From fcf40649feb6046fb9b1ed984fb9b19422cd5463 Mon Sep 17 00:00:00 2001 From: aamir Date: Sat, 21 Jul 2018 19:30:40 -0700 Subject: adding the cutlass examples --- cutlass-example/cutlass/convert.h | 102 +++ cutlass-example/cutlass/coord.h | 287 +++++++ cutlass-example/cutlass/core_io.h | 44 + cutlass-example/cutlass/cutlass.h | 73 ++ cutlass-example/cutlass/fragment.h | 278 +++++++ cutlass-example/cutlass/fragment_load_store.h | 135 ++++ cutlass-example/cutlass/fragment_multiply_add.h | 149 ++++ cutlass-example/cutlass/gemm/clear_accumulators.h | 57 ++ cutlass-example/cutlass/gemm/dgemm_traits.h | 127 +++ cutlass-example/cutlass/gemm/gemm.h | 344 ++++++++ cutlass-example/cutlass/gemm/gemm_epilogue.h | 231 ++++++ .../cutlass/gemm/gemm_epilogue_traits.h | 331 ++++++++ cutlass-example/cutlass/gemm/gemm_global_stream.h | 182 +++++ cutlass-example/cutlass/gemm/gemm_global_tile.h | 541 +++++++++++++ cutlass-example/cutlass/gemm/gemm_operand.h | 141 ++++ cutlass-example/cutlass/gemm/gemm_shared_stream.h | 113 +++ cutlass-example/cutlass/gemm/gemm_shared_tile.h | 417 ++++++++++ cutlass-example/cutlass/gemm/gemm_traits.h | 818 +++++++++++++++++++ cutlass-example/cutlass/gemm/hgemm_global_tile.h | 90 +++ cutlass-example/cutlass/gemm/hgemm_multiply_add.h | 104 +++ cutlass-example/cutlass/gemm/hgemm_swizzle.h | 94 +++ cutlass-example/cutlass/gemm/hgemm_traits.h | 397 +++++++++ .../cutlass/gemm/identity_block_swizzle.h | 48 ++ cutlass-example/cutlass/gemm/igemm_epilogue.h | 320 ++++++++ cutlass-example/cutlass/gemm/igemm_global_tile.h | 161 ++++ cutlass-example/cutlass/gemm/igemm_multiply_add.h | 89 ++ cutlass-example/cutlass/gemm/igemm_swizzle.h | 115 +++ cutlass-example/cutlass/gemm/igemm_traits.h | 539 ++++++++++++ cutlass-example/cutlass/gemm/linear_scaling.h | 85 ++ cutlass-example/cutlass/gemm/sgemm_traits.h | 127 +++ cutlass-example/cutlass/gemm/thread_multiply_add.h | 84 ++ .../cutlass/gemm/wmma_gemm_epilogue_traits.h | 161 ++++ .../cutlass/gemm/wmma_gemm_global_tile.h | 211 +++++ .../cutlass/gemm/wmma_gemm_multiply_add.h | 108 +++ .../cutlass/gemm/wmma_gemm_shared_tile.h | 240 ++++++ cutlass-example/cutlass/gemm/wmma_gemm_traits.h | 574 +++++++++++++ cutlass-example/cutlass/iterator_access.h | 318 ++++++++ cutlass-example/cutlass/load_store.h | 222 +++++ cutlass-example/cutlass/matrix_traits.h | 48 ++ cutlass-example/cutlass/predicate_vector.h | 493 +++++++++++ cutlass-example/cutlass/reshape_tile.h | 58 ++ cutlass-example/cutlass/shape.h | 305 +++++++ cutlass-example/cutlass/tensor_ref.h | 151 ++++ cutlass-example/cutlass/tensor_view.h | 172 ++++ cutlass-example/cutlass/tile_iterator.h | 899 +++++++++++++++++++++ cutlass-example/cutlass/tile_traits_standard.h | 238 ++++++ cutlass-example/cutlass/util/cutlass_math.h | 131 +++ cutlass-example/cutlass/util/debug.h | 122 +++ cutlass-example/cutlass/util/platform.h | 801 ++++++++++++++++++ cutlass-example/cutlass/vector.h | 229 ++++++ cutlass-example/cutlass/wmma_matrix.h | 193 +++++ 51 files changed, 12297 insertions(+) create mode 100644 cutlass-example/cutlass/convert.h create mode 100644 cutlass-example/cutlass/coord.h create mode 100644 cutlass-example/cutlass/core_io.h create mode 100644 cutlass-example/cutlass/cutlass.h create mode 100644 cutlass-example/cutlass/fragment.h create mode 100644 cutlass-example/cutlass/fragment_load_store.h create mode 100644 cutlass-example/cutlass/fragment_multiply_add.h create mode 100644 cutlass-example/cutlass/gemm/clear_accumulators.h create mode 100644 cutlass-example/cutlass/gemm/dgemm_traits.h create mode 100644 cutlass-example/cutlass/gemm/gemm.h create mode 100644 cutlass-example/cutlass/gemm/gemm_epilogue.h create mode 100644 cutlass-example/cutlass/gemm/gemm_epilogue_traits.h create mode 100644 cutlass-example/cutlass/gemm/gemm_global_stream.h create mode 100644 cutlass-example/cutlass/gemm/gemm_global_tile.h create mode 100644 cutlass-example/cutlass/gemm/gemm_operand.h create mode 100644 cutlass-example/cutlass/gemm/gemm_shared_stream.h create mode 100644 cutlass-example/cutlass/gemm/gemm_shared_tile.h create mode 100644 cutlass-example/cutlass/gemm/gemm_traits.h create mode 100644 cutlass-example/cutlass/gemm/hgemm_global_tile.h create mode 100644 cutlass-example/cutlass/gemm/hgemm_multiply_add.h create mode 100644 cutlass-example/cutlass/gemm/hgemm_swizzle.h create mode 100644 cutlass-example/cutlass/gemm/hgemm_traits.h create mode 100644 cutlass-example/cutlass/gemm/identity_block_swizzle.h create mode 100644 cutlass-example/cutlass/gemm/igemm_epilogue.h create mode 100644 cutlass-example/cutlass/gemm/igemm_global_tile.h create mode 100644 cutlass-example/cutlass/gemm/igemm_multiply_add.h create mode 100644 cutlass-example/cutlass/gemm/igemm_swizzle.h create mode 100644 cutlass-example/cutlass/gemm/igemm_traits.h create mode 100644 cutlass-example/cutlass/gemm/linear_scaling.h create mode 100644 cutlass-example/cutlass/gemm/sgemm_traits.h create mode 100644 cutlass-example/cutlass/gemm/thread_multiply_add.h create mode 100644 cutlass-example/cutlass/gemm/wmma_gemm_epilogue_traits.h create mode 100644 cutlass-example/cutlass/gemm/wmma_gemm_global_tile.h create mode 100644 cutlass-example/cutlass/gemm/wmma_gemm_multiply_add.h create mode 100644 cutlass-example/cutlass/gemm/wmma_gemm_shared_tile.h create mode 100644 cutlass-example/cutlass/gemm/wmma_gemm_traits.h create mode 100644 cutlass-example/cutlass/iterator_access.h create mode 100644 cutlass-example/cutlass/load_store.h create mode 100644 cutlass-example/cutlass/matrix_traits.h create mode 100644 cutlass-example/cutlass/predicate_vector.h create mode 100644 cutlass-example/cutlass/reshape_tile.h create mode 100644 cutlass-example/cutlass/shape.h create mode 100644 cutlass-example/cutlass/tensor_ref.h create mode 100644 cutlass-example/cutlass/tensor_view.h create mode 100644 cutlass-example/cutlass/tile_iterator.h create mode 100644 cutlass-example/cutlass/tile_traits_standard.h create mode 100644 cutlass-example/cutlass/util/cutlass_math.h create mode 100644 cutlass-example/cutlass/util/debug.h create mode 100644 cutlass-example/cutlass/util/platform.h create mode 100644 cutlass-example/cutlass/vector.h create mode 100644 cutlass-example/cutlass/wmma_matrix.h (limited to 'cutlass-example/cutlass') diff --git a/cutlass-example/cutlass/convert.h b/cutlass-example/cutlass/convert.h new file mode 100644 index 0000000..933d68a --- /dev/null +++ b/cutlass-example/cutlass/convert.h @@ -0,0 +1,102 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! + \file + \brief Defines conversion operations among Fragments of different base type. +*/ +#pragma once + +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Convert {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Convert, Fragment > { + /// The input fragment. + typedef Fragment InputFragment; + /// The output fragment. + typedef Fragment OutputFragment; + + /// Ctor. + CUTLASS_DEVICE Convert() {} + + /// Transform a fragment. + CUTLASS_DEVICE void transform(InputFragment const& src, OutputFragment& dst) { + transform(src, 0, dst); + } + + /// Transform a fragment. + template + CUTLASS_DEVICE void transform(Fragment_ const& src, int offset, OutputFragment& dst) { + for (int i = 0; i < kScalars_; ++i) { + dst[i] = static_cast(src[i + offset]); + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Copy { + /// The input fragment. + typedef Fragment_ InputFragment; + /// The output fragment. + typedef Fragment_ OutputFragment; + + /// Ctor. + CUTLASS_DEVICE Copy() {} + + /// Transform a fragment. + CUTLASS_DEVICE void transform(Fragment_ const& src, Fragment_& dst) { transform(src, 0, dst); } + + /// Transform a fragment. + template + CUTLASS_DEVICE void transform(InputFragment_ const& src, int offset, Fragment_& dst) { + if (sizeof(typename Fragment_::Element) == 8) { + uint64_t const* src_ptr = reinterpret_cast(&src[offset]); + uint64_t* dst_ptr = reinterpret_cast(&dst[0]); + for (int i = 0; i < sizeof(Fragment_) / 8; ++i) { + dst_ptr[i] = src_ptr[i]; + } + } else { + uint32_t const* src_ptr = reinterpret_cast(&src[offset]); + uint32_t* dst_ptr = reinterpret_cast(&dst[0]); + for (int i = 0; i < sizeof(Fragment_) / 4; ++i) { + dst_ptr[i] = src_ptr[i]; + } + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/coord.h b/cutlass-example/cutlass/coord.h new file mode 100644 index 0000000..431c9bf --- /dev/null +++ b/cutlass-example/cutlass/coord.h @@ -0,0 +1,287 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief A Coord is a coordinate of arbitrary rank into a tensor or matrix +*/ + +#pragma once + +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Describes identity elements +struct Identity { + /// Enumeration describing identity elements. Value assignments are significant. + /// Feel free to add or multiply by these, respectively. + enum Kind { Additive = 0, Multiplicative = 1 }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Statically-sized array specifying Coords within a tensor +template +struct Coord { + // + // Type and constant definitions + // + + static int const N = N_; + + // + // Data members + // + + /// Indices + int idx[N]; + + // + // Methods + // + + /// Default ctor initializes uniformly + CUTLASS_HOST_DEVICE + Coord(int value = 0) { + for (int i = 0; i < N; ++i) { + idx[i] = value; + } + } + + /// Constructs from an array of integers + CUTLASS_HOST_DEVICE + Coord(int _idx[]) { + for (int i = 0; i < N; ++i) { + idx[i] = _idx[i]; + } + } + + /// Element-wise addition + CUTLASS_HOST_DEVICE + Coord operator+(Coord const& b) const { + Coord c; + for (int i = 0; i < N; ++i) { + c.idx[i] = idx[i] + b.idx[i]; + } + return c; + } + + /// Element-wise subtraction + CUTLASS_HOST_DEVICE + Coord operator-(Coord const& b) const { + Coord c; + for (int i = 0; i < N; ++i) { + c.idx[i] = idx[i] - b.idx[i]; + } + return c; + } + + /// Element-wise multiplication + CUTLASS_HOST_DEVICE + Coord operator*(Coord const& b) const { + Coord c; + for (int i = 0; i < N; ++i) { + c.idx[i] = idx[i] * b.idx[i]; + } + return c; + } + + /// Element-wise division + CUTLASS_HOST_DEVICE + Coord operator/(Coord const& b) const { + Coord c; + for (int i = 0; i < N; ++i) { + c.idx[i] = idx[i] / b.idx[i]; + } + return c; + } + + /// In-place addition + CUTLASS_HOST_DEVICE + Coord& operator+=(Coord const& b) { + for (int i = 0; i < N; ++i) { + idx[i] += b.idx[i]; + } + return *this; + } + + /// In-place subtraction + CUTLASS_HOST_DEVICE + Coord& operator-=(Coord const& b) { + for (int i = 0; i < N; ++i) { + idx[i] -= b.idx[i]; + } + return *this; + } + + /// In-place multiplication + CUTLASS_HOST_DEVICE + Coord& operator*=(Coord const& b) { + for (int i = 0; i < N; ++i) { + idx[i] *= b.idx[i]; + } + return *this; + } + + /// In-place division + CUTLASS_HOST_DEVICE + Coord& operator/=(Coord const& b) { + for (int i = 0; i < N; ++i) { + idx[i] /= b.idx[i]; + } + return *this; + } + + /// Member access operator + CUTLASS_HOST_DEVICE int& operator[](int dim) { return idx[dim]; } + + /// Member access operator + CUTLASS_HOST_DEVICE int const& operator[](int dim) const { return idx[dim]; } + + /// Computes the dot product of two Coord instances + template + CUTLASS_HOST_DEVICE T dot(Coord const& b, T sum) const { + for (int i = 0; i < N; ++i) { + sum += idx[i] * b.idx[i]; + } + return sum; + } + + /// Computes the dot product of two Coord instances + template + CUTLASS_HOST_DEVICE T dot(Coord const& b) const { + T sum = T(0); + for (int i = 0; i < N; ++i) { + sum += idx[i] * b.idx[i]; + } + return sum; + } + + /// Gets the index of a given Coord element + template + CUTLASS_HOST_DEVICE int& at() { + return idx[Dim]; + } + + /// Access via index; may limit unrolling potential + CUTLASS_HOST_DEVICE + int& at(int dim) { return idx[dim]; } + + /// Gets the index of a given Coord element + template + CUTLASS_HOST_DEVICE int const& at() const { + return idx[Dim]; + } + + /// Access via index; may limit unrolling potential + CUTLASS_HOST_DEVICE + int const& at(int dim) const { return idx[dim]; } + + /// Determines if two Coord<> objects are equal + CUTLASS_HOST_DEVICE + bool operator==(Coord const& b) const { + bool equal = true; + for (int i = 0; equal && i < N; ++i) { + equal = (idx[i] == b.idx[i]); + } + return equal; + } + + /// Not equal + CUTLASS_HOST_DEVICE + bool operator!=(Coord const& b) const { return !(*this == b); } + + /// Clamps a coordinate to a range specified by maximum and minimum values + CUTLASS_HOST_DEVICE + Coord& clamp(Coord const& max, Coord const& min = Coord()) { + for (int i = 0; i < N; ++i) { + idx[i] = __NV_STD_MAX(__NV_STD_MIN(idx[i], max.idx[i]), min.idx[i]); + } + return *this; + } + + /// Returns the product of all elements + CUTLASS_HOST_DEVICE + int count() const { + int product = idx[0]; + for (int i = 1; i < N; ++i) { + product *= idx[i]; + } + return product; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Helper to make a 2-element coordinate +CUTLASS_HOST_DEVICE +Coord<1> make_Coord(int _0) { + int values[1] = {_0}; + return Coord<1>(values); +} + +/// Helper to make a 2-element coordinate +CUTLASS_HOST_DEVICE +Coord<2> make_Coord(int _0, int _1) { + int values[2] = {_0, _1}; + return Coord<2>(values); +} + +/// Helper to make a 3-element coordinate +CUTLASS_HOST_DEVICE +Coord<3> make_Coord(int _0, int _1, int _2) { + int values[3] = {_0, _1, _2}; + return Coord<3>(values); +} + +/// Helper to make a 4-element coordinate +CUTLASS_HOST_DEVICE +Coord<4> make_Coord(int _0, int _1, int _2, int _3) { + int values[4] = {_0, _1, _2, _3}; + return Coord<4>(values); +} + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Getter +CUTLASS_HOST_DEVICE +Coord<2> get_Coord_hw(Coord<3> const& coord) { return make_Coord(coord[1], coord[2]); } + +/// Getter +CUTLASS_HOST_DEVICE +Coord<2> get_Coord_hw(Coord<4> const& coord) { return make_Coord(coord[1], coord[2]); } + +/// Getter +CUTLASS_HOST_DEVICE +Coord<3> get_Coord_hwc(Coord<4> const& coord) { return make_Coord(coord[1], coord[2], coord[3]); } + +/// Getter +CUTLASS_HOST_DEVICE +Coord<3> get_Coord_dhw(Coord<4> const& coord) { return make_Coord(coord[0], coord[1], coord[2]); } + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/core_io.h b/cutlass-example/cutlass/core_io.h new file mode 100644 index 0000000..cceea4c --- /dev/null +++ b/cutlass-example/cutlass/core_io.h @@ -0,0 +1,44 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +#pragma once + +/*! \file + \brief Helpers for printing cutlass/core objects +*/ + +#pragma once + +#include +#include + +#include + +template +std::ostream& operator<<(std::ostream& out, cutlass::Coord const& coord) { + for (int i = 0; i < Rank; ++i) { + out << (i ? ", " : "") << coord.idx[i]; + } + return out; +} diff --git a/cutlass-example/cutlass/cutlass.h b/cutlass-example/cutlass/cutlass.h new file mode 100644 index 0000000..19600ec --- /dev/null +++ b/cutlass-example/cutlass/cutlass.h @@ -0,0 +1,73 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ + +/*! \file + \brief Basic include for CUTLASS macros +*/ + +#pragma once + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +#define CUTLASS_MAJOR 1 +#define CUTLASS_MINOR 0 +#define CUTLASS_PATCH 1 +#define CUTLASS_VERSION ((CUTLASS_MAJOR)*100 + (CUTLASS_MINOR)*10 + CUTLASS_PATCH) + +#ifdef __NVCC__ +#define CUTLASS_HOST_DEVICE __forceinline__ __device__ __host__ +#define CUTLASS_DEVICE __forceinline__ __device__ +#elif defined(__CUDACC_RTC__) +#define CUTLASS_HOST_DEVICE __forceinline__ __device__ +#define CUTLASS_DEVICE __forceinline__ __device__ +#else +#define CUTLASS_HOST_DEVICE +// CUTLASS_DEVICE is an error if not compiling device code +#endif + +// CUTLASS_PRAGMA_UNROLL inserts a CUTLASS_PRAGMA_UNROLL if supported by the compiler +#if defined(__CUDA_ARCH__) +#if defined(_MSC_VER) +#define CUTLASS_PRAGMA_UNROLL __pragma("unroll") +#define CUTLASS_PRAGMA_NO_UNROLL __pragma("unroll 1") +#else +#define CUTLASS_PRAGMA_UNROLL _Pragma("unroll") +#define CUTLASS_PRAGMA_NO_UNROLL _Pragma("unroll 1") +#endif +#else +#define CUTLASS_PRAGMA_UNROLL +#define CUTLASS_PRAGMA_NO_UNROLL +#endif + +#define CUTLASS_ASSERT(x) assert(x) + +namespace cutlass { + +/// NVIDIA GPU Warp size +static const int kWarpSize = 32; + +} // namespace cutlass + +//////////////////////////////////////////////////////////////////////////////////////////////////// diff --git a/cutlass-example/cutlass/fragment.h b/cutlass-example/cutlass/fragment.h new file mode 100644 index 0000000..886b114 --- /dev/null +++ b/cutlass-example/cutlass/fragment.h @@ -0,0 +1,278 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines Fragment, a statically-sized array for storing parts of matrices within a + thread's registers. +*/ +#pragma once + +#include +#include +#include +#include + +namespace cutlass { + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup fragment_concept Fragment Concept +@{ + +\ref fragment_concept is a statically sized array for storing parts of tiles held by individual CUDA +threads. + +@par \ref fragment_concept + Types satisfying \ref fragment_concept define the following members + - Element - type of each access held within the fragment + - kElements - number of elements stored by the fragment + - clear() - overwrites the fragment storage with zeros + - Element & operator[](int i) - by-reference access of the ith element + - Element const & operator[](int i) const - const by-reference access of the ith element +@} +*/ + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup fragment_iterator_concept Fragment Iterator Concept +@{ + +\ref fragment_iterator_concept provides structured access to the elements within a fragment with an +optional bitcast to the desired access type + +@par \ref fragment_iterator_concept + Types satisfying \ref fragment_iterator_concept define the following members + - AccessType& operator[](int i) - provides access to the ith element of the fragment + - AccessType& at(int d, int h, int w, int c) - applies \ref layout_concept to fragment and +provides access to element at (d, h, w, c) + +@} +*/ + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct StorageType { + typedef uint64_t Type; +}; +template <> +struct StorageType<4> { + typedef uint32_t Type; +}; +template <> +struct StorageType<2> { + typedef uint16_t Type; +}; +template <> +struct StorageType<1> { + typedef uint8_t Type; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief A template defining \ref fragment_concept +* @concept{fragment_concept} +*/ +template +struct Fragment : public AlignedStruct { + /// Make sure the alignment makes sense wrt the size of elements. + static_assert(kAlignment_ == 16 || kAlignment_ >= sizeof(Element_), "Alignment is too small"); + /// Alignment must be a power of two + static_assert(is_pow2::value, "Alignment must be a power of two"); + + /// This class. + typedef Fragment This_; + /// The element. + typedef Element_ Element; + /// The number of elements. + static int const kElements = kElements_; + + /// Clear a fragment. + CUTLASS_DEVICE void clear() { + // Avoid element-wise access for sub 32b element type + if (kAlignment_ >= 8 && (kElements * sizeof(Element)) % 8 == 0) { + uint64_t* ptr = reinterpret_cast(storage); + for (int i = 0; i < (kElements * sizeof(Element)) / 8; ++i) { + ptr[i] = uint64_t(0); + } + } else if (kAlignment_ >= 4 && (kElements * sizeof(Element)) % 4 == 0) { + uint32_t* ptr = reinterpret_cast(storage); + for (int i = 0; i < (kElements * sizeof(Element)) / 4; ++i) { + ptr[i] = uint32_t(0); + } + } else if (kAlignment_ >= 2 && (kElements * sizeof(Element)) % 2 == 0) { + uint16_t* ptr = reinterpret_cast(storage); + for (int i = 0; i < (kElements * sizeof(Element)) / 2; ++i) { + ptr[i] = uint16_t(0); + } + } else { + for (int i = 0; i < kElements; ++i) { + storage[i] = 0; + } + } + } + + /// The accessor. + CUTLASS_DEVICE Element& operator[](int i) { + assert(i < kElements_); + return reinterpret_cast(storage)[i]; + } + + /// The accessor. + CUTLASS_DEVICE Element const& operator[](int i) const { + assert(i < kElements_); + return reinterpret_cast(storage)[i]; + } + + private: + /// Storage type to use for Elements + typedef typename StorageType::Type StorageType; + + /// Number of elements in the storage + static int const kStorageCount = + (sizeof(Element_) * kElements_ + sizeof(StorageType) - 1) / sizeof(StorageType); + /// The storage. + StorageType storage[kStorageCount]; + + /// Ensure that there's enough storage for all elements + static_assert(sizeof(StorageType) <= kAlignment_, "StorageType is too big for given alignment"); +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief A template defining \ref fragment_iterator_concept +* @concept{fragment_iterator_concept} +*/ +template +struct FragmentIterator { + /// This class. + typedef FragmentIterator This_; + /// The fragment. + typedef Fragment_ Fragment; + /// The number of iterations. + typedef Iterations_ Iterations; + /// The access type. + typedef AccessType_ AccessType; + + /// The element. + typedef typename Fragment::Element Element; + /// The number of elements per access. + static int const kElementsPerAccess = (int)(sizeof(AccessType) / sizeof(Element)); + /// The shape of the the fragment. + typedef typename ShapeMul >::Shape FragmentShape; + /// The linear strides for iterations. + typedef typename ShapeStrides::Shape Strides; + + /// Ctor. + template + CUTLASS_DEVICE FragmentIterator(OtherFragment_& fragment, int offset = 0) + : pointer(reinterpret_cast(&fragment[offset])) { + static_assert(OtherFragment_::kElements >= Fragment::kElements, ""); + } + + /// The accessor. + CUTLASS_DEVICE AccessType const& at(int d, int h, int w, int c = 0) const { + int const imm = ComputeOffsetFromStrides::get(d, h, w, c); + return reinterpret_cast(pointer[imm]); + } + + /// The accessor. + CUTLASS_DEVICE AccessType& at(int d, int h, int w, int c = 0) { + int const imm = ComputeOffsetFromStrides::get(d, h, w, c); + return reinterpret_cast(pointer[imm]); + } + + /// The accessor. + CUTLASS_DEVICE AccessType const& operator[](int i) const { + return reinterpret_cast(pointer[i * kElementsPerAccess]); + } + + /// The accessor. + CUTLASS_DEVICE AccessType& operator[](int i) { + return reinterpret_cast(pointer[i * kElementsPerAccess]); + } + + /// Is the iterator valid? + CUTLASS_DEVICE bool valid(int d, int h, int w, int c) const { return true; } + + /// The pointer. + Element* pointer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct FragmentConstIterator { + /// This class. + typedef FragmentIterator This_; + /// The fragment. + typedef Fragment_ Fragment; + /// The number of iterations. + typedef Iterations_ Iterations; + /// The access type. + typedef AccessType_ AccessType; + + /// The element. + typedef typename Fragment::Element Element; + /// The number of elements per access. + static int const kElementsPerAccess = (int)(sizeof(AccessType) / sizeof(Element)); + /// The shape of the the fragment. + typedef typename ShapeMul >::Shape FragmentShape; + /// The linear strides for iterations. + typedef typename ShapeStrides::Shape IterationsStrides; + + /// Ctor. + template + CUTLASS_DEVICE FragmentConstIterator(OtherFragment_& fragment, int offset = 0) + : pointer(reinterpret_cast(&fragment[offset])) { + static_assert(OtherFragment_::kElements >= Fragment::kElements, ""); + } + /// Create from non-constant FragmentIterator + CUTLASS_DEVICE FragmentConstIterator( + FragmentIterator const& rhs_) + : pointer(reinterpret_cast(rhs_.offset)) {} + + /// The accessor. + CUTLASS_DEVICE AccessType const& at(int d, int h, int w, int c = 0) const { + int const imm = ComputeOffsetFromStrides::get(d, h, w, c); + return reinterpret_cast(pointer[imm]); + } + + /// The accessor. + CUTLASS_DEVICE AccessType const& operator[](int i) const { + return reinterpret_cast(pointer[i * kElementsPerAccess]); + } + + /// Is the iterator valid? + CUTLASS_DEVICE bool valid(int d, int h, int w, int c) const { return true; } + + /// The pointer. + Element const* pointer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/fragment_load_store.h b/cutlass-example/cutlass/fragment_load_store.h new file mode 100644 index 0000000..a7d272e --- /dev/null +++ b/cutlass-example/cutlass/fragment_load_store.h @@ -0,0 +1,135 @@ +/*************************************************************************************************** + * Copyright (c) 2017, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines accessors for loading and storing fragments to memory efficiently. +*/ +#pragma once + +#include +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct FragmentLoad {}; + +template +struct FragmentLoad { + /// The output type. + typedef FragmentElement_ AccessType; + + /// The load function. + static CUTLASS_DEVICE void load(AccessType& value, Scalar_ const* pointer, int offset) { + value.load(&pointer[offset], kStride); + } +}; + +template +struct FragmentLoad { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The load function. + static CUTLASS_DEVICE void load(AccessType& value, Scalar_ const* pointer, int offset) { + Load::load(value, pointer, offset); + } +}; + +template +struct FragmentStore {}; + +template +struct FragmentStore { + /// The input type. + typedef FragmentElement_ AccessType; + + /// The store function. + static CUTLASS_DEVICE void store(AccessType const& value, Scalar_* pointer, int offset) { + value.store(&pointer[offset], kStride); + } +}; + +template +struct FragmentStore { + /// The input type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void store(AccessType const& value, Scalar_* pointer, int offset) { + Store::store(value, pointer, offset); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} /// namespace cutlass diff --git a/cutlass-example/cutlass/fragment_multiply_add.h b/cutlass-example/cutlass/fragment_multiply_add.h new file mode 100644 index 0000000..36a4d6f --- /dev/null +++ b/cutlass-example/cutlass/fragment_multiply_add.h @@ -0,0 +1,149 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines multiply-add operations on fragments within a thread. +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct FragmentMultiplyAdd { + /// The shape of the instruction. + typedef Shape<1, 1, 1, 1> InstructionShape; + /// The type for A. + typedef Scalar_ ScalarA; + /// The type for B. + typedef Scalar_ ScalarB; + /// The type for C and D. + typedef Scalar_ ScalarC; + + /// Ctor. + CUTLASS_DEVICE FragmentMultiplyAdd() {} + + /// Multiply : d = a*b. + template + CUTLASS_DEVICE void multiply(Scalar_ a, FragmentB_ const& b, FragmentCd_& d) { + int const kReduction = FragmentB_::kElements / FragmentCd_::kElements; + for (int j = 0; j < FragmentCd_::kElements; ++j) { + d[j] = a * b[j * kReduction + 0]; + for (int k = 1; k < kReduction; ++k) { + d[j] += a * b[j * kReduction + k]; + } + } + } + + /// Multiply : d = a*b + c. + template + CUTLASS_DEVICE void multiply_add(Scalar_ a, + FragmentB_ const& b, + FragmentCd_ const& c, + FragmentCd_& d) { + int const kReduction = FragmentB_::kElements / FragmentCd_::kElements; + for (int j = 0; j < FragmentCd_::kElements; ++j) { + d[j] = a * b[j * kReduction + 0] + c[j]; + for (int k = 1; k < kReduction; ++k) { + d[j] += a * b[j * kReduction + k]; + } + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +#if !defined(__CUDACC_RTC__) || defined(CUTLASS_NVRTC_HAS_FP16) +template <> +struct FragmentMultiplyAdd { + /// The shape of the instruction. + typedef Shape<1, 1, 2, 1> InstructionShape; + /// The type for A. + typedef half ScalarA; + /// The type for B. + typedef half ScalarB; + /// The type for C and D. + typedef half ScalarC; + + /// Ctor. + CUTLASS_DEVICE FragmentMultiplyAdd() {} + + /// Multiply : d = a*b. + template + CUTLASS_DEVICE void multiply(half a, FragmentB_ const& b, FragmentCd_& d) { +#if defined(__CUDACC__) && __CUDA_ARCH__ >= 530 + + // Assemble a half2 from a. + __half2 const a_half2 = __half2half2(a); + // The input. + __half2 const* b_half2 = reinterpret_cast<__half2 const*>(&b[0]); + // The output. + __half2* d_half2 = reinterpret_cast<__half2*>(&d[0]); + + int const kReduction = FragmentB_::kElements / FragmentCd_::kElements; + for (int j = 0; j < FragmentCd_::kElements / 2; ++j) { + d_half2[j] = __hmul2(a_half2, b_half2[j * kReduction + 0]); + for (int k = 1; k < kReduction; ++k) { + d_half2[j] = __hfma2(a_half2, b_half2[j * kReduction + k], d_half2[j]); + } + } +#endif + } + + /// Multiply : d = a*b + c. + template + CUTLASS_DEVICE void multiply_add(half a, + FragmentB_ const& b, + FragmentCd_ const& c, + FragmentCd_& d) { +#if defined(__CUDACC__) && __CUDA_ARCH__ >= 530 + // Assemble a half2 from a. + __half2 const a_half2 = __half2half2(a); + // The inputs. + __half2 const* b_half2 = reinterpret_cast<__half2 const*>(&b[0]); + __half2 const* c_half2 = reinterpret_cast<__half2 const*>(&c[0]); + // The output. + __half2* d_half2 = reinterpret_cast<__half2*>(&d[0]); + + int const kReduction = (FragmentB_::kElements / FragmentCd_::kElements); + for (int j = 0; j < FragmentCd_::kElements / 2; ++j) { + d_half2[j] = __hfma2(a_half2, b_half2[j * kReduction + 0], c_half2[j]); + for (int k = 1; k < kReduction; ++k) { + d_half2[j] = __hfma2(a_half2, b_half2[j * kReduction + k], d_half2[j]); + } + } +#endif + } +}; + +#endif + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/clear_accumulators.h b/cutlass-example/cutlass/gemm/clear_accumulators.h new file mode 100644 index 0000000..441370f --- /dev/null +++ b/cutlass-example/cutlass/gemm/clear_accumulators.h @@ -0,0 +1,57 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines abstractions for efficiently clearing accumulator tiles. +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ClearAccumulators { + /// The shared storage. + struct SharedStorage {}; + + /// Ctor. + CUTLASS_DEVICE ClearAccumulators() {} + /// Ctor. + CUTLASS_DEVICE ClearAccumulators(SharedStorage& shared_storage) {} + + /// Clear the fragment. + template + CUTLASS_DEVICE void clear(Fragment_& fragment) { + fragment.clear(); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/dgemm_traits.h b/cutlass-example/cutlass/gemm/dgemm_traits.h new file mode 100644 index 0000000..0bbc221 --- /dev/null +++ b/cutlass-example/cutlass/gemm/dgemm_traits.h @@ -0,0 +1,127 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines structural traits of double-precision GEMM. +*/ +#pragma once + +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The tile size for the GEMM KxNxM. + typename OutputTile_, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_, + /// The number of scalars per LDG for A. + int kScalarsPerLdgA_ = 1, + /// The number of scalars per LDG for B. + int kScalarsPerLdgB_ = 1> +struct DgemmConfig + : public GemmConfig< + /// The scalar type for A. + double, + /// The scalar type for B. + double, + /// The scalar type for C. + double, + /// The scalar type for D. + double, + /// The tile size for the GEMM KxNxM. + OutputTile_, + /// The functor to do the math in the main loop. + ThreadMultiplyAdd, double, double, double>, + /// The number of scalars per LDG for A. + kScalarsPerLdgA_, + /// The number of scalars per STS for A. + kScalarsPerLdgA_, + /// The number of scalars per LDS for A. + 2, + /// The number of scalars per LDG for B. + kScalarsPerLdgB_, + /// The number of scalars per STS for B. + kScalarsPerLdgB_, + /// The number of scalars per LDS for B. + 2, + /// The number of scalars per LDG for C and STG for D. + 1, + /// The number of scalars per STS for D. + 2, + /// The number of scalars per LDS for D. + 1, + /// The number of stages in shared memory. + 2> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The output tile. + typename OutputTile_ = Shape<8, 64, 128>, + /// The functor to use in the epilogue. + typename EpilogueFunctor_ = LinearScaling, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_ = Shape<8, 8, 8>, + /// The number of doubles loaded in one LDG for A. + int kScalarsPerLdgA_ = 1, + /// The number of doubles loaded in one LDG for B. + int kScalarsPerLdgB_ = 1, + /// The index. + typename Index_ = int, + /// The DGEMM config. + typename GemmConfig_ = + DgemmConfig, + /// The traits class for the epilogue. + typename GemmEpilogueTraits_ = + SimplifiedGemmEpilogueTraits > +struct DgemmTraits : public SimplifiedGemmTraits< + // The layout for A. + kLayoutA_, + // The layout for B. + kLayoutB_, + // The config. + GemmConfig_, + // The epilogue. + GemmEpilogue, + // The index. + Index_> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm.h b/cutlass-example/cutlass/gemm/gemm.h new file mode 100644 index 0000000..c50a3f0 --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm.h @@ -0,0 +1,344 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Implements a software-pipelined efficient GEMM. +*/ +#pragma once + +#if !defined(__CUDACC_RTC__) +#include +#endif + +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +__global__ /*__launch_bounds__(Gemm_::kThreads)*/ void gemm_kernel(typename Gemm_::Params params) { + // Declare shared memory. + __shared__ typename Gemm_::SharedStorage shared_storage; + + // Construct the GEMM object. + Gemm_ gemm(params, shared_storage); + // Run GEMM. + gemm.multiply_add(); +} + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmDesc { + /// The dimensions of the GEMM. + Index_ m, n, k; + /// The alpha/beta scaling values. + Scalar_ alpha, beta; + /// The source matrix A. + void const* d_a; + /// The stride for A. + Index_ lda; + /// The source matrix B. + void const* d_b; + /// The stride for B. + Index_ ldb; + /// The source matrix C. + void const* d_c; + /// The stride for C. + Index_ ldc; + /// The destination matrix D. + void* d_d; + /// The stride for D. + Index_ ldd; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Gemm { + /// This class. + typedef Gemm This_; + /// The traits. + typedef GemmTraits_ Traits; + /// The shared storage. + typedef typename Traits::SharedStorage SharedStorage; + + /// The scalar for A. + typedef typename Traits::ScalarA ScalarA; + /// The scalar for B. + typedef typename Traits::ScalarB ScalarB; + /// The scalar in the epilogue. + typedef typename Traits::Epilogue::Scalar ScalarEpilogue; + /// The scalar for C. + typedef typename Traits::Epilogue::ScalarC ScalarC; + /// The scalar for D. + typedef typename Traits::Epilogue::ScalarD ScalarD; + /// The index. + typedef typename Traits::Index Index; + + /// The number of threads. + static int const kThreads = Traits::GemmConfig::kThreads; + + /// The params. + struct Params : public Traits::Params { + CUTLASS_HOST_DEVICE int initialize(Index m, + Index n, + Index k, + ScalarEpilogue alpha, + ScalarA const* d_a, + Index lda, + ScalarB const* d_b, + Index ldb, + ScalarEpilogue beta, + ScalarC const* d_c, + Index ldc, + ScalarD* d_d, + Index ldd) { + GemmDesc desc; + desc.m = m; + desc.n = n; + desc.k = k; + desc.alpha = alpha; + desc.beta = beta; + desc.d_a = reinterpret_cast(d_a); + desc.lda = lda; + desc.d_b = reinterpret_cast(d_b); + desc.ldb = ldb; + desc.d_c = reinterpret_cast(d_c); + desc.ldc = ldc; + desc.d_d = reinterpret_cast(d_d); + desc.ldd = ldd; + return Traits::Params::initialize(desc); + } + }; + +#if !defined(__CUDACC_RTC__) + /// Launch the kernel. + static __host__ cudaError_t launch(Params const& params, + cudaStream_t stream = cudaStreamDefault) { + // Setup the grid. + dim3 grid; + grid.x = (params.m + Traits::OutputTile::kW - 1) / Traits::OutputTile::kW; + grid.y = (params.n + Traits::OutputTile::kH - 1) / Traits::OutputTile::kH; + + // The number of threads. + dim3 block; + block.x = kThreads; + + // Launch the kernel. + void const* params_ = reinterpret_cast(¶ms); + + return cudaLaunchKernel(reinterpret_cast(&gemm_kernel), + grid, + block, + const_cast(¶ms_), + 0, + stream); + } + + /// Launch the kernel. + static __host__ cudaError_t launch(CUfunction kernel, + Params const& params, + CUstream stream = CU_STREAM_LEGACY) { + // Setup the grid. + dim3 grid; + grid.x = (params.m + Traits::OutputTile::kW - 1) / Traits::OutputTile::kW; + grid.y = (params.n + Traits::OutputTile::kH - 1) / Traits::OutputTile::kH; + + // The number of threads. + dim3 block; + block.x = kThreads; + + // Launch the kernel. + void* params_[] = {const_cast(reinterpret_cast(¶ms))}; + + // return cudaLaunchKernel(reinterpret_cast(&gemm_kernel), grid, block, + // const_cast(¶ms_), 0, stream); + CUresult result = cuLaunchKernel( + kernel, grid.x, grid.y, grid.z, block.x, block.y, block.z, 0, stream, params_, 0); + + if (result != CUDA_SUCCESS) { + return cudaErrorLaunchFailure; + } + return cudaSuccess; + } + +#endif + + /// Ctor. + CUTLASS_DEVICE Gemm(Params const& params_, SharedStorage& shared_storage_) + : params(params_), shared_storage(shared_storage_) {} + + /// Consume a single iteration of the loop. + template + CUTLASS_DEVICE void consume_tile(typename Traits::GlobalLoadStream& global_stream, + typename Traits::SharedLoadStream& shared_load_stream, + typename Traits::MultiplyAdd::Accumulators& accumulators, + Index outer_k) { + // If that's the last "load iteration" update the predicates. + if (!kIsLastIteration) { + global_stream.move_to_residue(outer_k); + } + + // Load data for the next iteration of the main loop. + if (!kIsLastIteration) { + global_stream.copy(); + } + + // The unrolling steps for the main loop. + int const kUnrollingSteps = + Traits::MultiplyAdd::AccumulatorsPerWarp::kD / Traits::MultiplyAdd::InstructionShape::kD; + + CUTLASS_PRAGMA_UNROLL + for (int step = 0; step < kUnrollingSteps - 1; ++step) { + // Trigger the copy from shared memory for the next A/B values. + shared_load_stream.copy(step + 1); + // Make sure the values are available for the current iteration to do the multiply-add. + shared_load_stream.commit(step); + + // Do the math on the fragments of the current iteration. + typename Traits::MultiplyAdd multiply_add; + multiply_add.multiply_add(shared_load_stream.fragment_a(step), + shared_load_stream.fragment_b(step), + accumulators, + accumulators); + } + + // Make sure the data from shared memory has been