diff --git a/include/nbl/builtin/hlsl/bitonic_sort/common.hlsl b/include/nbl/builtin/hlsl/bitonic_sort/common.hlsl
new file mode 100644
index 0000000000..0b5bfb69ac
--- /dev/null
+++ b/include/nbl/builtin/hlsl/bitonic_sort/common.hlsl
@@ -0,0 +1,53 @@
+#ifndef _NBL_BUILTIN_HLSL_BITONIC_SORT_COMMON_INCLUDED_
+#define _NBL_BUILTIN_HLSL_BITONIC_SORT_COMMON_INCLUDED_
+
+#include <nbl/builtin/hlsl/cpp_compat.hlsl>
+#include <nbl/builtin/hlsl/concepts.hlsl>
+#include <nbl/builtin/hlsl/math/intutil.hlsl>
+#include <nbl/builtin/hlsl/utility.hlsl>
+
+namespace nbl
+{
+namespace hlsl
+{
+namespace bitonic_sort
+{
+
+template<typename KeyType, typename ValueType, typename Comparator>
+void compareExchangeWithPartner(
+	bool takeLarger,
+	NBL_REF_ARG(pair<KeyType, ValueType>) loPair,
+	NBL_CONST_REF_ARG(pair<KeyType, ValueType>) partnerLoPair,
+	NBL_REF_ARG(pair<KeyType, ValueType>) hiPair,
+	NBL_CONST_REF_ARG(pair<KeyType, ValueType>) partnerHiPair,
+	NBL_CONST_REF_ARG(Comparator) comp)
+{
+	const bool loSelfSmaller = comp(loPair.first, partnerLoPair.first);
+	const bool takePartnerLo = takeLarger ? loSelfSmaller : !loSelfSmaller;
+	if (takePartnerLo)
+		loPair = partnerLoPair;
+
+	const bool hiSelfSmaller = comp(hiPair.first, partnerHiPair.first);
+	const bool takePartnerHi = takeLarger ? hiSelfSmaller : !hiSelfSmaller;
+	if (takePartnerHi)
+		hiPair = partnerHiPair;
+}
+
+template<typename KeyType, typename ValueType, typename Comparator>
+void compareSwap(
+	bool ascending,
+	NBL_REF_ARG(pair<KeyType, ValueType>) loPair,
+	NBL_REF_ARG(pair<KeyType, ValueType>) hiPair,
+	NBL_CONST_REF_ARG(Comparator) comp)
+{
+	const bool shouldSwap = comp(hiPair.first, loPair.first);
+	const bool doSwap = (shouldSwap == ascending);
+
+	if (doSwap)
+		swap(loPair, hiPair);
+}
+}
+}
+}
+
+#endif
diff --git a/include/nbl/builtin/hlsl/concepts/accessors/bitonic_sort.hlsl b/include/nbl/builtin/hlsl/concepts/accessors/bitonic_sort.hlsl
new file mode 100644
index 0000000000..a39e91ffb3
--- /dev/null
+++ b/include/nbl/builtin/hlsl/concepts/accessors/bitonic_sort.hlsl
@@ -0,0 +1,31 @@
+#ifndef _NBL_BUILTIN_HLSL_CONCEPTS_ACCESSORS_BITONIC_SORT_INCLUDED_
+#define _NBL_BUILTIN_HLSL_CONCEPTS_ACCESSORS_BITONIC_SORT_INCLUDED_
+
+#include "nbl/builtin/hlsl/concepts/accessors/generic_shared_data.hlsl"
+
+namespace nbl
+{
+namespace hlsl
+{
+namespace workgroup
+{
+namespace bitonic_sort
+{
+// The SharedMemoryAccessor MUST provide the following methods:
+//  * void get(uint32_t index, NBL_REF_ARG(uint32_t) value);
+//  * void set(uint32_t index, in uint32_t value);
+//  * void workgroupExecutionAndMemoryBarrier();
+template<typename T, typename V = uint32_t, typename I = uint32_t>
+NBL_BOOL_CONCEPT BitonicSortSharedMemoryAccessor = concepts::accessors::GenericSharedMemoryAccessor<T, V, I>;
+
+// The Accessor MUST provide the following methods:
+//  * void get(uint32_t index, NBL_REF_ARG(pair<KeyType, ValueType>) value);
+//  * void set(uint32_t index, in pair<KeyType, ValueType> value);
+template<typename T, typename KeyType, typename ValueType, typename I = uint32_t>
+NBL_BOOL_CONCEPT BitonicSortAccessor = concepts::accessors::GenericDataAccessor<T, pair<KeyType, ValueType>, I>;
+
+}
+}
+}
+}
+#endif
diff --git a/include/nbl/builtin/hlsl/memory_accessor.hlsl b/include/nbl/builtin/hlsl/memory_accessor.hlsl
index 2194b1e917..cee8617e1a 100644
--- a/include/nbl/builtin/hlsl/memory_accessor.hlsl
+++ b/include/nbl/builtin/hlsl/memory_accessor.hlsl
@@ -22,18 +22,6 @@ namespace nbl
 {
 namespace hlsl
 {
-
-// TODO: flesh out and move to `nbl/builtin/hlsl/utility.hlsl`
-template<typename T1, typename T2>
-struct pair
-{
-    using first_type = T1;
-    using second_type = T2;
-
-    first_type first;
-    second_type second;
-};
-
