Internalize cuda/detail/core/util.h (#3505)

cub/cub/agent/agent_adjacent_difference.cuh

@@ -79,7 +79,7 @@ template <typename Policy,
           bool ReadLeft>
 struct AgentDifference
 {
-  using LoadIt = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, InputIteratorT>::type;
+  using LoadIt = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, InputIteratorT>::type;
 
   using BlockLoad  = typename cub::BlockLoadType<Policy, LoadIt>::type;
   using BlockStore = typename cub::BlockStoreType<Policy, OutputIteratorT, OutputT>::type;
@@ -119,7 +119,7 @@ struct AgentDifference
     OffsetT num_items)
       : temp_storage(temp_storage.Alias())
       , input_it(input_it)
-      , load_it(THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator(Policy(), input_it))
+      , load_it(THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator(Policy(), input_it))
       , first_tile_previous(first_tile_previous)
       , result(result)
       , difference_op(difference_op)

cub/cub/agent/agent_merge.cuh

@@ -64,10 +64,10 @@ struct agent_t
   using key_type  = typename ::cuda::std::iterator_traits<KeysIt1>::value_type;
   using item_type = typename ::cuda::std::iterator_traits<ItemsIt1>::value_type;
 
-  using keys_load_it1  = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, KeysIt1>::type;
-  using keys_load_it2  = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, KeysIt2>::type;
-  using items_load_it1 = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, ItemsIt1>::type;
-  using items_load_it2 = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, ItemsIt2>::type;
+  using keys_load_it1  = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, KeysIt1>::type;
+  using keys_load_it2  = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, KeysIt2>::type;
+  using items_load_it1 = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, ItemsIt1>::type;
+  using items_load_it2 = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, ItemsIt2>::type;
 
   using block_load_keys1  = typename BlockLoadType<Policy, keys_load_it1>::type;
   using block_load_keys2  = typename BlockLoadType<Policy, keys_load_it2>::type;

cub/cub/agent/agent_merge_sort.cuh

@@ -91,8 +91,10 @@ struct AgentBlockSort
 
   using BlockMergeSortT = BlockMergeSort<KeyT, Policy::BLOCK_THREADS, Policy::ITEMS_PER_THREAD, ValueT>;
 
-  using KeysLoadIt  = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, KeyInputIteratorT>::type;
-  using ItemsLoadIt = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, ValueInputIteratorT>::type;
+  using KeysLoadIt =
+    typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, KeyInputIteratorT>::type;
+  using ItemsLoadIt =
+    typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, ValueInputIteratorT>::type;
 
   using BlockLoadKeys  = typename cub::BlockLoadType<Policy, KeysLoadIt>::type;
   using BlockLoadItems = typename cub::BlockLoadType<Policy, ItemsLoadIt>::type;
@@ -438,10 +440,11 @@ struct AgentMerge
   //---------------------------------------------------------------------
   // Types and constants
   //---------------------------------------------------------------------
-  using KeysLoadPingIt  = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, KeyIteratorT>::type;
-  using ItemsLoadPingIt = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, ValueIteratorT>::type;
-  using KeysLoadPongIt  = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, KeyT*>::type;
-  using ItemsLoadPongIt = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<Policy, ValueT*>::type;
+  using KeysLoadPingIt = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, KeyIteratorT>::type;
+  using ItemsLoadPingIt =
+    typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, ValueIteratorT>::type;
+  using KeysLoadPongIt  = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, KeyT*>::type;
+  using ItemsLoadPongIt = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<Policy, ValueT*>::type;
 
   using KeysOutputPongIt  = KeyIteratorT;
   using ItemsOutputPongIt = ValueIteratorT;

cub/cub/agent/agent_sub_warp_merge_sort.cuh

@@ -183,8 +183,8 @@ public:
 
   using WarpMergeSortT = WarpMergeSort<KeyT, PolicyT::ITEMS_PER_THREAD, PolicyT::WARP_THREADS, ValueT>;
 
