[SYCL][CUDA] Implementation of matrix ext using new "unified" interface #7077
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,221 @@ | ||
| //===------- matrix-unified.hpp - SYCL matrix extension ----*- C++ -*------===// | ||
| // | ||
| // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. | ||
| // See https://llvm.org/LICENSE.txt for license information. | ||
| // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception | ||
| // | ||
| // ===--------------------------------------------------------------------=== // | ||
|
|
||
| #pragma once | ||
| #include <sycl/ext/oneapi/matrix/matrix-tensorcores.hpp> | ||
|
|
||
| namespace sycl { | ||
| __SYCL_INLINE_VER_NAMESPACE(_V1) { | ||
| namespace ext { | ||
| namespace oneapi { | ||
| namespace experimental { | ||
| namespace matrix { | ||
|
|
||
| template <typename Group, typename T, use Use, size_t Rows, size_t Cols, | ||
| layout Layout> | ||
| struct joint_matrix { | ||
|
|
||
| #if defined(__SYCL_DEVICE_ONLY__) && defined(__SPIR__) | ||
| // TODO: Intel case here: we use the ext_oneapi_cuda case also for the host, | ||
| // because the Intel SPIRV functions will not be host compilable. | ||
| #else | ||
| sycl::ext::oneapi::detail::joint_matrix_cuda<T, Use, Rows, Cols, Layout> | ||
| cuda_impl; | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) && defined(__SPIR__) | ||
|
|
||
| joint_matrix() { | ||
| #ifndef __SYCL_DEVICE_ONLY__ | ||
| throw runtime_error("joint matrix is not supported on host device.", | ||
| PI_ERROR_INVALID_DEVICE); | ||
| #endif | ||
| } | ||
| }; | ||
|
|
||
| template <typename Group, typename T, use Use, size_t Rows, size_t Cols, | ||
| layout Layout> | ||
| inline __SYCL_ALWAYS_INLINE wi_data<Group, T, Use, Rows, Cols, Layout> | ||
| get_wi_data(Group sg, joint_matrix<Group, T, Use, Rows, Cols, Layout> &jm) { | ||
| #if defined(__SYCL_DEVICE_ONLY__) | ||
| #if defined(__NVPTX__) | ||
| std::ignore = sg; | ||
| return wi_data(jm); | ||
| #else | ||
| // TODO add Intel impl. | ||
|
There was a problem hiding this comment. @AerialMantis @JackAKirk @dkhaldi |
||
| #endif // defined(__NVPTX__) | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) | ||
| } | ||
|
|
||
| template <typename Group, typename T, size_t NumRows, size_t NumCols, use Use, | ||
| layout Layout, typename T2> | ||
| inline __SYCL_ALWAYS_INLINE void | ||
| joint_matrix_fill(Group sg, | ||
|
There was a problem hiding this comment. I noticed you don't have any test with joint_matrix_fill. Is that on purpose? There was a problem hiding this comment. There was a problem hiding this comment. We only have device code tests in intel/llvm for functions that call nvptx builtins. |
||
| joint_matrix<Group, T, Use, NumRows, NumCols, Layout> &res, | ||
| const T2 &v) { | ||
| #if defined(__SYCL_DEVICE_ONLY__) | ||
| #if defined(__NVPTX__) | ||
| std::ignore = sg; | ||
| res.cuda_impl.wi_marray = v; | ||
| #endif // defined(__NVPTX__) | ||
| #else | ||
| std::ignore = sg; | ||
| std::ignore = res; | ||
| std::ignore = v; | ||
| throw runtime_error( | ||
| "This version of the matrix extension is only currently supported on " | ||
| "Nvidia devices", | ||