 namespace accessor_adaptors
 {
 namespace impl
@@ -227,4 +215,4 @@ struct Offset : impl::OffsetBase<IndexType,_Offset>
 }
 }
 }
-#endif
\ No newline at end of file
+#endif
diff --git a/include/nbl/builtin/hlsl/subgroup/bitonic_sort.hlsl b/include/nbl/builtin/hlsl/subgroup/bitonic_sort.hlsl
new file mode 100644
index 0000000000..64ca5e1ced
--- /dev/null
+++ b/include/nbl/builtin/hlsl/subgroup/bitonic_sort.hlsl
@@ -0,0 +1,83 @@
+#ifndef NBL_BUILTIN_HLSL_SUBGROUP_BITONIC_SORT_INCLUDED
+#define NBL_BUILTIN_HLSL_SUBGROUP_BITONIC_SORT_INCLUDED
+#include "nbl/builtin/hlsl/bitonic_sort/common.hlsl"
+#include "nbl/builtin/hlsl/glsl_compat/subgroup_basic.hlsl"
+#include "nbl/builtin/hlsl/glsl_compat/subgroup_shuffle.hlsl"
+#include "nbl/builtin/hlsl/functional.hlsl"
+namespace nbl
+{
+namespace hlsl
+{
+namespace subgroup
+{
+
+template<typename KeyType, typename ValueType, typename Comparator = less<KeyType> >
+struct bitonic_sort_config
+{
+	using key_t = KeyType;
+	using value_t = ValueType;
+	using comparator_t = Comparator;
+};
+
+template<typename Config, class device_capabilities = void>
+struct bitonic_sort;
+
+template<typename KeyType, typename ValueType, typename Comparator, class device_capabilities>
+struct bitonic_sort<bitonic_sort_config<KeyType, ValueType, Comparator>, device_capabilities>
+{
+	using config_t = bitonic_sort_config<KeyType, ValueType, Comparator>;
+	using key_t = typename config_t::key_t;
+	using value_t = typename config_t::value_t;
+	using comparator_t = typename config_t::comparator_t;
+
+	static void mergeStage(uint32_t stage, bool bitonicAscending, uint32_t invocationID,
+		NBL_REF_ARG(pair<key_t, value_t>) loPair, NBL_REF_ARG(pair<key_t, value_t>) hiPair)
+	{
+		comparator_t comp;
+
+		[unroll]
+		for (uint32_t pass = 0; pass <= stage; pass++)
+		{
+			const uint32_t stride = 1u << (stage - pass); // Element stride
+			const uint32_t threadStride = stride >> 1;
+			if (threadStride == 0)
+			{
+				// Local compare and swap for stage 0
+				nbl::hlsl::bitonic_sort::compareSwap(bitonicAscending, loPair, hiPair, comp);
+			}
+			else
+			{
+				// Shuffle from partner using XOR
+				const key_t pLoKey = glsl::subgroupShuffleXor<key_t>(loPair.first, threadStride);
+				const value_t pLoVal = glsl::subgroupShuffleXor<value_t>(loPair.second, threadStride);
+				const key_t pHiKey = glsl::subgroupShuffleXor<key_t>(hiPair.first, threadStride);
+				const value_t pHiVal = glsl::subgroupShuffleXor<value_t>(hiPair.second, threadStride);
+
+				const pair<key_t, value_t> partnerLoPair = make_pair(pLoKey, pLoVal);
+				const pair<key_t, value_t> partnerHiPair = make_pair(pHiKey, pHiVal);
+
+				const bool isUpper = bool(invocationID & threadStride);
+				const bool takeLarger = isUpper == bitonicAscending;
+
+				nbl::hlsl::bitonic_sort::compareExchangeWithPartner(takeLarger, loPair, partnerLoPair, hiPair, partnerHiPair, comp);
+			}
+		}
+	}
+
+	static void __call(bool ascending, NBL_REF_ARG(pair<key_t, value_t>) loPair, NBL_REF_ARG(pair<key_t, value_t>) hiPair)
+	{
+		const uint32_t invocationID = glsl::gl_SubgroupInvocationID();
+		const uint32_t subgroupSizeLog2 = glsl::gl_SubgroupSizeLog2();
+		[unroll]
+		for (uint32_t stage = 0; stage <= subgroupSizeLog2; stage++)
+		{
+			const bool bitonicAscending = (stage == subgroupSizeLog2) ? ascending : !bool(invocationID & (1u << stage));
+			mergeStage(stage, bitonicAscending, invocationID, loPair, hiPair);
+		}
+	}
+};
+
+}
+}
+}
+#endif
diff --git a/include/nbl/builtin/hlsl/utility.hlsl b/include/nbl/builtin/hlsl/utility.hlsl
index 21f1eb1909..07c4b10624 100644
--- a/include/nbl/builtin/hlsl/utility.hlsl
+++ b/include/nbl/builtin/hlsl/utility.hlsl
@@ -1,40 +1,70 @@
-// Copyright (C) 2024 - DevSH Graphics Programming Sp. z O.O.
