x20(%r14), %rax movq %rax, 0x20(%r13) vmovaps %ymm0, (%r14) movq 0x60(%rsp), %rax movq %rax, 0x20(%r14) movl 0x68(%rsp), %eax movl %eax, 0x28(%r14) addq $0x40, %r13 jmp 0x13d3602 subq %rdi, %r13 sarq $0x6, %r13 jmp 0x13d35d9 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb1830c(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd51415(%rip), %rsi # 0x2124c08 movq 0xd511ce(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb182da(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd513e3(%rip), %rsi # 0x2124c08 movq 0xd5119c(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13d3855 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13d3879 jmp 0x13d3885 jmp 0x13d3852 movq %rax, %rbx leaq 0x60(%rsp), %rdi callq 0x6aab0 jmp 0x13d3890 jmp 0x13d3885 jmp 0x13d3885 jmp 0x13d3876 movq %rax, %rbx movq %r14, %rdi callq 0x1ee612d jmp 0x13d389a jmp 0x13d3885 movq %rax, %rbx leaq 0x60(%rsp), %rdi callq 0x6aab0 jmp 0x13d3890 movq %rax, %rdi callq 0x8d6de8 movq %rax, %rbx leaq 0x28(%rsp), %rdi callq 0x13d38a2 movq %rbx, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/../builders/../../common/algorithms/parallel_partition.h
embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::~BVHNBuilderTwoLevel()
BVHNBuilderTwoLevel<N,Mesh,Primitive>::~BVHNBuilderTwoLevel () { }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx leaq 0xd434a2(%rip), %rax # 0x21188e8 movq %rax, (%rdi) movq 0x78(%rdi), %r14 movq 0x80(%rdi), %rdi testq %rdi, %rdi je 0x13d5479 movq %r14, %rsi shlq $0x5, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13d5474 movzbl 0x68(%rbx), %edx callq 0x1ee67ca jmp 0x13d5479 callq 0x1ee612d testq %r14, %r14 je 0x13d5496 movq 0x60(%rbx), %rdi shlq $0x5, %r14 negq %r14 movq (%rdi), %rax movq %r14, %rsi movl $0x1, %edx callq *(%rax) leaq 0x70(%rbx), %rax vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, (%rax) movq $0x0, 0x10(%rax) movq 0x50(%rbx), %r14 movq 0x58(%rbx), %rdi testq %rdi, %rdi je 0x13d54d7 movq %r14, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x13d54d2 movzbl 0x40(%rbx), %edx callq 0x1ee67ca jmp 0x13d54d7 callq 0x1ee612d testq %r14, %r14 je 0x13d54f4 movq 0x38(%rbx), %rdi shlq $0x6, %r14 negq %r14 movq (%rdi), %rax movq %r14, %rsi movl $0x1, %edx callq *(%rax) leaq 0x10(%rbx), %rdi vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x48(%rbx) movq $0x0, 0x58(%rbx) addq $0x8, %rsp popq %rbx popq %r14 jmp 0x13d53f0 jmp 0x13d5517 movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::NodeRefPtr<8> embree::avx::BVHBuilderBinnedOpenMergeSAH::build<embree::NodeRefPtr<8>, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(unsigned long)>(embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(unsigned long), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*, unsigned long, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&, embree::avx::GeneralBVHBuilder::Settings const&)
static ReductionTy build(CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, NodeOpenerFunc nodeOpenerFunc, ProgressMonitor progressMonitor, BuildRef* prims, const size_t extSize, const PrimInfo& pinfo, const Settings& settings) { typedef HeuristicArrayOpenMergeSAH<NodeOpenerFunc,BuildRef,NUM_OBJECT_BINS_HQ> Heuristic; Heuristic heuristic(nodeOpenerFunc,prims,settings.branchingFactor); return GeneralBVHBuilder::build<ReductionTy,Heuristic,Set,BuildRef>( heuristic, prims, PrimInfoExtRange(0,pinfo.size(),extSize,pinfo), createAlloc, createNode, updateNode, createLeaf, progressMonitor, settings); }
pushq %r14 pushq %rbx subq $0x98, %rsp movq %r8, %rax movq %rcx, %r10 movq %rsi, %r8 movq %rdi, %rcx movq 0xb0(%rsp), %rsi movq 0xb8(%rsp), %rbx leaq 0x10(%rsp), %r14 movq %rdx, (%r14) leaq 0x8(%rsp), %r11 movq %r10, (%r11) movq (%rbx), %rdx leaq 0x18(%rsp), %rdi movq %rax, (%rdi) movq %r14, 0x8(%rdi) movq %rdx, 0x10(%rdi) movq 0x48(%rsi), %r10 subq 0x40(%rsi), %r10 vmovaps (%rsi), %xmm0 leaq 0x30(%rsp), %rdx vmovaps %xmm0, (%rdx) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, 0x10(%rdx) vmovaps 0x20(%rsi), %xmm0 vmovaps %xmm0, 0x20(%rdx) vmovaps 0x30(%rsi), %xmm0 vmovaps %xmm0, 0x30(%rdx) movq $0x0, 0x40(%rdx) movq %r10, 0x48(%rdx) movq %r9, 0x50(%rdx) movq %rbx, (%rsp) movq %rax, %rsi movq %r11, %r9 callq 0x13d9f4c addq $0x98, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::__internal_two_level_builder__::MeshBuilder<8, embree::TriangleMesh, embree::TriangleM<4>>::operator()(void*, embree::TriangleMesh*, unsigned long, embree::Geometry::GTypeMask, bool, embree::Builder*&)
void operator () (void* bvh, Mesh* mesh, size_t geomID, Geometry::GTypeMask gtype, bool useMortonBuilder, Builder*& builder) { if(useMortonBuilder) { builder = MortonBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); return; } switch (mesh->quality) { case RTC_BUILD_QUALITY_LOW: builder = MortonBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; case RTC_BUILD_QUALITY_MEDIUM: case RTC_BUILD_QUALITY_HIGH: builder = SAHBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; case RTC_BUILD_QUALITY_REFIT: builder = RefitBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; default: throw_RTCError(RTC_ERROR_UNKNOWN,"invalid build quality"); } }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x28, %rsp testl %r9d, %r9d je 0x13d8bc1 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x1447902 jmp 0x13d8c77 movzwl 0x3e(%rdx), %eax andl $0x7, %eax leal -0x1(%rax), %edi cmpl $0x2, %edi jb 0x13d8c57 cmpl $0x3, %eax je 0x13d8c68 testl %eax, %eax je 0x13d8bad movl $0x30, %edi callq 0x6a3b0 movq %rax, %rbx leaq 0x18(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb13c28(%rip), %rsi # 0x1eec826 leaq 0xb13c36(%rip), %rdx # 0x1eec83b leaq 0x8(%rsp), %rdi callq 0x8d7230 leaq 0xd28c9a(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x1, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x8(%rsp), %rsi movq 0x10(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd28c08(%rip), %rsi # 0x2101850 leaq -0xb01e33(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x127a76d jmp 0x13d8c77 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x150259d movq 0x50(%rsp), %rcx movq %rax, (%rcx) addq $0x28, %rsp popq %rbx popq %r14 popq %r15 popq %rbp retq movq %rax, %r14 xorl %ebp, %ebp jmp 0x13d8c9f movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x8(%rsp), %rdi cmpq %r15, %rdi je 0x13d8cb6 callq 0x6a4f0 jmp 0x13d8cb6 movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x13d8cc3 movq %rbx, %rdi callq 0x6a8a0 movq %r14, %rdi callq 0x6a600 nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel_internal.h
embree::avx::GeneralBVHBuilder::BuilderT<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::NodeRefPtr<8>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(unsigned long)>::BuilderT(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>&, embree::BVHN<8>::CreateAlloc const&, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2 const&, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2 const&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::avx::PrimInfoExtRange> const&, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::avx::PrimInfoExtRange> const&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(unsigned long) const&, embree::avx::GeneralBVHBuilder::Settings const&)
BuilderT (PrimRef* prims, Heuristic& heuristic, const CreateAllocFunc& createAlloc, const CreateNodeFunc& createNode, const UpdateNodeFunc& updateNode, const CreateLeafFunc& createLeaf, const CanCreateLeafFunc& canCreateLeaf, const CanCreateLeafSplitFunc& canCreateLeafSplit, const ProgressMonitor& progressMonitor, const Settings& settings) : cfg(settings), prims(prims), heuristic(heuristic), createAlloc(createAlloc), createNode(createNode), updateNode(updateNode), createLeaf(createLeaf), canCreateLeaf(canCreateLeaf), canCreateLeafSplit(canCreateLeafSplit), progressMonitor(progressMonitor) { if (cfg.branchingFactor > MAX_BRANCHING_FACTOR) throw_RTCError(RTC_ERROR_UNKNOWN,"bvh_builder: branching factor too large"); }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x28, %rsp vmovups 0x50(%rsp), %ymm0 movq 0x70(%rsp), %rax vmovups (%rax), %ymm1 vmovups 0x20(%rax), %ymm2 vmovups %ymm1, (%rdi) vmovups %ymm2, 0x20(%rdi) movq %rsi, 0x40(%rdi) movq %rdx, 0x48(%rdi) movq %rcx, 0x50(%rdi) movq %r8, 0x58(%rdi) movq %r9, 0x60(%rdi) vmovups %ymm0, 0x68(%rdi) cmpq $0x11, (%rdi) jae 0x13da0aa addq $0x28, %rsp popq %rbx popq %r14 popq %r15 popq %rbp vzeroupper retq movl $0x30, %edi vzeroupper callq 0x6a3b0 movq %rax, %rbx leaq 0x18(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb12b82(%rip), %rsi # 0x1eecc4c leaq 0xb12ba2(%rip), %rdx # 0x1eecc73 leaq 0x8(%rsp), %rdi callq 0x8d7230 leaq 0xd277ce(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x1, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x8(%rsp), %rsi movq 0x10(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd2773c(%rip), %rsi # 0x2101850 leaq -0xb032ff(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rax, %r14 xorl %ebp, %ebp jmp 0x13da138 movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x8(%rsp), %rdi cmpq %r15, %rdi je 0x13da14f callq 0x6a4f0 jmp 0x13da14f movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x13da15c movq %rbx, %rdi callq 0x6a8a0 movq %r14, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::find(embree::avx::PrimInfoExtRange&, unsigned long)
__noinline const Split find(PrimInfoExtRange& set, const size_t logBlockSize) { /* single element */ if (set.size() <= 1) return Split(); /* disable opening if there is no overlap */ const size_t D = 4; if (unlikely(set.has_ext_range() && set.size() <= D)) { bool disjoint = true; for (size_t j=set.begin(); j<set.end()-1; j++) { for (size_t i=set.begin()+1; i<set.end(); i++) { if (conjoint(prims0[j].bounds(),prims0[i].bounds())) { disjoint = false; break; } } } if (disjoint) set.set_ext_range(set.end()); /* disables opening */ } std::pair<size_t,bool> p(0,false); /* disable opening when all primitives are from same geometry */ if (unlikely(set.has_ext_range())) { p = getProperties(set); #if EQUAL_GEOMID_STOP_CRITERIA == 1 if (p.second) set.set_ext_range(set.end()); /* disable opening */ #endif } /* open nodes when we have sufficient space available */ if (unlikely(set.has_ext_range())) { #if USE_LOOP_OPENING == 1 openNodesBasedOnExtendLoop(set,p.first); #else if (p.first <= set.ext_range_size()) openNodesBasedOnExtend(set); #endif /* disable opening when insufficient space for opening a node available */ if (set.ext_range_size() < max_open_size-1) set.set_ext_range(set.end()); /* disable opening */ } /* find best split */ return object_find(set,logBlockSize); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rcx, %r14 movq %rdx, %r15 movq %rdi, %rbx movq 0x40(%rdx), %rcx movq 0x48(%rdx), %rax movq %rax, %rdx subq %rcx, %rdx cmpq $0x1, %rdx ja 0x13dad00 vmovsd 0xb231a6(%rip), %xmm0 # 0x1efdea0 vmovaps %xmm0, (%rbx) jmp 0x13dad4d movq %rsi, %r12 cmpq $0x4, %rdx ja 0x13dad13 cmpq %rax, 0x50(%r15) jne 0x13dadaa cmpq %rax, 0x50(%r15) jne 0x13dad5f xorl %eax, %eax movq 0x50(%r15), %rcx subq 0x48(%r15), %rcx jne 0x13dad79 movq 0x48(%r15), %rax subq 0x40(%r15), %rax movq %rbx, %rdi movq %r12, %rsi movq %r15, %rdx movq %r14, %rcx cmpq $0x3ff, %rax # imm = 0x3FF ja 0x13dad48 callq 0x13ddaa6 jmp 0x13dad4d callq 0x13de204 movq %rbx, %rax addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %r12, %rdi movq %r15, %rsi callq 0x13dc2bc testb $0x1, %dl je 0x13dad1b movq 0x48(%r15), %rcx movq %rcx, 0x50(%r15) jmp 0x13dad1b cmpq %rcx, %rax ja 0x13dad89 movq %r12, %rdi movq %r15, %rsi callq 0x13dc4ce movq 0x48(%r15), %rax movq 0x50(%r15), %rcx subq %rax, %rcx movq 0x10(%r12), %rdx decq %rdx cmpq %rdx, %rcx jae 0x13dad25 movq %rax, 0x50(%r15) jmp 0x13dad25 leaq -0x1(%rax), %rdx cmpq %rdx, %rcx jae 0x13dae25 movq (%r12), %rsi movq %rcx, %rdi shlq $0x6, %rdi addq %rsi, %rdi addq $0x50, %rdi movb $0x1, %r8b movq %rcx, %r9 movq %r9, %r10 shlq $0x6, %r10 addq %rsi, %r10 movq %rdi, %r13 movq %rcx, %r11 incq %r11 cmpq %rax, %r11 jae 0x13dae13 vmovaps (%r10), %xmm0 vmovaps 0x10(%r10), %xmm1 vmaxps -0x10(%r13), %xmm0, %xmm0 vminps (%r13), %xmm1, %xmm1 addq $0x40, %r13 vcmpleps %xmm1, %xmm0, %xmm0 vmovmskps %xmm0, %ebp notb %bpl testb $0x7, %bpl jne 0x13daddb xorl %r8d, %r8d incq %r9 cmpq %rdx, %r9 jne 0x13dadcb testb $0x1, %r8b je 0x13dad13 movq %rax, 0x50(%r15) jmp 0x13dad13
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::openNodesBasedOnExtend(embree::avx::PrimInfoExtRange&)