entirely consumed. + Traits::shared_load_fence(true); + + // Commit the data in shared memory for A/B. + if (!kIsLastIteration) { + global_stream.commit(); + } + + // Make sure the data is in shared memory. + Traits::shared_store_fence(true); + + // Trigger the loads for the next iteration (if needed). + if (!kIsLastIteration) { + // Move to the next stage for the load (if it makes sense). + shared_load_stream.inc_stage(); + // Trigger the copy from shared memory for the next loop iteration. + shared_load_stream.copy(0); + } + + // Make sure the values are available for the current iteration to do the multiply-add. + shared_load_stream.commit(kUnrollingSteps - 1); + + // Do the math on the fragments of the current iteration. + typename Traits::MultiplyAdd multiply_add; + multiply_add.multiply_add(shared_load_stream.fragment_a(kUnrollingSteps - 1), + shared_load_stream.fragment_b(kUnrollingSteps - 1), + accumulators, + accumulators); + } + + /// Do the GEMM. + CUTLASS_DEVICE void multiply_add() { + // Swizzle the IDs of the block (to enable better cache behavior). + typename Traits::BlockSwizzle block_swizzle; + dim3 block = block_swizzle.swizzle(); + + // Scale the id. + block.x *= Traits::OutputTile::kW; + block.y *= Traits::OutputTile::kH; + + // We may want to use shared memory to clear the registers. + typedef typename Traits::ClearAccumulators ClearAccumulators; + + // The streams to read A/B from global memory to shared memory. + typename Traits::GlobalLoadStream global_stream(params, shared_storage, block); + + // Create the accumulator clear. + ClearAccumulators clear(shared_storage.main_loop.clear); + + // By how much we unroll the main loop. + Index const kUnroll = static_cast(Traits::OutputTile::kD); + + // If we do not have enough steps in the main loop, trigger the residue code. + global_stream.move_to_residue(params.k); + + // Fetch the fragments for A and B from global memory. + global_stream.copy(); + + // Copy the elements to shared memory (after transformation if needed). + global_stream.commit(); + + // Make sure the data is in shared memory. + Traits::shared_store_fence(false); + + // Rollback to the beginning of the GEMM-K dimension. It may have no impact. + global_stream.rollback(); + + // The unrolling steps for the main loop. + int const kUnrollingSteps = + Traits::MultiplyAdd::AccumulatorsPerWarp::kD / Traits::MultiplyAdd::InstructionShape::kD; + + // Make sure we have at least 2 unrolling steps or our pipeling is not going to work. + static_assert(kUnrollingSteps >= 2, "The pipelining assumes at least two steps"); + + // The stream of data from shared memory to fragments. + typename Traits::SharedLoadStream shared_load_stream(params, shared_storage); + + // Trigger the copy from shared memory for the 1st stream. + shared_load_stream.copy(0); + + // Allocate the accumulators. + typename Traits::MultiplyAdd::Accumulators accumulators; + // Clear the accumulators. + clear.clear(accumulators); + + // The loop index. + Index outer_k = params.k - kUnroll; + + // Enter the main loop and iterate. + for (; outer_k > 0; outer_k -= kUnroll) { + consume_tile(global_stream, shared_load_stream, accumulators, outer_k); + } + + // Residual loop. + for (; outer_k > -kUnroll; outer_k -= kUnroll) { + consume_tile(global_stream, shared_load_stream, accumulators, outer_k); + } + + // Epilogue. + typedef typename Traits::Epilogue Epilogue; + Epilogue epilogue(params.epilogue, shared_storage.epilogue, params.m, params.n); + epilogue.epilogue(cutlass::make_Coord(0, block.y, block.x), accumulators); + } + + /// The params. + Params const& params; + /// The shared storage. + SharedStorage& shared_storage; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_epilogue.h b/cutlass-example/cutlass/gemm/gemm_epilogue.h new file mode 100644 index 0000000..bc25307 --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_epilogue.h @@ -0,0 +1,231 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Implements the epilogue phase of the GEMM kernel that efficiently updates global memory + with + the computed matrix product. +*/ +#pragma once + +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +CUTLASS_DEVICE bool is_zero(T x) { + return x == T(0); +} + +#if !defined(__CUDACC_RTC__) || defined(CUTLASS_NVRTC_HAS_FP16) +CUTLASS_DEVICE bool is_zero(half x) { return reinterpret_cast(x) == int16_t(0); } +#endif + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmEpilogue { + /// The traits class. + typedef GemmEpilogueTraits_ Traits; + /// The params. + typedef typename Traits::Params Params; + /// The shared storage. + typedef typename Traits::SharedStorage SharedStorage; + + /// The output tile. + typedef typename Traits::OutputTile OutputTile; + /// The number of iterations. + typedef typename Traits::Iterations Iterations; + /// The accumulators. + typedef typename Traits::Accumulators Accumulators; + /// The scalar. + typedef typename Traits::Scalar Scalar; + /// The functor in charge of the math. + typedef typename Traits::Functor Functor; + + /// We do not support 3D or 4D shapes. + static_assert(Iterations::kD == 1 && Iterations::kC == 1, "Unsupported 3D/4D shapes"); + + /// The iterator for C in global memory. + typedef typename Traits::GlobalLoadIteratorC GlobalLoadIteratorC; + /// The transformer for C. + typedef typename Traits::GlobalTransformerC GlobalTransformerC; + /// The transformer for D. + typedef typename Traits::GlobalTransformerD GlobalTransformerD; + /// The iterator for D in global memory. + typedef typename Traits::GlobalStoreIteratorD GlobalStoreIteratorD; + /// The iterator to store D in shared memory. + typedef typename Traits::SharedStoreIteratorD SharedStoreIteratorD; + /// The shared store transformer for D. + typedef typename Traits::SharedStoreTransformerD SharedStoreTransformerD; + /// The iterator to load D in shared memory. + typedef typename Traits::SharedLoadIteratorD SharedLoadIteratorD; + /// The shared load transformer for D. + typedef Copy SharedLoadTransformerD; + + /// The index. + typedef typename Traits::Index Index; + + /// The scalar for C. + typedef typename GlobalLoadIteratorC::Scalar ScalarC; + /// The scalar for D. + typedef typename GlobalStoreIteratorD::Scalar ScalarD; + + /// Ctor. + CUTLASS_DEVICE GemmEpilogue(Params const& params_, + SharedStorage& shared_storage_, + Index m_, + Index n_) + : params(params_), shared_storage(shared_storage_), m(m_), n(n_) {} + + /// Execute the epilogue. + CUTLASS_DEVICE void epilogue(Coord<3> const& block, Accumulators& accumulators) { + if (is_zero(params.functor.beta)) { + epilogue_with_or_without_beta(block, accumulators); + } else { + epilogue_with_or_without_beta(block, accumulators); + } + } + + template + CUTLASS_DEVICE void epilogue_with_or_without_beta(Coord<3> const& block, + Accumulators& accumulators) { + + // The problem size. + Coord<3> const bounds = cutlass::make_Coord(0, n, m); + + // The functor. + Functor functor(params.functor); + // The C fragment. + typename GlobalLoadIteratorC::Fragment fragment_c; + // The transformed C fragment. + typename GlobalTransformerC::OutputFragment transformed_c; + + CUTLASS_PRAGMA_UNROLL + for (int h = 0; h < Iterations::kH; ++h) { + // Compute pointer and predicate offsets for C and D global iterators. + int const pointer_offset = + ((params.iterator_d.inc_h * (GlobalStoreIteratorD::Iterations::kH - 1) + + params.iterator_d.inc_advance) * + Iterations::kW + + params.stride_h) * + h; + int const predicate_offset = + ((params.iterator_d.predicate_inc_h * (GlobalStoreIteratorD::Iterations::kH - 1) + + params.iterator_d.predicate_inc_advance) * + Iterations::kW + + Traits::Delta::kH) * + h; + + // The iterator to load the elements of the C matrix. + GlobalLoadIteratorC global_load_iterator( + params.iterator_c, bounds, block, pointer_offset, predicate_offset); + // The transformer for C. + GlobalTransformerC transformer_c; + // The transformer for D. + GlobalTransformerD transformer_d; + // The iterator to store into the D matrix. + GlobalStoreIteratorD global_store_iterator( + params.iterator_d, bounds, block, pointer_offset, predicate_offset); + + // The transformer to transform before storing to shared memory. + SharedStoreTransformerD shared_store_transformer; + typename SharedStoreTransformerD::OutputFragment shared_store_transformed_d; + + // The iterator to store to shared memory. + SharedStoreIteratorD shared_store_iterator(params.shared_store_iterator_d, + shared_storage.shared_stream.store); + + // The iterator to load from shared memory. TODO: Use a stream. + SharedLoadIteratorD shared_load_iterator(params.shared_load_iterator_d, + shared_storage.shared_stream.load); + + CUTLASS_PRAGMA_UNROLL + for (int w = 0; w < Iterations::kW; ++w) { + // Load the C matrix into fragment. + if (!kBetaIsZero_) { + iterator_load(global_load_iterator, fragment_c); + } + + // Make sure we can write to shared memory. + shared_load_fence(); + + // Copy the accumulators to shared memory. + int const offset = (h * Iterations::kW + w) * SharedStoreIteratorD::Fragment::kElements; + + shared_store_transformer.transform(accumulators, offset, shared_store_transformed_d); + shared_iterator_store(shared_store_iterator, shared_store_transformed_d); + + // Make sure the data is in shared memory. + shared_store_fence(); + + // Copy the accumulators back to registers from shared memory. + typename SharedLoadIteratorD::Fragment fetched_d; + shared_iterator_load(shared_load_iterator, fetched_d); + + // Do the math. + typename GlobalTransformerD::InputFragment fragment_d; + + if (kBetaIsZero_) { + functor.evaluate(fetched_d, fragment_d); + } else { + // Transform C fragment. + transformer_c.transform(fragment_c, transformed_c); + // Do the math. + functor.evaluate(fetched_d, transformed_c, fragment_d); + } + + // Transform D fragment. + typename GlobalTransformerD::OutputFragment transformed_d; + transformer_d.transform(fragment_d, transformed_d); + + // Copy the results to global memory. + iterator_store(global_store_iterator, transformed_d); + } + } + } + + /// The memory fence for shared loads. + CUTLASS_DEVICE void shared_load_fence() { __syncthreads(); } + + /// The memory fence for shared stores. + CUTLASS_DEVICE void shared_store_fence() { __syncthreads(); } + + /// The params. + Params const& params; + /// The shared storage. + SharedStorage& shared_storage; + /// The dimensions of the GEMM. + Index m, n; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_epilogue_traits.h b/cutlass-example/cutlass/gemm/gemm_epilogue_traits.h new file mode 100644 index 0000000..c06fc25 --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_epilogue_traits.h @@ -0,0 +1,331 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines structural properties of the GEMM epilogue. +*/ +#pragma once + +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The output tile. + typename OutputTile_, + /// The accumulators. + typename Accumulators_, + /// The iterator to load C from global memory. + typename GlobalLoadIteratorC_, + /// The transformer for C. + typename GlobalTransformerC_, + /// The transformer for D. + typename GlobalTransformerD_, + /// The iterator to store D to global memory. + typename GlobalStoreIteratorD_, + /// The iterator to store D to shared memory. + typename SharedStoreIteratorD_, + /// The shared store transformer for D. + typename SharedStoreTransformerD_, + /// The iterator to load D from shared memory. + typename SharedLoadIteratorD_, + /// The number of iterations in the epilogue. + typename Iterations_, + /// The iterations strides. + typename Delta_, + /// The functor to be used in the epilogue. + typename Functor_, + /// The index. + typename Index_ = int> +struct GemmEpilogueTraits { + // + /// The output tile. + typedef OutputTile_ OutputTile; + /// The number of iterations. + /// The accumulators. + typedef Accumulators_ Accumulators; + /// The iterator for C in global memory. + typedef GlobalLoadIteratorC_ GlobalLoadIteratorC; + /// The transformer for C. + typedef GlobalTransformerC_ GlobalTransformerC; + /// The transformer for D. + typedef GlobalTransformerD_ GlobalTransformerD; + /// The iterator for D in global memory. + typedef GlobalStoreIteratorD_ GlobalStoreIteratorD; + /// The iterator to store D in shared memory. + typedef SharedStoreIteratorD_ SharedStoreIteratorD; + /// The shared store transformer for D. + typedef SharedStoreTransformerD_ SharedStoreTransformerD; + /// The iterator to store D in shared memory. + typedef SharedLoadIteratorD_ SharedLoadIteratorD; + /// typedef typename GemmConfig::EpilogueIterations Iterations; + typedef Iterations_ Iterations; + /// The iterations strides. + typedef Delta_ Delta; + + /// The functor in charge of the math. + typedef Functor_ Functor; + /// The index. + typedef Index_ Index; + + /// We do not support 3D or 4D shapes. + static_assert(Iterations::kD == 1 && Iterations::kC == 1, "Unsupported 3D/4D shapes"); + + /// The scalar. + typedef typename Functor::Scalar Scalar; + /// The scalar for C. + typedef typename GlobalLoadIteratorC::Scalar ScalarC; + /// The scalar for D. + typedef typename GlobalStoreIteratorD::Scalar ScalarD; + + /// The params. + struct Params { + /// The strides for H and W in the different iterations of the epilogue. + Index stride_h, stride_w; + /// The params for the C iterator. + typename GlobalLoadIteratorC::Params iterator_c; + /// The params for the D global iterator. + typename GlobalStoreIteratorD::Params iterator_d; + /// The params for the D shared store iterator. + typename SharedStoreIteratorD::Params shared_store_iterator_d; + /// The params for the D shared load iterator. + typename SharedLoadIteratorD::Params shared_load_iterator_d; + /// The functor params. + typename Functor::Params functor; + + /// Setup the params. + template + CUTLASS_HOST_DEVICE int initialize(GemmDesc_ const& desc) { + // The parameters for the functor. + int error_code = functor.initialize(desc); + if (error_code) { + return error_code; + } + + // At the end of the H iteration, we jump over a number of columns. + this->stride_h = desc.ldd * Delta::kH; + // Nothing to do here. + this->stride_w = 0; + + // Setup the params for the global memory iterator for C. + error_code = iterator_c.initialize( + reinterpret_cast(desc.d_c), desc.ldc, desc.n, stride_w, Delta::kW); + if (error_code) { + return error_code; + } + + // Setup the params for the global memory iterator for D. + return iterator_d.initialize( + reinterpret_cast(desc.d_d), desc.ldd, desc.n, stride_w, Delta::kW); + } + }; + + /// The shared memory storage to exchange data. + union StreamSharedStorage { + // The storage for the store iterator. + typename SharedStoreIteratorD::SharedStorage store; + // The storage for the store iterator. + typename SharedLoadIteratorD::SharedStorage load; + }; + + /// The shared memory to swizzle the data in the epilogue. + struct SharedStorage { + // The storage for the shared stream D. + StreamSharedStorage shared_stream; + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmEpilogueTraitsHelper { + /// The scalar. + typedef typename EpilogueFunctor_::Scalar Scalar; + /// The output tile. + typedef typename GemmConfig_::OutputTile OutputTile; + + /// The number of iterations in the epilogue. + typedef Shape<1, + GemmConfig_::MultiplyAdd::AccumulatorsPerThread::kH / + GemmConfig_::kAccumulatorsPerLdsB, + GemmConfig_::kAccumulatorsPerLdsB> + Iterations; + // The iteration strides in the H/W dimension. + typedef Shape<0, + GemmConfig_::kAccumulatorsPerLdsB*( + GemmConfig_::Warps::kH* GemmConfig_::MultiplyAdd::ThreadsPerWarp::kH - 1), + 0> + Delta; + /// The functor to do the math in the epilogue. + typedef EpilogueFunctor_ Functor; + + /// The traits class to build the iterator to store to shared memory for D. + typedef GemmSharedStoreTileDTraits< + // The pointer is float. + typename Functor::Scalar, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The number of scalars per STS. + GemmConfig_::kScalarsPerStsD, + // The skew -- 128 / sizeof(ScalarD) / kScalarsPerStsD is the number of threads involved in + // a single STS. We divide by 2 as our objective is to add a skew to the odd threads to + // avoid bank conflicts between odd and even threads. + 128 / sizeof(typename GemmConfig_::ScalarD) / GemmConfig_::kScalarsPerStsD / 2 * + GemmConfig_::kScalarsPerStsD> + SharedStoreTileTraits; + + /// The iterator to store D to shared memory. + typedef TileStoreIterator + SharedStoreIteratorD; + + /// The shared store transformer for D. + typedef Copy SharedStoreTransformerD; + + /// The traits class to build the iterator to load from shared memory for D. + typedef GemmSharedLoadTileDTraits< + // The pointer is float. + typename Functor::Scalar, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The number of columns of the output tile written by iteration. + GemmConfig_::OutputTile::kH / ShapeCount::kCount, + // The number of scalars per LDS. + GemmConfig_::kScalarsPerLdsD, + // The skew. + SharedStoreTileTraits::kSkew> + SharedLoadTileTraits; + + /// The iterator to load D from shared memory. + typedef TileLoadIterator + SharedLoadIteratorD; + + /// The traits class to build the iterator to load data from global memory for C^N. + typedef GemmGlobalTileCdTraits< + // The pointer is float const. + typename GemmConfig_::ScalarC const, + // The tile has size (N / Iterations)xM in GEMM's terminology. + Shape<1, + GemmConfig_::OutputTile::kH / ShapeCount::kCount, + GemmConfig_::OutputTile::kW>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // How many elements do we jump over at each iteration? + Iterations::kW, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgC> + GlobalLoadTileTraits; + + /// The iterator to load C. + typedef GemmGlobalIteratorCd GlobalLoadIteratorC; + /// The transformer for C. + typedef Copy GlobalTransformerC; + + /// The traits class to build the iterator to store data to global memory for D^N. + typedef GemmGlobalTileCdTraits< + // The pointer is float. + typename GemmConfig_::ScalarD, + // The tile has size (N / Iterations)xM in GEMM's terminology. + Shape<1, + GemmConfig_::OutputTile::kH / ShapeCount::kCount, + GemmConfig_::OutputTile::kW>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // How many elements do we jump over at each iteration? + Iterations::kW, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerStgD> + GlobalStoreTileTraits; + + /// The iterator to store D. + typedef GemmGlobalIteratorCd GlobalStoreIteratorD; + /// The transformer for D. + typedef Copy GlobalTransformerD; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The GEMM config. + typename GemmConfig_, + /// The epilogue functor to do the math in the epilogue. + typename EpilogueFunctor_, + /// The index. + typename Index_ = int, + /// The helper to create the traits class. + typename Helper_ = GemmEpilogueTraitsHelper > +struct SimplifiedGemmEpilogueTraits : public GemmEpilogueTraits< + // The output tile. + typename GemmConfig_::OutputTile, + // The accumulators. + typename GemmConfig_::Accumulators, + // The global iterator for C. + typename Helper_::GlobalLoadIteratorC, + // The transformer for C. + typename Helper_::GlobalTransformerC, + // The transformer for D. + typename Helper_::GlobalTransformerD, + // The global iterator for D. + typename Helper_::GlobalStoreIteratorD, + // The iterator to store D to shared memory. + typename Helper_::SharedStoreIteratorD, + // The shared store transformer for D. + typename Helper_::SharedStoreTransformerD, + // The iterator to load D from shared memory. + typename Helper_::SharedLoadIteratorD, + // The number of iterations. + typename Helper_::Iterations, + // The strides between iterations. + typename Helper_::Delta, + // The functor to be used in the epilogue. + EpilogueFunctor_, + // The index. + Index_> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_global_stream.h b/cutlass-example/cutlass/gemm/gemm_global_stream.h new file mode 100644 index 0000000..ec675a3 --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_global_stream.h @@ -0,0 +1,182 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Implements efficient loading of the thread block-level tile from global memory and + storing + to shared memory. +*/ +#pragma once + +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The load iterator. + typename LoadIterator_, + /// The store iterator to copy to shared memory. + typename StoreIterator_, + /// The transformer to be applied after the data has been copied from global memory. + typename Transformer_> + +struct GlobalLoadStreamBase { + /// The load iterator. + typedef LoadIterator_ LoadIterator; + /// The transformer. + typedef Transformer_ Transformer; + /// The store iterator to write to shared memory. + typedef StoreIterator_ StoreIterator; + + /// The fragment that is copied from shared memory. + typedef typename LoadIterator::Fragment FetchedFragment; + /// The fragment that is obtained after the transformation by the transformer. + typedef typename Transformer::OutputFragment TransformedFragment; + /// Make sure the fragments match. + static_assert((platform::is_same::value), + ""); + /// The output fragment. + typedef TransformedFragment Fragment; + /// Make sure the transformed fragment is the same as the store fragment. + static_assert((platform::is_same::value), + ""); + + /// The layout. + static MatrixLayout::Kind const kLayout = LoadIterator::kLayout; + /// The scalar type of the iterator. + typedef typename LoadIterator::Scalar Scalar; + /// The pointer. + typedef typename LoadIterator::Pointer Pointer; + /// The index. + typedef typename LoadIterator::Index Index; + + /// The params. + struct Params { + // The load iterator. + typename LoadIterator::Params load_iterator; + // The store iterator. + typename StoreIterator::Params store_iterator; + + /// Setup the params. + template + CUTLASS_HOST_DEVICE int initialize(GemmDesc_ const& desc, Pointer pointer, Index ld) { + int error_code = load_iterator.initialize(desc, pointer, ld); + if (error_code) { + return error_code; + } + + return store_iterator.initialize(); + } + }; + + /// The amount of storage in shared memory needed to store the tile. + typedef typename StoreIterator::SharedStorage SharedStoreStorage; + + /// The storage in shared memory needed by that stream. + union SharedStorage { + // The load iterator. + typename LoadIterator::SharedStorage load_iterator; + // The store iterator. + SharedStoreStorage store_iterator; + }; + + /// Ctor. + CUTLASS_DEVICE GlobalLoadStreamBase(Params const& params, + SharedStorage& shared_storage, + Coord<3> const bounds, + Coord<3> const& block) + : load_iterator(params.load_iterator, bounds, block), + transformer(), + store_iterator(params.store_iterator, shared_storage.store_iterator) + + { + fetched_fragment.clear(); + } + + /// Load the data from shared memory to the fetch fragment. + CUTLASS_DEVICE void copy() { iterator_load(load_iterator, fetched_fragment); } + + /// Commit the data. + CUTLASS_DEVICE void commit() { + transformer.transform(fetched_fragment, transformed_fragment); + iterator_store(store_iterator, transformed_fragment); + store_iterator.inc_stage(); + } + + /// Move to the beginning of the residue code. That's a new code path in CUTLASS 1.0.1. + CUTLASS_DEVICE void move_to_residue(Index k) { load_iterator.move_to_residue(k); } + + /// Execute the residue code. + CUTLASS_DEVICE void residue(Index k, bool skip_clear = false) { + load_iterator.residue(k); + if (!skip_clear) { + fetched_fragment.clear(); + } + } + + /// Rollback to the beginning of the GEMM-k dimension. + CUTLASS_DEVICE void rollback() { load_iterator.rollback(); } + + /// The iterator. + LoadIterator load_iterator; + /// The fragment to fetch from shared memory. + FetchedFragment fetched_fragment; + /// The transformer. + Transformer transformer; + /// The fragment to convert the data after it has been fetched from shared memory. + TransformedFragment transformed_fragment; + /// The store iterator. + StoreIterator store_iterator; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The load iterator. + typename LoadIterator_, + /// The store iterator to copy to shared memory. + typename StoreIterator_, + /// The transformer to be applied after the data has been copied from global memory. + typename Transformer_ = Copy > + +struct GlobalLoadStream : public GlobalLoadStreamBase { + /// The base class. + typedef GlobalLoadStreamBase Base; + + /// Ctor. + CUTLASS_DEVICE GlobalLoadStream(typename Base::Params const& params, + typename Base::SharedStorage& shared_storage, + Coord<3> const& bounds, + Coord<3> const& block) + : Base(params, shared_storage, bounds, block) {} +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_global_tile.h b/cutlass-example/cutlass/gemm/gemm_global_tile.h new file mode 100644 index 0000000..1cc3b33 --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_global_tile.h @@ -0,0 +1,541 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines iterators for efficiently loading and storing to global memory. +*/ +#pragma once + +#include +#include + +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +// The following functor reshapes a tile of threads to match a tile of data. The idea is that when +// the user wants to build the iterator traits, he/she may want to specify the tile independently +// from the number of scalars loaded/stored per instruction. For example, in the row-major version +// with a tile of size 128x8 - the user may want to that the iterator works with 32x8 threads if +// each thread loads 1 scalar per LDG. If the user changes to 4 scalars per LDG, then the tile of +// threads has to change. The code below detects that and correct the code automatically - it is +// a helper when the user does not specify the right configuration. + +template +struct ReshapeThreads { + typedef Threads_ Threads; +}; + +template +struct ReshapeThreads { + typedef Shape Threads; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmGlobalTileTraits { + /// Identity of the operand + static GemmOperand::Kind const kOperand = kOperand_; + /// The layout. + static MatrixLayout::Kind const kLayout = kLayout_; + /// The scalar. + typedef typename platform::remove_const::type Scalar; + /// The pointer. + typedef Scalar_* Pointer; + /// The number of scalars per LDG/STG. + static int const kAccessSize = kAccessSize_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kGlobal; + + /// The tile shape + typedef typename ReshapeTile::Tile Tile; + /// The threads shape + typedef typename ReshapeThreads::Threads Threads; + /// The relative offset between two elements in the H/W dimension in adjacent threads. + typedef Shape<1, 1, Tile::kC> ThreadsDelta; + + /// The strides in each dimension between different loads/stores. + typedef Shape<0, Threads::kH, Threads::kW * kAccessSize> Delta; + /// Strides for immediate offset computation + typedef Shape<0, 0, Threads::kW * ThreadsDelta::kW, kAccessSize> ImmediateOffsetStrides; + /// The number of iterations needed to load/store the tile. + typedef Shape<1, Tile::kH / Threads::kH, Tile::kW / Threads::kW, Tile::kC / kAccessSize> + Iterations; + + typedef GemmMultiplicandTraits MultiplicandTraits; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + int thread_offset_h = threadIdx.x / Threads::kW * ThreadsDelta::kH; + int thread_offset_w = threadIdx.x % Threads::kW * ThreadsDelta::kW; + + return make_Coord(0, thread_offset_h, thread_offset_w, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmGlobalTileCdTraits : public GemmGlobalTileTraits { + /// The base class. + typedef GemmGlobalTileTraits + Base; + + /// The stride in the H dimension. + static int const kStrideH = kStrideH_; + /// Override the strides in each dimension between different loads/stores. + typedef Shape<0, 0, Base::Delta::kW, Base::Delta::kC> Delta; + + typedef typename Base::Iterations Iterations; + + typedef typename Base::Threads Threads; + + typedef typename Base::ThreadsDelta ThreadsDelta; + + typedef typename Base::ImmediateOffsetStrides ImmediateOffsetStrides; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + int thread_offset_h = threadIdx.x / Threads::kW * kStrideH * Iterations::kH; + int thread_offset_w = threadIdx.x % Threads::kW * ThreadsDelta::kW; + + return make_Coord(0, thread_offset_h, thread_offset_w, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmGlobalIteratorAb + : public TileLoadIterator { + /// This class. + typedef GemmGlobalIteratorAb This_; /// The base class. + + typedef TileLoadIterator + Base; + /// The layout. + static MatrixLayout::Kind const kLayout = TileTraits_::kLayout; + /// Fragment type loaded by the iterator + typedef typename Base::Fragment Fragment; + /// The scalar. + typedef typename TileTraits_::Scalar Scalar; + /// The threads. + typedef typename TileTraits_::Threads Threads; + /// The index. + typedef Index_ Index; + /// The thread offset + typedef typename TileTraits_::ThreadOffset ThreadOffset; + /// Specifies in which dimension post-increment accesses advance. + static IteratorAdvance::Kind const kAdvance = Base::kAdvance; + + typedef cutlass::PredicateVector::kCount> PredicateVector; + + /// Iterator parameters type + typedef typename Base::Params BaseParams; + + struct Params : public BaseParams { + /// Initializes params to load a strip-mined tile, given pointer and stride_h. + template + CUTLASS_HOST_DEVICE int initialize(GemmDesc_ const& desc, Scalar const* ptr, Index stride_h) { + Index inc_d = 0; + Index inc_advance = 0; + // Move by some columns for each iteration in the H dimension. + Index inc_h = Base::Delta::kH * stride_h; + + // Move by some more columns in the number of iterations if the D dimension is > 1. + if (Base::Delta::kD > 0) { + inc_d = Base::Delta::kD * stride_h - (Base::Iterations::kH - 1) * inc_h; + } + + // Move to the beginning of the next iteration. + if (kAdvance == IteratorAdvance::kH && Base::Delta::kD > 0) { + inc_advance = inc_d; + } else if (kAdvance == IteratorAdvance::kH) { + inc_advance = inc_h; + } else if (Base::Delta::kD > 0) { + inc_advance = (Base::Iterations::kW + 0) * ShapeCount::kWc - + (Base::Iterations::kH - 1) * inc_h - + (Base::Iterations::kD - 1) * Base::Delta::kD * stride_h; + } else { + inc_advance = (Base::Iterations::kW + 0) * ShapeCount::kWc - + (Base::Iterations::kH - 1) * inc_h; + } + + // The dimensions of the tile. + int const kH = TileTraits_::Tile::kH; + int const kW = TileTraits_::Tile::kW * TileTraits_::kAccessSize; + + // Move to the residue. + Index const kBlock = kAdvance == IteratorAdvance::kH ? kH : kW; + // The jump in the gemm-k dimension. + Index const stride = kAdvance == IteratorAdvance::kH ? stride_h : 1; + + // Compute the offset to the residue and how to "come" back. + Index const kResidue = desc.k % kBlock; + if (kResidue > 0) { + move_to_residue_offset = (desc.k - kResidue) * stride; + } else { + move_to_residue_offset = (desc.k - kBlock) * stride; + } + + Base::Params::initialize(ptr, 0, stride_h, 1, inc_d, inc_h, 0, inc_advance); + return 0; + } + + // The extra offset to control moving to the residue. + Index move_to_residue_offset; + }; + + /// Ctor. + CUTLASS_DEVICE GemmGlobalIteratorAb(Params const& _params, + const Coord<3>& bounds, + const Coord<3>& block, + ThreadOffset thread_offset_func = ThreadOffset()) + : params(_params) { + thread_offset = thread_offset_func(); + // The column. + Index block_h = thread_offset[1]; + // The contiguous dimension. + Index block_w = thread_offset[2]; + + // Add the blocks indices. + if (kAdvance == IteratorAdvance::kH) { + block_h += block[1]; + block_w += block[2]; + + } else { + block_h += block[2]; + block_w += block[1]; + } + + // Setup the pointer. + params.pointer += (block_h * params.stride_h + block_w); + + // Initialize predicates + initialize_predicates(bounds, make_Coord(0, block_h, block_w)); + } + + /// The accessor. + CUTLASS_DEVICE void get(typename Base::AccessType& value, int d, int h, int w, int c) const { + int const imm = + ComputeOffsetFromStrides::get(0, 0, w, c); + Load::load(value, params.pointer, imm); + } + + /// Increment the pointer in the H dimension. + CUTLASS_DEVICE void inc_h() { params.pointer += params.inc_h; } + /// Increment the pointer in the D dimension. + CUTLASS_DEVICE void inc_d() { params.pointer += params.inc_d; } + /// Increment the pointer to move to the next iteration. + CUTLASS_DEVICE void inc_advance() { params.pointer += params.inc_advance; } + + /// Initialize the predicates. + CUTLASS_DEVICE void initialize_predicates(const Coord<3>& bounds, const Coord<3>& block) { + // Setup the masks to control loads. + predicates.fill(0); + + int bounds_h, bounds_w; + if (kAdvance == IteratorAdvance::kH) { + bounds_w = bounds[2] - block[2]; + bounds_h = bounds[1]; + + } else { + bounds_w = bounds[1]; + bounds_h = bounds[2] - block[1]; + } + + // Fill in the bits of the predicate vector. + for (int d = 0; d < Base::Iterations::kD; ++d) { + for (int h = 0; h < Base::Iterations::kH; ++h) { + for (int w = 0; w < Base::Iterations::kW; ++w) { + for (int c = 0; c < Base::Iterations::kC; ++c) { + bool flag = w * Base::Delta::kW < bounds_w; + if (kAdvance == IteratorAdvance::kH) { + flag = flag && (h * Base::Delta::kH + d * Base::Delta::kD) < bounds_h; + } else { + flag = flag && (h * Base::Delta::kH) < bounds_h; + } + int const bit = ComputeOffsetFromShape::get(d, h, w, c); + predicates.set(bit, flag); + } + } + } + } + } + + /// Move to residue portion. + CUTLASS_DEVICE void move_to_residue(Index k) { + // Store the pointer and the predicates. + stored_pointer = params.pointer; + stored_predicates = predicates; + + // Move the pointer to the residue. + params.pointer += params.move_to_residue_offset; + + // The dimensions of the tile. + int const kH = TileTraits_::Tile::kH; + int const kW = TileTraits_::Tile::kW * TileTraits_::kAccessSize; + + // The unrolling factor. + int const kUnroll = kAdvance == IteratorAdvance::kH ? kH : kW; + + // Clear the predicates for the residue. TODO: We can do something smarter. + int const kResidue = (int)(k % (Index)kUnroll); + if (kResidue > 0) { + residue(kResidue); + } + } + + /// That's the residue! Update the predicates. + CUTLASS_DEVICE void residue(Index k) { + // The coordinates of the thread. + Index block_h = thread_offset[1]; + // The contiguous dimension. + Index block_w = thread_offset[2]; + + // Update the predicate vector. + for (int d = 0; d < Base::Iterations::kD; ++d) { + for (int h = 0; h < Base::Iterations::kH; ++h) { + for (int w = 0; w < Base::Iterations::kW; ++w) { + for (int c = 0; c < Base::Iterations::kC; ++c) { + Index offset = 0; + if (kAdvance == IteratorAdvance::kH) { + offset += block_h + h * Base::Delta::kH + d * Base::Delta::kD; + } else { + offset += block_w + w * Base::Delta::kW; + } + + int const bit = ComputeOffsetFromShape::get(d, h, w, c); + if (offset >= k) { + predicates.set(bit, false); + } + } + } + } + } + } + + /// Rollback to beginning of first tile and initialize predicates. + CUTLASS_DEVICE void rollback() { + params.pointer = stored_pointer; + predicates = stored_predicates; + } + + /// Is the iterator valid? + CUTLASS_DEVICE bool valid(int d, int h, int w, int c) const { + int const bit = ComputeOffsetFromShape::get(d, h, w, c); + return predicates[bit]; + } + + /// Offset of an individual lane from the start of the tile + Coord<4> thread_offset; + /// The parameters + Params params; + /// The pointer. + typename Base::Scalar const* stored_pointer; + /// The predicates. + PredicateVector predicates, stored_predicates; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmGlobalIteratorCd : public TileIteratorBase { + /// This class. + typedef GemmGlobalIteratorCd This_; + /// The base class. + typedef TileIteratorBase + Base; + + /// The layout. + static MatrixLayout::Kind const kLayout = TileTraits_::kLayout; + + /// The scalar. + typedef typename TileTraits_::Scalar Scalar; + /// The pointer. + typedef typename TileTraits_::Pointer Pointer; + /// The threads. + typedef typename TileTraits_::Threads Threads; + /// The index. + typedef Index_ Index; + /// The thread offset + typedef typename TileTraits_::ThreadOffset ThreadOffset; + + /// The params. + struct Params { + /// The pointer. + Pointer pointer; + /// The stride in the H dimension to setup the thread in the block. + Index stride_h; + /// The strides to increment the pointer. + Index inc_advance, inc_h; + /// The strides to increment the predicate offset + Index predicate_inc_advance, predicate_inc_h; + /// The column offset to compute the predicate for the columns. + Index predicate_offset; + + /// Setup the params. + CUTLASS_HOST_DEVICE int initialize( + Pointer pointer, Index ld, Index bound, Index epilogue_stride_w, Index epilogue_delta_w) { + // The pointer. + this->pointer = pointer; + // Each column of the matrix. + stride_h = TileTraits_::ThreadsDelta::kH * ld; + // Each thread output 1 column per iteration. The stride between columns is given by the + // number of scalars that are loaded per LDS for B. + inc_h = ld * TileTraits_::kStrideH; + inc_advance = + (ld - ld * TileTraits_::kStrideH * (Base::Iterations::kH - 1)) + epilogue_stride_w; + + predicate_offset = bound; + predicate_inc_h = TileTraits_::kStrideH; + predicate_inc_advance = + -((TileTraits_::kStrideH * (Base::Iterations::kH - 1) - 1) + epilogue_delta_w); + + return 0; + } + }; + + Params params; + /// Offset of an individual lane from the start of the tile + Coord<4> thread_offset; + + /// Ctor. + CUTLASS_DEVICE GemmGlobalIteratorCd() {} + + /// Ctor. + CUTLASS_DEVICE GemmGlobalIteratorCd(Params const& params, + const Coord<3>& bounds, + const Coord<3>& block, + int offset = 0, + int pred_offset = 0, + ThreadOffset thread_offset_func = ThreadOffset()) + : params(params) { + thread_offset = thread_offset_func(); + // Each warp works on a different column of the tile. + int const h = thread_offset[1] + block[1]; + // Each lane