-// This file is part of the "Nabla Engine".
-// For conditions of distribution and use, see copyright notice in nabla.h
-#ifndef _NBL_BUILTIN_HLSL_UTILITY_INCLUDED_
-#define _NBL_BUILTIN_HLSL_UTILITY_INCLUDED_
-
-
-#include <nbl/builtin/hlsl/type_traits.hlsl>
-
-
-// for now we only implement declval
-namespace nbl
-{
-namespace hlsl
-{
-template<typename T>
-const static bool always_true = true;
-#ifndef __HLSL_VERSION
-
-template<class T>
-std::add_rvalue_reference_t<T> declval() noexcept
-{
-	static_assert(false,"Actually calling declval is ill-formed.");
-}
-
-#else
-
-namespace experimental
-{
-
-template<class T>
-T declval() {}
-
-}
-
-#endif
-}
-}
-
-#endif
+// Copyright (C) 2024 - DevSH Graphics Programming Sp. z O.O.
+// This file is part of the "Nabla Engine".
+// For conditions of distribution and use, see copyright notice in nabla.h
+#ifndef _NBL_BUILTIN_HLSL_UTILITY_INCLUDED_
+#define _NBL_BUILTIN_HLSL_UTILITY_INCLUDED_
+
+
+#include <nbl/builtin/hlsl/type_traits.hlsl>
+
+
+namespace nbl
+{
+namespace hlsl
+{
+
+template<typename T1, typename T2>
+struct pair
+{
+	using first_type = T1;
+	using second_type = T2;
+
+	first_type first;
+	second_type second;
+};
+
+template<typename T1, typename T2>
+pair<T1, T2> make_pair(T1 f, T2 s)
+{
+	pair<T1, T2> p;
+	p.first = f;
+	p.second = s;
+	return p;
+}
+
+template<typename T1, typename T2>
+void swap(NBL_REF_ARG(pair<T1, T2>) a, NBL_REF_ARG(pair<T1, T2>) b)
+{
+	T1 temp_first = a.first;
+	T2 temp_second = a.second;
+	a.first = b.first;
+	a.second = b.second;
+	b.first = temp_first;
+	b.second = temp_second;
+}
+
+template<typename T>
+const static bool always_true = true;
+#ifndef __HLSL_VERSION
+
+template<class T>
+std::add_rvalue_reference_t<T> declval() noexcept
+{
+	static_assert(false,"Actually calling declval is ill-formed.");
+}
+
+#else
+
+namespace experimental
+{
+
+template<class T>
+T declval() {}
+
+}
+
+#endif
+}
+}
+
+#endif
diff --git a/include/nbl/builtin/hlsl/workgroup/bitonic_sort.hlsl b/include/nbl/builtin/hlsl/workgroup/bitonic_sort.hlsl
new file mode 100644
index 0000000000..4ae393fbca
--- /dev/null
+++ b/include/nbl/builtin/hlsl/workgroup/bitonic_sort.hlsl
@@ -0,0 +1,253 @@
+#ifndef NBL_BUILTIN_HLSL_WORKGROUP_BITONIC_SORT_INCLUDED
+#define NBL_BUILTIN_HLSL_WORKGROUP_BITONIC_SORT_INCLUDED
+#include "nbl/builtin/hlsl/bitonic_sort/common.hlsl"
+#include "nbl/builtin/hlsl/memory_accessor.hlsl"
+#include "nbl/builtin/hlsl/functional.hlsl"
+#include "nbl/builtin/hlsl/subgroup/bitonic_sort.hlsl"
+#include "nbl/builtin/hlsl/bit.hlsl"
+#include "nbl/builtin/hlsl/workgroup/shuffle.hlsl"
+#include "nbl/builtin/hlsl/workgroup/basic.hlsl"
+#include "nbl/builtin/hlsl/concepts/accessors/bitonic_sort.hlsl"
+
+namespace nbl
+{
+namespace hlsl
+{
+namespace workgroup
+{
+namespace bitonic_sort
+{
+
+template<uint16_t _ElementsPerInvocationLog2, uint16_t _WorkgroupSizeLog2, typename KeyType, typename ValueType, typename Comparator = less<KeyType> NBL_PRIMARY_REQUIRES(_ElementsPerInvocationLog2 >= 1 && _WorkgroupSizeLog2 >= 5)
+struct bitonic_sort_config
+{
+    using key_t = KeyType;
+    using value_t = ValueType;
+    using comparator_t = Comparator;
+    NBL_CONSTEXPR_STATIC_INLINE uint16_t ElementsPerInvocationLog2 = _ElementsPerInvocationLog2;
+    NBL_CONSTEXPR_STATIC_INLINE uint16_t WorkgroupSizeLog2 = _WorkgroupSizeLog2;
+
+    NBL_CONSTEXPR_STATIC_INLINE uint32_t ElementsPerInvocation = 1u << ElementsPerInvocationLog2;
+    NBL_CONSTEXPR_STATIC_INLINE uint32_t WorkgroupSize = 1u << WorkgroupSizeLog2;
+    NBL_CONSTEXPR_STATIC_INLINE uint32_t SharedmemDWORDs = sizeof(pair<key_t, value_t>) / sizeof(uint32_t) * WorkgroupSize;
+
+};
+}
+
+template<typename Config, class device_capabilities = void>
+struct BitonicSort;
+
+// ==================== ElementsPerThreadLog2 = 1 Specialization (No Virtual Threading) ====================
+// This handles arrays of size WorkgroupSize * 2 using subgroup + workgroup operations
+template<uint16_t WorkgroupSizeLog2, typename KeyType, typename ValueType, typename Comparator, class device_capabilities>
+struct BitonicSort<bitonic_sort::bitonic_sort_config<1, WorkgroupSizeLog2, KeyType, ValueType, Comparator>, device_capabilities>
+{
+    using config_t = bitonic_sort::bitonic_sort_config<1, WorkgroupSizeLog2, KeyType, ValueType, Comparator>;
+    using key_t = KeyType;
+    using value_t = ValueType;
+    using comparator_t = Comparator;
+
+    using SortConfig = subgroup::bitonic_sort_config<key_t, value_t, comparator_t>;
+
+    template<typename Key, typename Value, typename key_adaptor_t, typename value_adaptor_t>
+    static void shuffleXor(NBL_REF_ARG(pair<Key, Value>) p, uint32_t ownedIdx, uint32_t mask, NBL_REF_ARG(key_adaptor_t) keyAdaptor, NBL_REF_ARG(value_adaptor_t) valueAdaptor)
+    {
+        keyAdaptor.template set<Key>(ownedIdx, p.first);
+        valueAdaptor.template set<Value>(ownedIdx, p.second);
+
+        // Wait until all writes are done before reading - only barrier on one adaptor here
+        keyAdaptor.workgroupExecutionAndMemoryBarrier();
+
+        keyAdaptor.template get<Key>(ownedIdx ^ mask, p.first);
+        valueAdaptor.template get<Value>(ownedIdx ^ mask, p.second);
+    }
+    
+
+    template<typename SharedMemoryAccessor>
+    static void mergeStage(NBL_REF_ARG(SharedMemoryAccessor) sharedmemAccessor, uint32_t stage, bool bitonicAscending, uint32_t invocationID,