__noinline void openNodesBasedOnExtend(PrimInfoExtRange& set) { const OpenHeuristic heuristic(set); const size_t ext_range_start = set.end(); if (false && set.size() < PARALLEL_THRESHOLD) { size_t extra_elements = 0; for (size_t i=set.begin(); i<set.end(); i++) { if (heuristic(prims0[i])) { PrimRef tmp[MAX_OPENED_CHILD_NODES]; const size_t n = nodeOpenerFunc(prims0[i],tmp); assert(extra_elements + n-1 <= set.ext_range_size()); for (size_t j=0; j<n; j++) set.extend_center2(tmp[j]); prims0[i] = tmp[0]; for (size_t j=1; j<n; j++) prims0[ext_range_start+extra_elements+j-1] = tmp[j]; extra_elements += n-1; } } set._end += extra_elements; } else { std::atomic<size_t> ext_elements; ext_elements.store(0); PrimInfo info = parallel_reduce( set.begin(), set.end(), CREATE_SPLITS_STEP_SIZE, PrimInfo(empty), [&](const range<size_t>& r) -> PrimInfo { PrimInfo info(empty); for (size_t i=r.begin(); i<r.end(); i++) if (heuristic(prims0[i])) { PrimRef tmp[MAX_OPENED_CHILD_NODES]; const size_t n = nodeOpenerFunc(prims0[i],tmp); const size_t ID = ext_elements.fetch_add(n-1); assert(ID + n-1 <= set.ext_range_size()); for (size_t j=0; j<n; j++) info.extend_center2(tmp[j]); prims0[i] = tmp[0]; for (size_t j=1; j<n; j++) prims0[ext_range_start+ID+j-1] = tmp[j]; } return info; }, [] (const PrimInfo& a, const PrimInfo& b) { return PrimInfo::merge(a,b); }); set.centBounds.extend(info.centBounds); assert(ext_elements.load() <= set.ext_range_size()); set._end += ext_elements.load(); } }
pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x1f8, %rsp # imm = 0x1F8 movq %rsi, %rbx vmovaps 0x10(%rsi), %xmm0 vsubps (%rsi), %xmm0, %xmm0 vbroadcastss 0xb449d3(%rip), %xmm1 # 0x1f20ec4 vandps %xmm1, %xmm0, %xmm1 vshufps $0xf5, %xmm1, %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] xorl %eax, %eax vucomiss %xmm2, %xmm1 setbe %al vmaxss %xmm2, %xmm1, %xmm2 vshufps $0x4e, %xmm1, %xmm1, %xmm1 # xmm1 = xmm1[2,3,0,1] vucomiss %xmm1, %xmm2 movl $0x2, %ecx cmovaq %rax, %rcx leaq 0x90(%rsp), %r14 vmovaps %xmm0, (%r14) leaq 0x50(%rsp), %rax movq %rcx, (%rax) vmovss 0xb101de(%rip), %xmm0 # 0x1eec714 vdivss 0x90(%rsp,%rcx,4), %xmm0, %xmm0 vmovss %xmm0, 0x8(%rax) movq 0x48(%rsi), %rcx leaq 0x48(%rsp), %rdx movq %rcx, (%rdx) xorl %ecx, %ecx leaq 0x28(%rsp), %rsi xchgq %rcx, (%rsi) vbroadcastss 0xb0f4bd(%rip), %xmm0 # 0x1eeba20 leaq 0x1a0(%rsp), %r15 vmovaps %xmm0, (%r15) vbroadcastss 0xb1060b(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%r15) vmovaps %xmm0, 0x20(%r15) vmovaps %xmm1, 0x30(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x40(%r15) leaq 0x110(%rsp), %r12 movq %rax, (%r12) movq %rdi, 0x8(%r12) movq %rsi, 0x10(%r12) movq %rdx, 0x18(%r12) movw $0x401, 0xc(%r14) # imm = 0x401 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) vpermilps $0x4e, 0x40(%rbx), %xmm0 # xmm0 = mem[2,3,0,1] vmovaps %xmm0, 0x10(%rsp) movq %r14, %rdi callq 0x6a660 leaq 0x70(%rsp), %rdi vmovaps 0x10(%rsp), %xmm0 vmovaps %xmm0, (%rdi) movq $0x80, 0x10(%rdi) leaq 0xe(%rsp), %rax leaq 0x60(%rsp), %rcx movq %rax, (%rcx) movq %r12, 0x8(%rcx) leaq 0x130(%rsp), %rsi movq %r15, (%rsi) movq %rcx, 0x8(%rsi) movq %rax, 0x10(%rsi) vmovups (%r15), %ymm0 vmovups 0x20(%r15), %ymm1 vmovups %ymm0, 0x20(%rsi) vmovups %ymm1, 0x40(%rsi) vmovaps 0x40(%r15), %xmm0 vmovaps %xmm0, 0x60(%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx vzeroupper callq 0x13dcfca vmovaps 0x170(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rsp) vmovaps 0x180(%rsp), %xmm0 vmovaps %xmm0, 0x10(%rsp) leaq 0x90(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13dc6b9 leaq 0x90(%rsp), %rdi callq 0x6aab0 vmovaps 0x20(%rbx), %xmm0 vminps 0x30(%rsp), %xmm0, %xmm0 vmovaps 0x30(%rbx), %xmm1 vmaxps 0x10(%rsp), %xmm1, %xmm1 vmovaps %xmm0, 0x20(%rbx) vmovaps %xmm1, 0x30(%rbx) movq 0x28(%rsp), %rax addq %rax, 0x48(%rbx) addq $0x1f8, %rsp # imm = 0x1F8 popq %rbx popq %r12 popq %r14 popq %r15 retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb0f423(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd4852c(%rip), %rsi # 0x2124c08 movq 0xd482e5(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13dc6ff jmp 0x13dc714 jmp 0x13dc6fc movq %rax, %rbx leaq 0x90(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
unsigned long embree::parallel_partitioning<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>, embree::EmptyTy, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&), embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&), embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&)>(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*, unsigned long, unsigned long, embree::EmptyTy const&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&) const&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef const&) const&, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleM<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&) const&, unsigned long)
__noinline size_t parallel_partitioning(T* array, const size_t begin, const size_t end, const Vi &identity, V &leftReduction, V &rightReduction, const IsLeft& is_left, const Reduction_T& reduction_t, const Reduction_V& reduction_v, size_t BLOCK_SIZE = 128) { /* fall back to single threaded partitioning for small N */ if (unlikely(end-begin < BLOCK_SIZE)) return serial_partitioning(array,begin,end,leftReduction,rightReduction,is_left,reduction_t); /* otherwise use parallel code */ else { typedef parallel_partition_task<T,V,Vi,IsLeft,Reduction_T,Reduction_V> partition_task; std::unique_ptr<partition_task> p(new partition_task(&array[begin],end-begin,identity,is_left,reduction_t,reduction_v,BLOCK_SIZE)); return begin+p->partition(leftReduction,rightReduction); } }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x20, %rsp subq $0x100, %rsp # imm = 0x100 movq %r9, %r15 movq %r8, %r12 movq 0x18(%rbp), %r8 movq 0x10(%rbp), %r9 movq %rdx, %rbx subq %rsi, %rbx cmpq 0x28(%rbp), %rbx jb 0x13e0f94 movq %rdi, %r13 movl $0x34c0, %edi # imm = 0x34C0 movq %rsi, 0x8(%rsp) movl $0x40, %esi movq %rcx, 0x38(%rsp) callq 0x1ee60ac movq %rax, %r14 movq 0x8(%rsp), %rax shlq $0x6, %rax addq %rax, %r13 movq %r13, (%r14) movq %rbx, 0x8(%r14) movq 0x10(%rbp), %rax movq %rax, 0x10(%r14) movq 0x18(%rbp), %rax movq %rax, 0x18(%r14) movq 0x20(%rbp), %rax movq %rax, 0x20(%r14) movq 0x38(%rsp), %rax movq %rax, 0x28(%r14) xorl %edi, %edi callq 0x6ab80 movl %eax, %ecx movq 0x28(%rbp), %rsi leaq (%rbx,%rsi), %rax decq %rax xorl %edx, %edx divq %rsi cmpl $0x40, %ecx movl $0x40, %ebx cmovbl %ecx, %ebx cmpq %rbx, %rax cmovbq %rax, %rbx movq %rbx, 0x30(%r14) movq %r14, 0x28(%rsp) movq %r14, 0x30(%rsp) leaq 0x60(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) callq 0x6a660 leaq 0x30(%rsp), %rax leaq 0x40(%rsp), %rcx movq %rax, (%rcx) leaq 0x10(%rsp), %r8 leaq 0x60(%rsp), %r9 movl $0x1, %edx xorl %edi, %edi movq %rbx, %rsi callq 0x13e12b2 leaq 0x60(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13e1176 leaq 0x60(%rsp), %rdi callq 0x6aab0 movq 0x8(%r14), %rax movq 0x30(%r14), %rcx movq %rax, 0x40(%r14,%rcx,8) movq 0x30(%r14), %rax movq $0x0, 0x280(%r14,%rax,8) movq 0x30(%r14), %rdx testq %rdx, %rdx je 0x13e0d0f leaq 0x2108(%r14), %rax xorl %ecx, %ecx vmovaps (%r12), %xmm0 vmovaps 0x10(%r12), %xmm1 vmovaps 0x20(%r12), %xmm2 vminps -0x1448(%rax), %xmm0, %xmm0 vmovaps 0x30(%r12), %xmm3 vmovaps %xmm0, (%r12) vmaxps -0x1438(%rax), %xmm1, %xmm0 vmovaps %xmm0, 0x10(%r12) vminps -0x1428(%rax), %xmm2, %xmm0 vmovaps %xmm0, 0x20(%r12) vmaxps -0x1418(%rax), %xmm3, %xmm0 vmovaps %xmm0, 0x30(%r12) movq -0x1408(%rax), %rdx addq %rdx, 0x40(%r12) movq -0x1400(%rax), %rdx addq %rdx, 0x48(%r12) vmovaps (%r15), %xmm0 vmovaps 0x10(%r15), %xmm1 vmovaps 0x20(%r15), %xmm2 vminps -0x48(%rax), %xmm0, %xmm0 vmovaps 0x30(%r15), %xmm3 vmovaps %xmm0, (%r15) vmaxps -0x38(%rax), %xmm1, %xmm0 vmovaps %xmm0, 0x10(%r15) vminps -0x28(%rax), %xmm2, %xmm0 vmovaps %xmm0, 0x20(%r15) vmaxps -0x18(%rax), %xmm3, %xmm0 vmovaps %xmm0, 0x30(%r15) movq -0x8(%rax), %rdx addq %rdx, 0x40(%r15) movq (%rax), %rdx addq %rdx, 0x48(%r15) incq %rcx movq 0x30(%r14), %rdx addq $0x50, %rax cmpq %rdx, %rcx jb 0x13e0c3c movq 0x280(%r14), %r13 cmpq $0x2, %rdx jb 0x13e0e0c leaq 0x2(%rdx), %rax andq $-0x4, %rax addq $-0x2, %rdx vmovq %r13, %xmm0 vmovq %rdx, %xmm1 vpshufd $0x44, %xmm1, %xmm1 # xmm1 = xmm1[0,1,0,1] vinsertf128 $0x1, %xmm1, %ymm1, %ymm1 leaq 0x288(%r14), %rcx xorl %edx, %edx vmovaps 0xb410ee(%rip), %ymm3 # 0x1f21e40 vextractf128 $0x1, %ymm1, %xmm4 vmovddup 0xb41128(%rip), %xmm2 # xmm2 = mem[0,0] vxorps %xmm2, %xmm4, %xmm4 vpcmpeqd %xmm5, %xmm5, %xmm5 vxorps %xmm2, %xmm1, %xmm6 movq 0x8(%rsp), %r15 vmovq %rdx, %xmm7 vpshufd $0x44, %xmm7, %xmm7 # xmm7 = xmm7[0,1,0,1] vinsertf128 $0x1, %xmm7, %ymm7, %ymm7 vorps %ymm3, %ymm7, %ymm7 vextractf128 $0x1, %ymm7, %xmm8 vxorps %xmm2, %xmm8, %xmm8 vpcmpgtq %xmm4, %xmm8, %xmm8 vpxor %xmm5, %xmm8, %xmm9 vxorps %xmm2, %xmm7, %xmm10 vpcmpgtq %xmm6, %xmm10, %xmm10 vpxor %xmm5, %xmm10, %xmm10 vinsertf128 $0x1, %xmm9, %ymm10, %ymm9 vmaskmovpd (%rcx,%rdx,8), %ymm9, %ymm11 vmovdqa %ymm0, %ymm9 vextractf128 $0x1, %ymm11, %xmm0 vextractf128 $0x1, %ymm9, %xmm10 vpaddq %xmm0, %xmm10, %xmm0 vpaddq %xmm9, %xmm11, %xmm11 vinsertf128 $0x1, %xmm0, %ymm11, %ymm0 addq $0x4, %rdx cmpq %rdx, %rax jne 0x13e0d71 vxorps %xmm2, %xmm1, %xmm1 vxorps %xmm2, %xmm7, %xmm2 vpcmpgtq %xmm1, %xmm2, %xmm1 vextractf128 $0x1, %ymm0, %xmm2 vblendvpd %xmm8, %xmm10, %xmm2, %xmm2 vblendvpd %xmm1, %xmm9, %xmm0, %xmm0 vpaddq %xmm2, %xmm0, %xmm0 vpshufd $0xee, %xmm0, %xmm1 # xmm1 = xmm0[2,3,2,3] vpaddq %xmm1, %xmm0, %xmm0 vmovq %xmm0, %r13 jmp 0x13e0e11 movq 0x8(%rsp), %r15 movq 0x8(%r14), %rax xorl %ecx, %ecx movq %rcx, 0x10(%rsp) movq %rcx, 0x20(%rsp) movq %rcx, 0x18(%rsp) movq 0x30(%r14), %rbx testq %rbx, %rbx je 0x13e0ed5 leaq 0x4c0(%r14), %rcx movq %r14, %rdx addq $0x8c0, %rdx # imm = 0x8C0 xorl %esi, %esi movq 0x40(%r14,%rsi,8), %rdi movq 0x280(%r14,%rsi,8), %r8 addq %rdi, %r8 testq %r8, %r8 movl $0x0, %r9d cmovgq %r8, %r9 movq 0x48(%r14,%rsi,8), %r10 cmpq %r10, %r13 cmovlq %r13, %r10 cmpq %rdi, %r13 cmovgq %r13, %rdi cmpq %r8, %rax cmovlq %rax, %r8 movq %r10, %r11 subq %r9, %r11 jle 0x13e0ea5 addq %r11, 0x18(%rsp) movq 0x10(%rsp), %r11 leaq 0x1(%r11), %rbx movq %rbx, 0x10(%rsp) shlq $0x4, %r11 movq %r9, (%rcx,%r11) movq %r10, 0x8(%rcx,%r11) cmpq %rdi, %r8 jle 0x13e0ec5 movq 0x20(%rsp), %r9 leaq 0x1(%r9), %r10 movq %r10, 0x20(%rsp) shlq $0x4, %r9 movq %rdi, (%rdx,%r9) movq %r8, 0x8(%rdx,%r9) incq %rsi movq 0x30(%r14), %rbx cmpq %rbx, %rsi jb 0x13e0e46 cmpq $0x0, 0x18(%rsp) je 0x13e0f6f leaq 0x18(%rsp), %rax movq %rax, 0x40(%rsp) movq %r14, 0x48(%rsp) leaq 0x10(%rsp), %rax movq %rax, 0x50(%rsp) leaq 0x20(%rsp), %rax movq %rax, 0x58(%rsp) leaq 0x60(%rsp), %rdi movw $0x401, 0xc(%rdi) # imm = 0x401 vpxor %xmm0, %xmm0, %xmm0 vmovdqu %xmm0, 0x20(%rdi) movq $0x8, 0x40(%rdi) vzeroupper callq 0x6a660 leaq 0x30(%rsp), %rcx leaq 0x40(%rsp), %rax movq %rax, (%rcx) leaq 0x7(%rsp), %r8 leaq 0x60(%rsp), %r9 movl $0x1, %edx xorl %edi, %edi movq %rbx, %rsi callq 0x13e1d22 leaq 0x60(%rsp), %rdi callq 0x6a770 testb %al, %al movq 0x8(%rsp), %r15 jne 0x13e11a8 leaq 0x60(%rsp), %rdi callq 0x6aab0 addq %r15, %r13 leaq 0x28(%rsp), %rdi vzeroupper callq 0x13e1248 movq %r13, %rax leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movq %rdx, %r14 shlq $0x6, %rsi addq %rdi, %rsi shlq $0x6, %r14 addq %rdi, %r14 movq %rsi, %r13 addq $-0x40, %r14 cmpq %r14, %r13 ja 0x13e1091 movq (%r9), %rax movq 0x8(%r9), %rcx movq 0x10(%r9), %rdx vmovaps (%r13), %xmm2 vmovaps 0x10(%r13), %xmm1 vaddps %xmm1, %xmm2, %xmm0 vsubps 0x20(%rax), %xmm0, %xmm3 vmulps 0x30(%rax), %xmm3, %xmm3 vroundps $0x1, %xmm3, %xmm3 vcvtps2dq %xmm3, %xmm3 vmovdqa (%rcx), %xmm4 vpcmpgtd %xmm3, %xmm4, %xmm3 vtestps (%rdx), %xmm3 je 0x13e1091 vmovaps (%r12), %xmm3 vmovaps 0x10(%r12), %xmm4 vmovaps 0x20(%r12), %xmm5 vmovaps 0x30(%r12), %xmm6 vminps %xmm2, %xmm3, %xmm2 vmovaps %xmm2, (%r12) vmaxps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0x10(%r12) vminps %xmm0, %xmm5, %xmm1 vmovaps %xmm1, 0x20(%r12) vmaxps %xmm0, %xmm6, %xmm0 vmovaps %xmm0, 0x30(%r12) incq 0x48(%r12) addq $0x40, %r13 jmp 0x13e0fac vmovaps (%r15), %xmm3 vmovaps 0x10(%r15), %xmm4 vmovaps 0x20(%r15), %xmm5 vmovaps 0x30(%r15), %xmm6 vminps %xmm2, %xmm3, %xmm2 vmovaps %xmm2, (%r15) vmaxps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0x10(%r15) vminps %xmm0, %xmm5, %xmm1 vmovaps %xmm1, 0x20(%r15) vmaxps %xmm0, %xmm6, %xmm0 vmovaps %xmm0, 0x30(%r15) incq 0x48(%r15) addq $-0x40, %r14 cmpq %r14, %r13 seta %al ja 0x13e116a movq (%r9), %rcx movq 0x8(%r9), %rdx movq 0x10(%r9), %rsi vmovaps (%r14), %xmm2 vmovaps 0x10(%r14), %xmm1 vaddps %xmm1, %xmm2, %xmm0 vsubps 0x20(%rcx), %xmm0, %xmm3 vmulps 0x30(%rcx), %xmm3, %xmm3 vroundps $0x1, %xmm3, %xmm3 vcvtps2dq %xmm3, %xmm3 vmovdqa (%rdx), %xmm4 vpcmpgtd %xmm3, %xmm4, %xmm3 vtestps (%rsi), %xmm3 je 0x13e104b testb $0x1, %al jne 0x13e116a movq %rdi, 0x8(%rsp) movq %r8, %rdi movq %r12, %rsi movq %r14, %rdx movq %r8, %rbx vzeroupper callq 0x13e126a movq %rbx, %rdi movq %r15, %rsi movq %r13, %rdx callq 0x13e126a movq 0x10(%rbp), %r9 movq %rbx, %r8 movq 0x8(%rsp), %rdi vmovaps (%r13), %ymm0 movl 0x28(%r13), %eax movl %eax, 0x68(%rsp) movq 0x20(%r13), %rax movq %rax, 0x60(%rsp) vmovaps (%r14), %ymm1 vmovaps %ymm1, (%r13) movl 0x28(%r14), %eax movl %eax, 0x28(%r13) movq 0x20(%r14), %rax movq %rax, 0x20(%r13) vmovaps %ymm0, (%r14) movq 0x60(%rsp), %rax movq %rax, 0x20(%r14) movl 0x68(%rsp), %eax movl %eax, 0x28(%r14) addq $0x40, %r13 jmp 0x13e0fa8 subq %rdi, %r13 sarq $0x6, %r13 jmp 0x13e0f7f movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb0a966(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd43a6f(%rip), %rsi # 0x2124c08 movq 0xd43828(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xb0a934(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd43a3d(%rip), %rsi # 0x2124c08 movq 0xd437f6(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13e11fb movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13e121f jmp 0x13e122b jmp 0x13e11f8 movq %rax, %rbx leaq 0x60(%rsp), %rdi callq 0x6aab0 jmp 0x13e1236 jmp 0x13e122b jmp 0x13e122b jmp 0x13e121c movq %rax, %rbx movq %r14, %rdi callq 0x1ee612d jmp 0x13e1240 jmp 0x13e122b movq %rax, %rbx leaq 0x60(%rsp), %rdi callq 0x6aab0 jmp 0x13e1236 movq %rax, %rdi callq 0x8d6de8 movq %rax, %rbx leaq 0x28(%rsp), %rdi callq 0x13e1248 movq %rbx, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/../builders/../../common/algorithms/parallel_partition.h
embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::deleteGeometry(unsigned long)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::deleteGeometry(size_t geomID) { if (geomID >= bvh->objects.size()) return; if (builders[geomID]) builders[geomID].reset(); delete bvh->objects [geomID]; bvh->objects [geomID] = nullptr; }