writes a different element. + int const w = thread_offset[2] + block[2]; + // Setup the pointer. + this->params.pointer += ((h * params.stride_h + w) + offset); + + // Prepare the vector of predicates. + for (int i = 0; i < Base::Iterations::kW; ++i) { + predicates.set(i, w + i * Base::Delta::kW < bounds[2]); + } + this->params.predicate_offset -= (h + pred_offset); + } + + /// The accessor. + CUTLASS_DEVICE void get(typename Base::AccessType& value, int d, int h, int w, int c) const { + int const imm = + ComputeOffsetFromStrides::get(0, 0, w, c); + Load::load(value, params.pointer, imm); + } + + /// Increment the pointer in the C dimension. + CUTLASS_DEVICE void inc_c() {} + /// Increment the pointer in the W dimension. + CUTLASS_DEVICE void inc_w() {} + /// Increment the pointer in the H dimension. + CUTLASS_DEVICE void inc_h() { + params.pointer += params.inc_h; + params.predicate_offset -= params.predicate_inc_h; + } + /// Increment the pointer in the D dimension. + CUTLASS_DEVICE void inc_d() {} + /// Increment the pointer to move to the next iteration. + CUTLASS_DEVICE void inc_advance() { + params.pointer += params.inc_advance; + this->params.predicate_offset -= params.predicate_inc_advance; + } + + /// The accessor. + CUTLASS_DEVICE void set(typename Base::AccessType const& value, int d, int h, int w, int c) { + int const imm = + ComputeOffsetFromStrides::get(0, 0, w, c); + Store::store( + value, params.pointer, imm); + } + + /// Test the validity of the iterator. + CUTLASS_DEVICE bool valid(int d, int h, int w, int c) const { + return predicates.at(w) && params.predicate_offset > 0; + } + + /// The predicates for the row. + cutlass::PredicateVector predicates; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_operand.h b/cutlass-example/cutlass/gemm/gemm_operand.h new file mode 100644 index 0000000..737f993 --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_operand.h @@ -0,0 +1,141 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines constant expressions for mapping GEMM problem size and strides onto pitch-linear + memory. +*/ +#pragma once + +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Helper to describe attributes of GEMM matrix operands +template +struct GemmOperandTraitsAb { + static const bool Congruous = + (kOperand_ == GemmOperand::kA ^ kLayout_ == MatrixLayout::kRowMajor); +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GetExtent; + +template +struct GetExtent { + static const int kExtent = Tile_::kW; +}; + +template +struct GetExtent { + static const int kExtent = Tile_::kH; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Determines the shape of a multiplicand tile in terms of strided (H) and contiguous (W) +/// dimensions +template +struct GemmMultiplicandTraits { + // Only defined for A or B + static_assert(Usage == GemmOperand::kA || Usage == GemmOperand::kB, + "MultiplicandTileShape defined only for A or B operands."); + + /// Shape of GEMM thread block tile (K, N, M) + typedef ThreadBlockTile_ ThreadBlockTile; + + /// Identifies multiplicand + static GemmOperand::Kind const kUsage = Usage; + + /// Layout of tile + static MatrixLayout::Kind const kLayout = Layout; + + // True if K is the strided dimension + static bool const kKstrided = (kUsage == GemmOperand::kA ^ kLayout == MatrixLayout::kRowMajor); + + /// Map the ThreadBlockShape onto (kH, kW) dimensions for A and B operand + typedef typename platform::conditional< + kKstrided, + Shape<1, ThreadBlockTile::kD, GetExtent::kExtent>, + Shape<1, GetExtent::kExtent, ThreadBlockTile::kD> >::type Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Project's a coordinate (K, N, M) onto inner and outer dimensions defined for each +/// operand. +template +struct ProjectOperand; + +/// Project A operand - (0, K, M) +template +struct ProjectOperand { + CUTLASS_HOST_DEVICE + static Coord<3> project(Coord<3> const &coord) { + if (Kstrided) { + return make_Coord(0, coord[0], coord[2]); + } else { + return make_Coord(0, coord[2], coord[0]); + } + } +}; + +/// Project B operand - (0, K, N) +template +struct ProjectOperand { + CUTLASS_HOST_DEVICE + static Coord<3> project(Coord<3> const &coord) { + if (Kstrided) { + return make_Coord(0, coord[0], coord[1]); + } else { + return make_Coord(0, coord[1], coord[0]); + } + } +}; + +/// Project C operand - (0, N, M) +template <> +struct ProjectOperand { + CUTLASS_HOST_DEVICE + static Coord<3> project(Coord<3> const &coord) { return make_Coord(0, coord[1], coord[2]); } +}; + +/// Project D operand - (0, N, M) +template <> +struct ProjectOperand { + CUTLASS_HOST_DEVICE + static Coord<3> project(Coord<3> const &coord) { return make_Coord(0, coord[1], coord[2]); } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_shared_stream.h b/cutlass-example/cutlass/gemm/gemm_shared_stream.h new file mode 100644 index 0000000..c6ff7bd --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_shared_stream.h @@ -0,0 +1,113 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines abstractions for managing loading and storing fragments to shared memory in the + efficient GEMM pipeline. +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The load iterator. + typename Iterator_, + /// The transformer to be applied after the data has been copied from shared memory. + typename Transformer_ = Copy > + +struct SharedLoadStream { + /// The load iterator. + typedef Iterator_ Iterator; + /// The transformer. + typedef Transformer_ Transformer; + + /// The fragment that is copied from shared memory. + typedef typename Iterator::Fragment FetchedFragment; + /// The fragment that is obtained after the transformation by the transformer. + typedef typename Transformer::OutputFragment TransformedFragment; + /// Make sure the fragments match. + static_assert((platform::is_same::value), + ""); + /// The output fragment. + typedef TransformedFragment Fragment; + + /// The params. + struct Params { + /// The iterator params. + typename Iterator::Params iterator; + + /// Setup the params. + CUTLASS_HOST_DEVICE int initialize() { return iterator.initialize(); } + }; + + /// The storage in shared memory needed by that stream. + typedef typename Iterator::Storage SharedStorage; + + /// Ctor. + CUTLASS_DEVICE SharedLoadStream() {} + + /// Ctor. + CUTLASS_DEVICE SharedLoadStream(Params const ¶ms, SharedStorage &shared_storage) { + this->initialize(params, shared_storage); + } + + /// Initialize the stream. + CUTLASS_DEVICE void initialize(Params const ¶ms, SharedStorage &shared_storage) { + // The iterator. + iterator = Iterator(params.iterator, shared_storage); + // The transformer. + transformer = Transformer(); + } + + /// Load the data from shared memory to the fetch fragment. + CUTLASS_DEVICE void copy(FetchedFragment &fetched) { shared_iterator_load(iterator, fetched); } + + /// Load the data from shared memory to the fetch fragment. + CUTLASS_DEVICE void copy(int d, FetchedFragment &fetched) { + shared_iterator_load(iterator, fetched, d); + } + + /// Commit the data. + CUTLASS_DEVICE void commit(FetchedFragment &fetched, TransformedFragment &transformed) { + transformer.transform(fetched, transformed); + } + + /// Increment the stage. + CUTLASS_DEVICE void inc_stage() { iterator.inc_stage(); } + + /// The iterator. + Iterator iterator; + /// The transformer. + Transformer transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_shared_tile.h b/cutlass-example/cutlass/gemm/gemm_shared_tile.h new file mode 100644 index 0000000..7c61e02 --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_shared_tile.h @@ -0,0 +1,417 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines iterators for efficiently loading and storing tiles to and from shared memory. +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmSharedStoreTileAbTraits { + /// The scalar. + typedef typename platform::remove_const::type Scalar; + /// The pointer. + typedef Scalar_* Pointer; + /// The tile. + typedef typename ReshapeTile::Tile Tile; + /// The threads. + typedef Threads_ Threads; + /// The strides to compute the base position of the thread. + typedef Shape<0, ShapeCount::kWc, Tile::kC, kScalarsPerSts_> ThreadsStrides; + /// The skew. + static int const kSkew = 0; + /// The number of scalars per LDG/STG. + static int const kAccessSize = kScalarsPerSts_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + + /// The number of iterations needed to load/store the tile. + typedef Shape<1, + Tile::kH / Threads::kH, + Tile::kW / Threads::kW, + Tile::kC / Threads::kC / kAccessSize> + Iterations; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, Threads::kH * ShapeCount::kWc, Threads::kW * kAccessSize> Delta; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, Threads::kH * ShapeCount::kWc, Threads::kW * kAccessSize> + ImmediateOffsetStrides; + + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + int offset = ComputeThreadOffsetFromStrides::get(); + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmSharedStoreWithSkewTileAbTraits { + /// The scalar. + typedef typename platform::remove_const::type Scalar; + /// The pointer. + typedef Scalar_* Pointer; + /// The tile without skews. + typedef typename ReshapeTile::Tile TileWithoutSkew; + /// The tile. + typedef typename ReshapeTile, + kScalarsPerSts_>::Tile Tile; + /// The threads. + typedef Threads_ Threads; + /// The skew. + static int const kSkew = kSkew_; + /// The number of scalars per STS. + static int const kAccessSize = kScalarsPerSts_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + + /// The number of iterations needed to load/store the tile. + typedef Shape<1, TileWithoutSkew::kH / Threads::kW, TileWithoutSkew::kW / Threads::kH> Iterations; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, ShapeCount::kWc, Threads::kH * kAccessSize> Delta; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, ShapeCount::kWc, Threads::kH * kAccessSize> ImmediateOffsetStrides; + + struct ThreadOffset { + CUTLASS_HOST_DEVICE Coord<4> operator()() const { + int offset = ComputeThreadOffsetFromStrides::get(); + return make_Coord(0, 0, offset, 0); + } + }; + + protected: + /// The strides to compute the base position of the thread. + typedef Shape<0, kScalarsPerSts_, ShapeCount::kHwc / Threads::kW> ThreadsStrides; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmSharedLoadTileATraits { + static GemmOperand::Kind const kOperand = GemmOperand::kA; + /// The scalar. + typedef typename platform::remove_const::type Scalar; + /// The pointer. + typedef Scalar_* Pointer; + /// The tile without skew. + typedef Shape::kExtent * InstructionShape_::kD> + TileWithoutSkew_; + /// The tile with skew. + typedef Shape TileWithSkew; + /// The tile without skew after reshaping. + typedef typename ReshapeTile::Tile TileWithoutSkew; + /// The tile. + typedef typename ReshapeTile::Tile Tile; + /// The number of warps. + typedef Warps_ Warps; + /// The threads in a warp. + typedef ThreadsPerWarp_ ThreadsPerWarp; + /// The number of scalars per LDG/STG. + // static int const kScalarsPerLds = kScalarsPerLds_; + static int const kAccessSize = kScalarsPerLds_; + /// The skew. + static int const kSkew = kSkew_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + + /// The number of warps. + static int const kWarps = GetExtent::kExtent; + /// The number of threads in one dimension of the warp. + static int const kThreadsPerWarp = GetExtent::kExtent; + + /// The number of iterations needed to load/store the tile. + typedef Shape<1, 1, TileWithoutSkew::kW / kWarps / kThreadsPerWarp /* / kScalarsPerLds*/> + Iterations; + /// The strides in each dimension between different loads/stores. + typedef Shape + ImmediateOffsetStrides; + typedef Shape Delta; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE Coord<4> operator()() const { + // Extract the warp. + int const warp = threadIdx.x / kWarpSize; + // Extract the slice. + int const slice = warp / (Warps::kH * Warps::kW); + // Compute the row offset for each warp. + int const warp_row = warp % Warps::kW; + // Compute the row offset for each thread. + int const lane_row = (threadIdx.x & 0x0e) / 2; + // The offset. + int const offset = + slice * Tile::kW * Tile::kC + (warp_row * ThreadsPerWarp::kW + lane_row) * kAccessSize; + // Embed the offset in a 4D coordinate vector. + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmSharedLoadTileBTraits { + static GemmOperand::Kind const kOperand = GemmOperand::kB; + /// The scalar. + typedef typename platform::remove_const::type Scalar; + /// The pointer. + typedef Scalar_* Pointer; + /// The tile without skew. + typedef Shape::kExtent * InstructionShape_::kD> + TileWithoutSkew_; + /// The tile with skew. + typedef Shape TileWithSkew; + /// The tile without skew after reshaping. + typedef typename ReshapeTile::Tile TileWithoutSkew; + /// The tile. + typedef typename ReshapeTile::Tile Tile; + /// The number of warps. + typedef Warps_ Warps; + /// The threads in a warp. + typedef ThreadsPerWarp_ ThreadsPerWarp; + /// The number of scalars per LDG/STG. + static int const kAccessSize = kScalarsPerLds_; + /// The skew. + static int const kSkew = kSkew_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + + /// The number of warps. + static int const kWarps = GetExtent::kExtent; + /// The number of threads in one dimension of the warp. + static int const kThreadsPerWarp = GetExtent::kExtent; + + /// The number of iterations needed to load/store the tile. + typedef Shape<1, 1, TileWithoutSkew::kW / kWarps / kThreadsPerWarp /* / kAccessSize*/> Iterations; + /// The strides in each dimension between different loads/stores. + typedef Shape + ImmediateOffsetStrides; + typedef Shape Delta; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE Coord<4> operator()() const { + // Extract the warp. + int const warp = threadIdx.x / kWarpSize; + // Extract the slice. + int const slice = warp / (Warps::kH * Warps::kW); + // The warp in the slice. + int const warp_in_slice = warp % (Warps::kH * Warps::kW); + // Compute the row offset for each warp. + int const warp_col = warp_in_slice / Warps::kW; + // Compute the row offset for each thread. + int const lane_col = (threadIdx.x & 0x10) / 8 + (threadIdx.x & 0x01); + // The offset. + int const offset = + slice * Tile::kW * Tile::kC + (warp_col * ThreadsPerWarp::kH + lane_col) * kAccessSize; + // Embed the offset in a 4D coordinate. + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmSharedStoreTileDTraits { + /// The scalar. + typedef typename platform::remove_const::type Scalar; + /// The pointer. + typedef Scalar_* Pointer; + /// The dimension of the output tile. + typedef OutputTile_ OutputTile; + /// The warps in the tile. + typedef Warps_ Warps; + /// The threads in the warps. + typedef ThreadsPerWarp_ ThreadsPerWarp; + /// The number of scalars per LDG/STG. + static int const kAccessSize = kScalarsPerSts_; + /// The skew. + static int const kSkew = kSkew_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + + /// The number of scalars per thread. + static int const kScalarsPerThread = OutputTile_::kW / Warps::kW / ThreadsPerWarp::kW; + /// The number of threads. + static int const kThreads = ShapeCount::kCount * kWarpSize; + /// The number of scalars per row. We build a tile with 2 rows (to avoid bank conflicts). + static int const kScalarsPerRow = kThreads / 2 * kScalarsPerThread + kSkew; + + /// The tile. + typedef Shape<1, 2, kScalarsPerRow / kAccessSize, kAccessSize> Tile; + /// The number of iterations needed to store the tile. + typedef Shape<1, 1, kScalarsPerThread / kAccessSize> Iterations; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, 0, Warps::kW * ThreadsPerWarp::kW * kAccessSize> Delta; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, 0, Warps::kW * ThreadsPerWarp::kW * kAccessSize> ImmediateOffsetStrides; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE Coord<4> operator()() const { + // The warp. + int const warp = threadIdx.x / kWarpSize; + + // The position of the warp in the 2D tile. + int const warp_row = warp % Warps::kW; + int const warp_col = warp / Warps::kW; + + // We assume that the elements are distributed in a warps as 4 columns of 8 elements. The + // columns are stored in threads col0=[0, 2, 4, 6, 8, 10, 12, 14], col1=[1, 3, 5, 7, .., 15], + // col2=[16, 18, 20, ..., 30] and col3=[17, 19, ..., 31]. + int hi_halfwarp_offset = ((threadIdx.x >> 4) & 0x1) * OutputTile::kW; + int lo_halfwarp_offset = ((threadIdx.x >> 1) & 0x7) + ThreadsPerWarp::kW * warp_row; + + // Odd threads go to the second half of shared memory. + int const row = threadIdx.x & 0x01; + int col = warp_col * (ThreadsPerWarp::kH / 2) * OutputTile::kW + + lo_halfwarp_offset * kAccessSize + hi_halfwarp_offset; + // Embed the offset in a 4D coords. + return make_Coord(0, 0, row * kScalarsPerRow + col, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmSharedLoadTileDTraits { + /// The scalar. + typedef typename platform::remove_const::type Scalar; + /// The pointer. + typedef Scalar_* Pointer; + /// The dimension of the output tile. + typedef OutputTile_ OutputTile; + /// The warps in the tile. + typedef Warps_ Warps; + /// The threads in the warps. + typedef ThreadsPerWarp_ ThreadsPerWarp; + /// The number of scalars per LDG/STG. + static int const kAccessSize = kScalarsPerLds_; + /// The skew. + static int const kSkew = kSkew_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + + /// The number of scalars per thread. + static int const kScalarsPerThread = OutputTile_::kW / Warps::kW / ThreadsPerWarp::kW; + /// The number of threads. + static int const kThreads = ShapeCount::kCount * kWarpSize; + /// The number of scalars per row. We build a tile with 2 rows (to avoid bank conflicts). + static int const kScalarsPerRow = kThreads / 2 * kScalarsPerThread + kSkew; + + /// The tile. We have 2 rows of scalars. We use those two rows to make sure we do not have bank + /// conflicts in the epilogue. + typedef Shape<1, 2, kScalarsPerRow / kAccessSize, kAccessSize> Tile; + + // Compute the number of iterations per warp in the Tile::kH dimension. + static int const kIterationsInHPerWarp = kTileH_ / ShapeCount::kCount; + + // As explained above, the shared memory tile is composed of 2 rows and each rows is made of + // kScalarsPerRow. A warp is expected to read from the 1st row, then move to the 2nd row and go + // back to the 1st row. To model that scheme we define the Iterations shape as Shape. + // However, in some cases, we have only 1 iteration per warp. In that case, we must define the + // shape as Shape<1, 1, ...>. The following code does that except that we hijack the kH dimension + // to keep the number of elements to reduce for split-K. + static int const kIterationsH = kIterationsInHPerWarp == 1 ? 1 : 2; + // As soon as we know kIterationsH, it is trivial to compute kIterationsD: + static int const kIterationsD = kIterationsInHPerWarp / kIterationsH; + + // If we have split-K enabled, we have to jump over the elements from the "odd/even" column of + // threads to grab the other elements. + static int const kSplitK = OutputTile::kW * ThreadsPerWarp::kH / 2 * Warps::kH; + + /// The number of iterations needed to store the tile. + typedef Shape + Iterations; + /// The strides in each dimension between different loads/stores. + typedef Shape + ImmediateOffsetStrides; + /// The strides in each dimension between different loads/stores. + typedef Shape Delta; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE Coord<4> operator()() const { + // Each warp works on a different column. + int const h = threadIdx.x / kWarpSize; + // Compute the row. + int const w = (threadIdx.x & (kWarpSize - 1)) * kAccessSize; + int offset = 0; + if (Iterations::kH == 1) { + int const row = h & 0x1; + int const col = h / 2; + offset = row * ShapeCount::kWc + col * OutputTile::kW * Iterations::kD + w; + } else { + offset = h * OutputTile::kW * Iterations::kD + w; + } + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/gemm_traits.h b/cutlass-example/cutlass/gemm/gemm_traits.h new file mode 100644 index 0000000..cb57c4d --- /dev/null +++ b/cutlass-example/cutlass/gemm/gemm_traits.h @@ -0,0 +1,818 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines structural properties of complete GEMM computation. +*/ +#pragma once + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The scalar type for A. + typename ScalarA_, + /// The scalar type for B. + typename ScalarB_, + /// The scalar type for C. + typename ScalarC_, + /// The scalar type for D. + typename ScalarD_, + /// The output tile size for the GEMM KxNxM. + typename OutputTile_, + /// The functor to do the math. + typename MultiplyAdd_, + /// The number of scalars per LDG for A. + int kScalarsPerLdgA_, + /// The number of scalars per STS for A. + int kScalarsPerStsA_, + /// The number of scalars per LDG for A. + int kScalarsPerLdsA_, + /// The number of scalars per LDG for B. + int kScalarsPerLdgB_, + /// The number of scalars per STS for B. + int kScalarsPerStsB_, + /// The number of scalars per LDS for B. + int kScalarsPerLdsB_, + /// The number of scalars per LDG for C and STG for D. + int kScalarsPerLdgCAndStgD_, + /// The number of scalars per STS for D. + int kScalarsPerStsD_, + /// The number of scalars per LDS for D. + int kScalarsPerLdsD_, + /// The number of stages in shared memory to do single/double/triple-buffering. + int kStages_, + /// Do we do the residue in the prologue? + bool kResidueInPrologue_ = false> + +struct GemmConfig { + // + /// The scalar for A. + typedef ScalarA_ ScalarA; + /// The scalar for B. + typedef ScalarB_ ScalarB; + /// The scalar for C. + typedef ScalarC_ ScalarC; + /// The scalar for D. + typedef ScalarD_ ScalarD; + + /// The tile. + typedef OutputTile_ OutputTile; + /// The functor to do D = A*B + C. + typedef MultiplyAdd_ MultiplyAdd; + /// The shape of the instruction. + typedef typename MultiplyAdd::InstructionShape InstructionShape; + /// The number of accumulators per warp. + typedef typename MultiplyAdd::AccumulatorsPerWarp AccumulatorsPerWarp; + /// The accumulators. + typedef typename MultiplyAdd::Accumulators Accumulators; + + /// The number of warps. + typedef typename ShapeDiv::Shape Warps; + /// The default warp size (32 threads per warp). + static int const kWarpSize = cutlass::kWarpSize; + /// The numnber of threads. + static int const kThreads = ShapeCount::kCount * kWarpSize; + + /// The number of scalars per LDG/STS/LDS for A. + static int const kScalarsPerLdgA = kScalarsPerLdgA_; + static int const kScalarsPerStsA = kScalarsPerStsA_; + static int const kScalarsPerLdsA = kScalarsPerLdsA_; + + /// The number of scalars per LDG/STS/LDS for B. + static int const kScalarsPerLdgB = kScalarsPerLdgB_; + static int const kScalarsPerStsB = kScalarsPerStsB_; + static int const kScalarsPerLdsB = kScalarsPerLdsB_; + + /// The number of scalars per LDG for C. + static int const kScalarsPerLdgC = kScalarsPerLdgCAndStgD_; + + /// The number of scalars per STS/LDS/STG for D. + static int const kScalarsPerStgD = kScalarsPerLdgCAndStgD_; + static int const kScalarsPerStsD = kScalarsPerStsD_; + static int const kScalarsPerLdsD = kScalarsPerLdsD_; + + /// The number of accumulators that are going to be fed from one LDS A/B. + static int const kAccumulatorsPerLdsA = kScalarsPerLdsA / InstructionShape::kD; + static int const kAccumulatorsPerLdsB = kScalarsPerLdsB / InstructionShape::kD; + + /// The number of stages in shared memory to implement double, triple, more-buffering. + static int const kStages = kStages_; + + /// Do we do the residue in the prologue? + static bool const kResidueInPrologue = kResidueInPrologue_; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmTileTraitsHelperA {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmTileTraitsHelperA { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kColumnMajor; + + /// The input scalar. + typedef typename GemmConfig_::ScalarA Scalar; + /// The scalar stored in shared memory. + typedef typename GemmConfig_::MultiplyAdd::ScalarA MultiplyAddScalar; + + /// The traits class to build the iterator to load data from global memory for A^N. + typedef GemmGlobalTileTraits< + // That's A. + GemmOperand::kA, + // A is column-major. + MatrixLayout::kColumnMajor, + // The pointer is float const. + Scalar const, + // The tile has size KxM in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kW>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgA> + GlobalTileTraits; + + /// The traits class to build the iterator to store data to shared memory for A^N. + typedef GemmSharedStoreTileAbTraits< + // The pointer is float. + MultiplyAddScalar, + // The tile has size KxM in GEMM's terminology. + Shape, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + GemmConfig_::kScalarsPerStsA> + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for A^N. + typedef GemmSharedLoadTileATraits< + // The pointer is float const. + MultiplyAddScalar const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + GemmConfig_::kScalarsPerLdsA, + // The skew. + 0> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmTileTraitsHelperA { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kRowMajor; + + /// The input scalar. + typedef typename GemmConfig_::ScalarA Scalar; + /// The scalar stored in shared memory. + typedef typename GemmConfig_::MultiplyAdd::ScalarA MultiplyAddScalar; + + /// The traits class to build the iterator to load data from global memory for A^T. + typedef GemmGlobalTileTraits< + // That's A. + GemmOperand::kA, + // A is row-major. + MatrixLayout::kRowMajor, + // The pointer is float const. + Scalar const, + // The tile has size MxK in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kW, GemmConfig_::OutputTile::kD>, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + Shape<1, GemmConfig_::kThreads / GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kD>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgA> + GlobalTileTraits; + + /// The number of scalars in 4B. + static int const kScalarsIn4B = sizeof(MultiplyAddScalar) > 4 ? 1 : 4 / sizeof(MultiplyAddScalar); + /// The skew for A. + static int const kSkewA = 128 / sizeof(MultiplyAddScalar) / GemmConfig_::kScalarsPerStsA / + GlobalTileTraits::Threads::kW * kScalarsIn4B; + + /// The traits class to build the iterator to store data to shared memory for A^T. + typedef GemmSharedStoreWithSkewTileAbTraits < + // The pointer is float. + MultiplyAddScalar, + // The tile has size KxM in GEMM's terminology. + Shape, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS. + GemmConfig_::kScalarsPerStsA, + // The skew to avoid bank conflicts added in the tile W dimension. + kSkewA + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for A^T. + typedef GemmSharedLoadTileATraits< + // The pointer is float const. + MultiplyAddScalar const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + GemmConfig_::kScalarsPerLdsA, + // The skew. + SharedStoreTileTraits::kSkew> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmTileTraitsHelperB {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmTileTraitsHelperB { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kColumnMajor; + + /// The input scalar. + typedef typename GemmConfig_::ScalarB Scalar; + /// The scalar stored in shared memory. + typedef typename GemmConfig_::MultiplyAdd::ScalarB MultiplyAddScalar; + + /// The traits class to build the iterator to load data from global memory for B^N. + typedef GemmGlobalTileTraits< + // That's B. + GemmOperand::kB, + // B is column-major. + MatrixLayout::kColumnMajor, + // The pointer is float const. + Scalar const, + // The tile has size MxK in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kH, GemmConfig_::OutputTile::kD>, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + Shape<1, GemmConfig_::kThreads / GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kD>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgB> + GlobalTileTraits; + + /// The number of scalars in 4B. + static int const kScalarsIn4B = sizeof(MultiplyAddScalar) > 4 ? 1 : 4 / sizeof(MultiplyAddScalar); + /// The skew for B. + static int const kSkewB = 128 / sizeof(MultiplyAddScalar) / GemmConfig_::kScalarsPerStsB / + GlobalTileTraits::Threads::kW * kScalarsIn4B; + + /// The traits class to build the iterator to store data to shared memory for B^N. + typedef GemmSharedStoreWithSkewTileAbTraits < + // The pointer is float. + MultiplyAddScalar, + // The tile has size KxN in GEMM's terminology. + Shape, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS. + GemmConfig_::kScalarsPerStsB, + // The skew to avoid bank conflicts added in the tile W dimension. + kSkewB + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for B^N. + typedef GemmSharedLoadTileBTraits< + // The pointer is float const. + MultiplyAddScalar const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + GemmConfig_::kScalarsPerLdsB, + // The skew. + SharedStoreTileTraits::kSkew> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmTileTraitsHelperB { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kRowMajor; + + /// The input scalar. + typedef typename GemmConfig_::ScalarB Scalar; + /// The scalar stored in shared memory. + typedef typename GemmConfig_::MultiplyAdd::ScalarB MultiplyAddScalar; + + /// The traits class to build the iterator to load data from global memory for B^T. + typedef GemmGlobalTileTraits< + // That's B. + GemmOperand::kB, + // B is row-major. + MatrixLayout::kRowMajor, + // The pointer is float const. + Scalar const, + // The tile has size KxN in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kH>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgB> + GlobalTileTraits; + + /// The traits class to build the iterator to store data to shared memory for B^T. + typedef GemmSharedStoreTileAbTraits< + // The pointer is float. + MultiplyAddScalar, + // The tile has size KxN in GEMM's terminology. + Shape, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + GemmConfig_::kScalarsPerStsB> + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for B^T. + typedef GemmSharedLoadTileBTraits< + // The pointer is float const. + MultiplyAddScalar const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + GemmConfig_::kScalarsPerLdsB, + // The skew. + 0> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmResidue { + /// Move to residue portion. + template + static CUTLASS_DEVICE void move_to_residue(typename GemmTraits_::GlobalLoadStreamA& stream_a, + typename GemmTraits_::GlobalLoadStreamB& stream_b, + typename GemmTraits_::Index k) { + // The new code path in CUTLASS 1.0.1: We treat the residue in the prologue so we can have + // complete main loops after that. It helps simplify the logic in the main loop. + if (kIsPrologue) { + stream_a.move_to_residue(k); + stream_b.move_to_residue(k); + } + } + + /// Rollback to beginning of first tile and initialize predicates. + static CUTLASS_DEVICE void rollback(typename GemmTraits_::GlobalLoadStreamA& stream_a, + typename GemmTraits_::GlobalLoadStreamB& stream_b) { + stream_a.rollback(); + stream_b.rollback(); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct GemmResidue { + /// Move to residue portion. + template + static CUTLASS_DEVICE void move_to_residue(typename GemmTraits_::GlobalLoadStreamA& stream_a, + typename GemmTraits_::GlobalLoadStreamB& stream_b, + typename GemmTraits_::Index k) { + // The index. + typedef typename GemmTraits_::Index Index; + // By how much we unroll the main loop. + Index const kUnroll = static_cast(GemmTraits_::OutputTile::kD); + + // Call the residue code. That's the same path as CUTLASS 1.0.0. + if (kIsPrologue && k < kUnroll) { + stream_a.residue(k, true); + stream_b.residue(k, true); + } else if (k <= kUnroll) { + stream_a.residue(k, false); + stream_b.residue(k, false); + } + } + + /// Rollback to beginning of first tile and initialize predicates. + static CUTLASS_DEVICE void rollback(typename GemmTraits_::GlobalLoadStreamA& stream_a, + typename GemmTraits_::GlobalLoadStreamB& stream_b) {} +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The GEMM configuration. + typename GemmConfig_, + /// The stream to load A from global memory to shared memory. + typename GlobalLoadStreamA_, + /// The stream to load B from global memory to shared memory. + typename GlobalLoadStreamB_, + /// The stream to load A from shared memory. + typename SharedLoadStreamA_, + /// The stream to load B from shared memory. + typename SharedLoadStreamB_, + /// The epilogue. + typename Epilogue_, + /// The block swizzle to reorganize the grid. + typename BlockSwizzle_ = IdentityBlockSwizzle, + /// The index. + typename Index_ = int, + /// The tool used to clear accumulators. + typename ClearAccumulators_ = ClearAccumulators > + +struct GemmTraits { + /// This class. + typedef GemmTraits + This_; + + /// The configuration. + typedef GemmConfig_ GemmConfig; + /// The output tile. + typedef typename GemmConfig::OutputTile OutputTile; + /// Is the residue treated in the prologue? + static bool const kResidueInPrologue = GemmConfig::kResidueInPrologue; + + /// The stream to load A from global memory to shared memory. + typedef GlobalLoadStreamA_ GlobalLoadStreamA; + /// The layout of A. + static MatrixLayout::Kind const kLayoutA = GlobalLoadStreamA::kLayout; + /// The scalar for A. + typedef typename GlobalLoadStreamA_::Scalar ScalarA; + + /// The stream to load B from global memory to shared memory. + typedef GlobalLoadStreamB_ GlobalLoadStreamB; + /// The layout of B. + static MatrixLayout::Kind const kLayoutB = GlobalLoadStreamB::kLayout; + /// The scalar for B. + typedef typename GlobalLoadStreamB_::Scalar ScalarB; + + /// The iterator for A to load from shared memory. + typedef SharedLoadStreamA_ SharedLoadStreamA; + /// The iterator for B to load from shared memory. + typedef SharedLoadStreamB_ SharedLoadStreamB; + + /// The multiply-add functor. + typedef typename GemmConfig::MultiplyAdd MultiplyAdd; + /// The epilogue. + typedef Epilogue_ Epilogue; + /// The scalars in the epilogue. + typedef typename Epilogue::ScalarC ScalarC; + typedef typename Epilogue::ScalarD ScalarD; + + /// The block swizzle to reorganize the grid. + typedef BlockSwizzle_ BlockSwizzle; + /// The index. + typedef Index_ Index; + /// Clear the accumulators. + typedef ClearAccumulators_ ClearAccumulators; + + /// The params. + struct Params { + /// The dimensions of the GEMM. + Index m, n, k; + /// The params for the A stream. + typename GlobalLoadStreamA::Params global_stream_a; + /// The params for the B stream. + typename GlobalLoadStreamB::Params global_stream_b; + /// The params for the A stream from shared memory. + typename SharedLoadStreamA::Params shared_stream_a; + /// The params for the B stream from shared memory. + typename SharedLoadStreamB::Params shared_stream_b; + /// The params for the epilogue. + typename Epilogue::Params epilogue; + + /// Initialize the parameters. + template + CUTLASS_HOST_DEVICE int initialize(GemmDesc_ const& desc) { + // Set the problem size. + this->m = desc.m; + this->n = desc.n; + this->k = desc.k; + + // Initialize the iterator for A. + int error_code = + global_stream_a.initialize(desc, reinterpret_cast(desc.d_a), desc.lda); + + if (error_code) { + return error_code; + } + + // Initialize the iterator for B. + error_code = + global_stream_b.initialize(desc, reinterpret_cast(desc.d_b), desc.ldb); + + if (error_code) { + return error_code; + } + + // The epilogue. + return epilogue.initialize(desc); + } + }; + + // The storage for A. + template + union StreamSharedStorage { + // The storage needed by the global stream. + typename GlobalLoadStream_::SharedStorage global; + // The storage needed by the shared stream. + typename SharedLoadStream_::SharedStorage shared; + }; + + // The storage for the main loop + prologue. + struct MainLoopSharedStorage { + // The storage to shuffle the A matrix in shared memory. + StreamSharedStorage stream_a; + // The storage to shuffle the B matrix in shared memory. + StreamSharedStorage stream_b; + // The storage to clear the accumulators if needed. + typename ClearAccumulators::SharedStorage clear; + }; + + /// The storage in shared memory. + union SharedStorage { + // The storage for the main loop. + MainLoopSharedStorage main_loop; + // The storage for the epilogue. + typename Epilogue::SharedStorage epilogue; + }; + + /// Assemble the global load streams for A/B. + struct GlobalLoadStream { + /// Ctor. + CUTLASS_DEVICE GlobalLoadStream(Params const& params, + SharedStorage& shared_storage, + dim3 const& block) + : stream_a(params.global_stream_a, + shared_storage.main_loop.stream_a.global, + cutlass::make_Coord(0, params.k, params.m), + cutlass::make_Coord(0, 0, block.x)), + stream_b(params.global_stream_b, + shared_storage.main_loop.stream_b.global, + cutlass::make_Coord(0, params.k, params.n), + make_Coord(0, 0, block.y)) {} + + /// Trigger the copies from shared memory to registers. + CUTLASS_DEVICE void copy() { + stream_a.copy(); + stream_b.copy(); + } + + /// Commit the data. + CUTLASS_DEVICE void commit() { + stream_a.commit(); + stream_b.commit(); + } + + /// Move to residue portion. + template + CUTLASS_DEVICE void move_to_residue(Index k) { + GemmResidue::move_to_residue(stream_a, stream_b, k); + } + + /// Rollback to beginning of first tile and initialize predicates. + CUTLASS_DEVICE void rollback() { GemmResidue::rollback(stream_a, stream_b); } + + /// The stream for A. + GlobalLoadStreamA stream_a; + /// The stream for B. + GlobalLoadStreamB stream_b; + }; + + /// Assemble the shared load stream for A/B. + struct SharedLoadStream { + /// Ctor. + CUTLASS_DEVICE SharedLoadStream(Params const& params, SharedStorage& shared_storage) { + stream_a.initialize(params.shared_stream_a, shared_storage.main_loop.stream_a.shared); + stream_b.initialize(params.shared_stream_b, shared_storage.main_loop.stream_b.shared); + } + + /// Trigger the copies from shared memory to registers. + CUTLASS_DEVICE void copy(int step) { + stream_a.copy(step, fetched_a[step % 2]); + stream_b.copy(step, fetched_b[step % 2]); + } + + /// Commit the data. + CUTLASS_DEVICE void commit(int step) { + stream_a.commit(fetched_a[step % 2], transformed_a[step % 2]); + stream_b.commit(fetched_b[step % 2], transformed_b[step % 2]); + } + + /// The fragment A. + CUTLASS_DEVICE typename SharedLoadStreamA::Fragment const& fragment_a(int step) const { + return transformed_a[step % 2]; + } + + /// The fragment B. + CUTLASS_DEVICE typename SharedLoadStreamB::Fragment const& fragment_b(int step) const { + return transformed_b[step % 2]; + } + + /// Increment the stage. + CUTLASS_DEVICE void inc_stage() { + stream_a.inc_stage(); + stream_b.inc_stage(); + } + + /// The stream for A. + SharedLoadStreamA stream_a; + /// The fragments to fetch A. + typename SharedLoadStreamA::FetchedFragment fetched_a[2]; + /// The fragments to transform A. + typename SharedLoadStreamA::TransformedFragment transformed_a[2]; + /// The stream for B. + SharedLoadStreamB stream_b; + /// The fragments to fetch B. + typename SharedLoadStreamB::FetchedFragment fetched_b[2]; + /// The fragments to transform B. + typename SharedLoadStreamB::TransformedFragment transformed_b[2]; + }; + + /// The memory fence for shared loads. + static CUTLASS_DEVICE void shared_load_fence(bool in_loop) { + if (SharedLoadStreamA::Iterator::kRequiresLoadFence || + SharedLoadStreamB::Iterator::kRequiresLoadFence) { + __syncthreads(); + } + } + + /// The memory fence for shared stores. + static CUTLASS_DEVICE void shared_store_fence(bool in_loop) { __syncthreads(); } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct SimplifiedGemmTraitsHelper { + /// The global iterator to load A from global memory. + typedef GemmGlobalIteratorAb + GlobalLoadIteratorA; + /// The data converter for A before storing to shared memory. + typedef Copy GlobalTransformerA; + /// The iterator to store A to shared memory. + typedef TileStoreIterator + SharedStoreIteratorA; + /// The stream to load A from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamA; + + /// The global iterator to load B from global memory. + typedef GemmGlobalIteratorAb + GlobalLoadIteratorB; + /// The data converter for B before storing to shared memory. + typedef Copy GlobalTransformerB; + /// The iterator to store B to shared memory. + typedef TileStoreIterator + SharedStoreIteratorB; + /// The stream to load B from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamB; + + /// The iterator to load A from shared memory. + typedef TileLoadIterator + SharedLoadIteratorA; + /// The stream to load A from shared memory. + typedef SharedLoadStream SharedLoadStreamA; + /// The iterator to load B from shared memory. + typedef TileLoadIterator + SharedLoadIteratorB; + /// The stream to load B from shared memory. + typedef SharedLoadStream SharedLoadStreamB; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The config for the GEMM. + typename GemmConfig_, + /// The epilogue. + typename Epilogue_, + /// The index. + typename Index_ = int, + // The configuration for the A matrix. + typename GemmTileTraitsHelperA_ = GemmTileTraitsHelperA, + // The configuration for the B matrix. + typename GemmTileTraitsHelperB_ = GemmTileTraitsHelperB, + // The helper class to create the streams and iterators. + typename Helper_ = + SimplifiedGemmTraitsHelper > +struct SimplifiedGemmTraits : public GemmTraits< + // The config. + GemmConfig_, + // The stream to load A from global memory to shared memory. + typename Helper_::GlobalLoadStreamA, + // The stream to load B from global memory to shared memory. + typename Helper_::GlobalLoadStreamB, + // The stream to load A from shared memory. + typename Helper_::SharedLoadStreamA, + // The stream to load B from shared memory. + typename Helper_::SharedLoadStreamB, + // The epilogue. + Epilogue_, + // The block swizzle to reorganize the grid. + IdentityBlockSwizzle, + // The index. + Index_, + // The tool used to clear accumulators. + ClearAccumulators > { +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/hgemm_global_tile.h b/cutlass-example/cutlass/gemm/hgemm_global_tile.h new file mode 100644 index 0000000..f14dbb3 --- /dev/null +++ b/cutlass-example/cutlass/gemm/hgemm_global_tile.h @@ -0,0 +1,90 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Tile traits used to construct global tile iterator for HGEMM. This is intended to + partition the thread block-level tile into 2D subtiles loaded by the threads and facilitate + memory accesses larger than 16 bits. +*/ +#pragma once + +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmCrosswiseGlobalTileTraits : public GemmGlobalTileTraits< + // Which GEMM operand? + kOperand_, + // The layout. + kLayout_, + // The scalar. + Scalar_, + // The tile. + Tile_, + // The threads. + Threads_, + // The number of scalars per LDG/STG. + kAccessSize_> { + /// The base class. + typedef GemmGlobalTileTraits Base; + /// The threads. + typedef typename Base::Threads Threads; + /// The threads strides. + typedef Shape<1, 2, Base::Tile::kC> ThreadsDelta; + /// The strides in each dimension between different loads/stores. + typedef Shape Delta; + /// The number of iterations needed to load/store the tile. + typedef Shape + Iterations; + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + int thread_offset_h = threadIdx.x / Threads::kW * ThreadsDelta::kH; + int thread_offset_w = threadIdx.x % Threads::kW * ThreadsDelta::kW; + + return make_Coord(0, thread_offset_h, thread_offset_w, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/hgemm_multiply_add.h b/cutlass-example/cutlass/gemm/hgemm_multiply_add.h new file mode 100644 index 0000000..ebbdd06 --- /dev/null +++ b/cutlass-example/cutlass/gemm/hgemm_multiply_add.h @@ -0,0 +1,104 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Specialization implementing multiply-add operation on half-precision floating point + fragments. +*/ +#pragma once + +#include + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Template performing matrix multiply-add operation within a thread +template +struct ThreadMultiplyAdd { + /// The shape of the instruction. + typedef Shape<1, 1, 2, 1> InstructionShape; + /// The number of accumulators per thread. + typedef AccumulatorsPerThread_ AccumulatorsPerThread; + /// The number of threads per warp. + typedef ThreadsPerWarp_ ThreadsPerWarp; + /// The number of accumulators per warp. + typedef typename ShapeMul::Shape AccumulatorsPerWarp; + /// The type for A. + typedef half ScalarA; + /// The fragment for A. + typedef Fragment FragmentA; + /// The type for B. + typedef half ScalarB; + /// The fragment for B. + typedef Fragment FragmentB; + /// The type for C and D. + typedef half ScalarC; + /// The accumulators. + typedef Fragment Accumulators; + + /// Make sure there's an even number of elements in both dimensions. + static_assert(AccumulatorsPerThread::kH % 2 == 0, "Invalid size"); + static_assert(AccumulatorsPerThread::kW % 2 == 0, "Invalid size"); + + /// Ctor. + CUTLASS_DEVICE ThreadMultiplyAdd() {} + + /// Multiply : d = a*b + c. + CUTLASS_DEVICE void multiply_add(FragmentA const& a, + FragmentB const& b, + Accumulators const& c, + Accumulators& d) { +#if defined(__CUDACC__) && __CUDA_ARCH__ >= 530 + // The inputs. + __half2 const* a_half2 = reinterpret_cast<__half2 const*>(&a[0]); + __half2 const* b_half2 = reinterpret_cast<__half2 const*>(&b[0]); + __half2 const* c_half2 = reinterpret_cast<__half2 const*>(&c[0]); + + // The output. + __half2* d_half2 = reinterpret_cast<__half2*>(&d[0]); + + for (int j = 0; j < AccumulatorsPerThread::kH / 2; ++j) { + for (int i = 0; i < AccumulatorsPerThread::kW / 2; ++i) { + // The offsets in the output fragment. + int const k0 = (2 * j + 0) * (AccumulatorsPerThread::kW / 2) + i; + int const k1 = (2 * j + 1) * (AccumulatorsPerThread::kW / 2) + i; + + // Compute the product a[i] * b[j].H0_H0. + d_half2[k0] = __hfma2(a_half2[i], __low2half2(b_half2[j]), c_half2[k0]); + // Compute the product a[i] * b[j].H1_H1. + d_half2[k1] = __hfma2(a_half2[i], __high2half2(b_half2[j]), c_half2[k1]); + } + } +#endif + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/hgemm_swizzle.h b/cutlass-example/cutlass/gemm/hgemm_swizzle.h new file mode 100644 index 0000000..ebec0d4 --- /dev/null +++ b/cutlass-example/cutlass/gemm/hgemm_swizzle.h @@ -0,0 +1,94 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Transposes a tile of 16b elements. Used by HGEMM to construct a K-strided layout in + shared memory for multiplicands. +*/ +#pragma once + +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmSwizzle { + /// The global iterator. + typedef GlobalIterator_ GlobalIterator; + /// The source fragment. + typedef typename GlobalIterator::Fragment Fragment; + /// The shape of the source fragment. + typedef typename GlobalIterator::FragmentShape FragmentShape; + + /// The input fragment. + typedef Fragment InputFragment; + /// The output fragment. + typedef Fragment OutputFragment; + + /// The src/dst must be half fragments. + static_assert((platform::is_same::value), "Works on half"); + + /// The number of elements must be a multiple of 2. + static_assert(FragmentShape::kH == 2 && ShapeCount::kWc == 2, "Not multiple of 2"); + + /// Ctor. + CUTLASS_DEVICE HgemmSwizzle() {} + + /// Transform a fragment. + CUTLASS_DEVICE void transform(Fragment const& src, Fragment& dst) { + // Expose src/dst as int arrays. + int const* src_int = reinterpret_cast(&src[0]); + int* dst_int = reinterpret_cast(&dst[0]); + + // Transpose the data. + for (int d = 0; d < FragmentShape::kD; ++d) { + // The indices to read two consecutive "rows". + int const i0 = 2 * d + 0; + int const i1 = 2 * d + 1; + + int a0 = src_int[i0]; + int a1 = src_int[i1]; + + int b0, b1; + asm volatile("prmt.b32 %0, %1, %2, 0x5410;" : "=r"(b0) : "r"(a0), "r"(a1)); + asm volatile("prmt.b32 %0, %1, %2, 0x7632;" : "=r"(b1) : "r"(a0), "r"(a1)); + + // The indices to store with "strides". + int const j0 = 0 * (ShapeCount::kDhw / 2) + d; + int const j1 = 1 * (ShapeCount::kDhw / 2) + d; + + dst_int[j0] = b0; + dst_int[j1] = b1; + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/hgemm_traits.h b/cutlass-example/cutlass/gemm/hgemm_traits.h new file mode 100644 index 0000000..b08645b --- /dev/null +++ b/cutlass-example/cutlass/gemm/hgemm_traits.h @@ -0,0 +1,397 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defies structural properties of half-precision GEMM computation. +*/ +#pragma once + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The tile size for the GEMM KxNxM. + typename OutputTile_, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_, + /// The number of scalars per LDG for A. + int kScalarsPerLdgA_ = 2, + /// The number of scalars per LDG for B. + int kScalarsPerLdgB_ = 2> +struct HgemmConfig + : public GemmConfig< + /// The scalar type for A. + half, + /// The scalar type for B. + half, + /// The scalar type for C. + half, + /// The scalar type for D. + half, + /// The tile size for the GEMM KxNxM. + OutputTile_, + /// The functor to do the math in the main loop. + ThreadMultiplyAdd, half, half, half>, + /// The number of scalars per LDG for A. + kScalarsPerLdgA_, + /// The number of scalars per STS for A. + kScalarsPerLdgA_, + /// The number of scalars per LDS for A. + 8, + /// The number of scalars per LDG for B. + kScalarsPerLdgB_, + /// The number of scalars per STS for B. + kScalarsPerLdgB_, + /// The number of scalars per LDS for B. + 8, + /// The number of scalars per LDG for C and STG for D. + 2, + /// The number of scalars per STS for D. + 8, + /// The number of scalars per LDS for D. + 2, + /// The number of stages in shared memory. + 2> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmTransformerA {}; + +template +struct HgemmTransformerA { + typedef Convert Transformer; +}; + +template +struct HgemmTransformerA { + typedef HgemmSwizzle Transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmTransformerB {}; + +template +struct HgemmTransformerB { + typedef Convert Transformer; +}; + +template +struct HgemmTransformerB { + typedef HgemmSwizzle Transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmTileTraitsHelperA : public GemmTileTraitsHelperA {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmTileTraitsHelperA + : public GemmTileTraitsHelperA { + /// The base config. + typedef GemmTileTraitsHelperA Base; + + /// The traits class to build the iterator to load data from global memory for A^T. + typedef HgemmCrosswiseGlobalTileTraits< + GemmOperand::kA, + // The layout. + MatrixLayout::kRowMajor, + // The pointer. + half const, + // The tile has size MxK in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kW, GemmConfig_::OutputTile::kD>, + // The threads are distributed as (threads / K ) x K (the traits may reorganize). + Shape<1, GemmConfig_::kThreads / GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kD>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc) + GemmConfig_::kScalarsPerLdgA> + GlobalTileTraits; + + /// The skew. + static int const kSkewA = 128 / sizeof(half) / GlobalTileTraits::Threads::kW / 2; + + /// The traits class to build the iterator to store data to shared memory for A^T. + typedef GemmSharedStoreWithSkewTileAbTraits < + // The pointer. + half, + // The tile has size KxM in GEMM's terminology. + Shape, + // The threads are distributed as warps x 32(the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + 2, + // The skew to avoid bank conflicts added in the tile W dimension. + kSkewA + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for A^T. + typedef GemmSharedLoadTileATraits< + // The pointer. + half const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + 8, + // The skew. + SharedStoreTileTraits::kSkew> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmTileTraitsHelperB : public GemmTileTraitsHelperB {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct HgemmTileTraitsHelperB + : public GemmTileTraitsHelperB { + /// The base config. + typedef GemmTileTraitsHelperB Base; + + /// The traits class to build the iterator to load data from global memory for B^N. + typedef HgemmCrosswiseGlobalTileTraits< + GemmOperand::kB, + // The layout. + MatrixLayout::kColumnMajor, + // The pointer. + half const, + // The tile has size KxN in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kH, GemmConfig_::OutputTile::kD>, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + Shape<1, GemmConfig_::kThreads / GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kD>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc) + GemmConfig_::kScalarsPerLdgB> + GlobalTileTraits; + + /// The skew for B. + static int const kSkewB = 128 / sizeof(half) / GlobalTileTraits::Threads::kW / 2; + + /// The traits class to build the iterator to store data to shared memory for B^N. + typedef GemmSharedStoreWithSkewTileAbTraits < + // The pointer. + half, + // The tile has size KxN in GEMM's terminology. + Shape, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + 2, + // The skew to avoid bank conflicts added in the tile W dimension. + kSkewB + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for B^N. + typedef GemmSharedLoadTileBTraits< + // The pointer. + half const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + 8, + // The skew. + SharedStoreTileTraits::kSkew> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The output tile. + typename OutputTile_, + /// The functor to do the math in the epilogue. + typename EpilogueFunctor_, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_ = Shape<8, 8, 16>, + /// The number of halfs loaded in one LDG for A. + int kScalarsPerLdgA_ = 2, + /// The number of halfs loaded in one LDG for B. + int kScalarsPerLdgB_ = 2, + /// The index. + typename Index_ = int> +struct HgemmTraitsHelper { + /// The HGEMM config. + typedef HgemmConfig + GemmConfig; + /// The GEMM config for A. + typedef HgemmTileTraitsHelperA GemmTileTraitsHelperA; + /// The GEMM config for B. + typedef HgemmTileTraitsHelperB GemmTileTraitsHelperB; + + /// The iterator to load A from global memory. + typedef GemmGlobalIteratorAb + GlobalLoadIteratorA; + /// The default transformer for A. + typedef typename HgemmTransformerA::Transformer GlobalTransformerA; + /// The iterator to store A to shared memory. + typedef TileStoreIterator + SharedStoreIteratorA; + /// The stream to load A from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamA; + + /// The iterator to load B from global memory. + typedef GemmGlobalIteratorAb + GlobalLoadIteratorB; + // The default transformer for B. + typedef typename HgemmTransformerB::Transformer GlobalTransformerB; + /// The iterator to store B to shared memory. + typedef TileStoreIterator + SharedStoreIteratorB; + /// The stream to load B from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamB; + + /// The iterator to load A from shared memory + typedef TileLoadIterator + SharedLoadIteratorA; + /// The stream to load A from shared memory. + typedef SharedLoadStream SharedLoadStreamA; + /// The iterator to load B from shared memory. + typedef TileLoadIterator + SharedLoadIteratorB; + /// The stream to load B from shared memory. + typedef SharedLoadStream SharedLoadStreamB; + + /// The functor to do the multiply-add in the main loop. + typedef typename GemmConfig::MultiplyAdd MultiplyAdd; + /// The object to clear accumulators. + typedef ClearAccumulators ClearAccumulators; + + /// The traits class for the epilogue. + typedef SimplifiedGemmEpilogueTraits GemmEpilogueTraits; + /// The epilogue. + typedef GemmEpilogue Epilogue; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The output tile. + typename OutputTile_ = Shape<8, 128, 128>, + /// The functor to do the math in the epilogue. + typename EpilogueFunctor_ = LinearScaling, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_ = Shape<8, 8, 16>, + /// The number of halfs loaded in one LDG for A. + int kScalarsPerLdgA_ = 2, + /// The number of halfs loaded in one LDG for B. + int kScalarsPerLdgB_ = 2, + /// The index. + typename Index_ = int, + /// The helper class. + typename Helper_ = HgemmTraitsHelper > +struct HgemmTraits : public GemmTraits< + // The config. + typename Helper_::GemmConfig, + // The stream to load A from global memory to shared memory. + typename Helper_::GlobalLoadStreamA, + // The stream to load B from global memory to shared memory. + typename Helper_::GlobalLoadStreamB, + // The stream to load A from shared memory. + typename Helper_::SharedLoadStreamA, + // The stream to load B from shared memory. + typename Helper_::SharedLoadStreamB, + // The epilogue. + typename Helper_::Epilogue, + // The block swizzle to reorganize the grid. + IdentityBlockSwizzle, + // The index. + Index_, + // The tool used to clear accumulators. + typename Helper_::ClearAccumulators> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/identity_block_swizzle.h b/cutlass-example/cutlass/gemm/identity_block_swizzle.h new file mode 100644 index 0000000..e1bdb2e --- /dev/null +++ b/cutlass-example/cutlass/gemm/identity_block_swizzle.h @@ -0,0 +1,48 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defies functors for mapping blockIdx to partitions of the GEMM computation. + + Currently, we only implement an identity mapping. +*/ +#pragma once + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +struct IdentityBlockSwizzle { + /// Ctor. + CUTLASS_DEVICE IdentityBlockSwizzle() {} + + /// Swizzle the block index. + CUTLASS_DEVICE dim3 swizzle() { return blockIdx; } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/igemm_epilogue.h b/cutlass-example/cutlass/gemm/igemm_epilogue.h new file mode 100644 index 0000000..0d69980 --- /dev/null +++ b/cutlass-example/cutlass/gemm/igemm_epilogue.h @@ -0,0 +1,320 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines the epilogue phase of the GEMM computation for IGEMM, supporting integer and + floating-point output matrix formats. +*/ +#pragma once + +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmFloatToInt8Converter { + /// The input fragment. + typedef Fragment InputFragment; + /// The output fragment. + typedef Fragment OutputFragment; + + // We are packing 4 floats into int32 registers so we need kElements to be multiple of 4. + static_assert(kElements_ % 4 == 0, "kElements must be multiple of 4"); + + /// Ctor. + CUTLASS_DEVICE IgemmFloatToInt8Converter() {} + + /// Transform a fragment. + CUTLASS_DEVICE void transform(InputFragment const& src, OutputFragment& dst) { + transform(src, 0, dst); + } + + /// Transform a fragment. + template + CUTLASS_DEVICE void transform(Fragment_ const& src, int offset, OutputFragment& dst) { + // The inputs. + float4 const* src_f4 = reinterpret_cast(&src[0]); + // The outputs. + int* dst_int = reinterpret_cast(&dst[0]); + + // Iterate over the floats and pack them together to produce ints. + for (int i = 0; i < kElements_ / 4; ++i) { + // Read the float4. + float4 f4 = src_f4[i]; + + // Clamp the 4 elements of the floats to the [-128, +127] range. + float x = fmaxf(-128.f, fminf(127.f, f4.x)); + float y = fmaxf(-128.f, fminf(127.f, f4.y)); + float z = fmaxf(-128.f, fminf(127.f, f4.z)); + float w = fmaxf(-128.f, fminf(127.f, f4.w)); + + // Convert to integers. + int ix = (int)x; + int iy = (int)y; + int iz = (int)z; + int iw = (int)w; + + // Extract the lower bytes to build an int32 with 4 int8. + asm volatile("prmt.b32 %0, %0, %1, 0x1140;" : "+r"(ix) : "r"(iy)); + asm volatile("prmt.b32 %0, %0, %1, 0x1140;" : "+r"(iz) : "r"(iw)); + asm volatile("prmt.b32 %0, %0, %1, 0x5410;" : "+r"(ix) : "r"(iz)); + + // Store the int. + dst_int[i] = ix; + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmGlobalStoreTransformer { + typedef Convert, OutputFragment_> Transformer; +}; + +template +struct IgemmGlobalStoreTransformer > { + typedef IgemmFloatToInt8Converter Transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmInt8ToFloatConverter { + /// The input fragment. + typedef Fragment InputFragment; + /// The output fragment. + typedef Fragment OutputFragment; + + // We are unpacking 4 int8s from int32. + static_assert(kElements_ % 4 == 0, "kElements must be multiple of 4"); + + /// Ctor. + CUTLASS_DEVICE IgemmInt8ToFloatConverter() {} + + /// Transform a fragment. + CUTLASS_DEVICE void transform(InputFragment const& src, OutputFragment& dst) { + transform(src, 0, dst); + } + + /// Transform a fragment. + template + CUTLASS_DEVICE void transform(Fragment_ const& src, int offset, OutputFragment& dst) { + // The inputs. + int const* src_int = reinterpret_cast(&src[0]); + // The outputs. + float4* dst_f4 = reinterpret_cast(&dst[0]); + + // Iterate over the int8 and unpack them together to produce floats. + for (int i = 0; i < kElements_ / 4; ++i) { + // Read the int. + int ix, iy, iz, iw = src_int[i]; + + // Extract the 4 bytes. + asm volatile("prmt.b32 %0, 0x0, %1, 0x4440;" : "=r"(ix) : "r"(iw)); + asm volatile("prmt.b32 %0, 0x0, %1, 0x4441;" : "=r"(iy) : "r"(iw)); + asm volatile("prmt.b32 %0, 0x0, %1, 0x4442;" : "=r"(iz) : "r"(iw)); + asm volatile("prmt.b32 %0, 0x0, %1, 0x4443;" : "=r"(iw) : "r"(iw)); + + // The floats. + float fx, fy, fz, fw; + + // Convert to floats (make sure we generate I2F.F32.S8). + asm volatile("cvt.rn.f32.s8 %0, %1;" : "=f"(fx) : "r"(ix)); + asm volatile("cvt.rn.f32.s8 %0, %1;" : "=f"(fy) : "r"(iy)); + asm volatile("cvt.rn.f32.s8 %0, %1;" : "=f"(fz) : "r"(iz)); + asm volatile("cvt.rn.f32.s8 %0, %1;" : "=f"(fw) : "r"(iw)); + + // Store the float4. + dst_f4[i] = make_float4(fx, fy, fz, fw); + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmGlobalLoadTransformer { + typedef Convert > Transformer; +}; + +template +struct IgemmGlobalLoadTransformer, float> { + typedef IgemmInt8ToFloatConverter Transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmSharedStoreTransformer { + typedef Convert, OutputFragment_> Transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmEpilogueTraitsHelper + : public GemmEpilogueTraitsHelper { + /// The base class. + typedef GemmEpilogueTraitsHelper Base; + /// The config. + typedef IgemmConfig_ IgemmConfig; + + /// The scalar type of the epilogue. + typedef typename Base::Scalar Scalar; + /// The iterations. + typedef typename Base::Iterations Iterations; + /// The iterations strides. + typedef typename Base::Delta Delta; + + /// The traits class for the iterator. + typedef typename Base::GlobalLoadTileTraits GlobalLoadTileTraits; + /// The iterator to store to shared memory. + typedef GemmGlobalIteratorCd GlobalLoadIteratorC; + /// The fragment that needs to be produced by the load iterator. + typedef typename GlobalLoadIteratorC::Fragment GlobalFragmentC; + /// The transformer from loaded data to math fragment. + typedef + typename IgemmGlobalLoadTransformer::Transformer GlobalTransformerC; + + /// The traits class for the iterator. + typedef typename Base::GlobalStoreTileTraits GlobalStoreTileTraits; + /// The iterator to store to shared memory. + typedef GemmGlobalIteratorCd GlobalStoreIteratorD; + /// The fragment that needs to be passed to that store iterator. + typedef typename GlobalStoreIteratorD::Fragment GlobalFragmentD; + /// The transformer from accumulators to shared memory fragments. + typedef + typename IgemmGlobalStoreTransformer::Transformer GlobalTransformerD; + + /// The traits class for the shared iterator to store D to shared memory. + typedef typename Base::SharedStoreTileTraits SharedStoreTileTraits; + /// The shared iterator to store D to shared memory. + typedef TileStoreIterator + SharedStoreIteratorD; + /// The fragment that needs to be passed to that store iterator. + typedef typename SharedStoreIteratorD::Fragment SharedStoreFragmentD; + /// The transformer from accumulators to shared memory fragments. + typedef typename IgemmSharedStoreTransformer::Transformer + SharedStoreTransformerD; + /// The traits class for the shared iterator to load D from shared memory. + typedef typename Base::SharedLoadTileTraits SharedLoadTileTraits; + /// The shared iterator to load D from shared memory. + typedef TileLoadIterator + SharedLoadIteratorD; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The config. + typename IgemmConfig_, + /// The functor to do the math in the epilogue. + typename EpilogueFunctor_, + /// The index. + typename Index_ = int, + /// The helper class to assemble the traits. + typename Helper_ = IgemmEpilogueTraitsHelper > +struct IgemmEpilogueTraits : public GemmEpilogueTraits< + // The output tile. + typename IgemmConfig_::OutputTile, + // The accumulators. + typename IgemmConfig_::Accumulators, + // The global iterator for C. + typename Helper_::GlobalLoadIteratorC, + // The transformer for C. + typename Helper_::GlobalTransformerC, + // The transformer for D. + typename Helper_::GlobalTransformerD, + // The global iterator for D. + typename Helper_::GlobalStoreIteratorD, + // The iterator to store D to shared memory. + typename Helper_::SharedStoreIteratorD, + // The shared store transformer for D. + typename Helper_::SharedStoreTransformerD, + // The iterator to load D from shared memory. + typename Helper_::SharedLoadIteratorD, + // The iterations. + typename Helper_::Iterations, + // The strides between iterations. + typename Helper_::Delta, + // The functor to be used in the epilogue. + EpilogueFunctor_, + // The index. + Index_> { + /// Do we output in int8? + static bool const kInt8Output = + platform::is_same::value != 0; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmEpilogue : public GemmEpilogue { + /// The base class. + typedef GemmEpilogue Base; + + /// Ctor. + CUTLASS_DEVICE IgemmEpilogue(typename Base::Params const& params_, + typename Base::SharedStorage& shared_storage_, + typename Base::Index m_, + typename Base::Index n_) + : Base(params_, shared_storage_, m_, n_) {} +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmEpilogue : public GemmEpilogue { + /// The base class. + typedef GemmEpilogue Base; + + /// Ctor. + CUTLASS_DEVICE IgemmEpilogue(typename Base::Params const& params_, + typename Base::SharedStorage& shared_storage_, + typename Base::Index m_, + typename Base::Index n_) + : Base(params_, shared_storage_, m_, n_) {} +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/igemm_global_tile.h b/cutlass-example/cutlass/gemm/igemm_global_tile.h new file mode 100644 index 0000000..3f594ac --- /dev/null +++ b/cutlass-example/cutlass/gemm/igemm_global_tile.h @@ -0,0 +1,161 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Implements tile iterators to partition the thread block tile into 2D subtiles and + efficiently load each. Applies permute transformation to construct 'interleaved K-strided' + data layout in which 4-element dot products from the same K index are arranged in consecutive + locations within shared memory. + + Supports efficient loads from shared memory to target the DP4A instruction. +*/ +#pragma once + +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmGlobalTileTraits : public GemmGlobalTileTraits< + // Which GEMM operand? + kOperand_, + // The layout. + kLayout_, + // The scalar. + Scalar_, + // The tile. + Tile_, + // The threads. + Threads_, + // The number of scalars per LDG/STG. + kAccessSize_> { + /// The base class. + typedef GemmGlobalTileTraits Base; + /// The threads. + typedef typename Base::Threads Threads; + /// The strides in each dimension between different loads/stores. + typedef Shape Delta; + /// The number of iterations needed to load/store the tile. + typedef Shape + Iterations; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + int thread_offset_h = threadIdx.x / Threads::kW * ThreadsDelta::kH; + int thread_offset_w = threadIdx.x % Threads::kW * ThreadsDelta::kW; + + return make_Coord(0, thread_offset_h, thread_offset_w, 0); + } + }; + + public: + /// The threads strides. + typedef Shape<1, 4, Base::Tile::kC> ThreadsDelta; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Deprecated. Please use IgemmGlobalTileTraits instead. + +template +struct IgemmContiguousGlobalTileTraits + : public IgemmGlobalTileTraits {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmGlobalIteratorAb : public GemmGlobalIteratorAb { + /// The base class. + typedef GemmGlobalIteratorAb Base; + /// The functor to compute the thread offset. + typedef typename TileTraits_::ThreadOffset ThreadOffset; + + /// Constructor. + CUTLASS_DEVICE IgemmGlobalIteratorAb(typename Base::Params const& _params, + const Coord<3>& bounds, + const Coord<3>& block, + ThreadOffset thread_offset_func = ThreadOffset()) + : Base(_params, bounds, block, thread_offset_func), in_residue_(false), mask_(0xffffffff) { + // The number of elements read in a single iteration. + int const kBlock = TileTraits_::Tile::kW * TileTraits_::kAccessSize; + // The residue. + int const kResidue = (int)(bounds[1] % kBlock); + + // Compute the number of elements that are valid. + int const left = kResidue - Base::thread_offset[2]; + if (left > 0 && left < 4) { + mask_ = (1u << (8 * left)) - 1u; + } + } + + /// The accessor. + CUTLASS_DEVICE void get(typename Base::AccessType& value, int d, int h, int w, int c) const { + Base::get(value, d, h, w, c); + if (in_residue_) { + reinterpret_cast(value) &= mask_; + } + } + + /// Move to residue portion. + CUTLASS_DEVICE void move_to_residue(typename Base::Index k) { + Base::move_to_residue(k); + in_residue_ = true; + } + + /// Move back to the beginning of the first tile. + CUTLASS_DEVICE void rollback() { + Base::rollback(); + in_residue_ = false; + } + + /// Are we in the residue? + bool in_residue_; + /// The mask to clean up the values. + uint32_t mask_; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/igemm_multiply_add.h b/cutlass-example/cutlass/gemm/igemm_multiply_add.h new file mode 100644 index 0000000..5a8baec --- /dev/null +++ b/cutlass-example/cutlass/gemm/igemm_multiply_add.h @@ -0,0 +1,89 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Implements matrix multiply accumulate operation of 8-bit integer data using DP4A + instruction. +*/ +#pragma once + +#include + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Template performing matrix multiply-add operation within a thread +template +struct ThreadMultiplyAdd { + /// The shape of the instruction. + typedef Shape<4, 1, 1> InstructionShape; + /// The number of accumulators per thread. + typedef AccumulatorsPerThread_ AccumulatorsPerThread; + /// The number of threads per warp. + typedef ThreadsPerWarp_ ThreadsPerWarp; + /// The number of accumulators per warp. + typedef typename ShapeMul::Shape AccumulatorsPerWarp; + /// The type for A. + typedef int8_t ScalarA; + /// The fragment for A. + typedef Fragment FragmentA; + /// The type for B. + typedef int8_t ScalarB; + /// The fragment for B. + typedef Fragment FragmentB; + /// The type for C and D. + typedef int ScalarC; + /// The accumulators. + typedef Fragment Accumulators; + + /// Ctor. + CUTLASS_DEVICE ThreadMultiplyAdd() {} + + /// Multiply : d = a*b + c. + CUTLASS_DEVICE void multiply_add(FragmentA const& a, + FragmentB const& b, + Accumulators const& c, + Accumulators& d) { + // The inputs. + int const* a_int = reinterpret_cast(&a[0]); + int const* b_int = reinterpret_cast(&b[0]); + + for (int j = 0; j < AccumulatorsPerThread::kH; ++j) { + for (int i = 0; i < AccumulatorsPerThread::kW; ++i) { + asm volatile("dp4a.s32.s32 %0, %1, %2, %3;" + : "=r"(d[j * AccumulatorsPerThread::kW + i]) + : "r"(a_int[i]), "r"(b_int[j]), "r"(c[j * AccumulatorsPerThread::kW + i])); + } + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/igemm_swizzle.h b/cutlass-example/cutlass/gemm/igemm_swizzle.h new file mode 100644 index 0000000..77cf711 --- /dev/null +++ b/cutlass-example/cutlass/gemm/igemm_swizzle.h @@ -0,0 +1,115 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Transposes a fragment of data containing packed 8-bit integer elements. +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmSwizzle { + /// The global iterator. + typedef GlobalIterator_ GlobalIterator; + /// The source fragment. + typedef typename GlobalIterator::Fragment Fragment; + /// The shape of the source fragment. + typedef typename GlobalIterator::FragmentShape FragmentShape; + + /// The source fragment. + typedef Fragment InputFragment; + /// The destination fragment. + typedef Fragment OutputFragment; + + /// The src/dst must be int8 fragments. + static_assert((platform::is_same::value), "Works on int8"); + + /// The number of elements must be a multiple of 4. + static_assert(FragmentShape::kH % 4 == 0 && ShapeCount::kWc % 4 == 0, + "Not multiple of 4"); + + /// Ctor. + CUTLASS_DEVICE IgemmSwizzle() {} + + /// Transform a fragment. + CUTLASS_DEVICE void transform(Fragment const& src, Fragment& dst) { + // Expose src/dst as int arrays. + int const* src_int = reinterpret_cast(&src[0]); + int* dst_int = reinterpret_cast(&dst[0]); + + // Transpose the data. + for (int d = 0; d < FragmentShape::kD; ++d) { + for (int h = 0; h < FragmentShape::kH / 4; ++h) { + for (int w = 0; w < ShapeCount::kWc / 4; ++w) { + int const i0 = d * (ShapeCount::kHwc / 4) + + (4 * h + 0) * (ShapeCount::kWc / 4) + w; + int const i1 = d * (ShapeCount::kHwc / 4) + + (4 * h + 1) * (ShapeCount::kWc / 4) + w; + int const i2 = d * (ShapeCount::kHwc / 4) + + (4 * h + 2) * (ShapeCount::kWc / 4) + w; + int const i3 = d * (ShapeCount::kHwc / 4) + + (4 * h + 3) * (ShapeCount::kWc / 4) + w; + + int a0 = src_int[i0]; + int a1 = src_int[i1]; + int a2 = src_int[i2]; + int a3 = src_int[i3]; + + int b0, b1, b2, b3, c0; + asm volatile("prmt.b32 %0, %1, %2, 0x0040;" : "=r"(b0) : "r"(a0), "r"(a1)); + asm volatile("prmt.b32 %0, %1, %2, 0x0040;" : "=r"(c0) : "r"(a2), "r"(a3)); + asm volatile("prmt.b32 %0, %1, %2, 0x5410;" : "=r"(b0) : "r"(b0), "r"(c0)); + + asm volatile("prmt.b32 %0, %1, %2, 0x0051;" : "=r"(b1) : "r"(a0), "r"(a1)); + asm volatile("prmt.b32 %0, %1, %2, 0x0051;" : "=r"(c0) : "r"(a2), "r"(a3)); + asm volatile("prmt.b32 %0, %1, %2, 0x5410;" : "=r"(b1) : "r"(b1), "r"(c0)); + + asm volatile("prmt.b32 %0, %1, %2, 0x0062;" : "=r"(b2) : "r"(a0), "r"(a1)); + asm volatile("prmt.b32 %0, %1, %2, 0x0062;" : "=r"(c0) : "r"(a2), "r"(a3)); + asm volatile("prmt.b32 %0, %1, %2, 0x5410;" : "=r"(b2) : "r"(b2), "r"(c0)); + + asm volatile("prmt.b32 %0, %1, %2, 0x0073;" : "=r"(b3) : "r"(a0), "r"(a1)); + asm volatile("prmt.b32 %0, %1, %2, 0x0073;" : "=r"(c0) : "r"(a2), "r"(a3)); + asm volatile("prmt.b32 %0, %1, %2, 0x5410;" : "=r"(b3) : "r"(b3), "r"(c0)); + + dst_int[i0] = b0; + dst_int[i1] = b1; + dst_int[i2] = b2; + dst_int[i3] = b3; + } + } + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/igemm_traits.h b/cutlass-example/cutlass/gemm/igemm_traits.h new file mode 100644 index 0000000..82f8de5 --- /dev/null +++ b/cutlass-example/cutlass/gemm/igemm_traits.h @@ -0,0 +1,539 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defies structural properties of mixed-precision integer GEMM. Multiplicands are assumed + to be packed 8bit integers, accumulators are assumed to be 32b signed integers, and output + formats vary. +*/ +#pragma once + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The tile size for the GEMM KxNxM. + typename OutputTile_, + /// The output type. + typename ScalarD_, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_> +struct IgemmConfig + : public GemmConfig< + /// The scalar type for A. + int8_t, + /// The scalar type for B. + int8_t, + /// The scalar type for C. + ScalarD_, + /// The scalar type for D. + ScalarD_, + /// The tile size for the GEMM KxNxM. + OutputTile_, + /// The functor to do the math in the main loop. + ThreadMultiplyAdd, int8_t, int8_t, int>, + /// The number of scalars per LDG for A. + 4, + /// The number of scalars per STS for A. + 4, + /// The number of scalars per LDS for A. + 16, + /// The number of scalars per LDG for B. + 4, + /// The number of scalars per STS for B. + 4, + /// The number of scalars per LDS for B. + 16, + /// The number of scalars per LDG for C and STG for D. + 1, + /// The number of scalars per STS for D. + 4, + /// The number of scalars per LDS for D. + 