+        NBL_REF_ARG(nbl::hlsl::pair<key_t, value_t>) lopair, NBL_REF_ARG(nbl::hlsl::pair<key_t, value_t>) hipair)
+    {
+        const uint32_t WorkgroupSize = config_t::WorkgroupSize;
+        const uint32_t subgroupSizeLog2 = glsl::gl_SubgroupSizeLog2();
+        comparator_t comp;
+
+
+        using key_adaptor_t = accessor_adaptors::StructureOfArrays<SharedMemoryAccessor, uint32_t, uint32_t, 1, WorkgroupSize>;
+        using value_adaptor_t = accessor_adaptors::StructureOfArrays<SharedMemoryAccessor, uint32_t, uint32_t, 1, WorkgroupSize, integral_constant<uint32_t, WorkgroupSize * sizeof(key_t) / sizeof(uint32_t)> >;
+
+        key_adaptor_t keyAdaptor;
+        keyAdaptor.accessor = sharedmemAccessor;
+        value_adaptor_t valueAdaptor;
+        valueAdaptor.accessor = sharedmemAccessor;
+
+        [unroll]
+        for (uint32_t pass = 0; pass <= stage; pass++)
+        {
+            if (pass)
+                sharedmemAccessor.workgroupExecutionAndMemoryBarrier();
+
+            const uint32_t stridePower = (stage - pass + 1) + subgroupSizeLog2;
+            const uint32_t stride = 1u << stridePower;
+            const uint32_t threadStride = stride >> 1;
+
+            nbl::hlsl::pair<key_t, value_t> plopair = lopair;
+            shuffleXor(plopair, invocationID, threadStride, keyAdaptor, valueAdaptor);
+
+            nbl::hlsl::pair<key_t, value_t> phipair = hipair;
+            shuffleXor(phipair, invocationID ^ threadStride, threadStride, keyAdaptor, valueAdaptor);
+
+            const bool isUpper = (invocationID & threadStride) != 0;
+            const bool takeLarger = isUpper == bitonicAscending;
+
+            nbl::hlsl::bitonic_sort::compareExchangeWithPartner(takeLarger, lopair, plopair, hipair, phipair, comp);
+        }
+    }
+
+    template<typename Accessor, typename SharedMemoryAccessor NBL_FUNC_REQUIRES(bitonic_sort::BitonicSortAccessor<Accessor, key_t, value_t>&& bitonic_sort::BitonicSortSharedMemoryAccessor<SharedMemoryAccessor>)
+    static void __call(NBL_REF_ARG(Accessor) accessor, NBL_REF_ARG(SharedMemoryAccessor) sharedmemAccessor)
+    {
+        const uint32_t WorkgroupSize = config_t::WorkgroupSize;
+
+        const uint32_t invocationID = glsl::gl_LocalInvocationID().x;
+        const uint32_t subgroupSizeLog2 = glsl::gl_SubgroupSizeLog2();
+        const uint32_t subgroupSize = 1u << subgroupSizeLog2;
+        const uint32_t subgroupID = glsl::gl_SubgroupID();
+        const uint32_t numSubgroups = WorkgroupSize / subgroupSize;
+        const uint32_t numSubgroupsLog2 = findMSB(numSubgroups);
+
+        const uint32_t loIdx = invocationID * 2;
+        const uint32_t hiIdx = loIdx | 1;
+
+        nbl::hlsl::pair<key_t, value_t> lopair, hipair;
+        accessor.template get<nbl::hlsl::pair<key_t, value_t> >(loIdx, lopair);
+        accessor.template get<nbl::hlsl::pair<key_t, value_t> >(hiIdx, hipair);
+
+        const bool subgroupAscending = (subgroupID & 1) == 0;
+        subgroup::bitonic_sort<SortConfig>::__call(subgroupAscending, lopair, hipair);
+
+        const uint32_t subgroupInvocationID = glsl::gl_SubgroupInvocationID();
+
+        [unroll]
+        for (uint32_t stage = 0; stage < numSubgroupsLog2; ++stage)
+        {
+            const bool bitonicAscending = !bool(invocationID & (subgroupSize << (stage + 1)));
+
+            mergeStage(sharedmemAccessor, stage, bitonicAscending, invocationID, lopair, hipair);
+
+            subgroup::bitonic_sort<SortConfig>::mergeStage(subgroupSizeLog2, bitonicAscending, subgroupInvocationID, lopair, hipair);
+        }
+
+        accessor.template set<nbl::hlsl::pair<key_t, value_t> >(loIdx, lopair);
+        accessor.template set<nbl::hlsl::pair<key_t, value_t> >(hiIdx, hipair);
+    }
+};
+
+// ==================== ElementsPerThreadLog2 > 1 Specialization (Virtual Threading) ====================
+// This handles larger arrays by combining global memory stages with recursive E=1 workgroup sorts
+template<uint16_t ElementsPerThreadLog2, uint16_t WorkgroupSizeLog2, typename KeyType, typename ValueType, typename Comparator, class device_capabilities>
+struct BitonicSort<bitonic_sort::bitonic_sort_config<ElementsPerThreadLog2, WorkgroupSizeLog2, KeyType, ValueType, Comparator>, device_capabilities>
+{
+    using config_t = bitonic_sort::bitonic_sort_config<ElementsPerThreadLog2, WorkgroupSizeLog2, KeyType, ValueType, Comparator>;
+    using simple_config_t = bitonic_sort::bitonic_sort_config<1, WorkgroupSizeLog2, KeyType, ValueType, Comparator>;
+
+    using key_t = KeyType;
+    using value_t = ValueType;
+    using comparator_t = Comparator;
+
+    template<typename Accessor, typename SharedMemoryAccessor>
+    static void __call(NBL_REF_ARG(Accessor) accessor, NBL_REF_ARG(SharedMemoryAccessor) sharedmemAccessor)
+    {
+        const uint32_t WorkgroupSize = config_t::WorkgroupSize;
+        const uint32_t ElementsPerThread = config_t::ElementsPerInvocation;
+        const uint32_t TotalElements = WorkgroupSize * ElementsPerThread;
+        const uint32_t ElementsPerSimpleSort = WorkgroupSize * 2;
+
+        const uint32_t threadID = glsl::gl_LocalInvocationID().x;
+        comparator_t comp;
+
+        // PHASE 1: Sub-sorts in chunks of size WorkgroupSize*2
+        accessor_adaptors::Offset<Accessor> offsetAccessor;
+        offsetAccessor.accessor = accessor;
+
+        const uint32_t numSub = TotalElements / ElementsPerSimpleSort;