movq 0x28(%rdi), %rax movq 0x208(%rax), %rcx subq 0x200(%rax), %rcx sarq $0x3, %rcx cmpq %rsi, %rcx jbe 0x13e3cf4 pushq %r14 pushq %rbx pushq %rax movq %rsi, %rbx movq %rdi, %r14 movq 0x10(%rdi), %rax movq (%rax,%rsi,8), %rdi testq %rdi, %rdi je 0x13e3cc0 movq $0x0, (%rax,%rbx,8) movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13e3cda movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq $0x0, (%rax,%rbx,8) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::NodeRefPtr<8> embree::avx::BVHBuilderBinnedOpenMergeSAH::build<embree::NodeRefPtr<8>, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(unsigned long)>(embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(unsigned long), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*, unsigned long, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&, embree::avx::GeneralBVHBuilder::Settings const&)
static ReductionTy build(CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, NodeOpenerFunc nodeOpenerFunc, ProgressMonitor progressMonitor, BuildRef* prims, const size_t extSize, const PrimInfo& pinfo, const Settings& settings) { typedef HeuristicArrayOpenMergeSAH<NodeOpenerFunc,BuildRef,NUM_OBJECT_BINS_HQ> Heuristic; Heuristic heuristic(nodeOpenerFunc,prims,settings.branchingFactor); return GeneralBVHBuilder::build<ReductionTy,Heuristic,Set,BuildRef>( heuristic, prims, PrimInfoExtRange(0,pinfo.size(),extSize,pinfo), createAlloc, createNode, updateNode, createLeaf, progressMonitor, settings); }
pushq %r14 pushq %rbx subq $0x98, %rsp movq %r8, %rax movq %rcx, %r10 movq %rsi, %r8 movq %rdi, %rcx movq 0xb0(%rsp), %rsi movq 0xb8(%rsp), %rbx leaq 0x10(%rsp), %r14 movq %rdx, (%r14) leaq 0x8(%rsp), %r11 movq %r10, (%r11) movq (%rbx), %rdx leaq 0x18(%rsp), %rdi movq %rax, (%rdi) movq %r14, 0x8(%rdi) movq %rdx, 0x10(%rdi) movq 0x48(%rsi), %r10 subq 0x40(%rsi), %r10 vmovaps (%rsi), %xmm0 leaq 0x30(%rsp), %rdx vmovaps %xmm0, (%rdx) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, 0x10(%rdx) vmovaps 0x20(%rsi), %xmm0 vmovaps %xmm0, 0x20(%rdx) vmovaps 0x30(%rsi), %xmm0 vmovaps %xmm0, 0x30(%rdx) movq $0x0, 0x40(%rdx) movq %r10, 0x48(%rdx) movq %r9, 0x50(%rdx) movq %rbx, (%rsp) movq %rax, %rsi movq %r11, %r9 callq 0x13e7752 addq $0x98, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::createMeshAccel(unsigned long, embree::Builder*&)
void createMeshAccel (size_t geomID, Builder*& builder) { bvh->objects[geomID] = new BVH(Primitive::type,scene); BVH* accel = bvh->objects[geomID]; auto mesh = scene->getSafe<Mesh>(geomID); if (nullptr == mesh) { throw_RTCError(RTC_ERROR_INVALID_ARGUMENT,"geomID does not return correct type"); return; } __internal_two_level_builder__::MeshBuilder<N,Mesh,Primitive>()(accel, mesh, geomID, this->gtype, this->useMortonBuilder_, builder); }
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x30, %rsp movq %rdx, %r14 movq %rsi, %rbx movq %rdi, %r15 movl $0x240, %edi # imm = 0x240 movl $0x10, %esi callq 0x1ee60ac movq %rax, %r12 movq 0x30(%r15), %rdx leaq 0xd3ef56(%rip), %rsi # 0x21251b8 movq %rax, %rdi callq 0x14ce5ec movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq %r12, (%rax,%rbx,8) movq 0x30(%r15), %rax movq 0x1e8(%rax), %rax movq (%rax,%rbx,8), %rax xorl %edx, %edx testq %rax, %rax je 0x13e6297 cmpb $0x14, 0x3c(%rax) cmoveq %rax, %rdx testq %rdx, %rdx je 0x13e62d8 movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rsi movq 0x98(%r15), %r8 movzbl 0xa0(%r15), %r9d movq %r14, (%rsp) leaq 0x10(%rsp), %rdi movq %rbx, %rcx callq 0x13e63a4 addq $0x30, %rsp popq %rbx popq %r12 popq %r14 popq %r15 popq %rbp retq movl $0x30, %edi callq 0x6a3b0 movq %rax, %rbx leaq 0x20(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb2c4ab(%rip), %rsi # 0x1f127a0 leaq 0xb2c4c7(%rip), %rdx # 0x1f127c3 leaq 0x10(%rsp), %rdi callq 0x8d7230 leaq 0xd1b5a3(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x2, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x10(%rsp), %rsi movq 0x18(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd1b511(%rip), %rsi # 0x2101850 leaq -0xb0f52a(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rax, %r14 xorl %ebp, %ebp jmp 0x13e6363 movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x10(%rsp), %rdi cmpq %r15, %rdi je 0x13e637a callq 0x6a4f0 jmp 0x13e637a movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x13e6394 movq %rbx, %rdi callq 0x6a8a0 jmp 0x13e6394 movq %rax, %r14 movq %r12, %rdi callq 0x1ee612d movq %r14, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.h
embree::avx::GeneralBVHBuilder::BuilderT<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, embree::NodeRefPtr<8>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(unsigned long)>::BuilderT(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, 32ul>&, embree::BVHN<8>::CreateAlloc const&, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2 const&, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2 const&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, embree::avx::PrimInfoExtRange> const&, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, embree::avx::PrimInfoExtRange> const&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(unsigned long) const&, embree::avx::GeneralBVHBuilder::Settings const&)
BuilderT (PrimRef* prims, Heuristic& heuristic, const CreateAllocFunc& createAlloc, const CreateNodeFunc& createNode, const UpdateNodeFunc& updateNode, const CreateLeafFunc& createLeaf, const CanCreateLeafFunc& canCreateLeaf, const CanCreateLeafSplitFunc& canCreateLeafSplit, const ProgressMonitor& progressMonitor, const Settings& settings) : cfg(settings), prims(prims), heuristic(heuristic), createAlloc(createAlloc), createNode(createNode), updateNode(updateNode), createLeaf(createLeaf), canCreateLeaf(canCreateLeaf), canCreateLeafSplit(canCreateLeafSplit), progressMonitor(progressMonitor) { if (cfg.branchingFactor > MAX_BRANCHING_FACTOR) throw_RTCError(RTC_ERROR_UNKNOWN,"bvh_builder: branching factor too large"); }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x28, %rsp vmovups 0x50(%rsp), %ymm0 movq 0x70(%rsp), %rax vmovups (%rax), %ymm1 vmovups 0x20(%rax), %ymm2 vmovups %ymm1, (%rdi) vmovups %ymm2, 0x20(%rdi) movq %rsi, 0x40(%rdi) movq %rdx, 0x48(%rdi) movq %rcx, 0x50(%rdi) movq %r8, 0x58(%rdi) movq %r9, 0x60(%rdi) vmovups %ymm0, 0x68(%rdi) cmpq $0x11, (%rdi) jae 0x13e78b0 addq $0x28, %rsp popq %rbx popq %r14 popq %r15 popq %rbp vzeroupper retq movl $0x30, %edi vzeroupper callq 0x6a3b0 movq %rax, %rbx leaq 0x18(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb0537c(%rip), %rsi # 0x1eecc4c leaq 0xb0539c(%rip), %rdx # 0x1eecc73 leaq 0x8(%rsp), %rdi callq 0x8d7230 leaq 0xd19fc8(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x1, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x8(%rsp), %rsi movq 0x10(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd19f36(%rip), %rsi # 0x2101850 leaq -0xb10b05(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rax, %r14 xorl %ebp, %ebp jmp 0x13e793e movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x8(%rsp), %rdi cmpq %r15, %rdi je 0x13e7955 callq 0x6a4f0 jmp 0x13e7955 movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x13e7962 movq %rbx, %rdi callq 0x6a8a0 movq %r14, %rdi callq 0x6a600
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, 32ul>::sequential_object_find(embree::avx::PrimInfoExtRange const&, unsigned long)
__noinline const Split sequential_object_find(const PrimInfoExtRange& set, const size_t logBlockSize) { Binner binner(empty); const BinMapping<OBJECT_BINS> mapping(set.centBounds); binner.bin(prims0,set.begin(),set.end(),mapping); return binner.best(mapping,logBlockSize); }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x40, %rsp subq $0x1300, %rsp # imm = 0x1300 movq %rdi, %rax movl $0xc00, %edi # imm = 0xC00 leaq 0x110(%rsp), %r8 vbroadcastss 0xb00743(%rip), %xmm2 # 0x1eeba20 vbroadcastss 0xb0189e(%rip), %xmm3 # 0x1eecb84 vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm2, -0x10(%r8) vmovaps %xmm3, (%r8) vmovaps %xmm2, -0x30(%r8) vmovaps %xmm3, -0x20(%r8) vmovaps %xmm2, -0x50(%r8) vmovaps %xmm3, -0x40(%r8) vmovaps %xmm0, 0xc0(%rsp,%rdi) addq $0x10, %rdi addq $0x60, %r8 cmpq $0xe00, %rdi # imm = 0xE00 jne 0x13eb2ea movq $0x20, 0x90(%rsp) vmovaps 0x20(%rdx), %xmm0 vmovaps 0x30(%rdx), %xmm1 vsubps %xmm0, %xmm1, %xmm1 vbroadcastss 0xb36b26(%rip), %xmm4 # 0x1f21e70 vmaxps %xmm1, %xmm4, %xmm1 vcmpnleps %xmm4, %xmm1, %xmm4 vbroadcastss 0xb4d6cc(%rip), %xmm5 # 0x1f38a28 vdivps %xmm1, %xmm5, %xmm1 vandps %xmm1, %xmm4, %xmm1 vmovaps %xmm1, 0xb0(%rsp) vmovaps %xmm0, 0xa0(%rsp) movq 0x40(%rdx), %rdi movq 0x48(%rdx), %r8 subq %rdi, %r8 je 0x13eb705 movq (%rsi), %rdx shlq $0x6, %rdi movq %r8, %r9 decq %r9 movq %rdx, 0x30(%rsp) je 0x13eb5e3 leaq 0xcc0(%rsp), %r10 leaq 0xcc4(%rsp), %r11 leaq 0xcc8(%rsp), %rbx leaq (%rdi,%rdx), %r14 addq $0x5c, %r14 xorl %esi, %esi vbroadcastss 0xb4d662(%rip), %xmm4 # 0x1f38a2c vxorps %xmm5, %xmm5, %xmm5 vmovaps -0x5c(%r14), %xmm9 vmovaps -0x4c(%r14), %xmm8 vmovaps -0x1c(%r14), %xmm7 vmovaps -0xc(%r14), %xmm6 vaddps %xmm8, %xmm9, %xmm10 vsubps %xmm0, %xmm10, %xmm10 vmulps %xmm1, %xmm10, %xmm10 vroundps $0x1, %xmm10, %xmm10 vcvtps2dq %xmm10, %xmm10 vpminsd %xmm4, %xmm10, %xmm10 vpmaxsd %xmm5, %xmm10, %xmm11 vaddps %xmm6, %xmm7, %xmm10 vsubps %xmm0, %xmm10, %xmm10 vmulps %xmm1, %xmm10, %xmm10 vroundps $0x1, %xmm10, %xmm10 vcvtps2dq %xmm10, %xmm10 vmovd %xmm11, %r15d leaq (%r15,%r15,2), %r13 shll $0x5, %r13d vmovaps 0xc0(%rsp,%r13), %xmm12 vmovaps 0xd0(%rsp,%r13), %xmm13 vminps %xmm9, %xmm12, %xmm12 vmovaps %xmm12, 0xc0(%rsp,%r13) vmaxps %xmm8, %xmm13, %xmm12 vpextrd $0x1, %xmm11, %r12d vmovaps %xmm12, 0xd0(%rsp,%r13) leaq (%r12,%r12,2), %r13 shll $0x5, %r13d vmovaps 0xe0(%rsp,%r13), %xmm12 vmovaps 0xf0(%rsp,%r13), %xmm13 vminps %xmm9, %xmm12, %xmm12 vmovaps %xmm12, 0xe0(%rsp,%r13) vmaxps %xmm8, %xmm13, %xmm12 vmovaps %xmm12, 0xf0(%rsp,%r13) vpextrd $0x2, %xmm11, %r13d leaq (,%r13,2), %rdx addq %r13, %rdx shll $0x5, %edx vmovaps 0x100(%rsp,%rdx), %xmm11 vmovaps 0x110(%rsp,%rdx), %xmm12 vminps %xmm9, %xmm11, %xmm9 vmovaps %xmm9, 0x100(%rsp,%rdx) vmaxps %xmm8, %xmm12, %xmm8 vmovaps %xmm8, 0x110(%rsp,%rdx) movl -0x40(%r14), %edx shlq $0x4, %r15 addl %edx, (%r10,%r15) shlq $0x4, %r12 addl %edx, (%r11,%r12) shlq $0x4, %r13 addl %edx, (%rbx,%r13) vpminsd %xmm4, %xmm10, %xmm8 vpmaxsd %xmm5, %xmm8, %xmm8 vmovd %xmm8, %r15d leaq (%r15,%r15,2), %rdx shll $0x5, %edx vmovaps 0xc0(%rsp,%rdx), %xmm9 vmovaps 0xd0(%rsp,%rdx), %xmm10 vminps %xmm7, %xmm9, %xmm9 vmovaps %xmm9, 0xc0(%rsp,%rdx) vmaxps %xmm6, %xmm10, %xmm9 vpextrd $0x1, %xmm8, %r12d vmovaps %xmm9, 0xd0(%rsp,%rdx) leaq (%r12,%r12,2), %rdx shll $0x5, %edx vmovaps 0xe0(%rsp,%rdx), %xmm9 vmovaps 0xf0(%rsp,%rdx), %xmm10 vminps %xmm7, %xmm9, %xmm9 vmovaps %xmm9, 0xe0(%rsp,%rdx) vmaxps %xmm6, %xmm10, %xmm9 vmovaps %xmm9, 0xf0(%rsp,%rdx) vpextrd $0x2, %xmm8, %edx leaq (%rdx,%rdx,2), %r13 shll $0x5, %r13d vmovaps 0x100(%rsp,%r13), %xmm8 vmovaps 0x110(%rsp,%r13), %xmm9 vminps %xmm7, %xmm8, %xmm7 vmovaps %xmm7, 0x100(%rsp,%r13) vmaxps %xmm6, %xmm9, %xmm6 vmovaps %xmm6, 0x110(%rsp,%r13) movl (%r14), %r13d shlq $0x4, %r15 addl %r13d, (%r10,%r15) shlq $0x4, %r12 addl %r13d, (%r11,%r12) shlq $0x4, %rdx addl %r13d, (%rbx,%rdx) addq $0x2, %rsi subq $-0x80, %r14 cmpq %r9, %rsi jb 0x13eb3ce jmp 0x13eb5e5 xorl %esi, %esi cmpq %r8, %rsi jae 0x13eb705 movq 0x30(%rsp), %rdx addq %rdi, %rdx shlq $0x6, %rsi vmovaps (%rdx,%rsi), %xmm5 vmovaps 0x10(%rdx,%rsi), %xmm4 vaddps %xmm4, %xmm5, %xmm6 vsubps %xmm0, %xmm6, %xmm6 vmulps %xmm6, %xmm1, %xmm6 vroundps $0x1, %xmm6, %xmm6 vcvtps2dq %xmm6, %xmm6 vbroadcastss 0xb4d408(%rip), %xmm7 # 0x1f38a2c vpminsd %xmm7, %xmm6, %xmm6 vpxor %xmm7, %xmm7, %xmm7 vpmaxsd %xmm7, %xmm6, %xmm6 movl 0x1c(%rdx,%rsi), %edx vmovd %xmm6, %edi leaq (%rdi,%rdi,2), %rsi shlq $0x4, %rdi addl %edx, 0xcc0(%rsp,%rdi) shlq $0x5, %rsi vmovaps 0xc0(%rsp,%rsi), %xmm7 vmovaps 0xd0(%rsp,%rsi), %xmm8 vminps %xmm5, %xmm7, %xmm7 vmovaps %xmm7, 0xc0(%rsp,%rsi) vmaxps %xmm4, %xmm8, %xmm7 vmovaps %xmm7, 0xd0(%rsp,%rsi) vpextrd $0x1, %xmm6, %edi movl %edi, %esi shll $0x4, %esi addl %edx, 0xcc4(%rsp,%rsi) leaq (%rdi,%rdi,2), %rsi shlq $0x5, %rsi vmovaps 0xe0(%rsp,%rsi), %xmm7 vmovaps 0xf0(%rsp,%rsi), %xmm8 vminps %xmm5, %xmm7, %xmm7 vmovaps %xmm7, 0xe0(%rsp,%rsi) vmaxps %xmm4, %xmm8, %xmm7 vmovaps %xmm7, 0xf0(%rsp,%rsi) vpextrd $0x2, %xmm6, %edi movl %edi, %esi shll $0x4, %esi addl %edx, 0xcc8(%rsp,%rsi) leaq (%rdi,%rdi,2), %rdx shlq $0x5, %rdx vmovaps 0x100(%rsp,%rdx), %xmm6 vmovaps 0x110(%rsp,%rdx), %xmm7 vminps %xmm5, %xmm6, %xmm5 vmovaps %xmm5, 0x100(%rsp,%rdx) vmaxps %xmm4, %xmm7, %xmm4 vmovaps %xmm4, 0x110(%rsp,%rdx) vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x70(%rsp) vxorps %xmm4, %xmm4, %xmm4 movl $0xbf0, %edx # imm = 0xBF0 movl $0xdf0, %esi # imm = 0xDF0 vmovaps %xmm3, %xmm5 vmovaps %xmm2, %xmm6 vmovaps %xmm3, %xmm7 vmovaps %xmm2, %xmm8 vpaddd 0xc0(%rsp,%rsi), %xmm4, %xmm4 vminps 0x70(%rsp,%rdx), %xmm2, %xmm2 vmaxps 0x80(%rsp,%rdx), %xmm3, %xmm3 vmovdqa %xmm4, 0x4c0(%rsp,%rsi) vsubps %xmm2, %xmm3, %xmm9 vminps 0x90(%rsp,%rdx), %xmm6, %xmm6 vmaxps 0xa0(%rsp,%rdx), %xmm5, %xmm5 vsubps %xmm6, %xmm5, %xmm10 vinsertps $0x4c, %xmm9, %xmm10, %xmm11 # xmm11 = xmm9[1],xmm10[1],zero,zero vshufpd $0x1, %xmm9, %xmm9, %xmm12 # xmm12 = xmm9[1,0] vinsertps $0x9c, %xmm10, %xmm12, %xmm12 # xmm12 = xmm12[0],xmm10[2],zero,zero vaddps %xmm12, %xmm11, %xmm13 vmulps %xmm12, %xmm11, %xmm11 vunpcklps %xmm10, %xmm9, %xmm9 # xmm9 = xmm9[0],xmm10[0],xmm9[1],xmm10[1] vmulps %xmm13, %xmm9, %xmm9 vaddps %xmm11, %xmm9, %xmm9 vmovlps %xmm9, 0x2c0(%rsp,%rsi) vminps 0xb0(%rsp,%rdx), %xmm8, %xmm8 vmaxps 0xc0(%rsp,%rdx), %xmm7, %xmm7 vsubps %xmm8, %xmm7, %xmm9 vmovshdup %xmm9, %xmm10 # xmm10 = xmm9[1,1,3,3] vshufpd $0x1, %xmm9, %xmm9, %xmm11 # xmm11 = xmm9[1,0] vaddss %xmm11, %xmm10, %xmm12 vmulss %xmm11, %xmm10, %xmm10 vmulss %xmm12, %xmm9, %xmm9 vaddss %xmm10, %xmm9, %xmm9 vmovss %xmm9, 0x2c8(%rsp,%rsi) movl $0x0, 0x2cc(%rsp,%rsi) addq $-0x60, %rdx addq $-0x10, %rsi cmpq $0x50, %rdx jne 0x13eb72f movl $0xffffffff, %edx # imm = 0xFFFFFFFF shll %cl, %edx notl %edx vmovd %edx, %xmm2 vpshufd $0x0, %xmm2, %xmm0 # xmm0 = xmm2[0,0,0,0] vmovdqa %xmm0, 0x30(%rsp) vbroadcastss 0xb00205(%rip), %xmm3 # 0x1eeba20 vmovaps %xmm3, 0x50(%rsp) vpxor %xmm4, %xmm4, %xmm4 vmovdqa %xmm4, 0x40(%rsp) vmovddup 0xb356b5(%rip), %xmm5 # xmm5 = mem[0,0] vbroadcastss 0xb01348(%rip), %xmm6 # 0x1eecb84 movl $0xc00, %edx # imm = 0xC00 leaq 0x110(%rsp), %rsi vmovddup 0xb3662f(%rip), %xmm0 # xmm0 = mem[0,0] vmovaps %xmm0, 0x80(%rsp) vmovaps %xmm6, %xmm10 vmovaps %xmm3, %xmm11 vmovaps %xmm6, %xmm12 vmovaps %xmm3, %xmm13 vmovaps %xmm3, %xmm14 vxorps %xmm15, %xmm15, %xmm15 vminps -0x50(%rsi), %xmm3, %xmm3 vmaxps -0x40(%rsi), %xmm6, %xmm6 vsubps %xmm3, %xmm6, %xmm1 vminps -0x30(%rsi), %xmm11, %xmm11 vmaxps -0x20(%rsi), %xmm10, %xmm10 vminps -0x10(%rsi), %xmm13, %xmm13 vmaxps (%rsi), %xmm12, %xmm12 vsubps %xmm11, %xmm10, %xmm0 vsubps %xmm13, %xmm12, %xmm9 vshufps $0x65, %xmm9, %xmm1, %xmm8 # xmm8 = xmm1[1,1],xmm9[2,1] vblendps $0x2, %xmm0, %xmm8, %xmm8 # xmm8 = xmm8[0],xmm0[1],xmm8[2,3] vshufps $0x96, %xmm9, %xmm1, %xmm2 # xmm2 = xmm1[2,1],xmm9[1,2] vinsertps $0x90, %xmm0, %xmm2, %xmm2 # xmm2 = xmm2[0],xmm0[2],xmm2[2,3] vaddps %xmm2, %xmm8, %xmm7 vmulps %xmm2, %xmm8, %xmm2 vshufps $0x0, %xmm9, %xmm1, %xmm1 # xmm1 = xmm1[0,0],xmm9[0,0] vpaddd 0xc0(%rsp,%rdx), %xmm4, %xmm4 vinsertps $0x10, %xmm0, %xmm1, %xmm0 # xmm0 = xmm1[0],xmm0[0],xmm1[2,3] vmulps %xmm7, %xmm0, %xmm0 vaddps %xmm2, %xmm0, %xmm0 vmovdqa 0x30(%rsp), %xmm7 vpaddd %xmm7, %xmm4, %xmm1 vmovd %ecx, %xmm2 vpsrld %xmm2, %xmm1, %xmm1 vpaddd 0x4d0(%rsp,%rdx), %xmm7, %xmm7 vpsrld %xmm2, %xmm7, %xmm2 vpsrad $0x1f, %xmm1, %xmm7 vmovdqa 0x80(%rsp), %xmm8 vpand %xmm7, %xmm8, %xmm7 vbroadcastss 0xb355af(%rip), %xmm9 # 0x1f20ec4 vpand %xmm1, %xmm9, %xmm1 vcvtdq2ps %xmm1, %xmm1 vaddps %xmm7, %xmm1, %xmm1 vmulps %xmm1, %xmm0, %xmm0 vpsrad $0x1f, %xmm2, %xmm1 vpand %xmm1, %xmm8, %xmm1 vpand %xmm2, %xmm9, %xmm2 vcvtdq2ps %xmm2, %xmm2 vaddps %xmm1, %xmm2, %xmm1 vmulps 0x2d0(%rsp,%rdx), %xmm1, %xmm1 vaddps %xmm1, %xmm0, %xmm0 vcmpltps %xmm14, %xmm0, %xmm1 vblendvps %xmm1, %xmm5, %xmm15, %xmm15 vminps %xmm14, %xmm0, %xmm14 vpsubd 0xb004c0(%rip), %xmm5, %xmm5 # 0x1eebe20 addq $0x60, %rsi addq $0x10, %rdx cmpq $0xdf0, %rdx # imm = 0xDF0 jne 0x13eb873 vmovaps %xmm14, 0x50(%rsp) vmovaps %xmm15, 0x40(%rsp) vmovss 0xb00097(%rip), %xmm2 # 0x1eeba20 movl $0xffffffff, %edx # imm = 0xFFFFFFFF xorl %ecx, %ecx vxorps %xmm3, %xmm3, %xmm3 xorl %esi, %esi vmovaps 0x60(%rsp), %xmm1 vmovss 0xb0(%rsp,%rcx,4), %xmm0 vucomiss %xmm3, %xmm0 jne 0x13eb9ad jnp 0x13eb9c9 vmovss 0x50(%rsp,%rcx,4), %xmm4 vucomiss %xmm4, %xmm2 jbe 0x13eb9c9 movl 0x40(%rsp,%rcx,4), %edi testl %edi, %edi je 0x13eb9c9 vmovaps %xmm4, %xmm2 movl %ecx, %edx movl %edi, %esi incq %rcx cmpq $0x3, %rcx jne 0x13eb99c vmovss %xmm2, (%rax) movl %edx, 0x4(%rax) movl %esi, 0x8(%rax) movl $0x0, 0xc(%rax) movq $0x20, 0x10(%rax) vmovaps 0x70(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rax) vmovaps %xmm1, 0x30(%rax) leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const&)::operator()(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMv<4>>::BuildRef const&) const
__forceinline operator Vec3fa () const { return Vec3fa(m128); }
vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vmovaps (%rsi), %xmm2 vmovaps 0x10(%rsi), %xmm3 vmovaps 0x20(%rsi), %xmm4 vmovaps 0x30(%rsi), %xmm5 vminps %xmm0, %xmm2, %xmm2 vmovaps %xmm2, (%rsi) vmaxps %xmm1, %xmm3, %xmm2 vmovaps %xmm2, 0x10(%rsi) vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm4, %xmm1 vmovaps %xmm1, 0x20(%rsi) vmaxps %xmm0, %xmm5, %xmm0 vmovaps %xmm0, 0x30(%rsi) incq 0x48(%rsi) retq
/embree[P]embree/kernels/bvh/../common/../../common/sys/../math/vec3fa.h
embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::deleteGeometry(unsigned long)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::deleteGeometry(size_t geomID) { if (geomID >= bvh->objects.size()) return; if (builders[geomID]) builders[geomID].reset(); delete bvh->objects [geomID]; bvh->objects [geomID] = nullptr; }
movq 0x28(%rdi), %rax movq 0x208(%rax), %rcx subq 0x200(%rax), %rcx sarq $0x3, %rcx cmpq %rsi, %rcx jbe 0x13f14fa pushq %r14 pushq %rbx pushq %rax movq %rsi, %rbx movq %rdi, %r14 movq 0x10(%rdi), %rax movq (%rax,%rsi,8), %rdi testq %rdi, %rdi je 0x13f14c6 movq $0x0, (%rax,%rbx,8) movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x13f14e0 movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq $0x0, (%rax,%rbx,8) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::setupSmallBuildRefBuilder(unsigned long, embree::TriangleMesh const*)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupSmallBuildRefBuilder (size_t objectID, Mesh const * const /*mesh*/) { if (builders[objectID] == nullptr || // new mesh dynamic_cast<RefBuilderSmall*>(builders[objectID].get()) == nullptr) // size change resulted in large->small change { builders[objectID].reset (new RefBuilderSmall(objectID)); } }
pushq %r14 pushq %rbx pushq %rax movq %rsi, %rbx movq 0x10(%rdi), %r14 movq (%r14,%rsi,8), %rdi testq %rdi, %rdi je 0x13f300e leaq 0xd26365(%rip), %rsi # 0x2119360 leaq 0xd2636e(%rip), %rdx # 0x2119370 xorl %ecx, %ecx callq 0x6a9a0 testq %rax, %rax jne 0x13f3040 movl $0x10, %edi callq 0x6a170 leaq 0xd26379(%rip), %rcx # 0x2119398 movq %rcx, (%rax) movq %rbx, 0x8(%rax) movq (%r14,%rbx,8), %rdi movq %rax, (%r14,%rbx,8) testq %rdi, %rdi je 0x13f3040 movq (%rdi), %rax addq $0x8, %rsp popq %rbx popq %r14 jmpq *0x8(%rax) addq $0x8, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::createMeshAccel(unsigned long, embree::Builder*&)
void createMeshAccel (size_t geomID, Builder*& builder) { bvh->objects[geomID] = new BVH(Primitive::type,scene); BVH* accel = bvh->objects[geomID]; auto mesh = scene->getSafe<Mesh>(geomID); if (nullptr == mesh) { throw_RTCError(RTC_ERROR_INVALID_ARGUMENT,"geomID does not return correct type"); return; } __internal_two_level_builder__::MeshBuilder<N,Mesh,Primitive>()(accel, mesh, geomID, this->gtype, this->useMortonBuilder_, builder); }
pushq %rbp pushq %r15 pushq %r14 pushq %r12 pushq %rbx subq $0x30, %rsp movq %rdx, %r14 movq %rsi, %rbx movq %rdi, %r15 movl $0x240, %edi # imm = 0x240 movl $0x10, %esi callq 0x1ee60ac movq %rax, %r12 movq 0x30(%r15), %rdx leaq 0xd3185a(%rip), %rsi # 0x21251c0 movq %rax, %rdi callq 0x14ce5ec movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq %r12, (%rax,%rbx,8) movq 0x30(%r15), %rax movq 0x1e8(%rax), %rax movq (%rax,%rbx,8), %rax xorl %edx, %edx testq %rax, %rax je 0x13f399b cmpb $0x14, 0x3c(%rax) cmoveq %rax, %rdx testq %rdx, %rdx je 0x13f39dc movq 0x28(%r15), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rsi movq 0x98(%r15), %r8 movzbl 0xa0(%r15), %r9d movq %r14, (%rsp) leaq 0x10(%rsp), %rdi movq %rbx, %rcx callq 0x13f3aa8 addq $0x30, %rsp popq %rbx popq %r12 popq %r14 popq %r15 popq %rbp retq movl $0x30, %edi callq 0x6a3b0 movq %rax, %rbx leaq 0x20(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xb1eda7(%rip), %rsi # 0x1f127a0 leaq 0xb1edc3(%rip), %rdx # 0x1f127c3 leaq 0x10(%rsp), %rdi callq 0x8d7230 leaq 0xd0de9f(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x2, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x10(%rsp), %rsi movq 0x18(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd0de0d(%rip), %rsi # 0x2101850 leaq -0xb1cc2e(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rax, %r14 xorl %ebp, %ebp jmp 0x13f3a67 movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x10(%rsp), %rdi cmpq %r15, %rdi je 0x13f3a7e callq 0x6a4f0 jmp 0x13f3a7e movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x13f3a98 movq %rbx, %rdi callq 0x6a8a0 jmp 0x13f3a98 movq %rax, %r14 movq %r12, %rdi callq 0x1ee612d movq %r14, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.h
embree::NodeRefPtr<8> embree::avx::GeneralBVHBuilder::build<embree::NodeRefPtr<8>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(unsigned long)>(embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, 32ul>&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*, embree::avx::PrimInfoExtRange const&, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(unsigned long) const&, embree::avx::GeneralBVHBuilder::Settings const&)
__noinline static ReductionTy build(Heuristic& heuristic, PrimRef* prims, const Set& set, CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, const ProgressMonitor& progressMonitor, const Settings& settings) { typedef BuildRecordT<Set,typename Heuristic::Split> BuildRecord; typedef BuilderT< BuildRecord, Heuristic, Set, PrimRef, ReductionTy, decltype(createAlloc()), CreateAllocFunc, CreateNodeFunc, UpdateNodeFunc, CreateLeafFunc, DefaultCanCreateLeafFunc<PrimRef, Set>, DefaultCanCreateLeafSplitFunc<PrimRef, Set>, ProgressMonitor> Builder; /* instantiate builder */ Builder builder(prims, heuristic, createAlloc, createNode, updateNode, createLeaf, DefaultCanCreateLeafFunc<PrimRef, Set>(), DefaultCanCreateLeafSplitFunc<PrimRef, Set>(), progressMonitor, settings); /* build hierarchy */ BuildRecord record(1,set); const ReductionTy root = builder.recurse(record,nullptr,true); _mm_mfence(); // to allow non-temporal stores during build return root; }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 movq %r9, %rax movq %r8, %r10 movq %rdx, %r14 movq %rdi, %rdx movq 0x160(%rsp), %r12 leaq 0x28(%rsp), %r11 movq %rcx, (%r11) subq $0x8, %rsp leaq 0x15(%rsp), %r13 leaq 0x38(%rsp), %r15 leaq 0xb0(%rsp), %rbx leaq 0x17(%rsp), %r8 leaq 0x16(%rsp), %r9 movq %rbx, %rdi movq %r11, %rcx pushq %r12 pushq %rax pushq %r13 pushq %r15 pushq %r10 callq 0x13f4f60 addq $0x30, %rsp movq $0x1, (%r15) movb $0x0, 0x8(%r15) vmovaps (%r14), %xmm0 vmovaps %xmm0, 0x10(%r15) vmovaps 0x10(%r14), %xmm0 vmovaps %xmm0, 0x20(%r15) vmovaps 0x20(%r14), %xmm0 vmovaps %xmm0, 0x30(%r15) vmovaps 0x30(%r14), %xmm0 vmovaps %xmm0, 0x40(%r15) vmovaps 0x40(%r14), %xmm0 vmovaps %xmm0, 0x50(%r15) movq 0x50(%r14), %rax movq %rax, 0x60(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm0, 0x10(%rsp) movq $0x0, 0x20(%rsp) subq $0x20, %rsp movq 0x40(%rsp), %rax movq %rax, 0x10(%rsp) vmovaps 0x30(%rsp), %xmm0 vmovups %xmm0, (%rsp) leaq 0x50(%rsp), %rsi movq %rbx, %rdi movl $0x1, %edx callq 0x13f506e addq $0x20, %rsp mfence addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq nop
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, 32ul>::find(embree::avx::PrimInfoExtRange&, unsigned long)