-  using KeysLoadItT  = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<PolicyT, const KeyT*>::type;
-  using ItemsLoadItT = typename THRUST_NS_QUALIFIER::cuda_cub::core::LoadIterator<PolicyT, const ValueT*>::type;
+  using KeysLoadItT  = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<PolicyT, const KeyT*>::type;
+  using ItemsLoadItT = typename THRUST_NS_QUALIFIER::cuda_cub::core::detail::LoadIterator<PolicyT, const ValueT*>::type;
 
   using WarpLoadKeysT = cub::WarpLoad<KeyT, PolicyT::ITEMS_PER_THREAD, PolicyT::LOAD_ALGORITHM, PolicyT::WARP_THREADS>;
   using WarpLoadItemsT =

cub/cub/device/dispatch/dispatch_merge.cuh

@@ -138,7 +138,7 @@ __launch_bounds__(
     CompareOp>::type;
   using MergePolicy = typename MergeAgent::policy;
 
-  using THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator;
+  using THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator;
   using vsmem_helper_t = vsmem_helper_impl<MergeAgent>;
   __shared__ typename vsmem_helper_t::static_temp_storage_t shared_temp_storage;
   auto& temp_storage = vsmem_helper_t::get_temp_storage(shared_temp_storage, global_temp_storage);

cub/cub/device/dispatch/kernels/merge_sort.cuh

@@ -19,12 +19,13 @@
 
 THRUST_NAMESPACE_BEGIN
 
-namespace cuda_cub::core
+namespace cuda_cub::core::detail
 {
 // We must forward declare here because make_load_iterator.h pulls in non NVRTC compilable code
 template <class PtxPlan, class It>
-typename LoadIterator<PtxPlan, It>::type _CCCL_DEVICE _CCCL_FORCEINLINE make_load_iterator(PtxPlan const&, It it);
-} // namespace cuda_cub::core
+typename detail::LoadIterator<PtxPlan, It>::type _CCCL_DEVICE _CCCL_FORCEINLINE
+make_load_iterator(PtxPlan const&, It it);
+} // namespace cuda_cub::core::detail
 
 THRUST_NAMESPACE_END
 
@@ -196,8 +197,8 @@ __launch_bounds__(
   AgentBlockSortT agent(
     ping,
     temp_storage,
-    THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator(ActivePolicyT(), keys_in),
-    THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator(ActivePolicyT(), items_in),
+    THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator(ActivePolicyT(), keys_in),
+    THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator(ActivePolicyT(), items_in),
     keys_count,
     keys_out,
     items_out,
@@ -302,10 +303,10 @@ __launch_bounds__(
   AgentMergeT agent(
     ping,
     temp_storage,
-    THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator(ActivePolicyT(), keys_ping),
-    THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator(ActivePolicyT(), items_ping),
-    THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator(ActivePolicyT(), keys_pong),
-    THRUST_NS_QUALIFIER::cuda_cub::core::make_load_iterator(ActivePolicyT(), items_pong),
+    THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator(ActivePolicyT(), keys_ping),
+    THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator(ActivePolicyT(), items_ping),
+    THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator(ActivePolicyT(), keys_pong),
+    THRUST_NS_QUALIFIER::cuda_cub::core::detail::make_load_iterator(ActivePolicyT(), items_pong),
     keys_count,
     keys_pong,
     items_pong,