+
+        [unroll]
+        for (uint32_t sub = 0; sub < numSub; sub++)
+        {
+            if (sub)
+                sharedmemAccessor.workgroupExecutionAndMemoryBarrier();
+
+                offsetAccessor.offset = sub * ElementsPerSimpleSort;
+
+                // Call E=1 workgroup sort
+                BitonicSort<simple_config_t, device_capabilities>::template __call(offsetAccessor, sharedmemAccessor);
+        }
+        sharedmemAccessor.workgroupExecutionAndMemoryBarrier();
+
+        // PHASE 2: Reverse odd-indexed chunks to form bitonic sequences
+        const uint32_t simpleLog = hlsl::findMSB(ElementsPerSimpleSort - 1) + 1u;
+        [unroll]
+        for (uint32_t sub = 1; sub < numSub; sub += 2)
+        {
+            offsetAccessor.offset = sub * ElementsPerSimpleSort;
+            [unroll]
+            for (uint32_t strideLog = simpleLog - 1u; strideLog + 1u > 0u; strideLog--)
+            {
+                const uint32_t stride = 1u << strideLog;
+                [unroll]
+                for (uint32_t virtualThreadID = threadID; virtualThreadID < ElementsPerSimpleSort / 2; virtualThreadID += WorkgroupSize)
+                {
+                    const uint32_t loIx = (((virtualThreadID & (~(stride - 1u))) << 1u) | (virtualThreadID & (stride - 1u))) + offsetAccessor.offset;
+                    const uint32_t hiIx = loIx | stride;
+
+                    nbl::hlsl::pair<key_t, value_t> lopair, hipair;
+                    accessor.template get<nbl::hlsl::pair<key_t, value_t> >(loIx, lopair);
+                    accessor.template get<nbl::hlsl::pair<key_t, value_t> >(hiIx, hipair);
+
+                    swap(lopair, hipair);
+
+                    accessor.template set<nbl::hlsl::pair<key_t, value_t> >(loIx, lopair);
+                    accessor.template set<nbl::hlsl::pair<key_t, value_t> >(hiIx, hipair);
+                }
+                sharedmemAccessor.workgroupExecutionAndMemoryBarrier();
+                }
+            }
+
+            // PHASE 3: Global memory bitonic merge
+            const uint32_t totalLog = hlsl::findMSB(TotalElements - 1) + 1u;
+            [unroll]
+            for (uint32_t blockLog = simpleLog + 1u; blockLog <= totalLog; blockLog++)
+            {
+                const uint32_t k = 1u << blockLog;
+                [unroll]
+                for (uint32_t strideLog = blockLog - 1u; strideLog + 1u > 0u; strideLog--)
+                {
+                    const uint32_t stride = 1u << strideLog;
+                    [unroll]
+                    for (uint32_t virtualThreadID = threadID; virtualThreadID < TotalElements / 2; virtualThreadID += WorkgroupSize)
+                    {
+                        const uint32_t loIx = ((virtualThreadID & (~(stride - 1u))) << 1u) | (virtualThreadID & (stride - 1u));
+                        const uint32_t hiIx = loIx | stride;
+
+                        const bool bitonicAscending = ((loIx & k) == 0u);
+
+                        nbl::hlsl::pair<key_t, value_t> lopair, hipair;
+                        accessor.template get<nbl::hlsl::pair<key_t, value_t> >(loIx, lopair);
+                        accessor.template get<nbl::hlsl::pair<key_t, value_t> >(hiIx, hipair);
+
+                        nbl::hlsl::bitonic_sort::compareSwap(bitonicAscending, lopair, hipair, comp);
+
+                        accessor.template set<nbl::hlsl::pair<key_t, value_t> >(loIx, lopair);
+                        accessor.template set<nbl::hlsl::pair<key_t, value_t> >(hiIx, hipair);
+                    }
+                        sharedmemAccessor.workgroupExecutionAndMemoryBarrier();
+                }
+            }
+    }
+};
+
+}
+}
+}
+
+#endif
diff --git a/src/nbl/builtin/CMakeLists.txt b/src/nbl/builtin/CMakeLists.txt
index cc81b093a2..c56a468112 100644
--- a/src/nbl/builtin/CMakeLists.txt
+++ b/src/nbl/builtin/CMakeLists.txt
@@ -15,7 +15,15 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "shader/loader/gltf/fragment_impl.g
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "shader/loader/gltf/uv.frag")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "shader/loader/gltf/color.frag")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "shader/loader/gltf/no_uv_color.frag")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "material/lambertian/singletexture/specialized_shader.vert")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "material/lambertian/singletexture/specialized_shader.frag")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "material/debug/vertex_color/specialized_shader.vert")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "material/debug/vertex_normal/specialized_shader.vert")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "material/debug/vertex_normal/specialized_shader.frag")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "material/debug/vertex_uv/specialized_shader.frag")
 # generic GLSL headers after this line
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/macros.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/algorithm.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ieee754.glsl")
 # barycentric
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/barycentric/extensions.glsl")
@@ -31,6 +39,30 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bda/legacy_bda_accessor.hlsl"
 # bump mapping
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bump_mapping/fragment.glsl") # TODO: rename to `frag.glsl`