__noinline const Split find(PrimInfoExtRange& set, const size_t logBlockSize) { /* single element */ if (set.size() <= 1) return Split(); /* disable opening if there is no overlap */ const size_t D = 4; if (unlikely(set.has_ext_range() && set.size() <= D)) { bool disjoint = true; for (size_t j=set.begin(); j<set.end()-1; j++) { for (size_t i=set.begin()+1; i<set.end(); i++) { if (conjoint(prims0[j].bounds(),prims0[i].bounds())) { disjoint = false; break; } } } if (disjoint) set.set_ext_range(set.end()); /* disables opening */ } std::pair<size_t,bool> p(0,false); /* disable opening when all primitives are from same geometry */ if (unlikely(set.has_ext_range())) { p = getProperties(set); #if EQUAL_GEOMID_STOP_CRITERIA == 1 if (p.second) set.set_ext_range(set.end()); /* disable opening */ #endif } /* open nodes when we have sufficient space available */ if (unlikely(set.has_ext_range())) { #if USE_LOOP_OPENING == 1 openNodesBasedOnExtendLoop(set,p.first); #else if (p.first <= set.ext_range_size()) openNodesBasedOnExtend(set); #endif /* disable opening when insufficient space for opening a node available */ if (set.ext_range_size() < max_open_size-1) set.set_ext_range(set.end()); /* disable opening */ } /* find best split */ return object_find(set,logBlockSize); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rcx, %r14 movq %rdx, %r15 movq %rdi, %rbx movq 0x40(%rdx), %rcx movq 0x48(%rdx), %rax movq %rax, %rdx subq %rcx, %rdx cmpq $0x1, %rdx ja 0x13f5c0a vmovsd 0xb0829c(%rip), %xmm0 # 0x1efdea0 vmovaps %xmm0, (%rbx) jmp 0x13f5c57 movq %rsi, %r12 cmpq $0x4, %rdx ja 0x13f5c1d cmpq %rax, 0x50(%r15) jne 0x13f5cb4 cmpq %rax, 0x50(%r15) jne 0x13f5c69 xorl %eax, %eax movq 0x50(%r15), %rcx subq 0x48(%r15), %rcx jne 0x13f5c83 movq 0x48(%r15), %rax subq 0x40(%r15), %rax movq %rbx, %rdi movq %r12, %rsi movq %r15, %rdx movq %r14, %rcx cmpq $0x3ff, %rax # imm = 0x3FF ja 0x13f5c52 callq 0x13f89b0 jmp 0x13f5c57 callq 0x13f910e movq %rbx, %rax addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %r12, %rdi movq %r15, %rsi callq 0x13f71c6 testb $0x1, %dl je 0x13f5c25 movq 0x48(%r15), %rcx movq %rcx, 0x50(%r15) jmp 0x13f5c25 cmpq %rcx, %rax ja 0x13f5c93 movq %r12, %rdi movq %r15, %rsi callq 0x13f73d8 movq 0x48(%r15), %rax movq 0x50(%r15), %rcx subq %rax, %rcx movq 0x10(%r12), %rdx decq %rdx cmpq %rdx, %rcx jae 0x13f5c2f movq %rax, 0x50(%r15) jmp 0x13f5c2f leaq -0x1(%rax), %rdx cmpq %rdx, %rcx jae 0x13f5d2f movq (%r12), %rsi movq %rcx, %rdi shlq $0x6, %rdi addq %rsi, %rdi addq $0x50, %rdi movb $0x1, %r8b movq %rcx, %r9 movq %r9, %r10 shlq $0x6, %r10 addq %rsi, %r10 movq %rdi, %r13 movq %rcx, %r11 incq %r11 cmpq %rax, %r11 jae 0x13f5d1d vmovaps (%r10), %xmm0 vmovaps 0x10(%r10), %xmm1 vmaxps -0x10(%r13), %xmm0, %xmm0 vminps (%r13), %xmm1, %xmm1 addq $0x40, %r13 vcmpleps %xmm1, %xmm0, %xmm0 vmovmskps %xmm0, %ebp notb %bpl testb $0x7, %bpl jne 0x13f5ce5 xorl %r8d, %r8d incq %r9 cmpq %rdx, %r9 jne 0x13f5cd5 testb $0x1, %r8b je 0x13f5c1d movq %rax, 0x50(%r15) jmp 0x13f5c1d
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::TriangleMesh, embree::TriangleMi<4>>::BuildRef, 32ul>::parallel_object_find(embree::avx::PrimInfoExtRange const&, unsigned long)
__noinline const Split parallel_object_find(const PrimInfoExtRange& set, const size_t logBlockSize) { Binner binner(empty); const BinMapping<OBJECT_BINS> mapping(set.centBounds); const BinMapping<OBJECT_BINS>& _mapping = mapping; // CLANG 3.4 parser bug workaround auto body = [&] (const range<size_t>& r) -> Binner { Binner binner(empty); binner.bin(prims0+r.begin(),r.size(),_mapping); return binner; }; auto reduction = [&] (const Binner& b0, const Binner& b1) -> Binner { Binner r = b0; r.merge(b1,_mapping.size()); return r; }; binner = parallel_reduce(set.begin(),set.end(),PARALLEL_FIND_BLOCK_SIZE,binner,body,reduction); return binner.best(mapping,logBlockSize); }
pushq %rbp movq %rsp, %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx andq $-0x40, %rsp subq $0x1f00, %rsp # imm = 0x1F00 movq %rcx, (%rsp) movq %rdi, %rbx movl $0xc00, %eax # imm = 0xC00 leaq 0x310(%rsp), %rcx vbroadcastss 0xaf28dd(%rip), %xmm1 # 0x1eeba20 vbroadcastss 0xaf3a38(%rip), %xmm2 # 0x1eecb84 vxorps %xmm0, %xmm0, %xmm0 vmovaps %xmm1, -0x10(%rcx) vmovaps %xmm2, (%rcx) vmovaps %xmm1, -0x30(%rcx) vmovaps %xmm2, -0x20(%rcx) vmovaps %xmm1, -0x50(%rcx) vmovaps %xmm2, -0x40(%rcx) vmovaps %xmm0, 0x2c0(%rsp,%rax) addq $0x10, %rax addq $0x60, %rcx cmpq $0xe00, %rax # imm = 0xE00 jne 0x13f9150 leaq 0x80(%rsp), %rax movq $0x20, (%rax) vmovaps 0x20(%rdx), %xmm0 vmovaps 0x30(%rdx), %xmm1 vsubps %xmm0, %xmm1, %xmm1 vbroadcastss 0xb28cc4(%rip), %xmm2 # 0x1f21e70 vmaxps %xmm1, %xmm2, %xmm1 vcmpnleps %xmm2, %xmm1, %xmm2 vbroadcastss 0xb3f86a(%rip), %xmm3 # 0x1f38a28 vdivps %xmm1, %xmm3, %xmm1 vandps %xmm1, %xmm2, %xmm1 vmovaps %xmm1, 0x20(%rax) vmovaps %xmm0, 0x10(%rax) leaq 0xb0(%rsp), %r12 movq %rsi, (%r12) movq %rax, 0x8(%r12) leaq 0x78(%rsp), %r15 movq %rax, (%r15) leaq 0xc0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) vpermilps $0x4e, 0x40(%rdx), %xmm0 # xmm0 = mem[2,3,0,1] vmovaps %xmm0, 0x10(%rsp) movq %r14, %rdi callq 0x6a660 leaq 0x30(%rsp), %rsi vmovdqa 0x10(%rsp), %xmm0 vmovdqa %xmm0, (%rsi) movq $0x200, 0x10(%rsi) # imm = 0x200 leaq 0x20(%rsp), %rcx movq %r15, (%rcx) movq %r12, 0x8(%rcx) leaq 0x10c0(%rsp), %r12 leaq 0x2c0(%rsp), %r13 movq %r12, %rdi movq %r13, %rdx movq %r15, %r8 movq %r14, %r9 callq 0x13f96c1 leaq 0xc0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x13f965e leaq 0xc0(%rsp), %r15 movq %r15, %rdi callq 0x6aab0 xorl %eax, %eax movl $0x10, %ecx vmovaps -0x10(%r12,%rcx), %xmm0 vmovaps %xmm0, -0x10(%r13,%rcx) vmovdqa (%r12,%rcx), %xmm0 vmovdqa %xmm0, (%r13,%rcx) addq $0x20, %rcx cmpq $0x70, %rcx jne 0x13f928f incq %rax addq $0x60, %r12 addq $0x60, %r13 cmpq $0x20, %rax jne 0x13f928a leaq 0x1cc0(%rsp), %rsi movl $0x200, %edx # imm = 0x200 leaq 0xec0(%rsp), %rdi callq 0x6a0f0 movq 0x80(%rsp), %rax movq %rax, %rdx decq %rdx vbroadcastss 0xaf272a(%rip), %xmm10 # 0x1eeba20 vbroadcastss 0xaf3885(%rip), %xmm11 # 0x1eecb84 je 0x13f93fd movq %rax, %rdi shlq $0x4, %rdi leaq (%rdi,%r15), %rcx addq $-0x10, %rcx leaq 0x10c0(%rsp), %rsi leaq -0x4(%rdi,%rsi), %rsi leaq 0x2c0(%rsp), %r8 addq %r8, %rdi addq $0xbf0, %rdi # imm = 0xBF0 leaq (%rax,%rax,2), %r9 shlq $0x5, %r9 addq %r9, %r8 addq $-0x10, %r8 vpxor %xmm0, %xmm0, %xmm0 xorl %r9d, %r9d movq %rdx, %r10 vmovaps %xmm11, %xmm1 vmovaps %xmm10, %xmm2 vmovaps %xmm11, %xmm3 vmovaps %xmm10, %xmm4 vpaddd (%rdi,%r9), %xmm0, %xmm0 vminps -0x50(%r8), %xmm10, %xmm10 vmaxps -0x40(%r8), %xmm11, %xmm11 vmovdqa %xmm0, (%rcx,%r9) vsubps %xmm10, %xmm11, %xmm5 vminps -0x30(%r8), %xmm2, %xmm2 vmaxps -0x20(%r8), %xmm1, %xmm1 vsubps %xmm2, %xmm1, %xmm6 vinsertps $0x4c, %xmm5, %xmm6, %xmm7 # xmm7 = xmm5[1],xmm6[1],zero,zero vshufpd $0x1, %xmm5, %xmm5, %xmm8 # xmm8 = xmm5[1,0] vinsertps $0x9c, %xmm6, %xmm8, %xmm8 # xmm8 = xmm8[0],xmm6[2],zero,zero vaddps %xmm7, %xmm8, %xmm9 vmulps %xmm7, %xmm8, %xmm7 vunpcklps %xmm6, %xmm5, %xmm5 # xmm5 = xmm5[0],xmm6[0],xmm5[1],xmm6[1] vmulps %xmm5, %xmm9, %xmm5 vaddps %xmm7, %xmm5, %xmm5 vmovlps %xmm5, -0xc(%rsi,%r9) vminps -0x10(%r8), %xmm4, %xmm4 vmaxps (%r8), %xmm3, %xmm3 vsubps %xmm4, %xmm3, %xmm5 vmovshdup %xmm5, %xmm6 # xmm6 = xmm5[1,1,3,3] vshufpd $0x1, %xmm5, %xmm5, %xmm7 # xmm7 = xmm5[1,0] vaddss %xmm7, %xmm6, %xmm8 vmulss %xmm7, %xmm6, %xmm6 vmulss %xmm5, %xmm8, %xmm5 vaddss %xmm6, %xmm5, %xmm5 vmovss %xmm5, -0x4(%rsi,%r9) movl $0x0, (%rsi,%r9) addq $-0x10, %r9 addq $-0x60, %r8 decq %r10 jne 0x13f935c movl $0xffffffff, %esi # imm = 0xFFFFFFFF movl $0xffffffff, %edi # imm = 0xFFFFFFFF movq (%rsp), %rcx shll %cl, %edi vbroadcastss 0xaf260a(%rip), %xmm10 # 0x1eeba20 vmovaps %xmm10, 0x30(%rsp) vpxor %xmm0, %xmm0, %xmm0 vmovdqa %xmm0, 0x20(%rsp) cmpq $0x2, %rax jb 0x13f95c9 notl %edi vmovd %edi, %xmm0 vpshufd $0x0, %xmm0, %xmm0 # xmm0 = xmm0[0,0,0,0] vmovdqa %xmm0, 0x60(%rsp) vpxor %xmm0, %xmm0, %xmm0 vmovddup 0xb27a9b(%rip), %xmm2 # xmm2 = mem[0,0] vbroadcastss 0xaf25ca(%rip), %xmm3 # 0x1eeba20 vbroadcastss 0xaf3725(%rip), %xmm4 # 0x1eecb84 vxorps %xmm5, %xmm5, %xmm5 leaq 0x310(%rsp), %r8 xorl %edi, %edi vmovd %ecx, %xmm1 vmovdqa %xmm1, (%rsp) vmovddup 0xb28a02(%rip), %xmm1 # xmm1 = mem[0,0] vmovaps %xmm1, 0x50(%rsp) vmovaps %xmm4, %xmm11 vmovaps %xmm3, %xmm12 vmovaps %xmm4, %xmm13 vmovaps %xmm3, %xmm14 vmovaps %xmm3, %xmm10 vminps -0x50(%r8), %xmm3, %xmm3 vmaxps -0x40(%r8), %xmm4, %xmm4 vmovaps %xmm4, 0x10(%rsp) vminps -0x30(%r8), %xmm12, %xmm12 vmaxps -0x20(%r8), %xmm11, %xmm11 vsubps %xmm3, %xmm4, %xmm15 vsubps %xmm12, %xmm11, %xmm9 vminps -0x10(%r8), %xmm14, %xmm14 vmaxps (%r8), %xmm13, %xmm13 vsubps %xmm14, %xmm13, %xmm8 vshufps $0x65, %xmm8, %xmm15, %xmm1 # xmm1 = xmm15[1,1],xmm8[2,1] vblendps $0x2, %xmm9, %xmm1, %xmm1 # xmm1 = xmm1[0],xmm9[1],xmm1[2,3] vshufps $0x96, %xmm8, %xmm15, %xmm6 # xmm6 = xmm15[2,1],xmm8[1,2] vinsertps $0x90, %xmm9, %xmm6, %xmm6 # xmm6 = xmm6[0],xmm9[2],xmm6[2,3] vaddps %xmm6, %xmm1, %xmm7 vmulps %xmm6, %xmm1, %xmm1 vshufps $0x0, %xmm8, %xmm15, %xmm6 # xmm6 = xmm15[0,0],xmm8[0,0] vpaddd 0xec0(%rsp,%rdi), %xmm5, %xmm5 vinsertps $0x10, %xmm9, %xmm6, %xmm6 # xmm6 = xmm6[0],xmm9[0],xmm6[2,3] vmulps %xmm7, %xmm6, %xmm6 vaddps %xmm1, %xmm6, %xmm1 vmovdqa 0x60(%rsp), %xmm7 vpaddd %xmm7, %xmm5, %xmm6 vmovdqa (%rsp), %xmm9 vpsrld %xmm9, %xmm6, %xmm6 vpaddd 0xd0(%rsp,%rdi), %xmm7, %xmm7 vpsrad $0x1f, %xmm6, %xmm8 vmovdqa 0x50(%rsp), %xmm15 vpand %xmm15, %xmm8, %xmm8 vmovaps %xmm3, %xmm4 vmovdqa %xmm5, %xmm3 vmovaps %xmm2, %xmm5 vmovdqa %xmm0, %xmm2 vbroadcastss 0xb27972(%rip), %xmm0 # 0x1f20ec4 vpand %xmm0, %xmm6, %xmm6 vcvtdq2ps %xmm6, %xmm6 vaddps %xmm6, %xmm8, %xmm6 vpsrld %xmm9, %xmm7, %xmm7 vmulps %xmm6, %xmm1, %xmm1 vpsrad $0x1f, %xmm7, %xmm6 vpand %xmm6, %xmm15, %xmm6 vpand %xmm0, %xmm7, %xmm7 vmovdqa %xmm2, %xmm0 vmovaps %xmm5, %xmm2 vmovdqa %xmm3, %xmm5 vmovaps %xmm4, %xmm3 vmovaps 0x10(%rsp), %xmm4 vcvtdq2ps %xmm7, %xmm7 vaddps %xmm6, %xmm7, %xmm6 vmulps 0x10d0(%rsp,%rdi), %xmm6, %xmm6 vaddps %xmm6, %xmm1, %xmm1 vcmpltps %xmm10, %xmm1, %xmm6 vblendvps %xmm6, %xmm2, %xmm0, %xmm0 vminps %xmm10, %xmm1, %xmm10 vpsubd 0xaf2868(%rip), %xmm2, %xmm2 # 0x1eebe20 addq $0x60, %r8 addq $0x10, %rdi decq %rdx jne 0x13f9498 vmovaps %xmm10, 0x30(%rsp) vmovdqa %xmm0, 0x20(%rsp) vmovss 0xaf2443(%rip), %xmm0 # 0x1eeba20 xorl %ecx, %ecx vxorps %xmm1, %xmm1, %xmm1 xorl %edx, %edx vmovss 0xa0(%rsp,%rcx,4), %xmm2 vucomiss %xmm1, %xmm2 jne 0x13f95f6 jnp 0x13f9612 vmovss 0x30(%rsp,%rcx,4), %xmm2 vucomiss %xmm2, %xmm0 jbe 0x13f9612 movl 0x20(%rsp,%rcx,4), %edi testl %edi, %edi je 0x13f9612 vmovaps %xmm2, %xmm0 movl %ecx, %esi movl %edi, %edx incq %rcx cmpq $0x3, %rcx jne 0x13f95e5 vmovss %xmm0, (%rbx) movl %esi, 0x4(%rbx) movl %edx, 0x8(%rbx) movl $0x0, 0xc(%rbx) movq %rax, 0x10(%rbx) vmovaps 0x90(%rsp), %xmm0 vmovaps %xmm0, 0x20(%rbx) vmovaps 0xa0(%rsp), %xmm0 vmovaps %xmm0, 0x30(%rbx) movq %rbx, %rax leaq -0x28(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xaf247e(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd2b587(%rip), %rsi # 0x2124c08 movq 0xd2b340(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x13f96a4 jmp 0x13f96b9 jmp 0x13f96a1 movq %rax, %rbx leaq 0xc0(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::NodeRefPtr<8> embree::avx::BVHBuilderBinnedOpenMergeSAH::build<embree::NodeRefPtr<8>, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(unsigned long)>(embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&) const&, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(unsigned long), embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef*, unsigned long, embree::PrimInfoT<embree::BBox<embree::Vec3fa>> const&, embree::avx::GeneralBVHBuilder::Settings const&)
static ReductionTy build(CreateAllocFunc createAlloc, CreateNodeFunc createNode, UpdateNodeFunc updateNode, const CreateLeafFunc& createLeaf, NodeOpenerFunc nodeOpenerFunc, ProgressMonitor progressMonitor, BuildRef* prims, const size_t extSize, const PrimInfo& pinfo, const Settings& settings) { typedef HeuristicArrayOpenMergeSAH<NodeOpenerFunc,BuildRef,NUM_OBJECT_BINS_HQ> Heuristic; Heuristic heuristic(nodeOpenerFunc,prims,settings.branchingFactor); return GeneralBVHBuilder::build<ReductionTy,Heuristic,Set,BuildRef>( heuristic, prims, PrimInfoExtRange(0,pinfo.size(),extSize,pinfo), createAlloc, createNode, updateNode, createLeaf, progressMonitor, settings); }
pushq %r14 pushq %rbx subq $0x98, %rsp movq %r8, %rax movq %rcx, %r10 movq %rsi, %r8 movq %rdi, %rcx movq 0xb0(%rsp), %rsi movq 0xb8(%rsp), %rbx leaq 0x10(%rsp), %r14 movq %rdx, (%r14) leaq 0x8(%rsp), %r11 movq %r10, (%r11) movq (%rbx), %rdx leaq 0x18(%rsp), %rdi movq %rax, (%rdi) movq %r14, 0x8(%rdi) movq %rdx, 0x10(%rdi) movq 0x48(%rsi), %r10 subq 0x40(%rsi), %r10 vmovaps (%rsi), %xmm0 leaq 0x30(%rsp), %rdx vmovaps %xmm0, (%rdx) vmovaps 0x10(%rsi), %xmm0 vmovaps %xmm0, 0x10(%rdx) vmovaps 0x20(%rsi), %xmm0 vmovaps %xmm0, 0x20(%rdx) vmovaps 0x30(%rsi), %xmm0 vmovaps %xmm0, 0x30(%rdx) movq $0x0, 0x40(%rdx) movq %r10, 0x48(%rdx) movq %r9, 0x50(%rdx) movq %rbx, (%rsp) movq %rax, %rsi movq %r11, %r9 callq 0x14026cc addq $0x98, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::__internal_two_level_builder__::MeshBuilder<8, embree::QuadMesh, embree::QuadMv<4>>::operator()(void*, embree::QuadMesh*, unsigned long, embree::Geometry::GTypeMask, bool, embree::Builder*&)
void operator () (void* bvh, Mesh* mesh, size_t geomID, Geometry::GTypeMask gtype, bool useMortonBuilder, Builder*& builder) { if(useMortonBuilder) { builder = MortonBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); return; } switch (mesh->quality) { case RTC_BUILD_QUALITY_LOW: builder = MortonBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; case RTC_BUILD_QUALITY_MEDIUM: case RTC_BUILD_QUALITY_HIGH: builder = SAHBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; case RTC_BUILD_QUALITY_REFIT: builder = RefitBuilder<N,Mesh,Primitive>()(bvh,mesh,geomID,gtype); break; default: throw_RTCError(RTC_ERROR_UNKNOWN,"invalid build quality"); } }