1, + /// The number of stages in shared memory. + 2, + /// Enable the code path that deals with the residue in epilogue. + true> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmConfig + : public GemmConfig< + /// The scalar type for A. + int8_t, + /// The scalar type for B. + int8_t, + /// The scalar type for C. + int8_t, + /// The scalar type for D. + int8_t, + /// The tile size for the GEMM KxNxM. + OutputTile_, + /// The functor to do the math in the main loop. + ThreadMultiplyAdd, int8_t, int8_t, int>, + /// The number of scalars per LDG for A. + 4, + /// The number of scalars per STS for A. + 4, + /// The number of scalars per LDS for A. + 16, + /// The number of scalars per LDG for B. + 4, + /// The number of scalars per STS for B. + 4, + /// The number of scalars per LDS for B. + 16, + /// The number of scalars per LDG for C and STG for D. + 4, + /// The number of scalars per STS for D. + 4, + /// The number of scalars per LDS for D. + 4, + /// The number of stages in shared memory. + 2, + /// Enable the code path that deals with the residue in epilogue. + true> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTileTraitsHelperA : public GemmTileTraitsHelperA {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTileTraitsHelperA + : public GemmTileTraitsHelperA { + /// The base config. + typedef GemmTileTraitsHelperA Base; + + /// The number of scalars per LDG/STS/LDS for A. + static int const kScalarsPerStsA = 16; + + /// The traits class to build the iterator to load data from global memory for A^N. + typedef IgemmGlobalTileTraits< + GemmOperand::kA, + // The layout. + MatrixLayout::kColumnMajor, + // The pointer is float const. + int8_t const, + // The tile has size KxM in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kW>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgA> + GlobalTileTraits; + + // The iterator. + typedef GemmGlobalIteratorAb GlobalLoadIterator; + + /// The traits class to build the iterator to store data to shared memory for A^N. + typedef GemmSharedStoreTileAbTraits< + // The pointer is float. + int8_t, + // The tile has size KxM in GEMM's terminology. + Shape, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + kScalarsPerStsA> + SharedStoreTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTileTraitsHelperA { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kRowMajor; + + /// The input scalar. + typedef int8_t Scalar; + /// The scalar stored in shared memory. + typedef int8_t MultiplyAddScalar; + + /// The number of scalars per LDG/STS/LDS for A. + static int const kScalarsPerStsA = 16; + + /// The traits class to build the iterator to load data from global memory for A^T. + typedef IgemmGlobalTileTraits< + GemmOperand::kA, + // The layout. + MatrixLayout::kRowMajor, + // The pointer is float const. + int8_t const, + // The tile has size NxK in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kW, GemmConfig_::OutputTile::kD>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgA> + GlobalTileTraits; + + // The iterator. + typedef IgemmGlobalIteratorAb GlobalLoadIterator; + + /// The traits class to build the iterator to store data to shared memory for A^N. + typedef GemmSharedStoreWithSkewTileAbTraits< + // The pointer is int8. + int8_t, + // The tile has size KxN in GEMM's terminology. + Shape, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS. + kScalarsPerStsA, + // The skew to avoid bank conflicts added in the tile W dimension. + 16> + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for A^N. + typedef GemmSharedLoadTileATraits< + // The pointer is float const. + int8_t const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + 16, + // The skew. + SharedStoreTileTraits::kSkew> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTileTraitsHelperB : public GemmTileTraitsHelperB {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTileTraitsHelperB { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kColumnMajor; + + /// The input scalar. + typedef int8_t Scalar; + /// The scalar stored in shared memory. + typedef int8_t MultiplyAddScalar; + + /// The number of scalars per LDG/STS/LDS for B. + static int const kScalarsPerStsB = 16; + + /// The traits class to build the iterator to load data from global memory for B^T. + typedef IgemmGlobalTileTraits< + GemmOperand::kB, + // The layout. + MatrixLayout::kColumnMajor, + // The pointer is float const. + int8_t const, + // The tile has size NxK in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kH, GemmConfig_::OutputTile::kD>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgB> + GlobalTileTraits; + + // The iterator. + typedef IgemmGlobalIteratorAb GlobalLoadIterator; + + /// The traits class to build the iterator to store data to shared memory for B^N. + typedef GemmSharedStoreWithSkewTileAbTraits< + // The pointer is int8. + int8_t, + // The tile has size KxN in GEMM's terminology. + Shape, + // The threads are distributed as (threads / K) x K (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS. + kScalarsPerStsB, + // The skew to avoid bank conflicts added in the tile W dimension. + 16> + SharedStoreTileTraits; + + /// The traits class to build the iterator to load from shared memory for B^N. + typedef GemmSharedLoadTileBTraits< + // The pointer is float const. + int8_t const, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The number of threads per warp. + typename GemmConfig_::MultiplyAdd::ThreadsPerWarp, + // The shape of the FMA instruction. + typename GemmConfig_::InstructionShape, + // The number of stages. + GemmConfig_::kStages, + // The number of scalars per LDS. + 16, + // The skew. + SharedStoreTileTraits::kSkew> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTileTraitsHelperB + : public GemmTileTraitsHelperB { + /// The base config. + typedef GemmTileTraitsHelperB Base; + + /// The number of scalars per LDG/STS/LDS for B. + static int const kScalarsPerStsB = 16; + + /// The traits class to build the iterator to load data from global memory for B^T. + typedef IgemmGlobalTileTraits< + GemmOperand::kB, + // The layout. + MatrixLayout::kRowMajor, + // The pointer is float const. + int8_t const, + // The tile has size KxM in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kH>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgB> + GlobalTileTraits; + + // The iterator. + typedef GemmGlobalIteratorAb GlobalLoadIterator; + + /// The traits class to build the iterator to store data to shared memory for B^N. + typedef GemmSharedStoreTileAbTraits< + // The pointer is float. + int8_t, + // The tile has size KxM in GEMM's terminology. + Shape, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + kScalarsPerStsB> + SharedStoreTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTransformerA {}; + +template +struct IgemmTransformerA { + typedef Copy Transformer; +}; + +template +struct IgemmTransformerA { + typedef IgemmSwizzle Transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmTransformerB {}; + +template +struct IgemmTransformerB { + typedef Copy Transformer; +}; + +template +struct IgemmTransformerB { + typedef IgemmSwizzle Transformer; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The output tile. + typename OutputTile_, + /// The output type. + typename ScalarD_, + /// The functor to do the math in the epilogue. + typename EpilogueFunctor_, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_ = Shape<32, 8, 8>, + /// The index. + typename Index_ = int> +struct IgemmTraitsHelper { + /// The IGEMM config. + typedef IgemmConfig GemmConfig; + /// The GEMM config for A. + typedef IgemmTileTraitsHelperA GemmTileTraitsHelperA; + /// The GEMM config for B. + typedef IgemmTileTraitsHelperB GemmTileTraitsHelperB; + + /// The iterator to load A from global memory. + typedef typename GemmTileTraitsHelperA::GlobalLoadIterator GlobalLoadIteratorA; + + /// The default transformer for A. + typedef typename IgemmTransformerA::Transformer GlobalTransformerA; + /// The iterator to store A to shared memory. + typedef TileStoreIterator + SharedStoreIteratorA; + /// The stream to load A from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamA; + + /// The iterator to load B from global memory. + typedef typename GemmTileTraitsHelperB::GlobalLoadIterator GlobalLoadIteratorB; + + // The default transformer for B. + typedef typename IgemmTransformerB::Transformer GlobalTransformerB; + /// The iterator to store B to shared memory. + typedef TileStoreIterator + SharedStoreIteratorB; + /// The stream to load B from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamB; + + /// The iterator to load A from shared memory. + typedef TileLoadIterator + SharedLoadIteratorA; + /// The stream to load A from shared memory. + typedef SharedLoadStream > + SharedLoadStreamA; + /// The iterator to load B from shared memory. + typedef TileLoadIterator + SharedLoadIteratorB; + /// The stream to load B from shared memory. + typedef SharedLoadStream > + SharedLoadStreamB; + + /// The multiply-add functor. + typedef typename GemmConfig::MultiplyAdd MultiplyAdd; + /// The object to clear accumulators. + typedef ClearAccumulators ClearAccumulators; + + /// The epilogue. + typedef IgemmEpilogue > Epilogue; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct IgemmEpilogueScalar { + typedef float Scalar; +}; + +template <> +struct IgemmEpilogueScalar { + typedef int Scalar; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The output tile. + typename OutputTile_ = Shape<32, 128, 128>, + /// The output type. + typename ScalarD_ = int, + /// The functor to do the math in the epilogue. + typename EpilogueFunctor_ = LinearScaling::Scalar>, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_ = Shape<32, 8, 8>, + /// The index. + typename Index_ = int, + /// The helper class. + typename Helper_ = IgemmTraitsHelper > +struct IgemmTraits : public GemmTraits< + // The config. + typename Helper_::GemmConfig, + // The stream to load A from global memory to shared memory. + typename Helper_::GlobalLoadStreamA, + // The stream to load B from global memory to shared memory. + typename Helper_::GlobalLoadStreamB, + // The stream to load A from shared memory. + typename Helper_::SharedLoadStreamA, + // The stream to load B from shared memory. + typename Helper_::SharedLoadStreamB, + // The epilogue. + typename Helper_::Epilogue, + // The block swizzle to reorganize the grid. + IdentityBlockSwizzle, + // The index. + Index_, + // The tool used to clear accumulators. + typename Helper_::ClearAccumulators> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/linear_scaling.h b/cutlass-example/cutlass/gemm/linear_scaling.h new file mode 100644 index 0000000..979c93f --- /dev/null +++ b/cutlass-example/cutlass/gemm/linear_scaling.h @@ -0,0 +1,85 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Implements the BLAS linear scaling function alpha*AB + beta*C +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Functor to compute linear combination of fragments +template > +struct LinearScaling { + // The scalar. + typedef Scalar_ Scalar; + // The adapater. + typedef FragmentMultiplyAdd_ FragmentMultiplyAdd; + + /// The parameters. + struct Params { + /// The alpha/beta scaling params. + Scalar alpha, beta; + + /// Initialize the parameters. + template + CUTLASS_HOST_DEVICE int initialize(GemmDesc_ const& desc) { + alpha = desc.alpha; + beta = desc.beta; + return 0; + } + }; + + /// Ctor. + CUTLASS_DEVICE LinearScaling(Params const& params) : alpha(params.alpha), beta(params.beta) {} + + /// Evaluate the functor. + template + CUTLASS_DEVICE void evaluate(FragmentA_ const& accum, FragmentB_& output) { + FragmentMultiplyAdd mad; + mad.multiply(alpha, accum, output); + } + + /// Evaluate the functor. + template + CUTLASS_DEVICE void evaluate(FragmentA_ const& accum, FragmentB_ const& old, FragmentB_& output) { + FragmentMultiplyAdd mad; + FragmentB_ tmp; + mad.multiply(beta, old, tmp); + mad.multiply_add(alpha, accum, tmp, output); + } + + /// The alpha/beta scaling factors. + Scalar alpha, beta; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/sgemm_traits.h b/cutlass-example/cutlass/gemm/sgemm_traits.h new file mode 100644 index 0000000..66b7677 --- /dev/null +++ b/cutlass-example/cutlass/gemm/sgemm_traits.h @@ -0,0 +1,127 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defies structural properties of single-precision GEMM. +*/ +#pragma once + +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The tile size for the GEMM KxNxM. + typename OutputTile_, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_, + /// The number of scalars per LDG for A. + int kScalarsPerLdgA_ = 1, + /// The number of scalars per LDG for B. + int kScalarsPerLdgB_ = 1> +struct SgemmConfig + : public GemmConfig< + /// The scalar type for A. + float, + /// The scalar type for B. + float, + /// The scalar type for C. + float, + /// The scalar type for D. + float, + /// The tile size for the GEMM KxNxM. + OutputTile_, + /// The functor to do the math in the main loop. + ThreadMultiplyAdd, float, float, float>, + /// The number of scalars per LDG for A. + kScalarsPerLdgA_, + /// The number of scalars per STS for A. + kScalarsPerLdgA_, + /// The number of scalars per LDS for A. + 4, + /// The number of scalars per LDG for B. + kScalarsPerLdgB_, + /// The number of scalars per STS for B. + kScalarsPerLdgB_, + /// The number of scalars per LDS for B. + 4, + /// The number of scalars per LDG for C and STG for D. + 1, + /// The number of scalars per STS for D. + 4, + /// The number of scalars per LDS for D. + 1, + /// The number of stages in shared memory. + 2> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The output tile. + typename OutputTile_ = Shape<8, 128, 128>, + /// The functor to use in the epilogue. + typename EpilogueFunctor_ = LinearScaling, + /// The number of accumulators per thread. + typename AccumulatorsPerThread_ = Shape<8, 8, 8>, + /// The number of floats loaded in one LDG for A. + int kScalarsPerLdgA_ = 1, + /// The number of floats loaded in one LDG for B. + int kScalarsPerLdgB_ = 1, + /// The index. + typename Index_ = int, + /// The SGEMM config. + typename GemmConfig_ = + SgemmConfig, + /// The traits class for the epilogue. + typename GemmEpilogueTraits_ = + SimplifiedGemmEpilogueTraits > +struct SgemmTraits : public SimplifiedGemmTraits< + // The layout for A. + kLayoutA_, + // The layout for B. + kLayoutB_, + // The config. + GemmConfig_, + // The epilogue. + GemmEpilogue, + // The index. + Index_> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/thread_multiply_add.h b/cutlass-example/cutlass/gemm/thread_multiply_add.h new file mode 100644 index 0000000..20dca15 --- /dev/null +++ b/cutlass-example/cutlass/gemm/thread_multiply_add.h @@ -0,0 +1,84 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Template implementing matrix multiply-add operations on fragments. +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Template performing matrix multiply-add operation within a thread +template +struct ThreadMultiplyAdd { + /// The shape of the instruction. + typedef Shape<1, 1, 1, 1> InstructionShape; + /// The number of accumulators per thread. + typedef AccumulatorsPerThread_ AccumulatorsPerThread; + /// The number of threads per warp. + typedef ThreadsPerWarp_ ThreadsPerWarp; + /// The number of accumulators per warp. + typedef typename ShapeMul::Shape AccumulatorsPerWarp; + /// The type for A. + typedef ScalarA_ ScalarA; + /// The fragment for A. + typedef Fragment FragmentA; + /// The type for B. + typedef ScalarB_ ScalarB; + /// The fragment for B. + typedef Fragment FragmentB; + /// The type for C and D. + typedef ScalarC_ ScalarC; + /// The accumulators. + typedef Fragment Accumulators; + + /// Ctor. + CUTLASS_DEVICE ThreadMultiplyAdd() {} + + /// Multiply : d = a*b + c. + CUTLASS_DEVICE void multiply_add(FragmentA const& a, + FragmentB const& b, + Accumulators const& c, + Accumulators& d) { + for (int j = 0; j < AccumulatorsPerThread::kH; ++j) { + for (int i = 0; i < AccumulatorsPerThread::kW; ++i) { + d[j * AccumulatorsPerThread::kW + i] = a[i] * b[j] + c[j * AccumulatorsPerThread::kW + i]; + } + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/wmma_gemm_epilogue_traits.h b/cutlass-example/cutlass/gemm/wmma_gemm_epilogue_traits.h new file mode 100644 index 0000000..0fafacf --- /dev/null +++ b/cutlass-example/cutlass/gemm/wmma_gemm_epilogue_traits.h @@ -0,0 +1,161 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines structural properties of WMMA GEMM's epilogue phase. +*/ +#pragma once + +#include +#ifdef CUTLASS_USE_WMMA_API + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmEpilogueTraitsHelper { + /// The scalar. + typedef typename EpilogueFunctor_::Scalar Scalar; + /// The output tile. + typedef typename GemmConfig_::OutputTile OutputTile; + + /// The number of WMMAs in the H dimension. + static int const kWmmasPerH = + GemmConfig_::AccumulatorsPerWarp::kH / GemmConfig_::InstructionShape::kH; + /// The number of iterations in the epilogue. That's the number of "horizontal" WMMAs. + typedef Shape<1, 1, kWmmasPerH> Iterations; + // The iteration strides in the H/W dimension. + typedef Shape<0, 0, 0> Delta; + /// The functor to do the math in the epilogue. + typedef EpilogueFunctor_ Functor; + + /// The traits class to build the iterator to store to shared memory for D. + typedef WmmaGemmSharedStoreTileDTraits< + // The output layout. + MatrixLayout::kColumnMajor, + // The pointer is float. + typename Functor::Scalar, + // The output tile size. + typename GemmConfig_::OutputTile, + // The number of warps. + typename GemmConfig_::Warps, + // The shape of the instruction. + typename GemmConfig_::InstructionShape> + SharedStoreTileTraits; + + typedef WmmaMatrix + WmmaMatrix; + + /// The iterator to store D to shared memory. + typedef TileStoreIterator + SharedStoreIteratorD; + + /// The shared store transformer for D. + typedef Copy SharedStoreTransformerD; + + /// The traits class to build the iterator to load from shared memory for D. + typedef WmmaGemmSharedLoadTileDTraits< + // The pointer. + typename Functor::Scalar, + // The tile size. + typename SharedStoreIteratorD::Tile, + // The number of threads. + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDS. + GemmConfig_::kScalarsPerLdsD> + SharedLoadTileTraits; + + /// The iterator to load D from shared memory. + typedef TileLoadIterator + SharedLoadIteratorD; + + /// The traits class to build the iterator to load data from global memory for C^N. + typedef WmmaGemmGlobalIteratorCdTraits< + // The pointer is float const. + typename GemmConfig_::ScalarC const, + // The tile has size (N / Iterations)xM in GEMM's terminology. + Shape<1, + GemmConfig_::OutputTile::kH / ShapeCount::kCount, + GemmConfig_::OutputTile::kW>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgC> + GlobalLoadTileTraits; + + /// The iterator to load C. + typedef WmmaGemmGlobalIteratorCd GlobalLoadIteratorC; + /// The transformer for C. + typedef Copy GlobalTransformerC; + + /// The traits class to build the iterator to store data to global memory for D^N. + typedef WmmaGemmGlobalIteratorCdTraits< + // The pointer is float. + typename GemmConfig_::ScalarD, + // The tile has size (N / Iterations)xM in GEMM's terminology. + Shape<1, + GemmConfig_::OutputTile::kH / ShapeCount::kCount, + GemmConfig_::OutputTile::kW>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, ShapeCount::kCount, GemmConfig_::kWarpSize>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerStgD> + GlobalStoreTileTraits; + + /// The iterator to store D. + typedef WmmaGemmGlobalIteratorCd GlobalStoreIteratorD; + /// The transformer for D. + typedef Copy GlobalTransformerD; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass + +#endif // defined CUTLASS_USE_WMMA_API diff --git a/cutlass-example/cutlass/gemm/wmma_gemm_global_tile.h b/cutlass-example/cutlass/gemm/wmma_gemm_global_tile.h new file mode 100644 index 0000000..dbd57f6 --- /dev/null +++ b/cutlass-example/cutlass/gemm/wmma_gemm_global_tile.h @@ -0,0 +1,211 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines tile iterator traits for loading thread block-level tile from global memory. +*/ +#pragma once + +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmGlobalIteratorCdTraits : public GemmGlobalTileTraits { + /// The base class. + typedef GemmGlobalTileTraits + Base; + + /// Override the strides in each dimension between different loads/stores. + typedef Shape<0, 0, Base::Delta::kW, Base::Delta::kC> Delta; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + int thread_offset_h = threadIdx.x / Base::Threads::kW; + int thread_offset_w = threadIdx.x % Base::Threads::kW * Base::ThreadsDelta::kW; + + return make_Coord(0, thread_offset_h, thread_offset_w, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmGlobalIteratorCd : public TileIteratorBase { + /// This class. + typedef WmmaGemmGlobalIteratorCd This_; + /// The traits. + typedef TileTraits_ Traits; + /// The base class. + typedef TileIteratorBase + Base; + /// Override the strides in each dimension between different loads/stores. + typedef Shape<0, 0, Base::Delta::kW, Base::Delta::kC> ImmediateOffsetStrides; + /// The layout. + static MatrixLayout::Kind const kLayout = TileTraits_::kLayout; + + /// The scalar. + typedef typename TileTraits_::Scalar Scalar; + /// The pointer. + typedef typename TileTraits_::Pointer Pointer; + /// The threads. + typedef typename TileTraits_::Threads Threads; + /// The index. + typedef Index_ Index; + /// The thread offset functor. + typedef typename TileTraits_::ThreadOffset ThreadOffset; + + /// The params. + struct Params { + /// The pointer. + Pointer pointer; + /// The stride in the H dimension to setup the thread in the block. + Index stride_h; + /// The strides to increment the pointer. + Index inc_h, inc_advance; + /// The column offset to compute the predicate for the columns. + Index predicate_offset; + /// The strides to increment the predicate offset. + Index predicate_inc_h, predicate_inc_advance; + + /// Setup the params. + CUTLASS_HOST_DEVICE int initialize( + Pointer pointer, Index ld, Index n, Index epilogue_stride_w, Index epilogue_delta_w) { + // The pointer. + this->pointer = pointer; + // Setup the base stride. One "group of threads" per column. + stride_h = ld; + // Each thread output 1 column per iteration. . + inc_h = ld * TileTraits_::Threads::kH; + inc_advance = inc_h + epilogue_stride_w; + + predicate_offset = n; + predicate_inc_h = TileTraits_::Threads::kH; + predicate_inc_advance = predicate_inc_h + epilogue_delta_w; + + // It worked. + return 0; + } + }; + + Params params; + + Coord<4> thread_offset; + + /// Ctor. + CUTLASS_DEVICE WmmaGemmGlobalIteratorCd() {} + + /// Ctor. + CUTLASS_DEVICE WmmaGemmGlobalIteratorCd(Params const& params, + const Coord<3>& bounds, + const Coord<3>& block, + int const pointer_offset = 0, + int const pred_offset = 0, + ThreadOffset thread_offset_func = ThreadOffset()) + + : params(params) { + thread_offset = thread_offset_func(); + // Each warp works on a different column of the tile. + int const h = thread_offset[1] + block[1]; + // Each lane writes a different element. + int const w = thread_offset[2] + block[2]; + // Setup the pointer. + this->params.pointer += ((h * params.stride_h + w) + pointer_offset); + + // Prepare the vector of predicates. + for (int i = 0; i < Base::Iterations::kW; ++i) { + predicates.set(i, w + i * Base::Delta::kW < bounds[2]); + } + this->params.predicate_offset -= (h + pred_offset); + } + + /// The accessor. + CUTLASS_DEVICE void get(typename Base::AccessType& value, int d, int h, int w, int c) const { + int const imm = + ComputeOffsetFromStrides::get(0, 0, w, c); + Load::load(value, params.pointer, imm); + } + + /// Increment the pointer in the C dimension. + CUTLASS_DEVICE void inc_c() {} + /// Increment the pointer in the W dimension. + CUTLASS_DEVICE void inc_w() {} + /// Increment the pointer in the H dimension. + CUTLASS_DEVICE void inc_h() { + params.pointer += params.inc_h; + params.predicate_offset -= params.predicate_inc_h; + } + /// Increment the pointer in the D dimension. + CUTLASS_DEVICE void inc_d() {} + /// Increment the pointer to move to the next iteration. + CUTLASS_DEVICE void inc_advance() { + params.pointer += params.inc_advance; + params.predicate_offset -= params.predicate_inc_advance; + } + + /// The accessor. + CUTLASS_DEVICE void set(typename Base::AccessType const& value, int d, int h, int w, int c) { + int const imm = + ComputeOffsetFromStrides::get(d, h, w, 0); + Store::store( + value, params.pointer, imm); + } + + /// Test the predicate. + CUTLASS_DEVICE bool valid(int d, int h, int w, int c) const { + return predicates.at(w) && params.predicate_offset > 0; + } + + /// The predicates for the row. + cutlass::PredicateVector predicates; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass diff --git a/cutlass-example/cutlass/gemm/wmma_gemm_multiply_add.h b/cutlass-example/cutlass/gemm/wmma_gemm_multiply_add.h new file mode 100644 index 0000000..5968350 --- /dev/null +++ b/cutlass-example/cutlass/gemm/wmma_gemm_multiply_add.h @@ -0,0 +1,108 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Implements warp-level matrix multiply-accumulate operation using CUDA WMMA API. +*/ +#pragma once + +#include +#ifdef CUTLASS_USE_WMMA_API +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmMultiplyAdd { + /// The shape of the instruction. + typedef InstructionShape_ InstructionShape; + /// The number of threads per warp. That's a dummy configuration. + typedef Shape<1, InstructionShape_::kH, InstructionShape_::kW> ThreadsPerWarp; + /// The dimensions. + typedef AccumulatorsPerWarp_ AccumulatorsPerWarp; + /// The type for A. + typedef ScalarA_ ScalarA; + /// The type for B. + typedef ScalarB_ ScalarB; + /// The type for C and D. + typedef ScalarC_ ScalarC; + /// The number of iterations. + typedef typename ShapeDiv::Shape Iterations; + + /// The element for A. + typedef WmmaMatrix ElementA; + /// The fragment for A. + typedef Fragment FragmentA; + + /// The element for B. + typedef WmmaMatrix ElementB; + /// The fragment for B. + typedef Fragment FragmentB; + + /// The element for C. + typedef WmmaMatrix ElementC; + /// The fragment for C. + typedef Fragment Accumulators; + + /// Ctor. + CUTLASS_DEVICE WmmaGemmMultiplyAdd() {} + + /// Multiply : d = a*b. + CUTLASS_DEVICE void multiply_add(FragmentA const& a, + FragmentB const& b, + Accumulators const& c, + Accumulators& d) { + for (int j = 0; j < Iterations::kH; ++j) { + for (int i = 0; i < Iterations::kW; ++i) { + // The input elements. + ElementA const& elt_a = a[i]; + ElementB const& elt_b = b[j]; + ElementC const& elt_c = c[j * Iterations::kW + i]; + + // The output element. + ElementC& elt_d = d[j * Iterations::kW + i]; + + // The wmma instruction. + nvcuda::wmma::mma_sync(elt_d, elt_a, elt_b, elt_c); + } + } + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass + +#endif // defined CUTLASS_USE_WMMA_API diff --git a/cutlass-example/cutlass/gemm/wmma_gemm_shared_tile.h b/cutlass-example/cutlass/gemm/wmma_gemm_shared_tile.h new file mode 100644 index 0000000..7d15b26 --- /dev/null +++ b/cutlass-example/cutlass/gemm/wmma_gemm_shared_tile.h @@ -0,0 +1,240 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines iterator traits for efficiently loading and storing fragment to and from shared + memory, specialized for WMMA GEMM. +*/ +#pragma once + +#include +#ifdef CUTLASS_USE_WMMA_API + +#include +#include + +namespace cutlass { +namespace gemm { + +template +struct Debug {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmSharedLoadTileATraits { + /// The operand. + static GemmOperand::Kind const kOperand = GemmOperand::kA; + /// The layout. + static MatrixLayout::Kind const kLayout = kLayout_; + /// The scalar. + typedef Scalar_ Scalar; + /// The pointer. + typedef Scalar const* Pointer; + /// The access size + static int const kAccessSize = 1; + /// The tile with skew. + typedef Tile_ Tile; + /// The number of warps. + typedef Warps_ Warps; + /// The warps strides. + static int const kWarpStride = kWarpStride_; + /// The number of iterations. + typedef Iterations_ Iterations; + /// The strides between iterations. + typedef Delta_ Delta; + /// The strides between iterations. + typedef Delta_ ImmediateOffsetStrides; + /// The shape of the WMMA instruction. + typedef WmmaShape_ WmmaShape; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + /// ThreadOffset + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + // The warp id. + int const warp = threadIdx.x / kWarpSize; + // The offset. + int const offset = warp % Warps::kW * kWarpStride; + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmSharedLoadTileBTraits { + /// The operand. + static GemmOperand::Kind const kOperand = GemmOperand::kB; + /// The layout. + static MatrixLayout::Kind const kLayout = kLayout_; + /// The scalar. + typedef Scalar_ Scalar; + /// The pointer. + typedef Scalar const* Pointer; + /// The access size + static int const kAccessSize = 1; + /// The tile with skew. + typedef Tile_ Tile; + /// The number of warps. + typedef Warps_ Warps; + /// The warps strides. + static int const kWarpStride = kWarpStride_; + /// The number of iterations. + typedef Iterations_ Iterations; + /// The strides between iterations. + typedef Delta_ Delta; + /// The strides between iterations. + typedef Delta_ ImmediateOffsetStrides; + /// The shape of the WMMA instruction. + typedef WmmaShape_ WmmaShape; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + /// ThreadOffset + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + // The warp id. + int const warp = threadIdx.x / kWarpSize; + // The offset. + int const offset = warp / Warps::kW * kWarpStride; + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmSharedStoreTileDTraits { + /// The operand. + static GemmOperand::Kind const kOperand = GemmOperand::kC; + /// The layout. + static MatrixLayout::Kind const kLayout = kLayout_; + /// The scalar. + typedef Scalar_ Scalar; + // The access size + static int const kAccessSize = 1; + /// The pointer. + typedef Scalar* Pointer; + /// The number of warps. + typedef Warps_ Warps; + /// The shape of the WMMA instruction. + typedef WmmaShape_ WmmaShape; + /// The skew. + static int const kSkew = kSkew_; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + /// The tile with skew. + typedef Shape<1, Warps_::kH * WmmaShape_::kH, OutputTile_::kW + kSkew_> Tile; + /// The number of iterations needed to store the tile. + typedef Shape<1, 1, OutputTile_::kW / Warps::kW / WmmaShape_::kW> Iterations; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, 0, Warps::kW * WmmaShape_::kW, 0> Delta; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, 0, Warps::kW * WmmaShape_::kW, 0> ImmediateOffsetStrides; + + /// ThreadOffset + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + // The warp id. + int const warp = threadIdx.x / kWarpSize; + // The starting column. + int const h = warp / Warps::kW * WmmaShape::kH; + // The w. + int const w = warp % Warps::kW * WmmaShape::kW; + // The offset. + int const offset = h * Tile::kW + w; + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmSharedLoadTileDTraits { + /// The scalar. + typedef Scalar_ Scalar; + /// The pointer. + typedef Scalar const* Pointer; + /// The access size + static int const kAccessSize = kScalarsPerLds_; + /// The tile. + typedef typename ReshapeTile::Tile Tile; + /// The threads. + typedef typename ReshapeThreads::Threads Threads; + /// The threads strides. + typedef Shape<1, Tile::kW * Tile::kC, Tile::kC> ThreadsStrides; + /// The memory space. + static MemorySpace::Kind const kMemorySpace = MemorySpace::kShared; + + /// The strides in each dimension between different loads/stores. + typedef Shape<0, Threads::kH * ShapeCount::kWc, Threads::kW * kScalarsPerLds_> Delta; + /// The strides in each dimension between different loads/stores. + typedef Shape<0, Threads::kH * ShapeCount::kWc, Threads::kW * kScalarsPerLds_> + ImmediateOffsetStrides; + /// The number of iterations needed to load/store the tile. + typedef Shape<1, Tile::kH / Threads::kH, Tile::kW / Threads::kW, Tile::kC / kScalarsPerLds_> + Iterations; + + /// ThreadOffset + struct ThreadOffset { + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + // The offset. + int const offset = ComputeThreadOffsetFromStrides::get(); + return make_Coord(0, 0, offset, 0); + } + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass + +#endif // defined CUTLASS_USE_WMMA_API diff --git a/cutlass-example/cutlass/gemm/wmma_gemm_traits.h b/cutlass-example/cutlass/gemm/wmma_gemm_traits.h new file mode 100644 index 0000000..7901201 --- /dev/null +++ b/cutlass-example/cutlass/gemm/wmma_gemm_traits.h @@ -0,0 +1,574 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defies structural properties of GEMM targeting WMMA API in CUDA. +*/ +#pragma once + +#include +#ifdef CUTLASS_USE_WMMA_API + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace cutlass { +namespace gemm { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The tile size for the GEMM KxNxM. + typename OutputTile_, + /// The output type. + typename ScalarC_, + /// The accumulator type. + typename Accumulator_, + /// The number of accumulators per warp. + typename AccumulatorsPerWarp_, + /// The shape of the WMMA instruction. + typename InstructionShape_, + /// The number of scalars per LDG for A. + int kScalarsPerLdgA_, + /// The number of scalars per LDG for B. + int kScalarsPerLdgB_> +struct WmmaGemmConfig : public GemmConfig< + /// The scalar type for A. + half, + /// The scalar type for B. + half, + /// The scalar type for C. + ScalarC_, + /// The scalar type for D. + ScalarC_, + /// The tile size for the GEMM KxNxM. + OutputTile_, + /// The functor to do the math in the main loop. + WmmaGemmMultiplyAdd, + /// The number of scalars per LDG for A. + kScalarsPerLdgA_, + /// The number of scalars per STS for A. + kScalarsPerLdgA_, + /// The number of scalars per LDS for A. + 8, + /// The number of scalars per LDG for B. + kScalarsPerLdgB_, + /// The number of scalars per STS for B. + kScalarsPerLdgB_, + /// The number of scalars per LDS for B. + 8, + /// The number of scalars per LDG for C and STG for D. + 16 / sizeof(ScalarC_), + /// The number of scalars per STS for D. + 16 / sizeof(ScalarC_), + /// The number of scalars per LDS for D. + 16 / sizeof(ScalarC_), + /// The number of stages in shared memory. + 1> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmTileTraitsHelperA {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmTileTraitsHelperA + : public GemmTileTraitsHelperA { + /// The base config. + typedef GemmTileTraitsHelperA Base; + + /// The skew. + static int const kSkew = 16 / sizeof(typename Base::MultiplyAddScalar); + /// The shared tile size. + typedef Shape + Tile; + + /// WMMA matrix + typedef WmmaMatrix + WmmaMatrix; + + /// The traits class to build the iterator to store data to shared memory for A^N. + typedef GemmSharedStoreTileAbTraits< + // The pointer. + typename Base::MultiplyAddScalar, + // The tile has size KxM in GEMM's terminology. + Tile, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename Base::GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + GemmConfig_::kScalarsPerStsA> + SharedStoreTileTraits; + + /// The number of elements loaded in one LDG. + static int const kScalarsPerW = GemmConfig_::InstructionShape::kW * GemmConfig_::Warps::kW; + /// The number of scalars loaded per iteration. + static int const kScalarsPerIteration = Tile::kW * GemmConfig_::InstructionShape::kD; + /// The traits class to build the iterator to load from shared memory for A. + typedef WmmaGemmSharedLoadTileATraits< + // The layout of the matrix. + MatrixLayout::kColumnMajor, + // The pointer. + typename Base::MultiplyAddScalar, + // The output tile size. + Tile, + // The number of warps. + typename GemmConfig_::Warps, + // The strides between warps. + GemmConfig_::InstructionShape::kW, + // The number of iterations to load the data. + Shape<1, 1, GemmConfig_::OutputTile::kW / kScalarsPerW>, + // The stride between iterations. + Shape, + // The shape of the instruction. + typename GemmConfig_::InstructionShape> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmTileTraitsHelperA { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kRowMajor; + + /// The input scalar. + typedef typename GemmConfig_::ScalarA Scalar; + /// The scalar stored in shared memory. + typedef typename GemmConfig_::MultiplyAdd::ScalarA MultiplyAddScalar; + + /// WMMA matrix + typedef WmmaMatrix + WmmaMatrix; + + /// The traits class to build the iterator to load data from global memory for A^T. + typedef GemmGlobalTileTraits< + // That's A. + GemmOperand::kA, + // A is row-major. + MatrixLayout::kRowMajor, + // The pointer is float const. + Scalar const, + // The tile has size KxM in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kW, GemmConfig_::OutputTile::kD>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, GemmConfig_::kThreads / GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kD>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgA> + GlobalTileTraits; + + /// The skew. + static int const kSkew = 16 / sizeof(MultiplyAddScalar); + /// The tile. + typedef Shape + Tile; + + /// The traits class to build the iterator to store data to shared memory for A^N. + typedef GemmSharedStoreTileAbTraits< + // The pointer. + MultiplyAddScalar, + // The tile has size KxM in GEMM's terminology. + Tile, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + GemmConfig_::kScalarsPerStsA> + SharedStoreTileTraits; + + /// The number of elements loaded in one LDG. + static int const kScalarsPerW = GemmConfig_::InstructionShape::kW * GemmConfig_::Warps::kW; + /// The traits class to build the iterator to load from shared memory for A. + typedef WmmaGemmSharedLoadTileATraits< + // The layout of the matrix. + MatrixLayout::kRowMajor, + // The