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bump_mapping/utils.glsl")
+# bxdf
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/common.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/common_samples.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/fresnel.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/ndf/common.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/ndf/blinn_phong.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/ndf/beckmann.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/ndf/ggx.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/geom/smith/common.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/geom/smith/beckmann.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/geom/smith/ggx.glsl")
+# brdf
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/brdf/diffuse/lambert.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/brdf/diffuse/oren_nayar.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/brdf/specular/blinn_phong.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/brdf/specular/beckmann.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/brdf/specular/ggx.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/brdf/diffuse/fresnel_correction.glsl")
+# bsdf
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/bsdf/diffuse/lambert.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/bsdf/specular/common.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/bsdf/specular/dielectric.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/bsdf/specular/beckmann.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/bxdf/bsdf/specular/ggx.glsl")
 # colorspace
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/colorspace/EOTF.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/colorspace/OETF.glsl")
@@ -55,15 +87,19 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/format/constants.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/loader/mtl/common.glsl")
 # LoD Library
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/lod_library/structs.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/lod_library/descriptor_set.glsl")
 # math and limits
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/math/constants.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/math/complex.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/math/functions.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/math/quaternions.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/math/typeless_arithmetic.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/limits/numeric.glsl")
 # material_compiler
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/material_compiler/common.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/material_compiler/common_declarations.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/material_compiler/common_invariant_declarations.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/material_compiler/rasterization/impl.glsl")
 # property pool
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/property_pool/transfer.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/property_pool/copy.comp")
@@ -73,7 +109,6 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/random/xoroshiro.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/random/pcg.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/random/lcg.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/random/tea.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/random/dim_adaptor_recursive.hlsl")
 # sampling
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/sampling/bilinear.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/sampling/box_muller_transform.glsl")
@@ -84,6 +119,14 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/sampling/projected_spherical_
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/sampling/spherical_rectangle.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/sampling/spherical_triangle.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/sampling/quantized_sequence.glsl")
+# global exclusive scan
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scan/direct.comp")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scan/declarations.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scan/descriptors.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scan/default_scheduler.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scan/indirect.comp")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scan/parameters_struct.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scan/virtual_workgroup.glsl")
 # faster and easier scan
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/scanning_append/scanning_append.glsl")
 # scene
@@ -102,11 +145,30 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/skinning/debug.vert")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/skinning/linear.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/skinning/render_descriptor_set.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/skinning/update_descriptor_set.glsl")
+# subgroup emulation