| PI_ERROR_INVALID_DEVICE); | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) | ||
| } | ||
|
|
||
| template < | ||
| typename Group, typename S, typename T, size_t NumRows, size_t NumCols, | ||
| access::address_space Space, access::decorated IsDecorated, | ||
| std::enable_if_t<std::is_same<S, std::remove_const_t<T>>::value, bool> = | ||
| true> | ||
| inline __SYCL_ALWAYS_INLINE void joint_matrix_load( | ||
| Group sg, | ||
| joint_matrix<Group, S, use::accumulator, NumRows, NumCols, | ||
| sycl::ext::oneapi::experimental::matrix::layout::dynamic> &res, | ||
| multi_ptr<T, Space, IsDecorated> src, size_t stride, | ||
| sycl::ext::oneapi::experimental::matrix::layout Layout) { | ||
| #if defined(__SYCL_DEVICE_ONLY__) | ||
| #if defined(__NVPTX__) | ||
| std::ignore = sg; | ||
| sycl::ext::oneapi::detail::load_accumulator_cuda(res.cuda_impl, src, stride, | ||
| Layout); | ||
| #endif // defined(__NVPTX__) | ||
| #else | ||
| std::ignore = sg; | ||
| std::ignore = res; | ||
| std::ignore = src; | ||
| std::ignore = stride; | ||
| throw runtime_error( | ||
| "This version of the matrix extension is only currently supported on " | ||
| "Nvidia devices", | ||
| PI_ERROR_INVALID_DEVICE); | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) | ||
| } | ||
|
|
||
| template < | ||
| typename Group, typename S, typename T, use Use, size_t NumRows, | ||
| size_t NumCols, matrix::layout Layout, access::address_space Space, | ||
| access::decorated IsDecorated, | ||
| std::enable_if_t<std::is_same<S, std::remove_const_t<T>>::value || | ||
| (std::is_same<S, precision::tf32>::value && | ||
| std::is_same<std::remove_const_t<T>, float>::value), | ||
| bool> = true> | ||
| inline __SYCL_ALWAYS_INLINE void | ||
| joint_matrix_load(Group sg, | ||
| joint_matrix<Group, S, Use, NumRows, NumCols, Layout> &res, | ||
| multi_ptr<T, Space, IsDecorated> src, size_t stride) { | ||
| #if defined(__SYCL_DEVICE_ONLY__) | ||
| #if defined(__NVPTX__) | ||
| std::ignore = sg; | ||
| sycl::ext::oneapi::detail::load_multiplicand_cuda<S, T, NumRows, NumCols, Use, | ||
| Layout, Space>( | ||
| res.cuda_impl, src, stride); | ||
| #endif // defined(__NVPTX__) | ||
| #else | ||
| std::ignore = sg; | ||
| std::ignore = res; | ||
| std::ignore = src; | ||
| std::ignore = stride; | ||
| throw runtime_error( | ||
| "This version of the matrix extension is only currently supported on " | ||
| "Nvidia devices", | ||
| PI_ERROR_INVALID_DEVICE); | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) | ||
| } | ||
|
|
||
| template <typename Group, typename T, size_t NumRows, size_t NumCols, | ||
| access::address_space Space, access::decorated IsDecorated> | ||
| inline __SYCL_ALWAYS_INLINE void joint_matrix_store( | ||
| Group sg, | ||
| joint_matrix<Group, T, use::accumulator, NumRows, NumCols, | ||
| sycl::ext::oneapi::experimental::matrix::layout::dynamic> &src, | ||
| multi_ptr<T, Space, IsDecorated> dst, size_t stride, | ||
| sycl::ext::oneapi::experimental::matrix::layout Layout) { | ||
| #if defined(__SYCL_DEVICE_ONLY__) | ||
| #if defined(__NVPTX__) | ||
| std::ignore = sg; | ||
| sycl::ext::oneapi::detail::joint_matrix_store_cuda<T, NumRows, NumCols, | ||