pointer. + MultiplyAddScalar, + // The tile in shared memory. + Tile, + // The number of warps. + typename GemmConfig_::Warps, + // The strides between warps. + GemmConfig_::InstructionShape::kW * Tile::kW, + // The number of iterations to load the data. + Shape<1, 1, GemmConfig_::OutputTile::kW / kScalarsPerW>, + // The stride between iterations. + Shape, + // The shape of the instruction. + typename GemmConfig_::InstructionShape> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmTileTraitsHelperB {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmTileTraitsHelperB + : public GemmTileTraitsHelperB { + /// The base config. + typedef GemmTileTraitsHelperB Base; + + /// The skew. + static int const kSkew = 16 / sizeof(typename Base::MultiplyAddScalar); + /// The shared tile size. + typedef Shape + Tile; + + /// WMMA matrix + typedef WmmaMatrix + WmmaMatrix; + + /// The traits class to build the iterator to store data to shared memory for B^T. + typedef GemmSharedStoreTileAbTraits< + // The pointer. + typename Base::MultiplyAddScalar, + // The tile has size KxM in GEMM's terminology. + Tile, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename Base::GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + GemmConfig_::kScalarsPerStsB> + SharedStoreTileTraits; + + /// The number of elements loaded in one LDG. + static int const kScalarsPerW = GemmConfig_::InstructionShape::kH * GemmConfig_::Warps::kH; + /// The number of scalars loaded per iteration. + static int const kScalarsPerIteration = Tile::kW * GemmConfig_::InstructionShape::kD; + /// The traits class to build the iterator to load from shared memory for B. + typedef WmmaGemmSharedLoadTileBTraits< + // The layout of the matrix. + MatrixLayout::kRowMajor, + // The pointer. + typename Base::MultiplyAddScalar, + // The output tile size. + Tile, + // The number of warps. + typename GemmConfig_::Warps, + // The strides between warps. + GemmConfig_::InstructionShape::kH, + // The number of iterations to load the data. + Shape<1, 1, GemmConfig_::OutputTile::kH / kScalarsPerW>, + // The stride between iterations. + Shape, + // The shape of the instruction. + typename GemmConfig_::InstructionShape> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct WmmaGemmTileTraitsHelperB { + /// The layout. + static MatrixLayout::Kind const kLayout = MatrixLayout::kColumnMajor; + + /// The input scalar. + typedef typename GemmConfig_::ScalarB Scalar; + /// The scalar stored in shared memory. + typedef typename GemmConfig_::MultiplyAdd::ScalarB MultiplyAddScalar; + + /// WMMA matrix + typedef WmmaMatrix + WmmaMatrix; + + /// The traits class to build the iterator to load data from global memory for B^N. + typedef GemmGlobalTileTraits< + // That's B. + GemmOperand::kB, + // A is row-major. + MatrixLayout::kColumnMajor, + // The pointer is float const. + Scalar const, + // The tile has size KxM in GEMM's terminology. + Shape<1, GemmConfig_::OutputTile::kH, GemmConfig_::OutputTile::kD>, + // The threads are distributed as warps x 32 (the traits may reorganize). + Shape<1, GemmConfig_::kThreads / GemmConfig_::OutputTile::kD, GemmConfig_::OutputTile::kD>, + // The number of scalars per LDG (LDG.32 or LDG.128, etc). + GemmConfig_::kScalarsPerLdgB> + GlobalTileTraits; + + /// The skew. + static int const kSkew = 16 / sizeof(MultiplyAddScalar); + /// The tile. + typedef Shape + Tile; + + /// The traits class to build the iterator to store data to shared memory for B^N. + typedef GemmSharedStoreTileAbTraits< + // The pointer. + MultiplyAddScalar, + // The tile has size KxM in GEMM's terminology. + Tile, + // The threads are distributed as warps x 32 (the traits may reorganize). + typename GlobalTileTraits::Threads, + // The number of scalars per STS (STS.32 or STS.128, etc). + GemmConfig_::kScalarsPerStsB> + SharedStoreTileTraits; + + /// The number of elements loaded in one LDG. + static int const kScalarsPerW = GemmConfig_::InstructionShape::kH * GemmConfig_::Warps::kH; + /// The traits class to build the iterator to load from shared memory for B. + typedef WmmaGemmSharedLoadTileBTraits< + // The layout of the matrix. + MatrixLayout::kColumnMajor, + // The pointer. + MultiplyAddScalar, + // The tile in shared memory. + Tile, + // The number of warps. + typename GemmConfig_::Warps, + // The strides between warps. + GemmConfig_::InstructionShape::kH * Tile::kW, + // The number of iterations to load the data. + Shape<1, 1, GemmConfig_::OutputTile::kH / kScalarsPerW>, + // The stride between iterations. + Shape, + // The shape of the instruction. + typename GemmConfig_::InstructionShape> + SharedLoadTileTraits; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The output tile. + typename OutputTile_, + /// The output type. + typename ScalarC_, + /// The accumulator type. + typename Accumulator_, + /// The functor to do the math in the epilogue. + typename EpilogueFunctor_, + /// The number of accumulators per warp. + typename AccumulatorsPerWarp_, + /// The shape of the WMMA instruction. + typename InstructionShape_, + /// The number of halfs loaded in one LDG for A. + int kScalarsPerLdgA_, + /// The number of halfs loaded in one LDG for B. + int kScalarsPerLdgB_, + /// The index. + typename Index_> +struct WmmaGemmTraitsHelper { + /// The WMMA GEMM config. + typedef WmmaGemmConfig + GemmConfig; + + /// The GEMM config for A. + typedef WmmaGemmTileTraitsHelperA GemmTileTraitsHelperA; + /// The GEMM config for B. + typedef WmmaGemmTileTraitsHelperB GemmTileTraitsHelperB; + + /// The iterator to load A from global memory. + typedef GemmGlobalIteratorAb + GlobalLoadIteratorA; + /// The default transformer for A. + typedef Copy GlobalTransformerA; + /// The iterator to store A to shared memory. + typedef TileStoreIterator + SharedStoreIteratorA; + /// The stream to load A from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamA; + + /// The iterator to load B from global memory. + typedef GemmGlobalIteratorAb + GlobalLoadIteratorB; + // The default transformer for B. + typedef Copy GlobalTransformerB; + /// The iterator to store B to shared memory. + typedef TileStoreIterator + SharedStoreIteratorB; + /// The stream to load B from global memory to shared memory. + typedef GlobalLoadStream + GlobalLoadStreamB; + + /// The iterator to load A from shared memory. + typedef TileLoadIterator + SharedLoadIteratorA; + /// The stream to load A from shared memory. + typedef SharedLoadStream SharedLoadStreamA; + /// The iterator to load B from shared memory. + typedef TileLoadIterator + SharedLoadIteratorB; + /// The stream to load B from shared memory. + typedef SharedLoadStream SharedLoadStreamB; + + /// The functor to do the multiply-add in the main loop. + typedef typename GemmConfig::MultiplyAdd MultiplyAdd; + /// The object to clear accumulators. + typedef ClearAccumulators ClearAccumulators; + + /// The helper to create the epilogue traits. + typedef WmmaGemmEpilogueTraitsHelper EpilogueTraitsHelper; + /// The traits class for the epilogue. + typedef SimplifiedGemmEpilogueTraits + GemmEpilogueTraits; + /// The epilogue. + typedef GemmEpilogue Epilogue; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template > +struct WmmaGemmAccumulatorsPerWarp { + typedef typename ShapeMin::Shape Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template < + /// The layout for A. + MatrixLayout::Kind kLayoutA_, + /// The layout for B. + MatrixLayout::Kind kLayoutB_, + /// The tile size for the GEMM KxNxM. + typename OutputTile_ = Shape<64, 128, 128>, + /// The output type. + typename ScalarC_ = float, + /// The functor to do the math in the epilogue. + typename EpilogueFunctor_ = LinearScaling, + /// The accumulator type. + typename Accumulator_ = ScalarC_, + /// The number of accumulators per warp. + typename AccumulatorsPerWarp_ = typename WmmaGemmAccumulatorsPerWarp::Shape, + /// The shape of the WMMA instruction. + typename InstructionShape_ = Shape<16, 16, 16>, + /// The number of scalars per LDG for A. + int kScalarsPerLdgA_ = 8, + /// The number of scalars per LDG for B. + int kScalarsPerLdgB_ = 8, + /// The index. + typename Index_ = int, + /// The helper class. + typename Helper_ = WmmaGemmTraitsHelper > +struct WmmaGemmTraits : public GemmTraits< + // The config. + typename Helper_::GemmConfig, + // The stream to load A from global memory to shared memory. + typename Helper_::GlobalLoadStreamA, + // The stream to load B from global memory to shared memory. + typename Helper_::GlobalLoadStreamB, + // The stream to load A from shared memory. + typename Helper_::SharedLoadStreamA, + // The stream to load B from shared memory. + typename Helper_::SharedLoadStreamB, + // The epilogue. + typename Helper_::Epilogue, + // The block swizzle to reorganize the grid. + IdentityBlockSwizzle, + // The index. + Index_, + // The tool used to clear accumulators. + typename Helper_::ClearAccumulators> {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace gemm +} // namespace cutlass + +#endif // defined CUTLASS_USE_WMMA_API diff --git a/cutlass-example/cutlass/iterator_access.h b/cutlass-example/cutlass/iterator_access.h new file mode 100644 index 0000000..e94beb7 --- /dev/null +++ b/cutlass-example/cutlass/iterator_access.h @@ -0,0 +1,318 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Free functions for loading and storing to implementations of tile iteartor concepts. +*/ +#pragma once + +#include +#include +#include +#include + +namespace cutlass { + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Loads a fragment from an input iterator +template +CUTLASS_HOST_DEVICE void iterator_load(InputIterator &iterator, Fragment &fragment) { + typename InputIterator::FragmentIterator frag_iterator(fragment); + for (int d = 0; d < InputIterator::Iterations::kD; ++d) { + for (int h = 0; h < InputIterator::Iterations::kH; ++h) { + for (int w = 0; w < InputIterator::Iterations::kW; ++w) { + for (int c = 0; c < InputIterator::Iterations::kC; ++c) { + if (iterator.valid(d, h, w, c)) { + iterator.get(reinterpret_cast( + frag_iterator.at(d, h, w, c)), + d, + h, + w, + c); + } + } + if (w < InputIterator::Iterations::kW - 1) { + iterator.inc_w(); + } + } + if (h < InputIterator::Iterations::kH - 1) { + iterator.inc_h(); + } + } + if (d < InputIterator::Iterations::kD - 1) { + iterator.inc_d(); + } + } + iterator.inc_advance(); +} + +/// Loads a fragment from a shared memory input iterator +template +CUTLASS_DEVICE void shared_iterator_load(InputIterator &iterator, Fragment &fragment) { + typename InputIterator::FragmentIterator frag_iterator(fragment); + for (int d = 0; d < InputIterator::Iterations::kD; ++d) { + for (int h = 0; h < InputIterator::Iterations::kH; ++h) { + for (int w = 0; w < InputIterator::Iterations::kW; ++w) { + for (int c = 0; c < InputIterator::Iterations::kC; ++c) { + int const offset = + ComputeOffsetFromStrides::get( + d, h, w, c); + + FragmentLoad::load(frag_iterator.at(d, h, w, c), + iterator.data(), + offset); + } + } + } + } +} + +/// Loads a fragment from a shared memory input iterator +template +CUTLASS_DEVICE void shared_iterator_load(InputIterator &iterator, Fragment &fragment, int d) { + typename InputIterator::FragmentIterator frag_iterator(fragment); + for (int h = 0; h < InputIterator::Iterations::kH; ++h) { + for (int w = 0; w < InputIterator::Iterations::kW; ++w) { + for (int c = 0; c < InputIterator::Iterations::kC; ++c) { + int const offset = + ComputeOffsetFromStrides::get( + d, h, w, c); + + FragmentLoad::load(frag_iterator.at(0, h, w, c), + iterator.data(), + offset); + } + } + } +} + +/// Loads a fragment from an input iterator, masked by a predicate iterator +template +CUTLASS_HOST_DEVICE void iterator_load_post_increment(InputIterator &iterator, + Fragment &fragment, + typename InputIterator::Index offset, + ConstPredicateAdapter predicate_adapter) { + for (int d = 0; d < InputIterator::Iterations::kD; ++d, iterator.inc_d()) { + for (int h = 0; h < InputIterator::Iterations::kH; ++h, iterator.inc_h()) { + for (int w = 0; w < InputIterator::Iterations::kW; ++w, iterator.inc_w()) { + if (predicate_adapter.at(d, h, w, 0)) { + int idx = InputIterator::Tile::kC * + (w + InputIterator::Iterations::kW * (h + InputIterator::Iterations::kH * d)); + + Load:: + load(reinterpret_cast(fragment[idx]), + iterator.data(), + offset); + } + } + } + } +} + +/// Loads a fragment from an input iterator +template +CUTLASS_HOST_DEVICE void iterator_load_post_increment(InputIterator &iterator, + Fragment &fragment, + typename InputIterator::Index offset = 0) { + TrivialPredicateTileAdapter pred; + iterator_load_post_increment(iterator, fragment, offset, pred); +} + +/// Loads a fragment from an input iterator +template +CUTLASS_HOST_DEVICE void iterator_load_post_increment(InputIterator &iterator, + Fragment &fragment, + ConstPredicateAdapter pred_it) { + iterator_load_post_increment(iterator, fragment, 0, pred_it); +} + +template +CUTLASS_HOST_DEVICE void iterator_load(InputIterator const &_iterator, + Fragment &fragment, + typename InputIterator::Index offset, + ConstPredicateAdapter predicate_adapter) { + InputIterator iterator(_iterator); + iterator_load_post_increment(iterator, fragment, offset, predicate_adapter); +} + +/// Loads a fragment from an input iterator +template +CUTLASS_HOST_DEVICE void iterator_load(InputIterator const &iterator, + Fragment &fragment, + typename InputIterator::Index offset = 0) { + TrivialPredicateTileAdapter pred; + iterator_load(iterator, fragment, offset, pred); +} + +/// Loads a fragment from an input iterator +template +CUTLASS_HOST_DEVICE void iterator_load(InputIterator const &iterator, + Fragment &fragment, + ConstPredicateAdapter pred_it) { + iterator_load(iterator, fragment, 0, pred_it); +} + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Stores a fragment to an output iterator +template +CUTLASS_HOST_DEVICE void iterator_store(OutputIterator &iterator, Fragment &fragment) { + typename OutputIterator::FragmentIterator frag_iterator(fragment); + for (int d = 0; d < OutputIterator::Iterations::kD; ++d) { + for (int h = 0; h < OutputIterator::Iterations::kH; ++h) { + for (int w = 0; w < OutputIterator::Iterations::kW; ++w) { + if (iterator.valid(d, h, w, 0)) { + iterator.set(reinterpret_cast( + frag_iterator.at(d, h, w, 0)), + d, + h, + w, + 0); + } + if (w < OutputIterator::Iterations::kW - 1) { + iterator.inc_w(); + } + } + if (h < OutputIterator::Iterations::kH - 1) { + iterator.inc_h(); + } + } + if (d < OutputIterator::Iterations::kD - 1) { + iterator.inc_d(); + } + } + iterator.inc_advance(); +} + +/// Stores a fragment to a shared memory output iterator +template +CUTLASS_DEVICE void shared_iterator_store(OutputIterator &iterator, Fragment const &fragment) { + typename OutputIterator::FragmentConstIterator frag_iterator(fragment); + for (int d = 0; d < OutputIterator::Iterations::kD; ++d) { + for (int h = 0; h < OutputIterator::Iterations::kH; ++h) { + for (int w = 0; w < OutputIterator::Iterations::kW; ++w) { + for (int c = 0; c < OutputIterator::Iterations::kC; ++c) { + int const offset = + ComputeOffsetFromStrides::get( + d, h, w, c); + + FragmentStore::store(frag_iterator.at(d, h, w, c), + iterator.data(), + offset); + } + } + } + } +} + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Stores a fragment to an output iterator, masked by a predicate iterator +template +CUTLASS_HOST_DEVICE void iterator_store_post_increment(OutputIterator &iterator, + Fragment const &fragment, + typename OutputIterator::Index offset, + ConstPredicateAdapter predicate_adapter) { + for (int d = 0; d < OutputIterator::Iterations::kD; ++d, iterator.inc_d()) { + for (int h = 0; h < OutputIterator::Iterations::kH; ++h, iterator.inc_h()) { + for (int w = 0; w < OutputIterator::Iterations::kW; ++w, iterator.inc_w()) { + if (predicate_adapter.at(d, h, w, 0)) { + int idx = OutputIterator::Tile::kC * + (w + OutputIterator::Iterations::kW * (h + OutputIterator::Iterations::kH * d)); + + Store:: + store(reinterpret_cast(fragment[idx]), + iterator.data(), + offset); + } + } + } + } +} + +/// Stores a fragment to an output iterator +template +CUTLASS_HOST_DEVICE void iterator_store_post_increment(OutputIterator &iterator, + Fragment const &fragment, + typename OutputIterator::Index offset = 0) { + TrivialPredicateTileAdapter pred; + iterator_store_post_increment(iterator, fragment, offset, pred); +} + +/// Stores a fragment to an output iterator +template +CUTLASS_HOST_DEVICE void iterator_store_post_increment(OutputIterator &iterator, + Fragment const &fragment, + ConstPredicateAdapter pred_it) { + iterator_store_post_increment(iterator, fragment, 0, pred_it); +} + +/// Stores a fragment to an output iterator, masked by a predicate iterator +template +CUTLASS_HOST_DEVICE void iterator_store(OutputIterator const &_iterator, + Fragment const &fragment, + typename OutputIterator::Index offset, + ConstPredicateAdapter predicate_adapter) { + OutputIterator iterator(_iterator); + iterator_store_post_increment(iterator, fragment, offset, predicate_adapter); +} + +/// Stores a fragment to an output iterator +template +CUTLASS_HOST_DEVICE void iterator_store(OutputIterator const &iterator, + Fragment const &fragment, + typename OutputIterator::Index offset = 0) { + TrivialPredicateTileAdapter pred; + iterator_store(iterator, fragment, offset, pred); +} + +/// Stores a fragment to an output iterator +template +CUTLASS_HOST_DEVICE void iterator_store(OutputIterator const &iterator, + Fragment const &fragment, + ConstPredicateAdapter pred_it) { + iterator_store(iterator, fragment, 0, pred_it); +} + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/load_store.h b/cutlass-example/cutlass/load_store.h new file mode 100644 index 0000000..5cb5eb6 --- /dev/null +++ b/cutlass-example/cutlass/load_store.h @@ -0,0 +1,222 @@ +/*************************************************************************************************** + * Copyright (c) 2017, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines abstractions for efficiently loading and storing vectors to memory. +*/ +#pragma once + +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Enum to specify which memory space data resides in. +*/ +struct MemorySpace { + enum Kind { + kGeneric, // Data accessed through pointer dereferencing + kShared, // Data resides in shared memory + kGlobal // Data resides in global memory + }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template 1), + size_t = (sizeof(Scalar_) * Lanes_)> +struct Load { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The load function. + static CUTLASS_DEVICE void load(AccessType& dst, Scalar_ const* pointer, int offset) { + dst = reinterpret_cast(&pointer[offset])[0]; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Load { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void load(AccessType& dst, Scalar_ const* pointer, int offset) { + dst.registers[0] = reinterpret_cast(&pointer[offset])[0]; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Load { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void load(AccessType& dst, Scalar_ const* pointer, int offset) { + uint2 tmp = reinterpret_cast(&pointer[offset])[0]; + dst.registers[0] = tmp.x; + dst.registers[1] = tmp.y; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Load { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void load(AccessType& dst, double const* pointer, int offset) { + double2 tmp = reinterpret_cast(&pointer[offset])[0]; + dst[0] = tmp.x; + dst[1] = tmp.y; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +#if defined(__CUDACC_VERSION_MAJOR) && __CUDACC_VERSION_MAJOR < 10 +// WAR bug in NVCC where the upper and lower half of the register end up being the same +template +struct Load { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void load(AccessType& dst, half const* pointer, int offset) { + int2 tmp = reinterpret_cast(&pointer[offset])[0]; + dst.registers[0] = tmp.x; + dst.registers[1] = tmp.y; + + tmp = reinterpret_cast(&pointer[offset + 4])[0]; + dst.registers[2] = tmp.x; + dst.registers[3] = tmp.y; + } +}; + +#endif + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Load { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void load(AccessType& dst, Scalar_ const* pointer, int offset) { + uint4 tmp = reinterpret_cast(&pointer[offset])[0]; + dst.registers[0] = tmp.x; + dst.registers[1] = tmp.y; + dst.registers[2] = tmp.z; + dst.registers[3] = tmp.w; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template 1), + size_t = (sizeof(Scalar_) * Lanes_)> +struct Store { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void store(AccessType const& src, Scalar_* pointer, int offset) { + pointer[offset] = src; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Store { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void store(AccessType const& src, Scalar_* pointer, int offset) { + uint32_t* addr = reinterpret_cast(&pointer[offset]); + addr[0] = src.registers[0]; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Store { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void store(AccessType const& src, Scalar_* pointer, int offset) { + uint2* addr = reinterpret_cast(&pointer[offset]); + addr[0] = make_uint2(src.registers[0], src.registers[1]); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Store { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void store(AccessType const& src, double* pointer, int offset) { + double2* addr = reinterpret_cast(&pointer[offset]); + addr[0] = make_double2(src[0], src[1]); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Store { + /// The output type. + typedef typename Vectorize::Type AccessType; + + /// The store function. + static CUTLASS_DEVICE void store(AccessType const& src, Scalar_* pointer, int offset) { + uint4* addr = reinterpret_cast(&pointer[offset]); + addr[0] = make_uint4(src.registers[0], src.registers[1], src.registers[2], src.registers[3]); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/matrix_traits.h b/cutlass-example/cutlass/matrix_traits.h new file mode 100644 index 0000000..77e8b70 --- /dev/null +++ b/cutlass-example/cutlass/matrix_traits.h @@ -0,0 +1,48 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines properties of matrices used to denote layout and operands to GEMM kernels. +*/ +#pragma once + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Describes layouts of matrices +struct MatrixLayout { + enum Kind { kRowMajor, kColumnMajor }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Gemm operand - D = A * B + C +struct GemmOperand { + enum Kind { kA, kB, kC, kD }; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/predicate_vector.h b/cutlass-example/cutlass/predicate_vector.h new file mode 100644 index 0000000..8166857 --- /dev/null +++ b/cutlass-example/cutlass/predicate_vector.h @@ -0,0 +1,493 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines container classes and iterators for managing a statically sized vector + of boolean predicates. +*/ +#pragma once + +#include + +#include +#include + +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup predicate_vector_concept Predicate Vector Concept +@{ + +Implementations of \ref predicate_vector_concept contain an ordered set of boolean predicates which +may be used as conditionals in other device-side operations. Both random access and iterators +offering sequential access are provided. + +@par Predicate Vector + A \ref predicate_vector_concept satisfies the following expressions + - at(int idx) - returns the value of the indexed predicate + - set(int idx, bool value) - sets the value of the indexed predicate + - begin() - returns a \ref predicate_iterator_concept pointing to the first predicate + +@} +*/ + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup predicate_iterator_concept Predicate Iterator Concept +@{ + +Implementations of \ref predicate_iterator_concept enables accessing and traversing elements of a +bit vector. + +@par Const Predicate Iterator + A const \ref predicate_iterator_concept satisfies the following expressions + - ++it increments the iterator to the next predicate + - *it returns the value of the currently pointed-to predicate + +@par Mutable Predicate Iterator + A \ref predicate_iterator_concept that is non-const also satisfies the following expressions + - it.set(bool value) sets the value of the currently pointed-to predicate + +@} +*/ + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup predicate_tile_adapter Predicate Tile Adapter Concept +@{ + +Implementations of \ref predicate_tile_adapter provide a mapping between a the elements of a \ref +tile_traits_concept and a \ref predicate_vector_concept. + +@par Predicate Tile Adapter + A \ref predicate_tile_adapter satisfies the following expressions + - at(int d, int h, int w, int c) - returns the value of a predicate corresponding to the + access (d, h, w, c) within the tile. + +@} +*/ + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Statically sized array of bits implementing @concept{predicate_vector_concept}. +template < + /// Number of predicates conatined in predicate vector + int kPredicates_, + /// Number of predicates contained in each byte of internal storage + int kPredicatesPerByte_ = 4, + /// Location of first predicate within byte of internal storage + int kPredicateStart_ = 0> +struct PredicateVector { + /// Number of bits stored by the PredicateVector + static int const kPredicates = kPredicates_; + + /// Number of bits stored within each byte of the predicate bit vector + static int const kPredicatesPerByte = kPredicatesPerByte_; + + /// First bit withing each byte containing predicates + static int const kPredicateStart = kPredicateStart_; + + // Make sure no one tries to put more than 8 bits in a byte :) + static_assert(kPredicatesPerByte <= 8, "kPredicatesPerByte must fit within an actual byte"); + // Make sure the "offsetted" bits fit in one byte. + static_assert(kPredicateStart + kPredicatesPerByte < 8, + "The offsetted predicates must fit within an actual byte."); + + /// Storage type of individual elements + typedef uint32_t Storage; + + /// Number of bytes needed + static int const kBytes = (kPredicates + kPredicatesPerByte - 1) / kPredicatesPerByte; + + /// Number of storage elements needed + static int const kWordCount = (kBytes + sizeof(Storage) - 1) / sizeof(Storage); + + private: + // + // Data members + // + + /// Words of bit vector + Storage storageData[kWordCount]; + + // + // Methods + // + + /// Computes the word and bit corresponding to a logical predicate index + CUTLASS_HOST_DEVICE void computeStorageOffset(int &word, int &bit, int idx) const { + CUTLASS_ASSERT(idx < kPredicates); + + int byte = (idx / kPredicatesPerByte); + int bit_offset = (idx % kPredicatesPerByte); + + word = byte / sizeof(Storage); + int byte_offset = (byte % sizeof(Storage)); + + bit = byte_offset * 8 + bit_offset + kPredicateStart; + } + + /// Accesses a given word with optional assertions + CUTLASS_HOST_DEVICE Storage &storage(int word) { + CUTLASS_ASSERT(word < kWordCount); + return storageData[word]; + } + + /// Accesses a given word with optional assertions + CUTLASS_HOST_DEVICE Storage const &storage(int word) const { + CUTLASS_ASSERT(word < kWordCount); + return storageData[word]; + } + + public: + // + // Iterator + // + + /** + * @brief A const iterator implementing \ref predicate_iterator_concept enabling sequential + * read-only access to prediactes. + * @concept{predicate_iterator_concept} + */ + class ConstIterator { + /// Reference to PredicateVector instance + PredicateVector const &vec_; + + /// Index into PredicateVector + int bit_; + + public: + /// Copy constructor + CUTLASS_HOST_DEVICE + ConstIterator(ConstIterator const &it) : vec_(it.vec_), bit_(it.bit_) {} + + /// + CUTLASS_HOST_DEVICE + ConstIterator(PredicateVector const &_vec, int _start = 0) : vec_(_vec), bit_(_start) {} + + /// Pre-increment + CUTLASS_HOST_DEVICE + ConstIterator &operator++() { + ++bit_; + return *this; + } + + /// Pre-decrement + CUTLASS_HOST_DEVICE + ConstIterator &operator--() { + --bit_; + return *this; + } + + /// Post-increment + CUTLASS_HOST_DEVICE + ConstIterator operator++(int) { + ConstIterator ret(*this); + ret.bit_++; + return ret; + } + + /// Post-decrement + CUTLASS_HOST_DEVICE + ConstIterator operator--(int) { + ConstIterator ret(*this); + ret.bit_--; + return ret; + } + + /// Returns true if iterators point to the same bit + CUTLASS_HOST_DEVICE + bool operator==(ConstIterator const &it) const { return bit_ == it.bit_; } + + /// Returns false if iterators point to the same bit + CUTLASS_HOST_DEVICE + bool operator!=(ConstIterator const &it) const { return bit_ != it.bit_; } + + /// Dereferences iterator + CUTLASS_HOST_DEVICE + bool operator*() const { return vec_[bit_]; } + }; + + /** + * @brief An iterator implementing \ref predicate_iterator_concept enabling sequential + * read and write access to predicates. + * @concept{predicate_iterator_concept} + */ + class Iterator { + /// Reference to PredicateVector instance + PredicateVector &vec_; + + /// Index into PredicateVector + int bit_; + + public: + /// Copy constructor + CUTLASS_HOST_DEVICE + Iterator(Iterator const &it) : vec_(it.vec_), bit_(it.bit_) {} + + /// Constructs an iterator from a PredicateVector + CUTLASS_HOST_DEVICE + Iterator(PredicateVector &_vec, int _start = 0) : vec_(_vec), bit_(_start) {} + + /// Pre-increment + CUTLASS_HOST_DEVICE + Iterator &operator++() { + ++bit_; + return *this; + } + + /// Pre-decrement + CUTLASS_HOST_DEVICE + Iterator &operator--() { + --bit_; + return *this; + } + + /// Post-increment + CUTLASS_HOST_DEVICE + Iterator operator++(int) { + Iterator ret(*this); + ret.bit_++; + return ret; + } + + /// Post-decrement + CUTLASS_HOST_DEVICE + Iterator operator--(int) { + Iterator ret(*this); + ret.bit_--; + return ret; + } + + /// Returns true if iterators point to the same bit + CUTLASS_HOST_DEVICE + bool operator==(Iterator const &it) const { return bit_ == it.bit_; } + + /// Returns false if iterators point to the same bit + CUTLASS_HOST_DEVICE + bool operator!=(Iterator const &it) const { return bit_ != it.bit_; } + + /// Gets the bit at the pointed to location + CUTLASS_HOST_DEVICE + bool get() { return vec_[bit_]; } + + /// Dereferences iterator + CUTLASS_HOST_DEVICE + bool operator*() const { return vec_[bit_]; } + + /// Sets the bit at the pointed to location + CUTLASS_HOST_DEVICE + void set(bool value = true) { vec_.set(bit_, value); } + }; + + /// Iterator that always returns true + struct TrivialIterator { + /// Constructor + CUTLASS_HOST_DEVICE + TrivialIterator() {} + + /// Copy constructor + CUTLASS_HOST_DEVICE + TrivialIterator(Iterator const &it) {} + + /// Constructs an iterator from a PredicateVector + CUTLASS_HOST_DEVICE + TrivialIterator(PredicateVector const &_vec) {} + + /// Pre-increment + CUTLASS_HOST_DEVICE + TrivialIterator &operator++() { return *this; } + + /// Post-increment + CUTLASS_HOST_DEVICE + TrivialIterator operator++(int) { return *this; } + + /// Dereferences iterator + CUTLASS_HOST_DEVICE + bool operator*() const { return true; } + }; + + public: + // + // Methods + // + + /// Initialize the predicate vector + CUTLASS_HOST_DEVICE PredicateVector(bool value = true) { fill(value); } + + /// Fills all predicates with a given value + CUTLASS_HOST_DEVICE void fill(bool value = true) { + Storage item = (value ? ~Storage(0) : Storage(0)); + + CUTLASS_PRAGMA_UNROLL + for (int i = 0; i < kWordCount; ++i) { + storage(i) = item; + } + } + + /// Accesses a bit within the predicate vector. + CUTLASS_HOST_DEVICE bool operator[](int idx) const { return at(idx); } + + /// Accesses a bit within the predicate vector. + CUTLASS_HOST_DEVICE bool at(int idx) const { + int bit, word; + computeStorageOffset(word, bit, idx); + + return ((storage(word) >> bit) & 1); + } + + /// Set a bit within the predicate vector. + CUTLASS_HOST_DEVICE void set(int idx, bool value = true) { + int bit, word; + computeStorageOffset(word, bit, idx); + + Storage disable_mask = (~(Storage(1) << bit)); + Storage enable_mask = (Storage(value) << bit); + + storage(word) = ((storage(word) & disable_mask) | enable_mask); + } + + /// Computes the intersection of two identical predicate vectors. + CUTLASS_HOST_DEVICE PredicateVector &operator&=(PredicateVector const &predicates) { + CUTLASS_PRAGMA_UNROLL + for (int i = 0; i < kWordCount; ++i) { + storage(i) = (storage(i) & predicates.storage(i)); + } + return *this; + } + + /// Computes the union of two identical predicate vectors. + CUTLASS_HOST_DEVICE PredicateVector &operator|=(PredicateVector const &predicates) { + CUTLASS_PRAGMA_UNROLL + for (int i = 0; i < kWordCount; ++i) { + storage(i) = (storage(i) | predicates.storage(i)); + } + return *this; + } + + /// Returns true if entire predicate array is zero. + CUTLASS_HOST_DEVICE bool is_zero() const { + Storage mask(0); + for (int byte = 0; byte < sizeof(Storage); ++byte) { + Storage byte_mask = (((1 << kPredicatesPerByte) - 1) << kPredicateStart); + mask |= (byte_mask << (byte * 8)); + } + uint32_t result = 0; + for (int word = 0; word < kWordCount; ++word) { + result |= storage(word); + } + return result == 0; + } + + /// Returns an iterator to the start of the bit vector + CUTLASS_DEVICE + Iterator begin() { return Iterator(*this); } + + /// Returns an iterator + CUTLASS_DEVICE + Iterator end() { return Iterator(*this, kPredicates); } + + /// Returns a ConstIterator + CUTLASS_DEVICE + ConstIterator const_begin() const { return ConstIterator(*this); } + + /// Returns a ConstIterator + CUTLASS_DEVICE + ConstIterator const_end() const { return ConstIterator(*this, kPredicates); } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Always returns true predicate. +struct TrivialPredicateTileAdapter { + /// Ctor. + CUTLASS_HOST_DEVICE TrivialPredicateTileAdapter() {} + + /// The value at location (d, h, w, c). + CUTLASS_HOST_DEVICE bool at(int, int, int, int) const { return true; } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Adapter to enable random access to predicates via logical coordinate within a tile. +template +struct PredicateTileAdapter { + /// The vector of predicates. + typedef PredicateVector_ PredicateVector; + /// The iterations. + typedef Iterations_ Iterations; + + private: + /// The predicates. + PredicateVector &predicates; + + public: + /// Ctor. + CUTLASS_DEVICE PredicateTileAdapter(PredicateVector &predicates_) : predicates(predicates_) {} + + /// Get the value at location (d, h, w, c). + CUTLASS_DEVICE bool at(int d, int h, int w, int c) const { + int const bit = ComputeOffsetFromShape::get(d, h, w, c); + return predicates.at(bit); + } + + /// Set the value at location (d, h, w, c). + CUTLASS_DEVICE void set(int d, int h, int w, int c, bool value) { + int const bit = ComputeOffsetFromShape::get(d, h, w, c); + predicates.set(bit, value); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Adapter to enable random access to predicates via logical coordinate within a tile. +template +struct ConstPredicateTileAdapter { + /// The vector of predicates. + typedef PredicateVector_ PredicateVector; + /// The iterations. + typedef Iterations_ Iterations; + + private: + /// The predicates. + PredicateVector const &predicates; + + public: + /// Ctor. + CUTLASS_DEVICE ConstPredicateTileAdapter(PredicateVector const &predicates_) + : predicates(predicates_) {} + + /// Get the value at location (d, h, w, c). + CUTLASS_DEVICE bool at(int d, int h, int w, int c) const { + int const bit = ComputeOffsetFromShape::get(d, h, w, c); + return predicates.at(bit); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/reshape_tile.h b/cutlass-example/cutlass/reshape_tile.h new file mode 100644 index 0000000..55aebfc --- /dev/null +++ b/cutlass-example/cutlass/reshape_tile.h @@ -0,0 +1,58 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines a type for restructuring a tile. +*/ +#pragma once + +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +// The following functor reshapes a tile of data. The goal is to have at least kAccessSize in +// the inner-most dimension. If the user respects that constraint, there is nothing to be done. If +// that's not the case, this functor will correct that and "extract" the right number of elements +// from the next dimension. + +template +struct ReshapeTile { + typedef Tile_ Tile; +}; + +template +struct ReshapeTile { + // Make sure the W dimension of the tile is large enough. + static_assert(Tile_::kW >= kAccessSize_, "The W dimension is too small"); + // Make sure the dimension can be divided by the number of scalars. + static_assert(Tile_::kW % kAccessSize_ == 0, "Not supported"); + // Collapse the W dimension. + typedef Shape Tile; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/shape.h b/cutlass-example/cutlass/shape.h new file mode 100644 index 0000000..4f6b222 --- /dev/null +++ b/cutlass-example/cutlass/shape.h @@ -0,0 +1,305 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines Shape implementing the Layout concept for representing a 4D hypercube of objects. +*/ +#pragma once + +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup layout_concept Layout Concept +* @{ +* @par Implementations of \ref layout_concept are used to describe a cube with DxHxW elements and C +scalars per element. + A HxW slice of a cube is called an image and a cube consists of D images. +* +* @par Notations +* Let Layout be an implementation of the \ref layout_concept. +* +* @par Valid Expressions +* - Layout::D specifies the depth of a cube +* - Layout::H specifies the height of a cube +* - Layout::W specifies the height of a cube +* - Layout::C specifies the number of channels of each element in a cube +* - Layout::W_c specifies the number of scalars of each row in one image of a cube. +* - Layout::H_w specifies the number of elements in an image slice. +* - Layout::H_w_c_specifies the number of scalars in an image slice. +* - Layout::D_h_w specifies the