pushq %rbp pushq %r15 pushq %r14 pushq %rbx subq $0x28, %rsp testl %r9d, %r9d je 0x1401341 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x1447afa jmp 0x14013f7 movzwl 0x3e(%rdx), %eax andl $0x7, %eax leal -0x1(%rax), %edi cmpl $0x2, %edi jb 0x14013d7 cmpl $0x3, %eax je 0x14013e8 testl %eax, %eax je 0x140132d movl $0x30, %edi callq 0x6a3b0 movq %rax, %rbx leaq 0x18(%rsp), %r15 movq %r15, -0x10(%r15) leaq 0xaeb4a8(%rip), %rsi # 0x1eec826 leaq 0xaeb4b6(%rip), %rdx # 0x1eec83b leaq 0x8(%rsp), %rdi callq 0x8d7230 leaq 0xd0051a(%rip), %rax # 0x21018b0 movq %rax, (%rbx) movl $0x1, 0x8(%rbx) leaq 0x10(%rbx), %rdi movq %rbx, %rax addq $0x20, %rax movq %rax, 0x10(%rbx) movq 0x8(%rsp), %rsi movq 0x10(%rsp), %rdx addq %rsi, %rdx callq 0x8d7100 leaq 0xd00488(%rip), %rsi # 0x2101850 leaq -0xb2a5b3(%rip), %rdx # 0x8d6e1c movq %rbx, %rdi callq 0x6a5d0 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x127b463 jmp 0x14013f7 movq %rsi, %rdi movq %rdx, %rsi movl %ecx, %edx xorl %ecx, %ecx callq 0x1502729 movq 0x50(%rsp), %rcx movq %rax, (%rcx) addq $0x28, %rsp popq %rbx popq %r14 popq %r15 popq %rbp retq movq %rax, %r14 xorl %ebp, %ebp jmp 0x140141f movq %rax, %r14 movq %rbx, %rdi callq 0x6a0e0 movb $0x1, %bpl movq 0x8(%rsp), %rdi cmpq %r15, %rdi je 0x1401436 callq 0x6a4f0 jmp 0x1401436 movq %rax, %r14 movb $0x1, %bpl testb %bpl, %bpl je 0x1401443 movq %rbx, %rdi callq 0x6a8a0 movq %r14, %rdi callq 0x6a600 nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel_internal.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef, 32ul>::find(embree::avx::PrimInfoExtRange&, unsigned long)
__noinline const Split find(PrimInfoExtRange& set, const size_t logBlockSize) { /* single element */ if (set.size() <= 1) return Split(); /* disable opening if there is no overlap */ const size_t D = 4; if (unlikely(set.has_ext_range() && set.size() <= D)) { bool disjoint = true; for (size_t j=set.begin(); j<set.end()-1; j++) { for (size_t i=set.begin()+1; i<set.end(); i++) { if (conjoint(prims0[j].bounds(),prims0[i].bounds())) { disjoint = false; break; } } } if (disjoint) set.set_ext_range(set.end()); /* disables opening */ } std::pair<size_t,bool> p(0,false); /* disable opening when all primitives are from same geometry */ if (unlikely(set.has_ext_range())) { p = getProperties(set); #if EQUAL_GEOMID_STOP_CRITERIA == 1 if (p.second) set.set_ext_range(set.end()); /* disable opening */ #endif } /* open nodes when we have sufficient space available */ if (unlikely(set.has_ext_range())) { #if USE_LOOP_OPENING == 1 openNodesBasedOnExtendLoop(set,p.first); #else if (p.first <= set.ext_range_size()) openNodesBasedOnExtend(set); #endif /* disable opening when insufficient space for opening a node available */ if (set.ext_range_size() < max_open_size-1) set.set_ext_range(set.end()); /* disable opening */ } /* find best split */ return object_find(set,logBlockSize); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx pushq %rax movq %rcx, %r14 movq %rdx, %r15 movq %rdi, %rbx movq 0x40(%rdx), %rcx movq 0x48(%rdx), %rax movq %rax, %rdx subq %rcx, %rdx cmpq $0x1, %rdx ja 0x1403480 vmovsd 0xafaa26(%rip), %xmm0 # 0x1efdea0 vmovaps %xmm0, (%rbx) jmp 0x14034cd movq %rsi, %r12 cmpq $0x4, %rdx ja 0x1403493 cmpq %rax, 0x50(%r15) jne 0x140352a cmpq %rax, 0x50(%r15) jne 0x14034df xorl %eax, %eax movq 0x50(%r15), %rcx subq 0x48(%r15), %rcx jne 0x14034f9 movq 0x48(%r15), %rax subq 0x40(%r15), %rax movq %rbx, %rdi movq %r12, %rsi movq %r15, %rdx movq %r14, %rcx cmpq $0x3ff, %rax # imm = 0x3FF ja 0x14034c8 callq 0x1406226 jmp 0x14034cd callq 0x1406984 movq %rbx, %rax addq $0x8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq movq %r12, %rdi movq %r15, %rsi callq 0x1404a3c testb $0x1, %dl je 0x140349b movq 0x48(%r15), %rcx movq %rcx, 0x50(%r15) jmp 0x140349b cmpq %rcx, %rax ja 0x1403509 movq %r12, %rdi movq %r15, %rsi callq 0x1404c4e movq 0x48(%r15), %rax movq 0x50(%r15), %rcx subq %rax, %rcx movq 0x10(%r12), %rdx decq %rdx cmpq %rdx, %rcx jae 0x14034a5 movq %rax, 0x50(%r15) jmp 0x14034a5 leaq -0x1(%rax), %rdx cmpq %rdx, %rcx jae 0x14035a5 movq (%r12), %rsi movq %rcx, %rdi shlq $0x6, %rdi addq %rsi, %rdi addq $0x50, %rdi movb $0x1, %r8b movq %rcx, %r9 movq %r9, %r10 shlq $0x6, %r10 addq %rsi, %r10 movq %rdi, %r13 movq %rcx, %r11 incq %r11 cmpq %rax, %r11 jae 0x1403593 vmovaps (%r10), %xmm0 vmovaps 0x10(%r10), %xmm1 vmaxps -0x10(%r13), %xmm0, %xmm0 vminps (%r13), %xmm1, %xmm1 addq $0x40, %r13 vcmpleps %xmm1, %xmm0, %xmm0 vmovmskps %xmm0, %ebp notb %bpl testb $0x7, %bpl jne 0x140355b xorl %r8d, %r8d incq %r9 cmpq %rdx, %r9 jne 0x140354b testb $0x1, %r8b je 0x1403493 movq %rax, 0x50(%r15) jmp 0x1403493
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::QuadMesh, embree::QuadMv<4>>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)
__noinline std::pair<size_t,size_t> parallel_object_split(const Split& split, const PrimInfoExtRange& set, PrimInfoExtRange& lset, PrimInfoExtRange& rset) { const size_t begin = set.begin(); const size_t end = set.end(); PrimInfo left(empty); PrimInfo right(empty); const unsigned int splitPos = split.pos; const unsigned int splitDim = split.dim; const unsigned int splitDimMask = (unsigned int)1 << splitDim; const vint4 vSplitPos(splitPos); const vbool4 vSplitMask( (int)splitDimMask ); auto isLeft = [&] (const PrimRef& ref) { return split.mapping.bin_unsafe(ref,vSplitPos,vSplitMask); }; const size_t center = parallel_partitioning( prims0,begin,end,EmptyTy(),left,right,isLeft, [] (PrimInfo& pinfo,const PrimRef& ref) { pinfo.add_center2(ref); }, [] (PrimInfo& pinfo0,const PrimInfo& pinfo1) { pinfo0.merge(pinfo1); }, PARALLEL_PARTITION_BLOCK_SIZE); new (&lset) PrimInfoExtRange(begin,center,center,left); new (&rset) PrimInfoExtRange(center,end,end,right); assert(area(lset.geomBounds) >= 0.0f); assert(area(rset.geomBounds) >= 0.0f); return std::pair<size_t,size_t>(left.size(),right.size()); }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0xf8, %rsp movq %r8, %rbx movq %rcx, %r13 movq 0x40(%rdx), %rbp movq 0x48(%rdx), %r15 vbroadcastss 0xae2924(%rip), %xmm0 # 0x1eeba20 leaq 0xa0(%rsp), %r14 vmovaps %xmm0, (%r14) vbroadcastss 0xae3a72(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x10(%r14) vmovaps %xmm0, 0x20(%r14) vmovaps %xmm1, 0x30(%r14) vxorps %xmm2, %xmm2, %xmm2 vmovaps %xmm2, 0x40(%r14) leaq 0x50(%rsp), %r12 vmovaps %xmm0, (%r12) vmovaps %xmm1, 0x10(%r12) vmovaps %xmm0, 0x20(%r12) vmovaps %xmm1, 0x30(%r12) vmovaps %xmm2, 0x40(%r12) movb 0x4(%rsi), %cl movl $0x1, %eax shll %cl, %eax vpermilps $0x0, 0x8(%rsi), %xmm0 # xmm0 = mem[0,0,0,0] leaq 0x20(%rsp), %rcx vmovaps %xmm0, (%rcx) cltq shlq $0x4, %rax leaq 0xd46e04(%rip), %rdx # 0x214ff80 vmovaps (%rdx,%rax), %xmm0 leaq 0x10(%rsp), %rax vmovaps %xmm0, (%rax) leaq 0x38(%rsp), %r10 movq %rsi, (%r10) movq %rcx, 0x8(%r10) movq %rax, 0x10(%r10) movq (%rdi), %rdi leaq 0xd(%rsp), %rax leaq 0xe(%rsp), %r11 leaq 0xf(%rsp), %rcx movq %rbp, %rsi movq %r15, %rdx movq %r14, %r8 movq %r12, %r9 pushq $0x80 pushq %rax pushq %r11 pushq %r10 callq 0x1409264 addq $0x20, %rsp vmovaps (%r14), %xmm0 vmovaps %xmm0, (%r13) vmovaps 0x10(%r14), %xmm0 vmovaps %xmm0, 0x10(%r13) vmovaps 0x20(%r14), %xmm0 vmovaps %xmm0, 0x20(%r13) vmovaps 0x30(%r14), %xmm0 vmovaps %xmm0, 0x30(%r13) movq %rbp, 0x40(%r13) movq %rax, 0x48(%r13) movq %rax, 0x50(%r13) vmovaps (%r12), %xmm0 vmovaps %xmm0, (%rbx) vmovaps 0x10(%r12), %xmm0 vmovaps %xmm0, 0x10(%rbx) vmovaps 0x20(%r12), %xmm0 vmovaps %xmm0, 0x20(%rbx) vmovaps 0x30(%r12), %xmm0 vmovaps %xmm0, 0x30(%rbx) movq %rax, 0x40(%rbx) movq %r15, 0x48(%rbx) movq %r15, 0x50(%rbx) movq 0x48(%r14), %rax subq 0x40(%r14), %rax movq 0x48(%r12), %rdx subq 0x40(%r12), %rdx addq $0xf8, %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp retq
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::deleteGeometry(unsigned long)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::deleteGeometry(size_t geomID) { if (geomID >= bvh->objects.size()) return; if (builders[geomID]) builders[geomID].reset(); delete bvh->objects [geomID]; bvh->objects [geomID] = nullptr; }
movq 0x28(%rdi), %rax movq 0x208(%rax), %rcx subq 0x200(%rax), %rcx sarq $0x3, %rcx cmpq %rsi, %rcx jbe 0x140c480 pushq %r14 pushq %rbx pushq %rax movq %rsi, %rbx movq %rdi, %r14 movq 0x10(%rdi), %rax movq (%rax,%rsi,8), %rdi testq %rdi, %rdi je 0x140c44c movq $0x0, (%rax,%rbx,8) movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq (%rax,%rbx,8), %rdi testq %rdi, %rdi je 0x140c466 movq (%rdi), %rax callq *0x8(%rax) movq 0x28(%r14), %rax movq 0x200(%rax), %rax movq $0x0, (%rax,%rbx,8) addq $0x8, %rsp popq %rbx popq %r14 retq nop
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::setupSmallBuildRefBuilder(unsigned long, embree::UserGeometry const*)
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::setupSmallBuildRefBuilder (size_t objectID, Mesh const * const /*mesh*/) { if (builders[objectID] == nullptr || // new mesh dynamic_cast<RefBuilderSmall*>(builders[objectID].get()) == nullptr) // size change resulted in large->small change { builders[objectID].reset (new RefBuilderSmall(objectID)); } }
pushq %r14 pushq %rbx pushq %rax movq %rsi, %rbx movq 0x10(%rdi), %r14 movq (%r14,%rsi,8), %rdi testq %rdi, %rdi je 0x140df8a leaq 0xd0bd39(%rip), %rsi # 0x2119cb0 leaq 0xd0bd42(%rip), %rdx # 0x2119cc0 xorl %ecx, %ecx callq 0x6a9a0 testq %rax, %rax jne 0x140dfbc movl $0x10, %edi callq 0x6a170 leaq 0xd0bd4d(%rip), %rcx # 0x2119ce8 movq %rcx, (%rax) movq %rbx, 0x8(%rax) movq (%r14,%rbx,8), %rdi movq %rax, (%r14,%rbx,8) testq %rdi, %rdi je 0x140dfbc movq (%rdi), %rax addq $0x8, %rsp popq %rbx popq %r14 jmpq *0x8(%rax) addq $0x8, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp
embree::avx::GeneralBVHBuilder::BuilderT<embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>, embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef, 32ul>, embree::avx::PrimInfoExtRange, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef, embree::NodeRefPtr<8>, embree::FastAllocator::CachedAllocator, embree::BVHN<8>::CreateAlloc, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Create2, embree::AABBNode_t<embree::NodeRefPtr<8>, 8>::Set2, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef const*, embree::range<unsigned long> const&, embree::FastAllocator::CachedAllocator const&), embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::GeneralBVHBuilder::DefaultCanCreateLeafSplitFunc<embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef, embree::avx::PrimInfoExtRange>, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::build()::'lambda'(unsigned long)>::recurse(embree::avx::GeneralBVHBuilder::BuildRecordT<embree::avx::PrimInfoExtRange, embree::avx::BinSplit<32ul>>&, embree::FastAllocator::CachedAllocator, bool)
const ReductionTy recurse(BuildRecord& current, Allocator alloc, bool toplevel) { /* get thread local allocator */ if (!alloc) alloc = createAlloc(); /* call memory monitor function to signal progress */ if (toplevel && current.size() <= cfg.singleThreadThreshold) progressMonitor(current.size()); /*! find best split */ auto split = heuristic.find(current.prims,cfg.logBlockSize); /*! compute leaf and split cost */ const float leafSAH = cfg.intCost*current.prims.leafSAH(cfg.logBlockSize); const float splitSAH = cfg.travCost*halfArea(current.prims.geomBounds)+cfg.intCost*split.splitSAH(); assert((current.prims.size() == 0) || ((leafSAH >= 0) && (splitSAH >= 0))); /*! create a leaf node when threshold reached or SAH tells us to stop */ if (current.prims.size() <= cfg.minLeafSize || current.depth+MIN_LARGE_LEAF_LEVELS >= cfg.maxDepth || (current.prims.size() <= cfg.maxLeafSize && leafSAH <= splitSAH)) { heuristic.deterministic_order(current.prims); return createLargeLeaf(current,alloc); } /*! perform initial split */ Set lprims,rprims; heuristic.split(split,current.prims,lprims,rprims); /*! initialize child list with initial split */ ReductionTy values[MAX_BRANCHING_FACTOR]; BuildRecord children[MAX_BRANCHING_FACTOR]; children[0] = BuildRecord(current.depth+1,lprims); children[1] = BuildRecord(current.depth+1,rprims); size_t numChildren = 2; /*! split until node is full or SAH tells us to stop */ while (numChildren < cfg.branchingFactor) { /*! find best child to split */ float bestArea = neg_inf; ssize_t bestChild = -1; for (size_t i=0; i<numChildren; i++) { /* ignore leaves as they cannot get split */ if (children[i].prims.size() <= cfg.minLeafSize) continue; /* find child with largest surface area */ if (halfArea(children[i].prims.geomBounds) > bestArea) { bestChild = i; bestArea = halfArea(children[i].prims.geomBounds); } } if (bestChild == -1) break; /* perform best found split */ BuildRecord& brecord = children[bestChild]; BuildRecord lrecord(current.depth+1); BuildRecord rrecord(current.depth+1); auto split = heuristic.find(brecord.prims,cfg.logBlockSize); heuristic.split(split,brecord.prims,lrecord.prims,rrecord.prims); children[bestChild ] = lrecord; children[numChildren] = rrecord; numChildren++; } /* set barrier for primrefarrayalloc */ if (unlikely(current.size() > cfg.primrefarrayalloc)) for (size_t i=0; i<numChildren; i++) children[i].alloc_barrier = children[i].size() <= cfg.primrefarrayalloc; /* sort buildrecords for faster shadow ray traversal */ std::sort(&children[0],&children[numChildren],std::greater<BuildRecord>()); /*! create an inner node */ auto node = createNode(children,numChildren,alloc); /* spawn tasks */ if (current.size() > cfg.singleThreadThreshold) { /*! parallel_for is faster than spawning sub-tasks */ parallel_for(size_t(0), numChildren, [&] (const range<size_t>& r) { // FIXME: no range here for (size_t i=r.begin(); i<r.end(); i++) { values[i] = recurse(children[i],nullptr,true); _mm_mfence(); // to allow non-temporal stores during build } }); return updateNode(current,children,node,values,numChildren); } /* recurse into each child */ else { for (size_t i=0; i<numChildren; i++) values[i] = recurse(children[i],alloc,false); return updateNode(current,children,node,values,numChildren); } }