| Space>(src.cuda_impl, dst, | ||
| stride, Layout); | ||
| #endif // defined(__NVPTX__) | ||
| #else | ||
| std::ignore = sg; | ||
| std::ignore = src; | ||
| std::ignore = dst; | ||
| std::ignore = stride; | ||
| throw runtime_error( | ||
| "This version of the matrix extension is only currently supported on " | ||
| "Nvidia devices", | ||
| PI_ERROR_INVALID_DEVICE); | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) | ||
| } | ||
|
|
||
| template <typename Group, typename Ta, typename Tb, typename Tc, std::size_t M, | ||
| std::size_t K, std::size_t N, layout LayoutA, layout LayoutB> | ||
| inline __SYCL_ALWAYS_INLINE | ||
| joint_matrix<Group, Tc, use::accumulator, M, N, | ||
| sycl::ext::oneapi::experimental::matrix::layout::dynamic> | ||
| joint_matrix_mad( | ||
| Group sg, joint_matrix<Group, Ta, use::a, M, K, LayoutA> &A, | ||
| joint_matrix<Group, Tb, use::b, K, N, LayoutB> &B, | ||
| joint_matrix<Group, Tc, use::accumulator, M, N, | ||
| sycl::ext::oneapi::experimental::matrix::layout::dynamic> | ||
| &C) { | ||
| #if defined(__SYCL_DEVICE_ONLY__) | ||
| #if defined(__NVPTX__) | ||
| std::ignore = sg; | ||
| if constexpr (std::is_same<Ta, Tb>::value) { | ||
| joint_matrix<Group, Tc, use::accumulator, M, N, | ||
| sycl::ext::oneapi::experimental::matrix::layout::dynamic> | ||
| D; | ||
| sycl::ext::oneapi::detail::joint_matrix_mad_cuda<Ta, Tc, M, K, N, LayoutA, | ||
| LayoutB>( | ||
| D.cuda_impl, A.cuda_impl, B.cuda_impl, C.cuda_impl); | ||
| return D; | ||
| } else { | ||
| assert(false && "Ta != Tb : In the CUDA backend joint_matrix_mad " | ||
| "requires that joint_matrix data types Ta and Tb match"); | ||
| } | ||
| #endif // defined(__NVPTX__) | ||
| #else | ||
| std::ignore = sg; | ||
| std::ignore = A; | ||
| std::ignore = B; | ||
| std::ignore = C; | ||
| throw runtime_error( | ||
| "This version of the matrix extension is only currently supported on " | ||
| "Nvidia devices", | ||
| PI_ERROR_INVALID_DEVICE); | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) | ||
| } | ||
|
|
||
| // This function rounds the bottom 13 bits up or down, and then zeros out the | ||
| // bottom bits | ||
| inline __SYCL_ALWAYS_INLINE float round_to_tf32(float &a) { | ||
| #if defined(__SYCL_DEVICE_ONLY__) && defined(__NVPTX__) | ||
| int32_t tmp_int = __nvvm_f2tf32_rna(a); | ||
| return __nvvm_bitcast_i2f(tmp_int); | ||
| #else | ||
| uint32_t tmp_uint = reinterpret_cast<uint32_t &>(a); | ||
| tmp_uint += 0x1000u; | ||
| tmp_uint &= 0xFFFFE000u; | ||
| float ret = reinterpret_cast<float &>(tmp_uint); | ||
| return ret; | ||
| #endif // defined(__SYCL_DEVICE_ONLY__) && defined(__NVPTX__) | ||
| } | ||
|
|
||
| } // namespace matrix | ||
| } // namespace experimental | ||
| } // namespace oneapi | ||
| } // namespace ext | ||
| } // __SYCL_INLINE_VER_NAMESPACE(_V1) | ||
| } // namespace sycl | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@AerialMantis @JackAKirk
sorry, i remember previously it is:
in intel side, we can't let host compilation use sycl::ext::oneapi::detail::joint_matrix_cuda. so i go back to the previous code and i can still get passed in cuda's testcases.