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/subgroup/arithmetic_portability.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/subgroup/arithmetic_portability_impl.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/subgroup/basic_portability.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/subgroup/fft.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/subgroup/shared_arithmetic_portability.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/subgroup/shared_shuffle_portability.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/subgroup/shuffle_portability.glsl")
 # utilities
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/utils/common.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/utils/culling.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/utils/compressed_normal_matrix_t.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/utils/normal_decode.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/utils/normal_encode.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/utils/transform.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/utils/morton.glsl")
+# workgroup "intrinsics"
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/workgroup/arithmetic.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/workgroup/basic.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/workgroup/ballot.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/workgroup/fft.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/workgroup/shared_arithmetic.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/workgroup/shared_ballot.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/workgroup/shared_fft.glsl")
 #transform_tree
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/transform_tree/global_transform_update.comp")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/transform_tree/global_transform_and_normal_matrix_update.comp")
@@ -120,7 +182,14 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/transform_tree/relative_trans
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/transform_tree/relative_transform_update_common.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/transform_tree/relative_transform_update_descriptor_set.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/transform_tree/debug.vert")
-#
+# ext shouldn't be built into the engine, but there's no harm including some non-dynamic GLSL source to make life easier
+#LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/.glsl")
+# radix sort
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/FFT/default_compute_fft.comp")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/FFT/fft.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/FFT/parameters_struct.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/FFT/parameters.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/FFT/types.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/LumaMeter/common.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/LumaMeter/impl.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/ToneMapper/operators.glsl")
@@ -133,6 +202,10 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/MitsubaLoader/material_co
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/OIT/oit.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/OIT/insert_node.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/OIT/resolve.frag")
+# virtual geometry
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/virtual_geometry/descriptors.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/virtual_geometry/virtual_attribute.glsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/virtual_geometry/virtual_attribute_fetch.glsl")
 # depth pyramid generator
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/DepthPyramidGenerator/common.glsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/DepthPyramidGenerator/push_constants_struct_common.h")
@@ -141,7 +214,6 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "glsl/ext/DepthPyramidGenerator/vir
 
 # HLSL
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/macros.h")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/indirect_commands.hlsl")
 # emulated
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/emulated/float64_t.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/emulated/float64_t_impl.hlsl")
@@ -195,6 +267,7 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/limits.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/type_traits.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/tuple.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/utility.hlsl")
+
 #metaprogramming
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/mpl.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/member_test_macros.hlsl")
@@ -218,12 +291,10 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/format/shared_exp.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/format.hlsl")
 #linear algebra
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/linalg/fast_affine.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/linalg/transform.hlsl")
 # TODO: rename `equations` to `polynomials` probably
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/functions.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/geometry.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/intutil.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/angle_adding.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/equations/quadratic.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/equations/cubic.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/math/equations/quartic.hlsl")