pushq %rbp pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x998, %rsp # imm = 0x998 movl %edx, %ebp movq %rsi, %r14 movq %rdi, %rbx leaq 0x9d0(%rsp), %r13 cmpq $0x0, (%r13) jne 0x141000d movq %r14, 0x20(%rsp) movq 0x50(%rbx), %rax movq (%rax), %r14 leaq 0xd14e18(%rip), %rdi # 0x2124d38 callq 0x6a480 movq (%rax), %r15 testq %r15, %r15 jne 0x140ffe8 movq %rax, %r12 movl $0xc0, %edi movl $0x40, %esi callq 0x1ee60ac movq %rax, %r15 movq %rax, %rdi callq 0x1ee7b24 movq $0x0, 0x8(%r15) movq %r15, 0x40(%r15) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x48(%r15) vmovups %ymm0, 0x58(%r15) movq %r15, 0x80(%r15) vmovups %ymm0, 0x88(%r15) vmovups %ymm0, 0x98(%r15) movq %r15, (%r12) leaq 0xd168e7(%rip), %rdi # 0x2126878 movq %rdi, 0x1d0(%rsp) movb $0x1, 0x1d8(%rsp) vzeroupper callq 0x1ee7bb6 leaq 0x40(%rsp), %rsi movq %r15, (%rsi) leaq 0xd168c8(%rip), %rdi # 0x2126880 callq 0x90d91e movq 0x40(%rsp), %rsi testq %rsi, %rsi je 0x140ffd1 leaq 0x40(%rsp), %rdi callq 0x90da94 cmpb $0x1, 0x1d8(%rsp) jne 0x140ffe8 movq 0x1d0(%rsp), %rdi callq 0x1ee7c24 leaq 0x40(%r15), %rax subq $-0x80, %r15 cmpb $0x0, 0x10a(%r14) cmovneq %rax, %r15 movq %r14, (%r13) movq %rax, 0x8(%r13) movq %r15, 0x10(%r13) movq 0x20(%rsp), %r14 testb %bpl, %bpl je 0x141003b movq 0x58(%r14), %rax subq 0x50(%r14), %rax cmpq 0x30(%rbx), %rax ja 0x141003b movq 0x80(%rbx), %rax movq (%rax), %rax movq 0x28(%rax), %rax movq 0x68(%rax), %rdi vxorps %xmm0, %xmm0, %xmm0 callq 0x91839e movq 0x10(%rbx), %rcx movq 0x48(%rbx), %rsi leaq 0x10(%r14), %r15 leaq 0x190(%rsp), %rdi movq %r15, %rdx callq 0x1410a46 movb 0x10(%rbx), %cl movq 0x58(%r14), %rax movq 0x50(%r14), %rdx movq $-0x1, %rsi shlq %cl, %rsi notq %rsi addq %rax, %rsi subq %rdx, %rsi shrq %cl, %rsi testq %rsi, %rsi js 0x1410088 vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rsi, %xmm1, %xmm0 jmp 0x14100a1 movq %rsi, %rcx shrq %rcx andl $0x1, %esi orq %rcx, %rsi vxorps %xmm1, %xmm1, %xmm1 vcvtsi2ss %rsi, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 subq %rdx, %rax cmpq 0x18(%rbx), %rax jbe 0x141010b movq (%r14), %rcx addq $0x8, %rcx cmpq 0x8(%rbx), %rcx jae 0x141010b cmpq 0x20(%rbx), %rax ja 0x1410152 vmovaps 0x20(%r14), %xmm1 vsubps 0x10(%r14), %xmm1, %xmm1 vmovshdup %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] vshufpd $0x1, %xmm1, %xmm1, %xmm3 # xmm3 = xmm1[1,0] vaddss %xmm3, %xmm2, %xmm4 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm4, %xmm1, %xmm1 vaddss %xmm2, %xmm1, %xmm1 vmovss 0x2c(%rbx), %xmm2 vmulss %xmm0, %xmm1, %xmm0 vmulss 0x190(%rsp), %xmm2, %xmm3 vmulss 0x28(%rbx), %xmm1, %xmm1 vmulss %xmm0, %xmm2, %xmm0 vaddss %xmm3, %xmm1, %xmm1 vucomiss %xmm0, %xmm1 jb 0x1410152 leaq 0x50(%r14), %rsi movq 0x48(%rbx), %rdi callq 0x1410baa movq 0x10(%r13), %rax movq %rax, 0x10(%rsp) vmovups (%r13), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r14, %rsi callq 0x1410bfe movq %rax, %r12 movq %r12, %rax addq $0x998, %rsp # imm = 0x998 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp vzeroupper retq movq 0x48(%rbx), %rdi leaq 0x190(%rsp), %rsi leaq 0x930(%rsp), %r13 leaq 0x8d0(%rsp), %r12 movq %r15, %rdx movq %r13, %rcx movq %r12, %r8 callq 0x1411872 movq %r14, 0x20(%rsp) movq (%r14), %rax incq %rax vmovups (%r13), %ymm0 vmovups 0x20(%r13), %ymm1 movq 0x50(%r13), %rcx leaq 0x228(%rsp), %rdx movq %rax, -0x58(%rdx) xorl %esi, %esi movb %sil, -0x50(%rdx) vmovups %ymm0, -0x48(%rdx) vmovups %ymm1, -0x28(%rdx) vmovaps 0x40(%r13), %xmm0 vmovaps %xmm0, -0x8(%rdx) movq %rcx, 0x8(%rdx) vmovups (%r12), %ymm0 vmovups 0x20(%r12), %ymm1 movq 0x50(%r12), %rcx movq %rax, 0x18(%rdx) movb %sil, 0x20(%rdx) vmovups %ymm0, 0x28(%rdx) vmovups %ymm1, 0x48(%rdx) vmovaps 0x40(%r12), %xmm0 vmovaps %xmm0, 0x68(%rdx) movq %rcx, 0x78(%rdx) movl $0x2, %r15d cmpq (%rbx), %r15 jae 0x141045e testq %r15, %r15 je 0x1410271 movq 0x18(%rbx), %rax movq $-0x1, %rbp leaq 0x228(%rsp), %rcx vmovss 0xadc95f(%rip), %xmm0 # 0x1eecb84 xorl %edx, %edx movq (%rcx), %rsi subq -0x8(%rcx), %rsi cmpq %rax, %rsi jbe 0x1410263 vmovaps -0x38(%rcx), %xmm1 vsubps -0x48(%rcx), %xmm1, %xmm1 vmovshdup %xmm1, %xmm2 # xmm2 = xmm1[1,1,3,3] vshufpd $0x1, %xmm1, %xmm1, %xmm3 # xmm3 = xmm1[1,0] vaddss %xmm3, %xmm2, %xmm4 vmulss %xmm3, %xmm2, %xmm2 vmulss %xmm4, %xmm1, %xmm1 vaddss %xmm2, %xmm1, %xmm1 vucomiss %xmm0, %xmm1 jbe 0x1410263 movq %rdx, %rbp vmovaps %xmm1, %xmm0 incq %rdx addq $0x70, %rcx cmpq %rdx, %r15 jne 0x1410227 jmp 0x1410278 movq $-0x1, %rbp cmpq $-0x1, %rbp je 0x1410454 imulq $0x70, %rbp, %rax movq 0x20(%rsp), %rcx movq (%rcx), %rcx incq %rcx movq %rcx, 0x40(%rsp) xorl %edx, %edx movb %dl, 0x48(%rsp) vbroadcastss 0xadb77b(%rip), %xmm0 # 0x1eeba20 vmovaps %xmm0, 0x50(%rsp) vbroadcastss 0xadc8d0(%rip), %xmm1 # 0x1eecb84 vmovaps %xmm1, 0x60(%rsp) vmovaps %xmm0, 0x70(%rsp) vmovaps %xmm1, 0x80(%rsp) leaq 0x90(%rsp), %r14 movq %rdx, 0x10(%r14) vxorps %xmm2, %xmm2, %xmm2 vmovaps %xmm2, (%r14) movq %rcx, 0xc0(%rsp) movb %dl, 0xc8(%rsp) vmovaps %xmm0, 0xd0(%rsp) vmovaps %xmm1, 0xe0(%rsp) vmovaps %xmm0, 0xf0(%rsp) vmovaps %xmm1, 0x100(%rsp) movq %rbx, %r14 leaq 0x110(%rsp), %rbx movq %rdx, 0x10(%rbx) vmovaps %xmm2, (%rbx) movq 0x10(%r14), %rcx movq 0x48(%r14), %rsi leaq 0x1e0(%rsp,%rax), %r12 leaq 0x150(%rsp), %r13 movq %r13, %rdi movq %r12, %rdx vzeroupper callq 0x1410a46 movq 0x48(%r14), %rdi movq %r13, %rsi movq %r12, %rdx leaq 0x50(%rsp), %rcx leaq 0xd0(%rsp), %r8 callq 0x1411872 movb 0x48(%rsp), %al movb %al, -0x8(%r12) movq 0x40(%rsp), %rax movq %rax, -0x10(%r12) vmovaps 0x50(%rsp), %xmm0 vmovaps %xmm0, (%r12) vmovaps 0x60(%rsp), %xmm0 vmovaps %xmm0, 0x10(%r12) vmovaps 0x70(%rsp), %xmm0 vmovaps %xmm0, 0x20(%r12) vmovaps 0x80(%rsp), %xmm0 vmovaps %xmm0, 0x30(%r12) leaq 0x90(%rsp), %rcx movq 0x10(%rcx), %rax movq %rax, 0x50(%r12) vmovaps (%rcx), %xmm0 vmovaps %xmm0, 0x40(%r12) imulq $0x70, %r15, %rax movq 0xc0(%rsp), %rcx movq %rcx, 0x1d0(%rsp,%rax) movb 0xc8(%rsp), %cl movb %cl, 0x1d8(%rsp,%rax) vmovaps 0xd0(%rsp), %xmm0 vmovaps %xmm0, 0x1e0(%rsp,%rax) vmovaps 0xe0(%rsp), %xmm0 vmovaps %xmm0, 0x1f0(%rsp,%rax) vmovaps 0xf0(%rsp), %xmm0 vmovaps %xmm0, 0x200(%rsp,%rax) vmovaps 0x100(%rsp), %xmm0 vmovaps %xmm0, 0x210(%rsp,%rax) vmovaps (%rbx), %xmm0 vmovaps %xmm0, 0x220(%rsp,%rax) movq 0x10(%rbx), %rcx movq %r14, %rbx movq %rcx, 0x230(%rsp,%rax) incq %r15 cmpq $-0x1, %rbp jne 0x14101fc movq 0x20(%rsp), %r14 movq 0x58(%r14), %rax subq 0x50(%r14), %rax movq 0x38(%rbx), %rdx cmpq %rdx, %rax ja 0x1410876 testq %r15, %r15 je 0x14104ba imulq $0x70, %r15, %rax leaq (%rsp,%rax), %r12 addq $0x1d0, %r12 # imm = 0x1D0 bsrq %r15, %rdx xorl $0x3f, %edx addl %edx, %edx xorq $0x7e, %rdx leaq 0x1d0(%rsp), %r13 movq %r13, %rdi movq %r12, %rsi vzeroupper callq 0x1383896 movq %r13, %rdi movq %r12, %rsi callq 0x1383954 leaq 0x9d0(%rsp), %rax movq (%rax), %r13 movq 0x8(%rax), %rbp movq $0x100, 0x28(%rsp) # imm = 0x100 movq (%rbp), %r12 movq 0x8(%r12), %rax cmpq %rax, %r13 je 0x1410647 movq %r12, 0xc0(%rsp) movb $0x1, 0xc8(%rsp) movq %r12, %rdi vzeroupper callq 0x1ee7bb6 movq 0x8(%r12), %rax testq %rax, %rax je 0x1410567 movq 0xa8(%r12), %rax addq 0x68(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x118(%rcx) movq 0x58(%r12), %rax addq 0x98(%r12), %rax movq 0x50(%r12), %rcx addq 0x90(%r12), %rcx subq %rcx, %rax movq 0x8(%r12), %rcx lock addq %rax, 0x120(%rcx) movq 0xb0(%r12), %rax addq 0x70(%r12), %rax movq 0x8(%r12), %rcx lock addq %rax, 0x128(%rcx) vxorps %xmm0, %xmm0, %xmm0 vmovups %ymm0, 0x58(%r12) vmovups %ymm0, 0x48(%r12) testq %r13, %r13 je 0x14105a9 movq 0x10(%r13), %rax movq %rax, 0x60(%r12) vmovups %ymm0, 0x88(%r12) vmovups %ymm0, 0x98(%r12) movq 0x10(%r13), %rax movq %rax, 0xa0(%r12) jmp 0x14105bd vmovups %ymm0, 0x98(%r12) vmovups %ymm0, 0x88(%r12) movq %r13, %rax xchgq %rax, 0x8(%r12) movq %r12, 0x150(%rsp) leaq 0xd162a4(%rip), %rdi # 0x2126878 movq %rdi, 0x40(%rsp) movb $0x1, 0x48(%rsp) vzeroupper callq 0x1ee7bb6 movq 0x138(%r13), %rsi cmpq 0x140(%r13), %rsi je 0x141060b movq 0x150(%rsp), %rax movq %rax, (%rsi) addq $0x8, 0x138(%r13) jmp 0x141061f leaq 0x130(%r13), %rdi leaq 0x150(%rsp), %rdx callq 0x90b95a cmpb $0x1, 0x48(%rsp) jne 0x1410630 movq 0x40(%rsp), %rdi callq 0x1ee7c24 cmpb $0x1, 0xc8(%rsp) jne 0x1410647 movq 0xc0(%rsp), %rdi callq 0x1ee7c24 movq 0x28(%rsp), %rax addq %rax, 0x28(%rbp) movq 0x10(%rbp), %rcx movl %ecx, %edx negl %edx andl $0x1f, %edx leaq (%rcx,%rax), %r12 addq %rdx, %r12 movq %r12, 0x10(%rbp) cmpq 0x18(%rbp), %r12 ja 0x1410896 addq %rdx, 0x30(%rbp) subq %rax, %r12 addq 0x8(%rbp), %r12 vbroadcastss 0xadb39c(%rip), %ymm0 # 0x1eeba20 vmovaps %ymm0, 0xc0(%r12) vmovaps %ymm0, 0x80(%r12) vmovaps %ymm0, 0x40(%r12) vbroadcastss 0xadc4dc(%rip), %ymm0 # 0x1eecb84 vmovaps %ymm0, 0xe0(%r12) vmovaps %ymm0, 0xa0(%r12) vmovaps %ymm0, 0x60(%r12) xorl %eax, %eax vbroadcastsd 0xb117aa(%rip), %ymm0 # 0x1f21e78 vmovups %ymm0, (%r12,%rax,8) addq $0x4, %rax cmpq $0x8, %rax jne 0x14106ce testq %r15, %r15 je 0x141073c leaq 0x1f0(%rsp), %rax xorl %ecx, %ecx vmovaps -0x10(%rax), %xmm0 vmovaps (%rax), %xmm1 vmovss %xmm0, 0x40(%r12,%rcx,4) vextractps $0x1, %xmm0, 0x80(%r12,%rcx,4) vextractps $0x2, %xmm0, 0xc0(%r12,%rcx,4) vmovss %xmm1, 0x60(%r12,%rcx,4) vextractps $0x1, %xmm1, 0xa0(%r12,%rcx,4) vextractps $0x2, %xmm1, 0xe0(%r12,%rcx,4) incq %rcx addq $0x70, %rax cmpq %rcx, %r15 jne 0x14106ed movq 0x58(%r14), %rax subq 0x50(%r14), %rax cmpq 0x30(%rbx), %rax jbe 0x1410808 leaq 0xc0(%rsp), %rax leaq 0x28(%rsp), %r14 movq %rax, (%r14) movq %rbx, 0x8(%r14) leaq 0x1d0(%rsp), %rax movq %rax, 0x10(%r14) leaq 0x40(%rsp), %rbx movw $0x401, 0xc(%rbx) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%rbx) movq $0x8, 0x40(%rbx) movq %rbx, %rdi vzeroupper callq 0x6a660 leaq 0x150(%rsp), %rdi movq %r15, (%rdi) movq $0x0, 0x8(%rdi) movq $0x1, 0x10(%rdi) leaq 0x148(%rsp), %rsi movq %r14, (%rsi) leaq 0x1f(%rsp), %rdx movq %rbx, %rcx callq 0x14183fc leaq 0x40(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1410925 leaq 0x40(%rsp), %rdi callq 0x6aab0 testq %r15, %r15 je 0x141013a xorl %eax, %eax movq 0xc0(%rsp,%rax,8), %rcx movq %rcx, (%r12,%rax,8) incq %rax cmpq %rax, %r15 jne 0x14107ef jmp 0x141013a testq %r15, %r15 je 0x141013a leaq 0x1d0(%rsp), %r14 xorl %r13d, %r13d leaq 0x9d0(%rsp), %rbp movq 0x10(%rbp), %rax movq %rax, 0x10(%rsp) vmovups (%rbp), %xmm0 vmovups %xmm0, (%rsp) movq %rbx, %rdi movq %r14, %rsi xorl %edx, %edx vzeroupper callq 0x140fee0 movq %rax, 0xc0(%rsp,%r13,8) incq %r13 addq $0x70, %r14 cmpq %r13, %r15 jne 0x1410824 xorl %eax, %eax movq 0xc0(%rsp,%rax,8), %rcx movq %rcx, (%r12,%rax,8) incq %rax cmpq %rax, %r15 jne 0x141085d jmp 0x141013a leaq 0x1d0(%rsp), %rsi movq %r15, %rdi vzeroupper callq 0x1ff6ba testb $0x1, %al je 0x1410478 jmp 0x14104ba movq %rcx, 0x10(%rbp) shlq $0x2, %rax movq 0x20(%rbp), %rcx cmpq %rcx, %rax jbe 0x14108c3 leaq 0x28(%rsp), %rsi movl $0x40, %edx movq %r13, %rdi xorl %ecx, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 jmp 0x141091b leaq 0x40(%rsp), %r14 movq %rcx, (%r14) movl $0x40, %edx movq %r13, %rdi movq %r14, %rsi movl $0x1, %ecx vzeroupper callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbp) movq 0x18(%rbp), %rax subq 0x10(%rbp), %rax addq 0x30(%rbp), %rax movq %rax, 0x30(%rbp) movq $0x0, 0x10(%rbp) movq (%r14), %rcx movq %rcx, 0x18(%rbp) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbp) cmpq %rcx, %rdx ja 0x1410957 movq %rax, 0x30(%rbp) movq 0x20(%rsp), %r14 jmp 0x141067b movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xadb1b7(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd142c0(%rip), %rsi # 0x2124c08 movq 0xd14079(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 xorl %r14d, %r14d movq %r14, 0x10(%rbp) movq 0x20(%rbp), %rax leaq 0x40(%rsp), %rsi movq %rax, (%rsi) movl $0x40, %edx movq %r13, %rdi movq %rsi, %r13 xorl %ecx, %ecx callq 0x90b5ee movq %rax, %r12 movq %rax, 0x8(%rbp) movq 0x18(%rbp), %rax subq 0x10(%rbp), %rax addq 0x30(%rbp), %rax movq %rax, 0x30(%rbp) movq %r14, 0x10(%rbp) movq (%r13), %rcx movq %rcx, 0x18(%rbp) movq 0x28(%rsp), %rdx movq %rdx, 0x10(%rbp) cmpq %rcx, %rdx jbe 0x1410917 movq $0x0, 0x10(%rbp) xorl %r12d, %r12d jmp 0x141091b movq %rax, %rbx leaq 0x40(%rsp), %rdi callq 0x8d6eda jmp 0x14109ea movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x14109f7 jmp 0x1410a3e jmp 0x1410a3e jmp 0x1410a3e jmp 0x14109f4 jmp 0x1410a3e movq %rax, %rbx leaq 0xc0(%rsp), %rdi jmp 0x1410a24 movq %rax, %rbx leaq 0x40(%rsp), %rdi callq 0x6aab0 jmp 0x1410a36 jmp 0x1410a3e movq %rax, %rbx movq 0x40(%rsp), %rsi testq %rsi, %rsi je 0x1410a1c leaq 0x40(%rsp), %rdi callq 0x90da94 leaq 0x1d0(%rsp), %rdi callq 0x8d6eda jmp 0x1410a36 movq %rax, %rbx movq %r15, %rdi callq 0x1ee612d movq %rbx, %rdi callq 0x6a600 movq %rax, %rdi callq 0x8d6de8