number of elements in a cube. +* - Layout::D_h_w_c specifies the number of scalars in a cube. +* - Layout::Strides is a \ref layout_concept specifying the strides. +* @} +*/ + +/** +* @brief A Shape implementing \ref layout_concept describing the dimensions of a cube. +* @concept{layout_concept} +*/ +template +struct Shape { + /// The depth of the cube. + static int const kD = kD_; + /// The height of the cube. + static int const kH = kH_; + /// The width of the cube. + static int const kW = kW_; + /// The number of scalars per element. + static int const kC = kC_; +}; + +/** +* @brief Compute derived counted of a \ref layout_concept based class +*/ +template +struct ShapeCount { + /// The number of elements per row. + static int const kWc = Shape::kW * Shape::kC; + /// The number of pixels per image. + static int const kHw = Shape::kH * Shape::kW; + /// The number of elements per image. + static int const kHwc = Shape::kH * kWc; + /// The number of pixels per cube. + static int const kDhw = Shape::kD * kHw; + /// The number of elements in the 4D space. + static int const kDhwc = Shape::kD * kHwc; + /// The number of elements in the 4D space. + static int const kCount = kDhwc; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeScale { + typedef Shape Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeAdd { + typedef Shape Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeSub { + typedef Shape Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeMul { + typedef Shape Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeDiv { + typedef Shape Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeMax { + typedef Shape<(A_::kD > B_::kD ? A_::kD : B_::kD), + (A_::kH > B_::kH ? A_::kH : B_::kH), + (A_::kW > B_::kW ? A_::kW : B_::kW), + (A_::kC > B_::kC ? A_::kC : B_::kC)> + Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeMin { + typedef Shape<(A_::kD < B_::kD ? A_::kD : B_::kD), + (A_::kH < B_::kH ? A_::kH : B_::kH), + (A_::kW < B_::kW ? A_::kW : B_::kW), + (A_::kC < B_::kC ? A_::kC : B_::kC)> + Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct ShapeStrides { + typedef Shape + Shape; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Compute the offset for the given coordinates in a cube +* @tparam A \ref layout_concept where each dimension of the cube specifies the corresponding stride. +*/ +template +struct ComputeOffsetFromShape { + static CUTLASS_DEVICE int get(int d, int h, int w, int c) { + // clang-format off + return d * Shape_::kH * Shape_::kW * Shape_::kC + + h * Shape_::kW * Shape_::kC + + w * Shape_::kC + + c; + // clang-format on + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Compute the offset for the given coordinates in a cube with a depth of 1 +* @tparam kSh Elements in the H dimension +* @tparam kSw Elements in the W dimension +* @tparam kSc Separation between two elements in "elements" +*/ +template +struct ComputeOffsetFromShape > { + static CUTLASS_DEVICE int get(int d, int h, int w, int c) { + return h * kSw_ * kSc_ + w * kSc_ + c; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Compute the offset for the given coordinates in a cube with one channel and a depth of 1 +* @tparam kSh Elements in the H dimension +* @tparam kSw Elements in the W dimension +*/ +template +struct ComputeOffsetFromShape > { + static CUTLASS_DEVICE int get(int d, int h, int w, int c) { return h * kSw_ + w; } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Compute the offset for the given coordinates in a cube +* @tparam A \ref layout_concept where each dimension of the cube specifies the corresponding stride. +*/ +template +struct ComputeOffsetFromStrides { + static CUTLASS_DEVICE int get(int d, int h, int w, int c) { + return d * Strides_::kD + h * Strides_::kH + w * Strides_::kW + c * Strides_::kC; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Compute the offset for the given coordinates in a cube with a depth of 1 +* @tparam S_h Stride in the H dimension in scalars +* @tparam S_w Stride in the W dimension in scalars +* @tparam S_c Stride between two scalars. +*/ +template +struct ComputeOffsetFromStrides > { + static CUTLASS_DEVICE int get(int d, int h, int w, int c) { + return h * S_h_ + w * S_w_ + c * S_c_; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Compute the offset for the given coordinates in a cube with one channel and a depth of 1 +* @tparam S_h Stride in the H dimension in scalars +* @tparam S_w Stride in the W dimension in scalars +*/ +template +struct ComputeOffsetFromStrides > { + static CUTLASS_DEVICE int get(int d, int h, int w, int c) { return h * S_h_ + w * S_w_; } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief Decompose threadId.x into coordinate of a cube whose dimensions are specified by Threads_. +* Afterwards compute the offset of those coordinates using Strides_ +* @tparam Threads_ The dimension of the cube the threadIdx.x value is mapped on +* @tparam Strides_ The strides to use when compute the offsets based on the coordinates of the cube. +*/ +template +struct ComputeThreadOffsetFromStrides { + static CUTLASS_DEVICE int get() { + // Decompose the thread index. + int c = threadIdx.x % Threads_::kC; + int w = threadIdx.x / Threads_::kC % Threads_::kW; + int h = threadIdx.x / Threads_::kC / Threads_::kW % Threads_::kH; + int d = threadIdx.x / Threads_::kC / Threads_::kW / Threads_::kH; + + // Compute the offset. + return d * Strides_::kD + h * Strides_::kH + w * Strides_::kW + c * Strides_::kC; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// +/** +*@brief Specialization for D=1 +*/ +template +struct ComputeThreadOffsetFromStrides, Shape<1, S_h_, S_w_, S_c_> > { + static CUTLASS_DEVICE int get() { + // Decompose the thread index. + int c = threadIdx.x % T_c_; + int w = threadIdx.x / T_c_ % T_w_; + int h = threadIdx.x / T_c_ / T_w_ % T_h_; + + // Compute the offset. + return h * S_h_ + w * S_w_ + c * S_c_; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +*@brief Specialization for D=1 and C=1 +*/ +template +struct ComputeThreadOffsetFromStrides, Shape<1, S_h_, S_w_, 1> > { + static CUTLASS_DEVICE int get() { + // Decompose the thread index. + int w = threadIdx.x % T_w_; + int h = threadIdx.x / T_w_; + + // Compute the offset. + return h * S_h_ + w * S_w_; + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/tensor_ref.h b/cutlass-example/cutlass/tensor_ref.h new file mode 100644 index 0000000..8ef31e3 --- /dev/null +++ b/cutlass-example/cutlass/tensor_ref.h @@ -0,0 +1,151 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines a structure containing strides, bounds, and a pointer to tensor data. +*/ +#pragma once + +#include + +#include +#include +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Structure modeling a pointer and stride into a tensor +template +class TensorRef { + public: + /// Data type of individual access + typedef Storage_ Storage; + + /// Rank of tensor + static int const Rank = Rank_; + + private: + // + // Data members + // + + /// Pointer to storage element + Storage* ptr_; + + /// Stride information + Coord stride_; + + public: + // + // Methods + // + + /// Default ctor + CUTLASS_HOST_DEVICE + TensorRef() : ptr_(nullptr) {} + + /// Constructs from a pointer, size, and stride + CUTLASS_HOST_DEVICE + TensorRef(Storage* ptr, Coord stride) : ptr_(ptr), stride_(stride) {} + + /// Updates the pointer, stride, and location within a TensorRef + CUTLASS_HOST_DEVICE + void reset(Storage* ptr = nullptr, Coord stride = Coord(0)) { + ptr_ = ptr; + stride_ = stride; + } + + /// Conversion function + template + TensorRef convert() { + Coord converted_stride; + for (int i = 0; i < Rank - 1; ++i) { + converted_stride[i] = stride_[i] * Extent::kValue / Extent::kValue; + } + converted_stride[Rank - 1] = stride_[Rank - 1]; + + return TensorRef(reinterpret_cast(ptr_), converted_stride); + } + + /// Returns true if the TensorRef may be safely accessed + CUTLASS_HOST_DEVICE + bool good() const { return ptr_ != nullptr; } + + /// Returns the pointer to referenced data + CUTLASS_HOST_DEVICE + Storage* data() const { return ptr_; } + + /// Returns the stride of the tensor + CUTLASS_HOST_DEVICE + Coord const& stride() const { return stride_; } + + /// Returns the stride of the tensor in the given dimension + CUTLASS_HOST_DEVICE + int const& stride(int dim) const { return stride_.at(dim); } + + /// Returns the maximum stride element as the 'leading dimension' + CUTLASS_HOST_DEVICE + int leading_dim() const { return __NV_STD_MAX(stride_[1], stride_[2]); } + + /// Computes the offset of an index from the origin of the tensor + CUTLASS_HOST_DEVICE + long long offset(Coord const& coord) const { + return stride_.template dot(coord); + } + + /// Returns a reference to the element at a given Coord + CUTLASS_HOST_DEVICE + Storage& at(Coord const& coord) const { return ptr_[offset(coord)]; } + + /// Element-wise accessor + Storage& operator[](Coord const& coord) const { return at(coord); } + + /// Returns a reference to the element at a given Coord + CUTLASS_HOST_DEVICE + Storage& at(int idx) const { return ptr_[idx]; } + + /// Element-wise accessor + Storage& operator[](int idx) const { return at(idx); } + + /// Adds an offset to the pointer + CUTLASS_HOST_DEVICE + TensorRef& advance(Coord const& b) { + ptr_ += offset(b); + return *this; + } + + /// Returns a TensorRef offset by a given amount + CUTLASS_HOST_DEVICE + TensorRef operator+(Coord const& b) const { return TensorRef(ptr_ + offset(b), stride_); } + + /// Returns a TensorRef offset by a given amount + CUTLASS_HOST_DEVICE + TensorRef operator-(Coord const& b) const { return TensorRef(ptr_ - offset(b), stride_); } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/tensor_view.h b/cutlass-example/cutlass/tensor_view.h new file mode 100644 index 0000000..89c6bd5 --- /dev/null +++ b/cutlass-example/cutlass/tensor_view.h @@ -0,0 +1,172 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines a structure containing strides and a pointer to tensor data. +*/ + +#pragma once + +#include + +#include +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Host-side reference implementation of tensor operations +template +class TensorView : public TensorRef { + public: + /// Reference and stride + typedef TensorRef Base; + + /// Reference and stride + typedef Base TensorRef_t; + + /// Reference to constant type + typedef TensorRef ConstTensorRef_t; + + /// Rank of tensor + static int const Rank = TensorRef_t::Rank; + + /// Type used to compute the offset of an element to the base of a tensor + typedef int Offset_t; + + /// Coordinate into tensor + typedef Coord Coord_t; + + private: + // + // Data members + // + + /// Pointer to pitch-linear memory + TensorRef_t ref_; + + /// Dimensions of coordinate (independent of stride) + Coord_t size_; + + public: + // + // Device and Host Methods + // + + /// Default constructor + CUTLASS_HOST_DEVICE + TensorView() {} + + /// Constructs a Tensor_view from a TensorRef and size + CUTLASS_HOST_DEVICE + TensorView(TensorRef_t const& _ref, Coord_t const& _size) : Base(_ref), size_(_size) {} + + /// Returns true if the Tensor_view is bound to some memory + CUTLASS_HOST_DEVICE + bool good() const { return ref().good(); } + + /// Returns a pointer to data + CUTLASS_HOST_DEVICE + T* data() const { return ref().data(); } + + /// Updates the reference and size of a Tensor_view object + CUTLASS_HOST_DEVICE + void reset(TensorRef_t const& _ref = TensorRef_t(0), Coord_t const& _size = Coord_t()) { + Base::operator=(_ref); + size_ = _size; + } + + /// Accesses the tensor reference pointing to data + CUTLASS_HOST_DEVICE + TensorRef_t& ref() { return *this; } + + /// + CUTLASS_HOST_DEVICE + ConstTensorRef_t const_ref() { return ConstTensorRef_t(data(), stride()); } + + /// Accesses the tensor reference pointing to data + CUTLASS_HOST_DEVICE + TensorRef_t const& ref() const { return *this; } + + /// Accesses the size + CUTLASS_HOST_DEVICE + Coord_t const& size() const { return size_; } + + /// Accesses the size + CUTLASS_HOST_DEVICE + int size(int dim) const { return size_.at(dim); } + + /// Accesses the stride + CUTLASS_HOST_DEVICE + Coord_t const& stride() const { return ref().stride(); } + + /// Accesses the stride + CUTLASS_HOST_DEVICE + int const& stride(int dim) const { return ref().stride(dim); } + + /// Assigns the Tensor_view + CUTLASS_HOST_DEVICE + TensorView& operator=(TensorView const& _tensor) { + Base::operator=(_tensor._ref); + size_ = _tensor.size_; + return *this; + } + + /// Returns the index of an element + CUTLASS_HOST_DEVICE + Offset_t offset(Coord_t const& coord) const { return ref().offset(coord); } + + /// Determines whether a location is within a tensor + CUTLASS_HOST_DEVICE + bool contains(Coord_t const& coord) const { + for (int dim = 0; dim < Rank; ++dim) { + if (coord.at(dim) >= size_.at(dim)) { + return false; + } + } + return true; + } + + /// Element-wise accessor + CUTLASS_HOST_DEVICE + T& at(Coord_t const& coord) const { return ref().at(coord); } + + /// Element-wise accessor + T& operator[](Coord const& coord) const { return at(coord); } + + /// Element-wise accessor + CUTLASS_HOST_DEVICE + T& at(Offset_t idx) const { return ref().at(idx); } + + /// Returns a Tensor_view given location and size quantities + CUTLASS_HOST_DEVICE + TensorView subview(Coord_t const& location, Coord_t size) const { + return TensorView(ref() + location, size.clamp(size_ - location)); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/tile_iterator.h b/cutlass-example/cutlass/tile_iterator.h new file mode 100644 index 0000000..5d39c4f --- /dev/null +++ b/cutlass-example/cutlass/tile_iterator.h @@ -0,0 +1,899 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines the Tile Traits concept and iterators for loading and storing to tiles + efficiently. +*/ +#pragma once + +#include +#include +#include +#include + +namespace cutlass { + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup tile_traits_concept Tile Traits Concept +@{ + +\ref tile_traits_concept is a type definining the shape of a tile and the distribution of accesses +by individual entities, either threads or other. + +@par Tile Traits Concept + Types satisfying \ref tile_traits_concept define the following members + - Tile - a type satisfying \ref layout_concept describing the dimensions of the tile + - Delta - a type satisfying \ref layout_concept describing the increments between accesses +along each dimension + - Iterations - a type satisfying \ref layout_concept describing the number of accesses +along each dimension + - Offset - the type of a functor computing the offset of each participating entity +as a Coord<4>. +@} +*/ + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Specifies dimension in which post-increment accesses advance +struct IteratorAdvance { + enum Kind { kD, kH, kW }; +}; + +/// Specifies whether iterator storage fragment consists of Scalar values or WMMA matrix +struct IteratorFragment { + enum Kind { kScalar, kWmmaMatrix }; +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief A template defining \ref tile_traits_concept +* @concept{tile_traits_concept} +*/ +template +struct TileTraits { + /// Shape of the tile + typedef Tile_ Tile; + + /// Number of steps between accesses along each dimension + typedef Delta_ Delta; + + /// Number of accesses performed + typedef Iterations_ Iterations; + + /// Functor that returns the logical coordinate of each entity's initial offset in the tile + typedef ThreadOffset_ ThreadOffset; +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Iterator for accessing a stripmined tile in memory +template > +struct TileIteratorBase { + /// concept TileTraits + typedef Traits_ Traits; + + /// Scalar element + typedef Scalar_ Scalar; + + /// Fragment element + typedef FragmentElement_ FragmentElement; + + /// Specifies dimension in which post-increment accesses advance. + static IteratorAdvance::Kind const kAdvance = Advance_; + + /// Specifies iterator storage fragment type (Scalar or WmmaMatrix) + static IteratorFragment::Kind const kIteratorFragment = IteratorFragment_; + + /// Source or destination memory space + static MemorySpace::Kind const kMemorySpace = MemorySpace; + + /// Index type + typedef Index_ Index; + + /// Skew quantity + typedef Skew_ Skew; + + /// Tile shape + typedef typename Traits::Tile Tile; + + /// Distance along each dimension + typedef typename Traits::Delta Delta; + + /// The strides in each dimension between different loads/stores. + typedef typename Traits::ImmediateOffsetStrides ImmediateOffsetStrides; + + /// Iterations + typedef typename Traits::Iterations Iterations; + + /// Thread offset + typedef typename Traits::ThreadOffset ThreadOffset; + + /// The number of scalars accessed per load/store. + static int const kAccessSize = Tile::kC; + + /// The elements loaded/store by one instruction. + typedef typename Vectorize::Type AccessType; + + /// The size of storage needed per fragment + static int const kFragmentSize = + (kIteratorFragment == IteratorFragment::kWmmaMatrix ? 16 : sizeof(AccessType)); + /// The storage. + typedef Fragment::kCount, kFragmentSize> Storage; + /// The fragment. + typedef Fragment::kCount * kAccessSize> Fragment; + /// The fragment iterator. + typedef FragmentIterator FragmentIterator; + /// The fragment const iterator. + typedef FragmentConstIterator FragmentConstIterator; + /// The shape of the fragment. + typedef typename FragmentIterator::FragmentShape FragmentShape; + + /// Default predicate mask type + typedef PredicateVector::kCount> PredicateVector; + + // + // Params struct + // + + /// Parameters to the iterator + struct Params { + Index stride_d; + Index stride_h; + Index stride_w; + + Index inc_d; + Index inc_h; + Index inc_w; + + Index inc_advance; + + /// Initializes params + CUTLASS_HOST_DEVICE + int initialize(Index _stride_d, + Index _stride_h, + Index _stride_w, + Index _inc_d, + Index _inc_h, + Index _inc_w, + Index _inc_advance) { + stride_d = _stride_d; + stride_h = _stride_h; + stride_w = _stride_w; + + inc_d = _inc_d; + inc_h = _inc_h; + inc_w = _inc_w; + inc_advance = _inc_advance; + + return 0; + } + + CUTLASS_HOST_DEVICE + int initialize(Index _stride_d, Index _stride_h, Index _stride_w) { + stride_d = _stride_d; + stride_h = _stride_h; + stride_w = _stride_w; + + inc_w = stride_w * Delta::kW; + inc_h = stride_h * Delta::kH - stride_w * Delta::kW * (Iterations::kW - 1); + + if (kAdvance == IteratorAdvance::kH) { + // Advance in the H dimension. + inc_d = 0; + } else if (kAdvance == IteratorAdvance::kW) { + // Advance in the W dimension. + inc_d = stride_w * Tile::kW - stride_h * Tile::kH; + } else { + // Advance in the D dimension. + inc_d = stride_d; + } + + inc_advance = 0; + + return 0; + } + + CUTLASS_HOST_DEVICE int initialize() { + stride_d = 0; + stride_h = 0; + stride_w = 1; + + inc_d = inc_h = inc_w = inc_advance = 0; + + return 0; + } + }; + + /// Is the iterator valid? + CUTLASS_DEVICE bool valid(int d, int h, int w, int c) const { return true; } + + // + // Static function members + // + + /// Initializes a predicate vector + template + CUTLASS_DEVICE static void initialize_predicates(PredicateIterator predicate_it, + Coord<3> const &bounds, + Coord<3> const &offset = make_Coord(0, 0, 0)) { + for (int d = 0; d < Iterations::kD; ++d) { + bool enable_d = (d * Delta::kD + offset[0] < bounds[0]); + for (int h = 0; h < Iterations::kH; ++h) { + bool enable_h = (h * Delta::kH + offset[1] < bounds[1]); + for (int w = 0; w < Iterations::kW; ++w) { + bool enable_w = (w * Tile::kC * Delta::kW + offset[2] < bounds[2]); + predicate_it.set(d, h, w, 0, enable_d && enable_h && enable_w); + } + } + } + } +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup tile_load_iterator_concept Tile Load Iterator Concept +@{ + +\ref tile_load_iterator_concept enables loading a tile from addressable memory into a fragment + +@par Tile Load Iterator Concept + Types satisfying \ref tile_load_iterator_concept define the following members + - PredicateVector - a \ref predicate_vector_concept with sufficient predicate storage for +each access implied by the tile traits + - Fragment - the destination fragment type satisfying \ref fragment_concept + - initialize_predicates(pred_it, bounds, block_offset) - function initializing a predicate +vector according to externally specified bounds + - load_post_increment(fragment, pred_it) - a method that loads a fragment and increments +the iterator to the next tile, guarded by a \ref predicate_iterator_concept + - load_post_increment(fragment) - a method that loads a fragment and increments the +iterator to the next tile + - load(fragment, pred_it) - a const method that loads a fragment, guarded by a \ref +predicate_iterator_concept + - load(fragment) - a method that loads a fragment + +@} +*/ + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief An iterator implementing \ref tile_load_iterator_concept for loading a tile from memory +* @concept{tile_load_iterator_concept} +*/ +template > +struct TileLoadIterator : public TileIteratorBase { + /// Base class + typedef TileIteratorBase + Base; + + /// concept TileTraits + typedef typename Base::Traits Traits; + + /// Scalar element + typedef typename Base::Scalar Scalar; + + /// Fragment element + typedef typename Base::FragmentElement FragmentElement; + + /// Specifies in which dimension post-increment accesses advance. + static IteratorAdvance::Kind const kAdvance = Base::kAdvance; + + /// Specifies type of iterator fragment storage (Salar or WmmaMatrix) + static IteratorFragment::Kind const kIteratorFragment = Base::kIteratorFragment; + + /// Source or destination memory space + static MemorySpace::Kind const kMemorySpace = Base::kMemorySpace; + + /// Index type + typedef typename Base::Index Index; + + /// Skew quantity + typedef typename Base::Skew Skew; + + /// Tile shape + typedef typename Base::Tile Tile; + + /// Delta + typedef typename Base::Delta Delta; + + /// Iterations + typedef typename Base::Iterations Iterations; + + /// ThreadOffset functor + typedef typename Base::ThreadOffset ThreadOffset; + + /// Fragment type + typedef typename Base::FragmentShape FragmentShape; + + /// Memory access type + typedef typename Base::AccessType AccessType; + + /// Fragment definition + typedef typename Base::Fragment Fragment; + + /// Fragment iterator definition + typedef typename Base::FragmentIterator FragmentIterator; + + /// Fragment const iterator definition + typedef typename Base::FragmentConstIterator FragmentConstIterator; + + /// Default predicate mask type + typedef typename Base::PredicateVector PredicateVector; + + /// Storage object that may be loaded from + typedef typename Base::Storage SharedStorage; + + /// IteratorBase parameters + typedef typename Base::Params BaseParams; + + /// Do we require a fence? + enum { kRequiresLoadFence = Tile::kD == 1 }; + + /// The pointer type + typedef Scalar const *Pointer; + + /// Parameters + struct Params : public BaseParams { + /// Pointer to memory + Scalar const *pointer; + + /// Initialize params to access storage object + CUTLASS_HOST_DEVICE + int initialize(SharedStorage const &storage) { + pointer = &storage[0]; + return 0; + } + + /// Initializes params to access a raw pointer + CUTLASS_HOST_DEVICE + int initialize(Scalar const *ptr, Index stride_d, Index stride_h, Index stride_w) { + Base::Params::initialize(stride_d, stride_h, stride_w); + pointer = ptr; + return 0; + } + + /// Initializes params + CUTLASS_HOST_DEVICE + int initialize(Scalar const *ptr, + Index _stride_d, + Index _stride_h, + Index _stride_w, + Index _inc_d, + Index _inc_h, + Index _inc_w, + Index _inc_advance) { + pointer = ptr; + Base::Params::initialize( + _stride_d, _stride_h, _stride_w, _inc_d, _inc_h, _inc_w, _inc_advance); + return 0; + } + + // Initializes params to default values + CUTLASS_HOST_DEVICE + int initialize() { return Base::Params::initialize(); } + }; + + // + // Data members + // + + /// Parameters structure + Params params; + + /// Offset of an individual lane from the start of the tile + Coord<4> thread_offset; + + /// Stage argument enables wrapping after some number of tiles have been loaded. + int stage; + + // + // Static member functions + // + + /// Initializes a predicate vector + template + CUTLASS_HOST_DEVICE void initialize_predicates(PredicateIterator predicate_it, + Coord<3> const &bounds, + Coord<3> const &block_offset = make_Coord(0, + 0, + 0)) { + Base::initialize_predicates( + predicate_it, + bounds, + block_offset + make_Coord(0, thread_offset[1], thread_offset[2] * Tile::kC)); + } + + // + // Methods + // + + /// Default constructor + CUTLASS_HOST_DEVICE + TileLoadIterator() {} + + /// Constructs a tile load iterator + CUTLASS_HOST_DEVICE + TileLoadIterator(Params const &_params, + Coord<3> const &block_offset = make_Coord(0, 0, 0), + ThreadOffset thread_offset_func = ThreadOffset()) + : params(_params), stage(0) { + thread_offset = thread_offset_func(); + + Index block_offset_h = 0; + Index block_offset_w = 0; + if (kAdvance == IteratorAdvance::kH) { + block_offset_h = block_offset[1]; + block_offset_w = block_offset[2]; + } else { + block_offset_h = block_offset[2]; + block_offset_w = block_offset[1]; + } + + params.pointer += block_offset[0] * params.stride_d + + (block_offset_h + thread_offset[1]) * params.stride_h + + (block_offset_w + thread_offset[2] * Tile::kC) / Tile::kC * params.stride_w; + } + + /// Constructs a tile load iterator + CUTLASS_HOST_DEVICE + TileLoadIterator(Params const &, + SharedStorage &shared_storage, + Coord<3> const &block_offset = make_Coord(0, 0, 0), + ThreadOffset thread_offset_func = ThreadOffset()) + : stage(0) { + int const offset = thread_offset_func()[2]; + params.pointer = &shared_storage[offset]; + } + + /// Returns the current pointer + CUTLASS_HOST_DEVICE + Scalar const *data() const { return params.pointer; } + + /// The accessor. + CUTLASS_DEVICE void get(AccessType &value, int d, int h, int w, int c) const { + int const imm = + ComputeOffsetFromStrides::get(d, h, w, c); + Load::load(value, params.pointer, imm); + } + + /// Increment in the D dimension + CUTLASS_HOST_DEVICE void inc_d() { params.pointer += params.inc_d; } + + /// Increment in the H dimension + CUTLASS_HOST_DEVICE void inc_h() { params.pointer += params.inc_h; } + + /// Increment in the W dimension + CUTLASS_HOST_DEVICE void inc_w() { params.pointer += params.inc_w; } + + /// Increment in the next dimension + CUTLASS_HOST_DEVICE void inc_advance() { params.pointer += params.inc_advance; } + + /// Increment the stage. + CUTLASS_DEVICE void inc_stage() { + if (Tile::kD > 1) { + int const kStageSize = Tile::kH * Tile::kW * Tile::kC; + if (stage == Tile::kD - 1) { + params.pointer -= (Tile::kD - 1) * kStageSize; + stage = 0; + } else { + params.pointer += kStageSize; + stage = stage + 1; + } + } + } + + public: + /// Loads a fragment and advances the iterator to the next tile. + template + CUTLASS_HOST_DEVICE void load_post_increment(Fragment &fragment, PredicateIterator pred_it) { + FragmentIterator frag_iterator(fragment); + + for (int d = 0; d < Iterations::kD; ++d) { + for (int h = 0; h < Iterations::kH; ++h) { + for (int w = 0; w < Iterations::kW; ++w, ++pred_it) { + if (*pred_it) { + Load::load( + reinterpret_cast(frag_iterator.at(d, h, w, 0)), data(), 0); + } + + if (w < Iterations::kW - 1) { + inc_w(); + } + } + if (h < Iterations::kH - 1) { + inc_h(); + } + } + if (d < Iterations::kD - 1) { + inc_d(); + } + } + inc_advance(); + } + + /// Loads a fragment and advances the iterator to the next tile. + template + CUTLASS_HOST_DEVICE void load_post_increment(Fragment &fragment) { + typename PredicateVector::TrivialIterator pred_it; + load_post_increment(fragment, pred_it); + } + + /// Loads a fragment without advancing the iterator.. + template + CUTLASS_HOST_DEVICE void load(Fragment &fragment, PredicateIterator pred_it) const { + TileLoadIterator _load_it(*this); + _load_it.load_post_increment(fragment, pred_it); + } + + /// Loads a fragment without advancing the iterator.. + template + CUTLASS_HOST_DEVICE void load(Fragment &fragment) const { + typename PredicateVector::TrivialIterator pred_it; + load(fragment, pred_it); + } +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/*!@defgroup tile_store_iterator_concept Tile Store Iterator Concept +@{ + +\ref tile_store_iterator_concept enables storing a tile to addressable memory + +@par Tile Store Iterator Concept + Types satisfying \ref tile_load_iterator_concept define the following members + - PredicateVector - a \ref predicate_vector_concept with sufficient predicate storage for +each access implied by the tile traits + - Fragment - the destination fragment type satisfying \ref fragment_concept + - initialize_predicates(pred_it, bounds, block_offset) - function initializing a predicate +vector according to externally specified bounds + - store_post_increment(fragment, pred_it) - a method that stores a fragment and increments +the iterator to the next tile, guarded by a \ref predicate_iterator_concept + - store_post_increment(fragment) - a method that stores a fragment and increments the +iterator to the next tile + - store(fragment, pred_it) - a const method that stores a fragment, guarded by a \ref +predicate_iterator_concept + - store(fragment) - a method that loads a fragment + +@} +*/ + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/** +* @brief An iterator implementing \ref tile_store_iterator_concept for storing a tile to memory +* @concept{tile_store_iterator_concept} +*/ +template > +struct TileStoreIterator : public TileIteratorBase { + /// Base class + typedef TileIteratorBase + Base; + + /// concept TileTraits + typedef typename Base::Traits Traits; + + /// Scalar element + typedef typename Base::Scalar Scalar; + + /// Fragment element + typedef typename Base::FragmentElement FragmentElement; + + /// Specifies in which dimension post-increment accesses advance. + static IteratorAdvance::Kind const kAdvance = Base::kAdvance; + + /// Specifies type of iterator fragment storage (Salar or WmmaMatrix) + static IteratorFragment::Kind const kIteratorFragment = Base::kIteratorFragment; + + /// Source or destination memory space + static MemorySpace::Kind const kMemorySpace = Base::kMemorySpace; + + /// Index type + typedef typename Base::Index Index; + + /// Skew quantity + typedef typename Base::Skew Skew; + + /// Tile shape + typedef typename Base::Tile Tile; + + /// Delta + typedef typename Base::Delta Delta; + + /// Iterations + typedef typename Base::Iterations Iterations; + + /// ThreadOffset functor + typedef typename Base::ThreadOffset ThreadOffset; + + /// Fragment type + typedef typename Base::FragmentShape FragmentShape; + + /// Memory access type + typedef typename Base::AccessType AccessType; + + /// Fragment definition + typedef typename Base::Fragment Fragment; + + /// Fragment iterator definition + typedef typename Base::FragmentIterator FragmentIterator; + + /// Fragment const iterator definition + typedef typename Base::FragmentConstIterator FragmentConstIterator; + + /// Default predicate mask type + typedef typename Base::PredicateVector PredicateVector; + + /// Storage object which may be stored to + typedef typename Base::Storage SharedStorage; + + /// IteratorBase parameters + typedef typename Base::Params BaseParams; + + /// Parameters + struct Params : public BaseParams { + /// Pointer to memory + Scalar *pointer; + + /// Initialize params to access storage object + CUTLASS_HOST_DEVICE + int initialize(SharedStorage &storage) { + pointer = &storage[0]; + return 0; + } + + /// Initializes params to access a raw pointer + CUTLASS_HOST_DEVICE + int initialize(Scalar *ptr, Index stride_d, Index stride_h, Index stride_w) { + Base::Params::initialize(stride_d, stride_h, stride_w); + pointer = ptr; + return 0; + } + + /// Initializes params + CUTLASS_HOST_DEVICE + int initialize(Scalar *ptr, + Index _stride_d, + Index _stride_h, + Index _stride_w, + Index _inc_d, + Index _inc_h, + Index _inc_w, + Index _inc_advance) { + pointer = ptr; + Base::Params::initialize( + _stride_d, _stride_h, _stride_w, _inc_d, _inc_h, _inc_w, _inc_advance); + return 0; + } + + /// Initializes params to default values + CUTLASS_HOST_DEVICE + int initialize() { return Base::Params::initialize(); } + }; + + // + // Data members + // + + /// Parameters structure + Params params; + + /// Offset of an individual lane from the start of the tile + Coord<4> thread_offset; + + /// The stage. + int stage; + + // + // Static member functions + // + + /// Initializes a predicate vector + template + CUTLASS_HOST_DEVICE void initialize_predicates(PredicateIterator predicate_it, + Coord<3> const &bounds, + Coord<3> const &block_offset = make_Coord(0, + 0, + 0)) { + Base::initialize_predicates( + predicate_it, + bounds, + block_offset + make_Coord(0, thread_offset[1], thread_offset[2] * Tile::kC)); + } + + // + // Methods + // + + /// Default constructor + CUTLASS_HOST_DEVICE + TileStoreIterator() {} + + /// Constructs a tile store iterator + CUTLASS_HOST_DEVICE + TileStoreIterator(Params const &_params, + Coord<3> const &block_offset = make_Coord(0, 0, 0), + ThreadOffset thread_offset_func = ThreadOffset()) + : params(_params), stage(0) { + thread_offset = thread_offset_func(); + + params.pointer += block_offset[0] * params.stride_d + + (block_offset[1] + thread_offset[1]) * params.stride_h + + (block_offset[2] + thread_offset[2] * Tile::kC) / Tile::kC * params.stride_w; + } + + /// Constructs a tile store iterator + CUTLASS_HOST_DEVICE + TileStoreIterator(Params const &, + SharedStorage &shared_storage, + Coord<3> const &block_offset = make_Coord(0, 0, 0), + ThreadOffset thread_offset_func = ThreadOffset()) + : stage(0) { + int const offset = thread_offset_func()[2]; + params.pointer = &shared_storage[offset]; + } + + /// Returns the current pointer + CUTLASS_HOST_DEVICE + Scalar *data() const { return params.pointer; } + + /// Increment in the D dimension + CUTLASS_HOST_DEVICE void inc_d() { params.pointer += params.inc_d; } + + /// Increment in the H dimension + CUTLASS_HOST_DEVICE void inc_h() { params.pointer += params.inc_h; } + + /// Increment in the W dimension + CUTLASS_HOST_DEVICE void inc_w() { params.pointer += params.inc_w; } + + /// Increment in the next dimension + CUTLASS_HOST_DEVICE void inc_advance() {} + + /// Increment the stage. + CUTLASS_DEVICE void inc_stage() { + if (Tile::kD > 1) { + int const kStageSize = Tile::kH * Tile::kW * Tile::kC; + if (stage == Tile::kD - 1) { + params.pointer -= (Tile::kD - 1) * kStageSize; + stage = 0; + } else { + params.pointer += kStageSize; + stage = stage + 1; + } + } + } + + /// The accessor. + CUTLASS_DEVICE void set(AccessType const &value, int d, int h, int w, int c) { + int const imm = + ComputeOffsetFromStrides::get(d, h, w, c); + Store::store(value, params.pointer, imm); + } + + public: + /// Stores a fragment and advances to the next tile. + template + CUTLASS_HOST_DEVICE void store_post_increment(Fragment &fragment, PredicateIterator pred_it) { + FragmentIterator frag_iterator(fragment); + + for (int d = 0; d < Iterations::kD; ++d) { + for (int h = 0; h < Iterations::kH; ++h) { + for (int w = 0; w < Iterations::kW; ++w, ++pred_it) { + if (*pred_it) { + Store::store( + reinterpret_cast(frag_iterator.at(d, h, w, 0)), data(), 0); + } + if (w < Iterations::kW - 1) { + inc_w(); + } + } + if (h < Iterations::kH - 1) { + inc_h(); + } + } + if (d < Iterations::kD - 1) { + inc_d(); + } + } + inc_advance(); + } + + /// Stores a fragment and advances to the next tile. + template + CUTLASS_HOST_DEVICE void store_post_increment(Fragment &fragment) { + typename PredicateVector::TrivialIterator pred_it; + store_post_increment(fragment, pred_it); + } + + /// Stores a fragment without advancing the iterator. + template + CUTLASS_HOST_DEVICE void store(Fragment &fragment, PredicateIterator pred_it) const { + TileStoreIterator _store_it(*this); + _store_it.store_post_increment(fragment, pred_it); + } + + /// Stores a fragment without advancing the iterator. + template + CUTLASS_HOST_DEVICE void store(Fragment &fragment) const { + typename PredicateVector::TrivialIterator pred_it; + store(fragment, pred_it); + } +}; +} diff --git a/cutlass-example/cutlass/tile_traits_standard.h b/cutlass-example/cutlass/tile_traits_standard.h new file mode 100644 index 0000000..14ecd01 --- /dev/null +++ b/cutlass-example/cutlass/tile_traits_standard.h @@ -0,0 +1,238 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines tile traits for several tile partitioning arrangements of threads expected to + achieve efficient streaming performance. +*/ +#pragma once + +#include + +namespace cutlass { + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Basic thread offset function computed from a thread shape +template +struct TiledThreadOffset { + /// Computes the logical coordinate from thread shape + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + Coord<4> thread_offset; + + int index = threadIdx.x; + + thread_offset[3] = (index % ThreadShape::kC); + index = (index / ThreadShape::kC); + + thread_offset[2] = (index % ThreadShape::kW); + index = (index / ThreadShape::kW); + + thread_offset[1] = (index % ThreadShape::kH); + index = (index / ThreadShape::kH); + + thread_offset[0] = index; + + return thread_offset; + } +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Tiling in which the number of threads is greater than the +/// contiguous dimension of the tile. +template +struct TileTraitsStrideMajor { + /// Shape of tile + typedef Tile_ Tile; + + /// Number of participating threads + static int const kThreads = Threads; + + // Static assertions + static_assert(!