thrust/thrust/system/cuda/detail/core/agent_launcher.h

@@ -62,7 +62,8 @@ namespace cuda_cub
 {
 namespace core
 {
-
+namespace detail
+{
 #  ifndef THRUST_DETAIL_KERNEL_ATTRIBUTES
 #    define THRUST_DETAIL_KERNEL_ATTRIBUTES CCCL_DETAIL_KERNEL_ATTRIBUTES
 #  endif
@@ -97,7 +98,7 @@ THRUST_DETAIL_KERNEL_ATTRIBUTES void _kernel_agent_vshmem(char*, Args... args)
 template <class Agent>
 struct AgentLauncher : Agent
 {
-  core::AgentPlan plan;
+  AgentPlan plan;
   size_t count;
   cudaStream_t stream;
   char const* name;
@@ -121,7 +122,7 @@ struct AgentLauncher : Agent
       , name(name_)
       , grid(static_cast<unsigned int>((count + plan.items_per_tile - 1) / plan.items_per_tile))
       , vshmem(nullptr)
-      , has_shmem((size_t) core::get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
+      , has_shmem((size_t) get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
       , shmem_size(has_shmem ? plan.shared_memory_size : 0)
   {
     assert(count > 0);
@@ -136,7 +137,7 @@ struct AgentLauncher : Agent
       , name(name_)
       , grid(static_cast<unsigned int>((count + plan.items_per_tile - 1) / plan.items_per_tile))
       , vshmem(vshmem)
-      , has_shmem((size_t) core::get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
+      , has_shmem((size_t) get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
       , shmem_size(has_shmem ? plan.shared_memory_size : 0)
   {
     assert(count > 0);
@@ -149,7 +150,7 @@ struct AgentLauncher : Agent
       , name(name_)
       , grid(plan.grid_size)
       , vshmem(nullptr)
-      , has_shmem((size_t) core::get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
+      , has_shmem((size_t) get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
       , shmem_size(has_shmem ? plan.shared_memory_size : 0)
   {
     assert(plan.grid_size > 0);
@@ -162,43 +163,19 @@ struct AgentLauncher : Agent
       , name(name_)
       , grid(plan.grid_size)
       , vshmem(vshmem)
-      , has_shmem((size_t) core::get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
+      , has_shmem((size_t) get_max_shared_memory_per_block() >= (size_t) plan.shared_memory_size)
       , shmem_size(has_shmem ? plan.shared_memory_size : 0)
   {
     assert(plan.grid_size > 0);
   }
 
-#  if 0
-    THRUST_RUNTIME_FUNCTION
-    AgentPlan static get_plan(cudaStream_t s, void* d_ptr = 0)
-    {
-      // in separable compilation mode, we have no choice
-      // but to call kernel to get agent_plan
-      // otherwise the risk is something may fail
-      // if user mix & match ptx versions in a separably compiled function
-      // http://nvbugs/1772071
-      // XXX may be it is too string of a requirements, consider relaxing it in
-      // the future
-#    ifdef __CUDACC_RDC__
-      return core::get_agent_plan<Agent>(s, d_ptr);
-#    else
-      return get_agent_plan<Agent>(core::get_ptx_version());
-#    endif
-    }
-    THRUST_RUNTIME_FUNCTION
-    AgentPlan static get_plan_default()
-    {
-      return get_agent_plan<Agent>(sm_arch<0>::type::ver);
-    }
-#  endif
-
-  THRUST_RUNTIME_FUNCTION typename core::get_plan<Agent>::type static get_plan(cudaStream_t, void* d_ptr = 0)
+  THRUST_RUNTIME_FUNCTION typename get_plan<Agent>::type static get_plan(cudaStream_t, void* d_ptr = 0)
   {
     THRUST_UNUSED_VAR(d_ptr);
-    return get_agent_plan<Agent>(core::get_ptx_version());
+    return get_agent_plan<Agent>(get_ptx_version());
   }
 
-  THRUST_RUNTIME_FUNCTION typename core::get_plan<Agent>::type static get_plan()
+  THRUST_RUNTIME_FUNCTION typename detail::get_plan<Agent>::type static get_plan()
   {
     return get_agent_plan<Agent>(lowest_supported_sm_arch::ver);
   }
@@ -227,7 +204,7 @@ struct AgentLauncher : Agent
   {
 #  if THRUST_DEBUG_SYNC_FLAG
     cuda_optional<int> occ = max_sm_occupancy(k);
-    const int ptx_version  = core::get_ptx_version();
+    const int ptx_version  = get_ptx_version();
     if (count > 0)
     {
       _CubLog(
@@ -305,6 +282,7 @@ struct AgentLauncher : Agent
   }
 };
 
+} // namespace detail
 } // namespace core
 } // namespace cuda_cub
 

thrust/thrust/system/cuda/detail/core/load_iterator.h

@@ -34,7 +34,7 @@
 
 THRUST_NAMESPACE_BEGIN
 
-namespace cuda_cub::core
+namespace cuda_cub::core::detail
 {
 
 // LoadIterator
@@ -52,6 +52,6 @@ struct LoadIterator
                                cub::CacheModifiedInputIterator<PtxPlan::LOAD_MODIFIER, value_type, size_type>,
                                It>;
 }; // struct Iterator
-} // namespace cuda_cub::core
+} // namespace cuda_cub::core::detail
 