/embree[P]embree/kernels/bvh/../builders/bvh_builder_sah.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef, 32ul>::split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)
__noinline void split(const Split& split, const PrimInfoExtRange& set_i, PrimInfoExtRange& lset, PrimInfoExtRange& rset) { PrimInfoExtRange set = set_i; /* valid split */ if (unlikely(!split.valid())) { deterministic_order(set); splitFallback(set,lset,rset); return; } std::pair<size_t,size_t> ext_weights(0,0); /* object split */ if (likely(set.size() < PARALLEL_THRESHOLD)) ext_weights = sequential_object_split(split,set,lset,rset); else ext_weights = parallel_object_split(split,set,lset,rset); /* if we have an extended range, set extended child ranges and move right split range */ if (unlikely(set.has_ext_range())) { setExtentedRanges(set,lset,rset,ext_weights.first,ext_weights.second); moveExtentedRange(set,lset,rset); } }
pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $0x130, %rsp # imm = 0x130 movq %r8, %rbx movq %rcx, %r15 movq %rdi, %r14 vmovaps (%rdx), %xmm0 vmovaps %xmm0, 0x50(%rsp) vmovaps 0x10(%rdx), %xmm0 vmovaps %xmm0, 0x60(%rsp) vmovaps 0x20(%rdx), %xmm0 vmovaps %xmm0, 0x70(%rsp) vmovaps 0x30(%rdx), %xmm0 vmovaps %xmm0, 0x80(%rsp) vmovdqa 0x40(%rdx), %xmm0 vmovdqa %xmm0, 0x90(%rsp) movq 0x50(%rdx), %rax movq %rax, 0xa0(%rsp) cmpl $-0x1, 0x4(%rsi) je 0x141192f movq 0x98(%rsp), %rax vmovq %xmm0, %rcx subq %rcx, %rax leaq 0x50(%rsp), %rdx movq %r14, %rdi movq %r15, %rcx movq %rbx, %r8 cmpq $0x3ff, %rax # imm = 0x3FF ja 0x14119c5 callq 0x1fe414 movq 0xa0(%rsp), %rcx subq 0x98(%rsp), %rcx jne 0x14119cf addq $0x130, %rsp # imm = 0x130 popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 retq leaq 0x90(%rsp), %r12 movq %r14, %rdi movq %r12, %rsi callq 0x1410baa movq (%r12), %rdx movq 0x8(%r12), %rax leaq (%rax,%rdx), %rcx shrq %rcx vbroadcastss 0xadb229(%rip), %xmm0 # 0x1eecb84 vbroadcastss 0xada0bc(%rip), %xmm1 # 0x1eeba20 movq %rcx, %rsi subq %rdx, %rsi jbe 0x14119e2 movq (%r14), %rdi movq %rdx, %r8 shlq $0x6, %r8 addq %r8, %rdi addq $0x10, %rdi vbroadcastss 0xada09a(%rip), %xmm2 # 0x1eeba20 vbroadcastss 0xadb1f5(%rip), %xmm3 # 0x1eecb84 vmovaps %xmm3, %xmm4 vmovaps %xmm2, %xmm5 vmovaps -0x10(%rdi), %xmm6 vmovaps (%rdi), %xmm7 vminps %xmm6, %xmm5, %xmm5 vmaxps %xmm7, %xmm4, %xmm4 vaddps %xmm7, %xmm6, %xmm6 vminps %xmm6, %xmm2, %xmm2 vmaxps %xmm6, %xmm3, %xmm3 addq $0x40, %rdi decq %rsi jne 0x1411997 movq %rcx, %rsi subq %rdx, %rsi jmp 0x14119f4 callq 0x14166d0 jmp 0x1411908 testq %rax, %rax js 0x1411acc vcvtsi2ss %rax, %xmm1, %xmm0 jmp 0x1411ae3 xorl %esi, %esi vmovaps %xmm1, %xmm5 vmovaps %xmm0, %xmm4 vmovaps %xmm1, %xmm2 vmovaps %xmm0, %xmm3 movq %rax, %rdi subq %rcx, %rdi jbe 0x1411a5c movq (%r14), %r8 movq %rcx, %r9 shlq $0x6, %r9 addq %r9, %r8 addq $0x10, %r8 vbroadcastss 0xadb16e(%rip), %xmm6 # 0x1eecb84 vbroadcastss 0xada001(%rip), %xmm7 # 0x1eeba20 vmovaps %xmm7, %xmm1 vmovaps %xmm6, %xmm0 vmovaps -0x10(%r8), %xmm8 vmovaps (%r8), %xmm9 vminps %xmm8, %xmm7, %xmm7 vmaxps %xmm9, %xmm6, %xmm6 vaddps %xmm9, %xmm8, %xmm8 vminps %xmm8, %xmm1, %xmm1 vmaxps %xmm8, %xmm0, %xmm0 addq $0x40, %r8 decq %rdi jne 0x1411a27 movq %rax, %rdi subq %rcx, %rdi jmp 0x1411a66 xorl %edi, %edi vmovaps %xmm0, %xmm6 vmovaps %xmm1, %xmm7 vmovaps %xmm5, (%r15) vmovaps %xmm4, 0x10(%r15) vmovaps %xmm2, 0x20(%r15) vmovaps %xmm3, 0x30(%r15) movq %rdx, 0x40(%r15) movq %rcx, 0x48(%r15) movq %rcx, 0x50(%r15) vmovaps %xmm7, (%rbx) vmovaps %xmm6, 0x10(%rbx) vmovaps %xmm1, 0x20(%rbx) vmovaps %xmm0, 0x30(%rbx) movq %rcx, 0x40(%rbx) movq %rax, 0x48(%rbx) movq %rax, 0x50(%rbx) movq 0xa0(%rsp), %rcx subq %rax, %rcx je 0x141191e testq %rsi, %rsi js 0x1411c34 vcvtsi2ss %rsi, %xmm10, %xmm0 jmp 0x1411c4b movq %rax, %rsi shrq %rsi movl %eax, %edi andl $0x1, %edi orq %rsi, %rdi vcvtsi2ss %rdi, %xmm1, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rdx, %rax js 0x1411aef vcvtsi2ss %rax, %xmm1, %xmm1 jmp 0x1411b04 movq %rax, %rdx shrq %rdx andl $0x1, %eax orq %rdx, %rax vcvtsi2ss %rax, %xmm1, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vdivss %xmm1, %xmm0, %xmm0 testq %rcx, %rcx js 0x1411b14 vcvtsi2ss %rcx, %xmm2, %xmm1 jmp 0x1411b2b movq %rcx, %rax shrq %rax movl %ecx, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm2, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmulss %xmm1, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rax movq %rax, %rdx sarq $0x3f, %rdx vsubss 0xada2fb(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rax, %rsi cmpq %rcx, %rsi cmovaeq %rcx, %rsi subq %rsi, %rcx addq 0x48(%r15), %rsi movq %rsi, 0x50(%r15) movq 0x48(%rbx), %r12 addq %r12, %rcx movq %rcx, 0x50(%rbx) movq 0x50(%r15), %r13 movq 0x48(%r15), %rcx movq %r13, %rax subq %rcx, %rax movq %rax, 0x10(%rsp) movq 0x40(%rbx), %r15 movq %r12, %rax subq %r15, %rax movq %rax, 0x18(%rsp) subq %rcx, %r13 je 0x141191e cmpq %rax, %r13 jae 0x1411d9c addq %r15, %r13 leaq 0x28(%rsp), %r12 movq %r14, (%r12) leaq 0x18(%rsp), %rax movq %rax, 0x8(%r12) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r13, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r12, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x14158b8 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1411ecf leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x1411ead movq %rsi, %rax shrq %rax movl %esi, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm10, %xmm0 vaddss %xmm0, %xmm0, %xmm0 addq %rsi, %rdi js 0x1411c57 vcvtsi2ss %rdi, %xmm10, %xmm1 jmp 0x1411c6c movq %rdi, %rax shrq %rax andl $0x1, %edi orq %rax, %rdi vcvtsi2ss %rdi, %xmm10, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vdivss %xmm1, %xmm0, %xmm0 testq %rcx, %rcx js 0x1411c7c vcvtsi2ss %rcx, %xmm10, %xmm1 jmp 0x1411c93 movq %rcx, %rax shrq %rax movl %ecx, %edx andl $0x1, %edx orq %rax, %rdx vcvtsi2ss %rdx, %xmm10, %xmm1 vaddss %xmm1, %xmm1, %xmm1 vmulss %xmm1, %xmm0, %xmm0 vroundss $0x9, %xmm0, %xmm0, %xmm0 vcvttss2si %xmm0, %rax movq %rax, %rdx sarq $0x3f, %rdx vsubss 0xada193(%rip), %xmm0, %xmm0 # 0x1eebe44 vcvttss2si %xmm0, %rsi andq %rdx, %rsi orq %rax, %rsi cmpq %rcx, %rsi cmovaeq %rcx, %rsi subq %rsi, %rcx addq 0x48(%r15), %rsi movq %rsi, 0x50(%r15) movq 0x48(%rbx), %r12 addq %r12, %rcx movq %rcx, 0x50(%rbx) movq 0x50(%r15), %r13 movq 0x48(%r15), %rcx movq %r13, %rax subq %rcx, %rax movq %rax, 0x10(%rsp) movq 0x40(%rbx), %r15 movq %r12, %rax subq %r15, %rax movq %rax, 0x18(%rsp) subq %rcx, %r13 je 0x141191e cmpq %rax, %r13 jae 0x1411e27 addq %r15, %r13 leaq 0x28(%rsp), %r12 movq %r14, (%r12) leaq 0x18(%rsp), %rax movq %rax, 0x8(%r12) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r13, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r12, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x14158b8 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1411f33 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x1411ead leaq 0x28(%rsp), %r13 movq %r14, (%r13) leaq 0x10(%rsp), %rax movq %rax, 0x8(%r13) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r13, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x1415fc4 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1411f01 leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x1411ead leaq 0x28(%rsp), %r13 movq %r14, (%r13) leaq 0x10(%rsp), %rax movq %rax, 0x8(%r13) leaq 0xb0(%rsp), %r14 movw $0x401, 0xc(%r14) # imm = 0x401 vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x20(%r14) movq $0x8, 0x40(%r14) movq %r14, %rdi callq 0x6a660 leaq 0x38(%rsp), %rdi movq %r12, (%rdi) movq %r15, 0x8(%rdi) movq $0x40, 0x10(%rdi) leaq 0x20(%rsp), %rsi movq %r13, (%rsi) leaq 0xf(%rsp), %rdx movq %r14, %rcx callq 0x1415fc4 leaq 0xb0(%rsp), %rdi callq 0x6a770 testb %al, %al jne 0x1411f65 leaq 0xb0(%rsp), %rdi callq 0x6aab0 movq 0x10(%rsp), %rax vmovq %rax, %xmm0 vpshufd $0x44, %xmm0, %xmm0 # xmm0 = xmm0[0,1,0,1] vpaddq 0x40(%rbx), %xmm0, %xmm0 vmovdqa %xmm0, 0x40(%rbx) addq %rax, 0x50(%rbx) jmp 0x141191e movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xad9c0d(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd12d16(%rip), %rsi # 0x2124c08 movq 0xd12acf(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xad9bdb(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd12ce4(%rip), %rsi # 0x2124c08 movq 0xd12a9d(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xad9ba9(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd12cb2(%rip), %rsi # 0x2124c08 movq 0xd12a6b(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movl $0x10, %edi callq 0x6a3b0 movq %rax, %r14 leaq 0xad9b77(%rip), %rsi # 0x1eebaf0 movq %rax, %rdi callq 0x6a230 movq 0xd12c80(%rip), %rsi # 0x2124c08 movq 0xd12a39(%rip), %rdx # 0x21249c8 movq %r14, %rdi callq 0x6a5d0 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1411fd6 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1411fea movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1412006 movq %rax, %rbx movq %r14, %rdi callq 0x6a8a0 jmp 0x1412018 jmp 0x141202f jmp 0x141202f jmp 0x1411fd3 jmp 0x1411fe7 movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x1412025 jmp 0x141202f movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x1412025 jmp 0x141202f jmp 0x141202f jmp 0x141202f jmp 0x1412003 jmp 0x1412015 movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 jmp 0x1412025 movq %rax, %rbx leaq 0xb0(%rsp), %rdi callq 0x6aab0 movq %rbx, %rdi callq 0x6a600 jmp 0x141202f movq %rax, %rdi callq 0x8d6de8 nop
/embree[P]embree/kernels/bvh/../builders/heuristic_openmerge_array.h
embree::avx::HeuristicArrayOpenMergeSAH<embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::build()::'lambda'(embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef&, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef*), embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef, 32ul>::parallel_object_split(embree::avx::BinSplit<32ul> const&, embree::avx::PrimInfoExtRange const&, embree::avx::PrimInfoExtRange&, embree::avx::PrimInfoExtRange&)::'lambda'(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef const&)::operator()(embree::PrimInfoT<embree::BBox<embree::Vec3fa>>&, embree::avx::BVHNBuilderTwoLevel<8, embree::UserGeometry, embree::Object>::BuildRef const&) const
__forceinline operator Vec3fa () const { return Vec3fa(m128); }
vmovaps (%rdx), %xmm0 vmovaps 0x10(%rdx), %xmm1 vmovaps (%rsi), %xmm2 vmovaps 0x10(%rsi), %xmm3 vmovaps 0x20(%rsi), %xmm4 vmovaps 0x30(%rsi), %xmm5 vminps %xmm0, %xmm2, %xmm2 vmovaps %xmm2, (%rsi) vmaxps %xmm1, %xmm3, %xmm2 vmovaps %xmm2, 0x10(%rsi) vaddps %xmm1, %xmm0, %xmm0 vminps %xmm0, %xmm4, %xmm1 vmovaps %xmm1, 0x20(%rsi) vmaxps %xmm0, %xmm5, %xmm0 vmovaps %xmm0, 0x30(%rsi) incq 0x48(%rsi) retq
/embree[P]embree/kernels/bvh/../common/../../common/sys/../math/vec3fa.h
embree::avx::BVHNBuilderTwoLevel<8, embree::Instance, embree::InstancePrimitive>::clear()
void BVHNBuilderTwoLevel<N,Mesh,Primitive>::clear() { for (size_t i=0; i<bvh->objects.size(); i++) if (bvh->objects[i]) bvh->objects[i]->clear(); for (size_t i=0; i<builders.size(); i++) if (builders[i]) builders[i].reset(); refs.clear(); }
pushq %r14 pushq %rbx pushq %rax movq %rdi, %rbx movq 0x28(%rdi), %rcx movq 0x200(%rcx), %rax cmpq %rax, 0x208(%rcx) je 0x1419acc xorl %r14d, %r14d movq (%rax,%r14,8), %rdi testq %rdi, %rdi je 0x1419aab movq (%rdi), %rax callq *0x28(%rax) incq %r14 movq 0x28(%rbx), %rcx movq 0x200(%rcx), %rax movq 0x208(%rcx), %rcx subq %rax, %rcx sarq $0x3, %rcx cmpq %rcx, %r14 jb 0x1419a9c movq 0x10(%rbx), %rax cmpq %rax, 0x18(%rbx) je 0x1419b07 xorl %r14d, %r14d movq (%rax,%r14,8), %rdi testq %rdi, %rdi je 0x1419af0 movq $0x0, (%rax,%r14,8) movq (%rdi), %rax callq *0x8(%rax) incq %r14 movq 0x10(%rbx), %rax movq 0x18(%rbx), %rcx subq %rax, %rcx sarq $0x3, %rcx cmpq %rcx, %r14 jb 0x1419ad9 movq 0x50(%rbx), %r14 movq 0x58(%rbx), %rdi testq %rdi, %rdi je 0x1419b34 movq %r14, %rsi shlq $0x6, %rsi cmpq $0x1c00000, %rsi # imm = 0x1C00000 jb 0x1419b2f movzbl 0x40(%rbx), %edx callq 0x1ee67ca jmp 0x1419b34 callq 0x1ee612d testq %r14, %r14 je 0x1419b51 movq 0x38(%rbx), %rdi shlq $0x6, %r14 negq %r14 movq (%rdi), %rax movq %r14, %rsi movl $0x1, %edx callq *(%rax) vxorps %xmm0, %xmm0, %xmm0 vmovups %xmm0, 0x48(%rbx) movq $0x0, 0x58(%rbx) addq $0x8, %rsp popq %rbx popq %r14 retq
/embree[P]embree/kernels/bvh/bvh_builder_twolevel.cpp