(ShapeCount::kDhw % kThreads), + "Tiling undefined if elements not divisible by threads."); + + static_assert(Tile::kW <= kThreads, + "This specialization assumes there are more threads than the contiguous dimension " + "of the tile."); + + /// Shape of threads + typedef Shape<1, kThreads / Tile::kW, Tile::kW, 1> ThreadShape; + + /// Delta along each dimension + typedef Shape<1, ThreadShape::kH, 1, 1> Delta; + + /// Number of iterations + typedef Shape<1, Tile::kH / ThreadShape::kH, 1, 1> Iterations; + + /// Computes the initial offset + typedef TiledThreadOffset ThreadOffset; +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Tiling in which the number of threads is fewer than the tile size +/// in the contiguous dimension. +template +struct TileTraitsContiguousMajor { + /// Shape of tile + typedef Tile_ Tile; + + /// Number of participating threads + static int const kThreads = Threads; + + // Static assertions + static_assert(Tile::kW >= kThreads, + "This specialization assumes there are more threads than the contiguous dimension " + "of the tile."); + + static_assert(!(ShapeCount::kDhw % kThreads), + "Tiling undefined if elements not divisible by threads."); + + static_assert(!(Tile::kW % kThreads), + "The contiguous size of the tile must be divisible by the number of threads."); + + /// Thread shape + typedef Shape<1, 1, kThreads> ThreadShape; + + /// Delta between each thread's access + typedef Shape<1, 1, kThreads> Delta; + + /// Number of iterations + typedef Shape<1, Tile::kH, Tile::kW / kThreads> Iterations; + + /// Computes the initial offset + typedef TiledThreadOffset ThreadOffset; +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Tiling in which warps rake across the contiguous dimension +template +struct TileTraitsWarpRake { + /// Shape of tile + typedef Tile_ Tile; + + /// Number of participating threads + static int const kThreads = Threads; + + /// Hard-coded warp size + static int const kWarpSize = 32; + + /// Number of participating warps + static int const kWarpCount = kThreads / kWarpSize; + + // Static assertions + static_assert(!(ShapeCount::kDhw % kThreads), + "Tiling undefined if elements not divisible by threads."); + + static_assert(!(kThreads % kWarpSize), "Number of threads must be divisible by the warp size."); + + static_assert(!(Tile::kW % kWarpSize), "Contiguous dimension must be divisible by the warp size"); + + /// Warps strip-mined across strided dimension + static int const kWarpsStrided = __NV_STD_MIN(kWarpCount, Tile::kH); + + /// Warps stripmined contiguous dimension + static int const kWarpsContiguous = kWarpCount / kWarpsStrided; + + /// Arrangement of threads + typedef Shape<1, kWarpsStrided, kWarpsContiguous * kWarpSize> ThreadShape; + + /// The same warp rakes along the contiguous dimension + typedef Shape<1, kWarpsStrided, kWarpSize> Delta; + + /// Number of iterations + typedef Shape<1, Tile::kH / Delta::kH, Tile::kW / ThreadShape::kW> Iterations; + + /// Computes the thread offset in (H, W) based on thread ID + struct ThreadOffset { + /// Basic thread offset function computed from a thread shape + CUTLASS_HOST_DEVICE + Coord<4> operator()() const { + int tid = threadIdx.x; + int warp = (tid / kWarpSize); + int lane = (tid % kWarpSize); + + static int const kWarpSpanContiguous = kWarpSize * Iterations::kW; + + int warp_w = (warp % kWarpsContiguous); + int warp_h = (warp / kWarpsContiguous); + + return make_Coord(0, warp_h, lane + kWarpSpanContiguous * warp_w, 0); + } + }; +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Chooses 'best' shape to enable warp raking along contiguous dimension if possible. +template +struct TileTraitsStandard { + /// Shape of tile + typedef Tile_ Tile; + + /// Number of participating threads + static int const kThreads = Threads; + + /// Hard-coded warp size + static int const kWarpSize = 32; + + /// Number of participating warps + static int const kWarpCount = kThreads / kWarpSize; + + // Static assertions + static_assert(!(ShapeCount::kDhw % kThreads), + "Tiling undefined if elements not divisible by threads."); + + /// Choose the stride-major contiguous tiling if the contiguous dimension is + /// smaller than the warp size. Otherwise, if it is divisible by the warp size, + /// choose the warp rake arrangement. + typedef typename platform::conditional < + Tile::kW, + typename platform::conditional, + TileTraitsContiguousMajor >::type>:: + type Traits; + + /// Delta between accesses + typedef typename Traits::Delta Delta; + + /// Delta between each thread's access + /// TODO MTA this is wrong for sure, but Delta is used for stride computation at the moment + typedef Delta ImmediateOffsetStrides; + + /// Number of accesses + typedef typename Traits::Iterations Iterations; + + /// Thread offset functor + typedef typename Traits::ThreadOffset ThreadOffset; +}; + +/////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/util/cutlass_math.h b/cutlass-example/cutlass/util/cutlass_math.h new file mode 100644 index 0000000..0ecdc43 --- /dev/null +++ b/cutlass-example/cutlass/util/cutlass_math.h @@ -0,0 +1,131 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ + +#pragma once + +/** + * \file + * \brief Math utilities + */ + +#include + +namespace cutlass { + +/****************************************************************************** + * Static math utilities + ******************************************************************************/ + +/** + * Statically determine if N is a power-of-two + */ +template +struct is_pow2 : platform::integral_constant {}; + +/** + * Statically determine log2(N), rounded down + */ +template +struct log2_down { + /// Static logarithm value + enum { value = log2_down> 1), Count + 1>::value }; +}; + +// Base case +template +struct log2_down { + enum { value = Count }; +}; + +/** + * Statically determine log2(N), rounded up + */ +template +struct log2_up { + /// Static logarithm value + enum { value = log2_up> 1), Count + 1>::value }; +}; + +// Base case +template +struct log2_up { + enum { value = ((1 << Count) < N) ? Count + 1 : Count }; +}; + +/** + * Statically estimate sqrt(N) to the nearest power-of-two + */ +template +struct sqrt_est { + enum { value = 1 << (log2_up::value / 2) }; +}; + +/** + * For performing a constant-division with a compile-time assertion that the + * Divisor evenly-divides the Dividend. + */ +template +struct divide_assert { + enum { value = Dividend / Divisor }; + + static_assert((Dividend % Divisor == 0), "Not an even multiple"); +}; + +/****************************************************************************** + * Rounding + ******************************************************************************/ + +/** + * Round dividend up to the nearest multiple of divisor + */ +template +CUTLASS_HOST_DEVICE dividend_t round_nearest(dividend_t dividend, divisor_t divisor) { + return ((dividend + divisor - 1) / divisor) * divisor; +} + +/** + * Greatest common divisor + */ +template +CUTLASS_HOST_DEVICE value_t gcd(value_t a, value_t b) { + for (;;) { + if (a == 0) return b; + b %= a; + if (b == 0) return a; + a %= b; + } +} + +/** + * Least common multiple + */ +template +CUTLASS_HOST_DEVICE value_t lcm(value_t a, value_t b) { + value_t temp = gcd(a, b); + + return temp ? (a / temp * b) : 0; +} + +} // namespace cutlass diff --git a/cutlass-example/cutlass/util/debug.h b/cutlass-example/cutlass/util/debug.h new file mode 100644 index 0000000..6055e3f --- /dev/null +++ b/cutlass-example/cutlass/util/debug.h @@ -0,0 +1,122 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ + +#pragma once + +/** + * \file + * \brief Debugging and logging functionality + */ + +#include + +namespace cutlass { + +/****************************************************************************** + * Debug and logging macros + ******************************************************************************/ + +/** + * Formats and prints the given message to stdout + */ +#if !defined(CUDA_LOG) +#if !defined(__CUDA_ARCH__) +#define CUDA_LOG(format, ...) printf(format, __VA_ARGS__) +#else +#define CUDA_LOG(format, ...) \ + printf("[block (%d,%d,%d), thread (%d,%d,%d)]: " format, \ + blockIdx.x, \ + blockIdx.y, \ + blockIdx.z, \ + threadIdx.x, \ + threadIdx.y, \ + threadIdx.z, \ + __VA_ARGS__); +#endif +#endif + +/** + * Formats and prints the given message to stdout only if DEBUG is defined + */ +#if !defined(CUDA_LOG_DEBUG) +#ifdef DEBUG +#define CUDA_LOG_DEBUG(format, ...) CUDA_LOG(format, __VA_ARGS__) +#else +#define CUDA_LOG_DEBUG(format, ...) +#endif +#endif + +/** + * \brief The corresponding error message is printed to \p stderr (or \p stdout in device code) + * along with the supplied source context. + * + * \return The CUDA error. + */ +__host__ CUTLASS_DEVICE cudaError_t cuda_perror_impl(cudaError_t error, + const char* filename, + int line) { + (void)filename; + (void)line; + if (error) { +#if !defined(__CUDA_ARCH__) + fprintf( + stderr, "CUDA error %d [%s, %d]: %s\n", error, filename, line, cudaGetErrorString(error)); + fflush(stderr); +#else + printf("CUDA error %d [%s, %d]\n", error, filename, line); +#endif + } + return error; +} + +/** + * \brief Perror macro + */ +#ifndef CUDA_PERROR +#define CUDA_PERROR(e) cuda_perror_impl((cudaError_t)(e), __FILE__, __LINE__) +#endif + +/** + * \brief Perror macro with exit + */ +#ifndef CUDA_PERROR_EXIT +#define CUDA_PERROR_EXIT(e) \ + if (cuda_perror_impl((cudaError_t)(e), __FILE__, __LINE__)) { \ + exit(1); \ + } +#endif + +/** + * \brief Perror macro only if DEBUG is defined + */ +#ifndef CUDA_PERROR_DEBUG +#ifdef DEBUG +#define CUDA_PERROR_DEBUG(e) CUDA_PERROR(e) +#else +#define CUDA_PERROR_DEBUG(e) (e) +#endif +#endif + +} // namespace cutlass diff --git a/cutlass-example/cutlass/util/platform.h b/cutlass-example/cutlass/util/platform.h new file mode 100644 index 0000000..2a44c10 --- /dev/null +++ b/cutlass-example/cutlass/util/platform.h @@ -0,0 +1,801 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ + +#pragma once + +/** + * \file + * \brief C++ features that may be otherwise unimplemented for CUDA device functions. + * + * This file has three components: + * + * (1) Macros: + * - Empty macro defines for C++ keywords not supported by the current + * version of C++. These simply allow compilation to proceed (but do + * not provide the added semantics). + * - \p noexcept + * - \p constexpr + * - \p nullptr + * - \p static_assert + * + * - Macro functions that we need in constant expressions because the + * C++ equivalents require constexpr compiler support. These are + * prefixed with \p __NV_STD_* + * - \p __NV_STD_MAX + * - \p __NV_STD_MIN + * + * (2) Re-implementations of STL functions and types: + * - C++ features that need the \p __device__ annotation. These are + * placed into the \p platform namespace. + * - \p plus + * - \p less + * - \p greater + * - \p min + * - \p max + * - \p methods on std::pair (==, !=, <, <=, >, >=, and make_pair()) + * + * (3) Stop-gap implementations of unsupported STL functions and types: + * - STL functions and types defined by C++ 11/14/17/etc. that are not + * provided by the current version of C++. These are placed into the + * \p platform namespace + * - \p integral_constant + * - \p nullptr_t + * - \p true_type + * - \p false_type + * - \p bool_constant + * - \p enable_if + * - \p conditional + * - \p is_same + * - \p is_base_of + * - \p remove_const + * - \p remove_volatile + * - \p remove_cv + * - \p is_volatile + * - \p is_pointer + * - \p is_void + * - \p is_integral + * - \p is_floating_point + * - \p is_arithmetic + * - \p is_fundamental + * - \p is_trivially_copyable + * - \p alignment_of + * - \p aligned_storage + * + * (4) Functions and types that are STL-like (but aren't in the STL): + * - \p TODO: min and max functors? + * + * The idea is that, as we drop support for older compilers, we can simply #define + * the \p __NV_STD_XYZ macros and \p platform namespace to alias their C++ + * counterparts (or trivially find-and-replace their occurrences in code text). + */ + +//----------------------------------------------------------------------------- +// Dependencies +//----------------------------------------------------------------------------- + +#include + +#if !defined(__CUDACC_RTC__) +//----------------------------------------------------------------------------- +// Include STL files that platform provides functionality for +//----------------------------------------------------------------------------- + +#include // Minimum/maximum operations +#include // nullptr_t +#include // Arithmetic operations +#include // For methods on std::pair +#if (!defined(_MSC_VER) && (__cplusplus >= 201103L)) || (defined(_MSC_VER) && (_MS_VER >= 1500)) +#include // For integral constants, conditional metaprogramming, and type traits +#endif + +#include + +#endif +/****************************************************************************** + * Macros + ******************************************************************************/ +//----------------------------------------------------------------------------- +// Keywords +//----------------------------------------------------------------------------- + +/// noexcept, constexpr +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1900)) +#ifndef noexcept +#define noexcept +#endif +#ifndef constexpr +#define constexpr +#endif +#endif + +/// nullptr +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1310)) +#ifndef nullptr +#define nullptr 0 +#endif +#endif + +/// static_assert +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1600)) +#ifndef static_assert +#define __platform_cat_(a, b) a##b +#define __platform_cat(a, b) __platform_cat_(a, b) +#define static_assert(__e, __m) typedef int __platform_cat(AsSeRt, __LINE__)[(__e) ? 1 : -1] +#endif +#endif + +//----------------------------------------------------------------------------- +// Functions +//----------------------------------------------------------------------------- + +/// Select maximum(a, b) +#ifndef __NV_STD_MAX +#define __NV_STD_MAX(a, b) (((b) > (a)) ? (b) : (a)) +#endif + +/// Select minimum(a, b) +#ifndef __NV_STD_MIN +#define __NV_STD_MIN(a, b) (((b) < (a)) ? (b) : (a)) +#endif + +/****************************************************************************** + * Re-implementations + ******************************************************************************/ +namespace cutlass { +namespace platform { + +//----------------------------------------------------------------------------- +// Arithmetic operations, comparisons +//----------------------------------------------------------------------------- + +/// platform::plus +template +struct plus { + CUTLASS_HOST_DEVICE constexpr T operator()(const T& lhs, const T& rhs) const { return lhs + rhs; } +}; + +/// std::less +template +struct less { + CUTLASS_HOST_DEVICE constexpr bool operator()(const T& lhs, const T& rhs) const { + return lhs < rhs; + } +}; + +/// std::greater +template +struct greater { + CUTLASS_HOST_DEVICE constexpr bool operator()(const T& lhs, const T& rhs) const { + return lhs > rhs; + } +}; + +//----------------------------------------------------------------------------- +// Minimum/maximum operations +//----------------------------------------------------------------------------- + +/// std::min +template +CUTLASS_HOST_DEVICE constexpr const T& min(const T& a, const T& b) { + return (b < a) ? b : a; +} + +/// std::max +template +CUTLASS_HOST_DEVICE constexpr const T& max(const T& a, const T& b) { + return (a < b) ? b : a; +} + +#if !defined(__CUDACC_RTC__) +//----------------------------------------------------------------------------- +// Methods on std::pair +//----------------------------------------------------------------------------- + +using std::pair; + +template +CUTLASS_HOST_DEVICE constexpr bool operator==(const pair& lhs, const pair& rhs) { + return (lhs.first == rhs.first) && (lhs.second == rhs.second); +} + +template +CUTLASS_HOST_DEVICE constexpr bool operator!=(const pair& lhs, const pair& rhs) { + return (lhs.first != rhs.first) && (lhs.second != rhs.second); +} + +template +CUTLASS_HOST_DEVICE constexpr bool operator<(const pair& lhs, const pair& rhs) { + return (lhs.first < rhs.first) ? true : (rhs.first < lhs.first) ? false + : (lhs.second < rhs.second); +} + +template +CUTLASS_HOST_DEVICE constexpr bool operator<=(const pair& lhs, const pair& rhs) { + return !(rhs < lhs); +} + +template +CUTLASS_HOST_DEVICE constexpr bool operator>(const pair& lhs, const pair& rhs) { + return (rhs < lhs); +} + +template +CUTLASS_HOST_DEVICE constexpr bool operator>=(const pair& lhs, const pair& rhs) { + return !(lhs < rhs); +} + +template +CUTLASS_HOST_DEVICE std::pair make_pair(T1 t, T2 u) { + std::pair retval; + retval.first = t; + retval.second = u; + return retval; +} +#endif + +} // namespace platform + +/****************************************************************************** + * Implementations of C++ 11/14/17/... STL features + ******************************************************************************/ + +namespace platform { + +//----------------------------------------------------------------------------- +// Integral constant helper types +//----------------------------------------------------------------------------- + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1500)) + +/// std::integral_constant +template +struct integral_constant; + +/// std::integral_constant +template +struct integral_constant { + static const value_t value = V; + + typedef value_t value_type; + typedef integral_constant type; + + CUTLASS_HOST_DEVICE operator value_type() const { return value; } + + CUTLASS_HOST_DEVICE const value_type operator()() const { return value; } +}; + +#else + +using std::integral_constant; +using std::pair; + +#endif + +/// The type used as a compile-time boolean with true value. +typedef integral_constant true_type; + +/// The type used as a compile-time boolean with false value. +typedef integral_constant false_type; + +#if (!defined(_MSC_VER) && (__cplusplus <= 201402L)) || (defined(_MSC_VER) && (_MSC_VER < 1900)) + +/// std::bool_constant +template +struct bool_constant : platform::integral_constant {}; + +#else + +using std::bool_constant; + +#endif + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1700)) + +/// std::nullptr_t +struct nullptr_t {}; + +#else + +using std::nullptr_t; + +#endif + +//----------------------------------------------------------------------------- +// Conditional metaprogramming +//----------------------------------------------------------------------------- + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1600)) + +/// std::enable_if (true specialization) +template +struct enable_if { + typedef T type; +}; + +/// std::enable_if (false specialization) +template +struct enable_if {}; + +/// std::conditional (true specialization) +template +struct conditional { + typedef T type; +}; + +/// std::conditional (false specialization) +template +struct conditional { + typedef F type; +}; + +#else + +using std::enable_if; +using std::conditional; + +#endif + +//----------------------------------------------------------------------------- +// Const/volatility specifiers +//----------------------------------------------------------------------------- + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1500)) + +/// std::remove_const (non-const specialization) +template +struct remove_const { + typedef T type; +}; + +/// std::remove_const (const specialization) +template +struct remove_const { + typedef T type; +}; + +/// std::remove_volatile (non-volatile specialization) +template +struct remove_volatile { + typedef T type; +}; + +/// std::remove_volatile (volatile specialization) +template +struct remove_volatile { + typedef T type; +}; + +/// std::remove_cv +template +struct remove_cv { + typedef typename remove_volatile::type>::type type; +}; + +#else + +using std::remove_const; +using std::remove_volatile; +using std::remove_cv; + +#endif + +//----------------------------------------------------------------------------- +// Type relationships +//----------------------------------------------------------------------------- + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1500)) + +/// std::is_same (false specialization) +template +struct is_same : false_type {}; + +/// std::is_same (true specialization) +template +struct is_same : true_type {}; + +/// Helper for std::is_base_of +template +struct is_base_of_helper { + typedef char (&yes)[1]; + typedef char (&no)[2]; + + template + struct dummy { + CUTLASS_HOST_DEVICE operator B*() const; + CUTLASS_HOST_DEVICE operator D*(); + }; + + template + CUTLASS_HOST_DEVICE static yes check(DerivedT*, T); + + CUTLASS_HOST_DEVICE static no check(BaseT*, int); + + static const bool value = sizeof(check(dummy(), int())) == sizeof(yes); +}; + +/// std::is_base_of +template +struct is_base_of + : integral_constant::type, + typename remove_cv::type>::value) || + (is_same::type, + typename remove_cv::type>::value)> {}; + +#else + +using std::is_same; +using std::is_base_of; + +#endif + +//----------------------------------------------------------------------------- +// Type properties +//----------------------------------------------------------------------------- + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1500)) + +/// std::is_volatile +template +struct is_volatile : false_type {}; +template +struct is_volatile : true_type {}; + +/// Helper for std::is_pointer (false specialization) +template +struct is_pointer_helper : false_type {}; + +/// Helper for std::is_pointer (true specialization) +template +struct is_pointer_helper : true_type {}; + +/// std::is_pointer +template +struct is_pointer : is_pointer_helper::type> {}; + +/// std::is_void +template +struct is_void : is_same::type> {}; + +/// std::is_integral +template +struct is_integral : false_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template <> +struct is_integral : true_type {}; +template +struct is_integral : is_integral {}; +template +struct is_integral : is_integral {}; +template +struct is_integral : is_integral {}; + +/// std::is_floating_point +template +struct is_floating_point + : integral_constant::type>::value || + is_same::type>::value)> {}; + +/// std::is_arithmetic +template +struct is_arithmetic + : integral_constant::value || is_floating_point::value)> {}; + +/// std::is_fundamental +template +struct is_fundamental + : integral_constant::value || is_void::value || + is_same::type>::value)> {}; + +#else + +using std::is_volatile; +using std::is_pointer; +using std::is_void; +using std::is_integral; +using std::is_floating_point; +using std::is_arithmetic; +using std::is_fundamental; + +#endif + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1800)) || \ + (defined(__GNUG__) && (__GNUC__ < 5)) + +/** + * std::is_trivially_copyable + * + * This implementation only evaluates true if T is fundamental or pointer + * + * Without help from partial template specializations provided by the user for + * a specific class or struct, this trait will never report that the specified + * class or struct is trivially-copyable ; this is always safe, + * if possibly sub-optimal. + */ +template +struct is_trivially_copyable + : integral_constant::value || is_pointer::value)> {}; + +#else + +using std::is_trivially_copyable; + +#endif + +//----------------------------------------------------------------------------- +// Alignment and layout utilities +//----------------------------------------------------------------------------- + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1500)) + +/// std::alignment_of +template +struct alignment_of { + struct pad { + value_t val; + char byte; + }; + + enum { value = sizeof(pad) - sizeof(value_t) }; +}; + +#else + +template +struct alignment_of : std::alignment_of {}; + +#endif + +/* 16B specializations where 32-bit Win32 host compiler disagrees with device compiler */ +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; +template <> +struct alignment_of { + enum { value = 16 }; +}; + +// Specializations for volatile/const qualified types +template +struct alignment_of : alignment_of {}; +template +struct alignment_of : alignment_of {}; +template +struct alignment_of : alignment_of {}; + +#if (!defined(_MSC_VER) && (__cplusplus < 201103L)) || (defined(_MSC_VER) && (_MSC_VER < 1800)) + +template +struct aligned_chunk; +template <> +struct __align__(1) aligned_chunk<1> { + uint8_t buff; +}; +template <> +struct __align__(2) aligned_chunk<2> { + uint16_t buff; +}; +template <> +struct __align__(4) aligned_chunk<4> { + uint32_t buff; +}; +template <> +struct __align__(8) aligned_chunk<8> { + uint32_t buff[2]; +}; +template <> +struct __align__(16) aligned_chunk<16> { + uint32_t buff[4]; +}; +template <> +struct __align__(32) aligned_chunk<32> { + uint32_t buff[8]; +}; +template <> +struct __align__(64) aligned_chunk<64> { + uint32_t buff[16]; +}; +template <> +struct __align__(128) aligned_chunk<128> { + uint32_t buff[32]; +}; +template <> +struct __align__(256) aligned_chunk<256> { + uint32_t buff[64]; +}; +template <> +struct __align__(512) aligned_chunk<512> { + uint32_t buff[128]; +}; +template <> +struct __align__(1024) aligned_chunk<1024> { + uint32_t buff[256]; +}; +template <> +struct __align__(2048) aligned_chunk<2048> { + uint32_t buff[512]; +}; +template <> +struct __align__(4096) aligned_chunk<4096> { + uint32_t buff[1024]; +}; + +/// std::aligned_storage +template +struct aligned_storage { + typedef aligned_chunk type[Len / sizeof(aligned_chunk)]; +}; + +#else + +using std::aligned_storage; + +#endif + +#if !defined(__CUDACC_RTC__) +/// Default deleter +template +struct default_delete { + void operator()(T* ptr) const { delete ptr; } +}; + +/// Partial specialization for deleting array types +template +struct default_delete { + void operator()(T* ptr) const { delete[] ptr; } +}; + +/// std::unique_ptr +template > +class unique_ptr { + public: + typedef T* pointer; + typedef T element_type; + typedef Deleter deleter_type; + + private: + /// Pointer to memory + pointer _ptr; + + /// Deleter + deleter_type _deleter; + + public: + unique_ptr() : _ptr(nullptr) {} + unique_ptr(pointer p) : _ptr(p) {} + + ~unique_ptr() { + if (_ptr) { + _deleter(_ptr); + } + } + /// Returns a pointer to the managed object or nullptr if no object is owned. + pointer get() const noexcept { return _ptr; } + + /// Releases ownership of the managed object, if any + pointer release() noexcept { + pointer p(_ptr); + _ptr = nullptr; + return p; + } + + /// Replaces the managed object, deleting the old object. + void reset(pointer p = pointer()) noexcept { + pointer old_ptr = _ptr; + _ptr = p; + if (old_ptr != nullptr) { + get_deleter()(old_ptr); + } + } + + /// Swaps the managed objects with *this and another unique_ptr + void swap(unique_ptr& other) noexcept { std::swap(_ptr, other._ptr); } + + /// Returns the deleter object + Deleter& get_deleter() noexcept { return _deleter; } + + /// Returns the deleter object + Deleter const& get_deleter() const noexcept { return _deleter; } + + /// Checks whether an object is owned + operator bool() const noexcept { return _ptr != nullptr; } + + /// Dereferences the unique_ptr + T& operator*() const { return *_ptr; } + + /// Returns a pointer to the managed object + pointer operator->() const noexcept { return _ptr; } + + /// Array access to managed object + T& operator[](size_t i) const { return _ptr[i]; } +}; + +/// Specializes the swap algorithm +template +void swap(unique_ptr& lhs, unique_ptr& rhs) noexcept { + lhs.swap(rhs); +} +#endif + +}; // namespace platform +}; // namespace cutlass diff --git a/cutlass-example/cutlass/vector.h b/cutlass-example/cutlass/vector.h new file mode 100644 index 0000000..a66dfde --- /dev/null +++ b/cutlass-example/cutlass/vector.h @@ -0,0 +1,229 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Defines a 1D vector of elements held in the registers of each thread. +*/ +#pragma once + +#if !defined(__CUDACC_RTC__) || defined(CUTLASS_NVRTC_HAS_FP16) +#include +#endif + +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct AlignedStruct {}; + +template <> +struct __align__(1) AlignedStruct<1>{}; +template <> +struct __align__(2) AlignedStruct<2>{}; +template <> +struct __align__(4) AlignedStruct<4>{}; +template <> +struct __align__(8) AlignedStruct<8>{}; +template <> +struct __align__(16) AlignedStruct<16>{}; +template <> +struct __align__(32) AlignedStruct<32>{}; +template <> +struct __align__(64) AlignedStruct<64>{}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +union Vector { + /// The scalar type. + typedef Scalar_ Scalar; + + /// The number of elements in the vector. + enum { kLanes = kLanes_ }; + /// The size of the vector. + enum { kVectorSize = kLanes * (int)sizeof(Scalar) }; + /// The number of registers needed to store the vector. + enum { kRegisters = kVectorSize < 4 ? 1 : kVectorSize / 4 }; + + // Make sure that the vector type makes sense. + static_assert(kVectorSize <= 16, "Vector type is too large"); + + /// The aligned storage to make sure we have good alignment. + AlignedStruct aligned_; + /// The associated array of scalars. + Scalar scalars[kLanes]; + /// The data in registers. + uint32_t registers[kRegisters]; + + /// Accessor to the ith lane. + CUTLASS_DEVICE Scalar const& operator[](uint32_t i) const { return scalars[i]; } + /// Accessor to the ith lane. + CUTLASS_DEVICE Scalar& operator[](uint32_t i) { return scalars[i]; } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +#if !defined(__CUDACC_RTC__) || defined(CUTLASS_NVRTC_HAS_FP16) + +template +union Vector { + /// The scalar type. + typedef half Scalar; + + /// The number of elements in the vector. + enum { kLanes = kLanes_ }; + /// The size of the vector. + enum { kVectorSize = kLanes * (int)sizeof(Scalar) }; + /// The number of registers needed to store the vector. + enum { kRegisters = kVectorSize < 4 ? 1 : kVectorSize / 4 }; + + // Make sure that the vector type makes sense. + static_assert(kVectorSize <= size_t(16), "Vector type is too large"); + + /// The aligned storage to make sure we have good alignment. + AlignedStruct aligned_; + /// The associated array of scalars. + uint16_t scalars[kLanes]; + /// The data in registers. + uint32_t registers[kRegisters]; + + /// Accessor to the ith lane. + CUTLASS_DEVICE Scalar const& operator[](uint32_t i) const { + return reinterpret_cast(scalars[i]); + } + /// Accessor to the ith lane. + CUTLASS_DEVICE Scalar& operator[](uint32_t i) { return reinterpret_cast(scalars[i]); } +}; + +#endif + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +CUTLASS_DEVICE void make_zero(Scalar_& x) { + x = Scalar_(0); +} + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Vectorize { + typedef Vector Type; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +struct Vectorize { + typedef Element_ Type; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +template +CUTLASS_DEVICE void make_zero(Vector& vec) { + for (int i = 0; i < Vector::kRegisters; ++i) { + vec.registers[i] = 0; + } +} + +//////////////////////////////////////////////////////////////////////////////////////////////////// +// +// cutlass::Extent similar to std::extent but applicable to CUTLASS types +// + +/// Returns the extent of a scalar or vector +template +struct Extent { + static size_t const kValue = 1; +}; + +/// Returns the number of lanes of a vector if need be +template +struct Extent > { + static size_t const kValue = Lanes; +}; + +/// Returns the number of lanes of a vector if need be +template +struct Extent const> { + static size_t const kValue = Lanes; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Traits describing properties of vectors and scalar-as-vectors +template +struct VectorTraits { + /// Scalar type + typedef T Scalar; + + /// Number of lanes of vector + static int const kLanes = 1; + + /// True if the type is actually a cutlass::Vector, otherwise false + static bool const IsVector = false; + + /// Type that is always a vector + typedef Vector Vector; +}; + +/// Partial specialization for actual cutlass::Vector +template +struct VectorTraits > { + /// Scalar type + typedef T Scalar; + + /// Number of lanes of vector + static int const kLanes = Lanes; + + /// Type is actually a cutlass::Vector + static bool const IsVector = true; + + /// Type that is always a Vector + typedef Vector Vector; +}; + +/// Partial specialization for actual cutlass::Vector +template +struct VectorTraits const> { + /// Scalar type + typedef T Scalar; + + /// Number of lanes of vector + static int const kLanes = Lanes; + + /// Type is actually a cutlass::Vector + static bool const IsVector = true; + + /// Type that is always a Vector + typedef Vector Vector; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass diff --git a/cutlass-example/cutlass/wmma_matrix.h b/cutlass-example/cutlass/wmma_matrix.h new file mode 100644 index 0000000..c4d8a0b --- /dev/null +++ b/cutlass-example/cutlass/wmma_matrix.h @@ -0,0 +1,193 @@ +/*************************************************************************************************** + * Copyright (c) 2017-2018, NVIDIA CORPORATION. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without modification, are permitted + * provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright notice, this list of + * conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright notice, this list of + * conditions and the following disclaimer in the documentation and/or other materials + * provided with the distribution. + * * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used + * to endorse or promote products derived from this software without specific prior written + * permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, + * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; + * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************************************/ +/*! \file + \brief Abstractions for loading and storing matrices using the CUDA WMMA API. +*/ +#pragma once + +#if defined(__CUDACC__) && (!defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 700) + +// Dependent header files should use the following macro to guard all code using +// nvcuda::wmma:: to enable compilation for CUDA Compute Capabilities < sm_70. +// Earlier shader models not support Tensor Cores. +#define CUTLASS_USE_WMMA_API + +#include "stdio.h" + +#include +#include +#include +#include +#include +#include + +namespace cutlass { + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Statically maps cutlass::MatrixLayout => nvcuda::wmma layout tags +template +struct WmmaLayout { + typedef nvcuda::wmma::col_major Layout; +}; + +/// Statically maps cutlass::MatrixLayout => nvcuda::wmma layout tags +template <> +struct WmmaLayout { + typedef nvcuda::wmma::row_major Layout; +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Adapter to nvcuda::wmma fragment load and store operations +template +struct WmmaMatrix {}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Adapter to nvcuda::wmma fragment accessors for A operand +template +struct WmmaMatrix + : public nvcuda::wmma::fragment< + /// The nvcuda::wmma operand name. + nvcuda::wmma::matrix_a, + /// The dimensions. + WmmaShape_::kW, + WmmaShape_::kH, + WmmaShape_::kD, + /// The scalar. + Scalar_, + /// The layout. + typename WmmaLayout::Layout> { + /// This type. + typedef WmmaMatrix This_; + + /// Fill-in the element. + CUTLASS_DEVICE This_& operator=(Scalar_ const& x) { + nvcuda::wmma::fill_fragment(*this, x); + return *this; + } + + /// Load from memory. + CUTLASS_DEVICE void load(Scalar_ const* pointer, int const stride) { + nvcuda::wmma::load_matrix_sync(*this, pointer, stride); + } + + /// Store to memory. + CUTLASS_DEVICE void store(Scalar_* pointer, int const stride) const { + nvcuda::wmma::store_matrix_sync(pointer, *this, stride); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Adapter to nvcuda::wmma fragment accessors for B operand +template +struct WmmaMatrix + : public nvcuda::wmma::fragment< + /// The nvcuda::wmma operand name. + nvcuda::wmma::matrix_b, + /// The dimensions. + WmmaShape_::kW, + WmmaShape_::kH, + WmmaShape_::kD, + /// The scalar. + Scalar_, + /// The layout. + typename WmmaLayout::Layout> { + /// This type. + typedef WmmaMatrix This_; + + /// Fill-in the element. + CUTLASS_DEVICE This_& operator=(Scalar_ const& x) { + nvcuda::wmma::fill_fragment(*this, x); + return *this; + } + + /// Load from memory. + CUTLASS_DEVICE void load(Scalar_ const* pointer, int const stride) { + nvcuda::wmma::load_matrix_sync(*this, pointer, stride); + } + + /// Store to memory. + CUTLASS_DEVICE void store(Scalar_* pointer, int const stride) const { + nvcuda::wmma::store_matrix_sync(pointer, *this, stride); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +/// Adapter to nvcuda::wmma fragment accessors for C operand +template +struct WmmaMatrix + : public nvcuda::wmma::fragment< + /// The nvcuda::wmma operand name. + nvcuda::wmma::accumulator, + /// The dimensions. + WmmaShape_::kW, + WmmaShape_::kH, + WmmaShape_::kD, + /// The scalar. + Scalar_> { + /// This type. + typedef WmmaMatrix This_; + /// The layout. + static MatrixLayout::Kind const kLayout = kLayout_; + + /// Fill-in the element. + CUTLASS_DEVICE This_& operator=(Scalar_ const& x) { + nvcuda::wmma::fill_fragment(*this, x); + return *this; + } + + /// Load from memory. + CUTLASS_DEVICE void load(Scalar_ const* pointer, int const stride) { + bool const kIsRowMajor = kLayout == MatrixLayout::kRowMajor; + nvcuda::wmma::load_matrix_sync( + *this, + pointer, + stride, + kIsRowMajor ? nvcuda::wmma::mem_row_major : nvcuda::wmma::mem_col_major); + } + + /// Store to memory. + CUTLASS_DEVICE void store(Scalar_* pointer, int const stride) const { + bool const kIsRowMajor = kLayout == MatrixLayout::kRowMajor; + nvcuda::wmma::store_matrix_sync( + pointer, + *this, + stride, + kIsRowMajor ? nvcuda::wmma::mem_row_major : nvcuda::wmma::mem_col_major); + } +}; + +//////////////////////////////////////////////////////////////////////////////////////////////////// + +} // namespace cutlass + +#endif // defined CUTLASS_USE_WMMA_API -- cgit v1.3