@@ -246,52 +317,24 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/shapes/circle.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/shapes/ellipse.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/shapes/line.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/shapes/beziers.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/shapes/aabb.hlsl")
-#sampling
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/sampling/concentric_mapping.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/sampling/cos_weighted_spheres.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/sampling/quotient_and_pdf.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/sampling/uniform_spheres.hlsl")
 #
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/ndarray_addressing.hlsl")
 #
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/shapes/util.hlsl")
 #FFT
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/fft/common.hlsl")
+#Bitonic_sort
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bitonic_sort/common.hlsl")
 #sort
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/sort/common.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/sort/counting.hlsl")
-#bxdf
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/common.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/fresnel.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/ndf.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/config.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/cook_torrance_base.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/reflection.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/transmission.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/bxdf_traits.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/ndf/beckmann.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/ndf/ggx.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/ndf/microfacet_to_light_transform.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/base/lambertian.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/base/oren_nayar.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/reflection/beckmann.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/reflection/ggx.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/reflection/lambertian.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/reflection/oren_nayar.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/reflection/delta_distribution.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/transmission/beckmann.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/transmission/ggx.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/transmission/lambertian.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/transmission/oren_nayar.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/transmission/smooth_dielectric.hlsl")
-LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/bxdf/transmission/delta_distribution.hlsl")
 #subgroup
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup/ballot.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup/basic.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup/arithmetic_portability.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup/arithmetic_portability_impl.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup/fft.hlsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup/bitonic_sort.hlsl")
 #subgroup2
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup2/ballot.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/subgroup2/arithmetic_params.hlsl")
@@ -305,6 +348,7 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/basic.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/ballot.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/broadcast.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/fft.hlsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/bitonic_sort.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/scratch_size.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/shared_scan.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/workgroup/shuffle.hlsl")
@@ -338,6 +382,7 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/concepts/accessors/mip_mapped
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/concepts/accessors/storable_image.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/concepts/accessors/generic_shared_data.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/concepts/accessors/fft.hlsl")
+LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/concepts/accessors/bitonic_sort.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/concepts/accessors/workgroup_arithmetic.hlsl")
 #tgmath
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/tgmath.hlsl")
@@ -347,4 +392,4 @@ LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/tgmath/output_structs.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/prefix_sum_blur/blur.hlsl")
 LIST_BUILTIN_RESOURCE(NBL_RESOURCES_TO_EMBED "hlsl/prefix_sum_blur/box_sampler.hlsl")
 
-ADD_CUSTOM_BUILTIN_RESOURCES(nblBuiltinResourceData NBL_RESOURCES_TO_EMBED "${NBL_ROOT_PATH}/include" "nbl/builtin" "nbl::builtin" "${NBL_ROOT_PATH_BINARY}/include" "${NBL_ROOT_PATH_BINARY}/src" "STATIC" "INTERNAL")
\ No newline at end of file
+ADD_CUSTOM_BUILTIN_RESOURCES(nblBuiltinResourceData NBL_RESOURCES_TO_EMBED "${NBL_ROOT_PATH}/include" "nbl/builtin" "nbl::builtin" "${NBL_ROOT_PATH_BINARY}/include" "${NBL_ROOT_PATH_BINARY}/src" "STATIC" "INTERNAL")