 THRUST_NAMESPACE_END

thrust/thrust/system/cuda/detail/core/make_load_iterator.h

@@ -33,7 +33,7 @@
 
 THRUST_NAMESPACE_BEGIN
 
-namespace cuda_cub::core
+namespace cuda_cub::core::detail
 {
 template <class PtxPlan, class It>
 typename LoadIterator<PtxPlan, It>::type _CCCL_DEVICE _CCCL_FORCEINLINE
@@ -55,6 +55,6 @@ typename LoadIterator<PtxPlan, It>::type _CCCL_DEVICE _CCCL_FORCEINLINE make_loa
   return make_load_iterator_impl<PtxPlan>(it, typename is_contiguous_iterator<It>::type());
 }
 
-} // namespace cuda_cub::core
+} // namespace cuda_cub::core::detail
 
 THRUST_NAMESPACE_END

thrust/thrust/system/cuda/detail/core/util.h

@@ -78,6 +78,8 @@ namespace core
 #  endif
 #endif
 
+namespace detail
+{
 /// Typelist - a container of types
 template <typename...>
 struct typelist;
@@ -458,22 +460,9 @@ THRUST_RUNTIME_FUNCTION inline size_t get_max_shared_memory_per_block()
   return static_cast<size_t>(i32value);
 }
 
-THRUST_RUNTIME_FUNCTION inline size_t virtual_shmem_size(size_t shmem_per_block)
-{
-  size_t max_shmem_per_block = core::get_max_shared_memory_per_block();
-  if (shmem_per_block > max_shmem_per_block)
-  {
-    return shmem_per_block;
-  }
-  else
-  {
-    return 0;
-  }
-}
-
 THRUST_RUNTIME_FUNCTION inline size_t vshmem_size(size_t shmem_per_block, size_t num_blocks)
 {
-  size_t max_shmem_per_block = core::get_max_shared_memory_per_block();
+  size_t max_shmem_per_block = get_max_shared_memory_per_block();
   if (shmem_per_block > max_shmem_per_block)
   {
     return shmem_per_block * num_blocks;
@@ -509,22 +498,6 @@ struct BlockLoad
                    get_arch<PtxPlan>::type::ver>;
 };
 
-// BlockStore
-// -----------
-// a helper metaprogram that returns type of a block loader
-template <class PtxPlan, class It, class T = typename iterator_traits<It>::value_type>
-struct BlockStore
-{
-  using type =
-    cub::BlockStore<T,
-                    PtxPlan::BLOCK_THREADS,
-                    PtxPlan::ITEMS_PER_THREAD,
-                    PtxPlan::STORE_ALGORITHM,
-                    1,
-                    1,
-                    get_arch<PtxPlan>::type::ver>;
-};
-
 // cuda_optional
 // --------------
 // used for function that return cudaError_t along with the result
@@ -619,16 +592,6 @@ THRUST_RUNTIME_FUNCTION inline int get_ptx_version()
   return ptx_version;
 }
 
-THRUST_RUNTIME_FUNCTION inline cudaError_t sync_stream(cudaStream_t stream)
-{
-  return cub::SyncStream(stream);
-}
-
-inline void _CCCL_DEVICE sync_threadblock()
-{
-  __syncthreads();
-}
-
 // Deprecated [Since 2.8]
 #define CUDA_CUB_RET_IF_FAIL(e)                \
   {                                            \
@@ -719,11 +682,6 @@ struct uninitialized_array
   }
 };
 
-_CCCL_HOST_DEVICE _CCCL_FORCEINLINE size_t align_to(size_t n, size_t align)
-{
-  return ((n + align - 1) / align) * align;
-}
-
 namespace host
 {
 inline cuda_optional<size_t> get_max_shared_memory_per_block()
@@ -753,9 +711,8 @@ THRUST_RUNTIME_FUNCTION cudaError_t alias_storage(
   return cub::AliasTemporaries(storage_ptr, storage_size, allocations, allocation_sizes);
 }
 
+} // namespace detail
 } // namespace core
-using core::sm52;
-using core::sm60;
 } // namespace cuda_cub
 
 THRUST_NAMESPACE_END