cmake list

client_example/25_tensor_transforms/tensor_transform.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2023, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <iostream>
+
+#include "ck/ck.hpp"
+
+#include "ck/utility/number.hpp"
+#include "ck/utility/tuple.hpp"
+#include "ck/utility/sequence.hpp"
+
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_description/multi_index_transform_helper.hpp"
+
+static constexpr auto I0 = ck::Number<0>{};
+static constexpr auto I1 = ck::Number<1>{};
+static constexpr auto I2 = ck::Number<2>{};
+
+using DataType = int;
+
+template <typename Desc>
+void Print1d(const Desc& desc)
+{
+    std::cout << "Print1d" << std::endl;
+    for(ck::index_t w = 0; w < desc.GetLength(I0); w++)
+    {
+        std::cout << desc.CalculateOffset(ck::make_tuple(w)) << " ";
+    }
+    std::cout << std::endl;
+}
+
+template <typename Desc>
+void Print2d(const Desc& desc)
+{
+    std::cout << "Print2d" << std::endl;
+    for(ck::index_t h = 0; h < desc.GetLength(I0); h++)
+    {
+        for(ck::index_t w = 0; w < desc.GetLength(I1); w++)
+        {
+            std::cout << desc.CalculateOffset(ck::make_tuple(h, w)) << " ";
+        }
+        std::cout << std::endl;
+    }
+}
+
+template <typename Desc>
+void Print3dCustom(const Desc& desc)
+{
+    std::cout << "Print3dCustom" << std::endl;
+    for(ck::index_t d = 0; d < desc.GetLength(I0); d++)
+    {
+        for(ck::index_t h = 0; h < desc.GetLength(I1); h++)
+        {
+            for(ck::index_t w = 0; w < desc.GetLength(I2); w++)
+            {
+                std::cout << desc.CalculateOffset(ck::make_tuple(d, h, w)) << " ";
+            }
+            std::cout << std::endl;
+        }
+        std::cout << std::endl;
+    }
+}
+
+int main()
+{
+    // Tensor descriptor traverse in row-major (need to reverse dims)
+    std::cout << "Note: Tensor descriptor traverse in row-major" << std::endl;
+    // Basic descriptor 0, 1, 2, ... 30, 31
+    // (dims:4,8 strides:1,4)
+    const auto desc_4x8_s1x4 =
+        ck::make_naive_tensor_descriptor(ck::make_tuple(ck::Number<4>{}, ck::Number<8>{}),
+                                         ck::make_tuple(ck::Number<1>{}, ck::Number<4>{}));
+    std::cout << "dims:4,8 strides:1,4" << std::endl;
+    Print2d(desc_4x8_s1x4);
+
+    using Cord1x1Type                = ck::Tuple<ck::Number<1>, ck::Number<1>>;
+    constexpr ck::index_t offset_1x1 = desc_4x8_s1x4.CalculateOffset(Cord1x1Type{});
+    std::cout << "Constexpr calculated [1, 1] offset:" << offset_1x1 << std::endl;
+
+    // Basic descriptor 0, 1, 8, 9, 16, 17, ... 30, 31 (compile-time descriptor)
+    // dims:4,(2,4) strides:2,(1,8)
+    const auto desc_4x2x4_s2x1x8 =
+        ck::make_naive_tensor_descriptor(ck::make_tuple(4, 2, 4), ck::make_tuple(2, 1, 8));
+    // Transform to 2d (column-major, need to to reverse dims)
+    const auto desc_4x2x4_s2x1x8_merged = ck::transform_tensor_descriptor(
+        desc_4x2x4_s2x1x8,
+        ck::make_tuple(ck::make_pass_through_transform(4),
+                       ck::make_merge_transform(ck::make_tuple(4, 2))),
+        ck::make_tuple(ck::Sequence<0>{}, ck::Sequence<2, 1>{}),
+        ck::make_tuple(ck::Sequence<0>{}, ck::Sequence<1>{}));
+
+    std::cout << "dims:4,(2,4) strides:2,(1,8)" << std::endl;
+    Print2d(desc_4x2x4_s2x1x8_merged);
+
+    // Basic descriptor 0, 1, 8, 9, 16, 17, ... 30, 31 (compile-time descriptor)
+    // dims:(2,2),(2,4) strides:((1,4),(2,8)
+    const auto desc_2x2x2x4_s1x4x2x8 =
+        ck::make_naive_tensor_descriptor(ck::make_tuple(2, 2, 2, 4), ck::make_tuple(1, 4, 2, 8));
+    // Transform to 2d
+    const auto desc_2x2x2x4_s1x4x2x8_double_merged_2d = ck::transform_tensor_descriptor(
+        desc_2x2x2x4_s1x4x2x8,
+        ck::make_tuple(ck::make_merge_transform(ck::make_tuple(2, 2)),
+                       ck::make_merge_transform(ck::make_tuple(4, 2))),
+        ck::make_tuple(ck::Sequence<1, 0>{}, ck::Sequence<3, 2>{}),
+        ck::make_tuple(ck::Sequence<0>{}, ck::Sequence<1>{}));
+    // Transform to 3d
+    const auto desc_2x2x2x4_s1x4x2x8_double_merged_3d = ck::transform_tensor_descriptor(
+        desc_2x2x2x4_s1x4x2x8,
+        ck::make_tuple(ck::make_pass_through_transform(2),
+                       ck::make_pass_through_transform(2),
+                       ck::make_merge_transform(ck::make_tuple(4, 2))),
+        ck::make_tuple(ck::Sequence<0>{}, ck::Sequence<1>{}, ck::Sequence<3, 2>{}),
+        ck::make_tuple(ck::Sequence<0>{}, ck::Sequence<1>{}, ck::Sequence<2>{}));
+
+    std::cout << "dims:(2,2),(2,4) strides:(1,4),(2,8)" << std::endl;
+    Print2d(desc_2x2x2x4_s1x4x2x8_double_merged_2d);
+    Print3dCustom(desc_2x2x2x4_s1x4x2x8_double_merged_3d);
+
+    // Basic descriptor 0, 1, 8, 9, 16, 17, ... 30, 31 (compile-time descriptor)
+    // dims:((2,2),2),4 strides:((1,4),2),8
+    // Transform to 2d
+    const auto desc_2x2x2x4_s1x4x2x8_nested =
+        ck::make_naive_tensor_descriptor(ck::make_tuple(2, 2, 2, 4), ck::make_tuple(1, 4, 2, 8));
+    const auto desc_2x2x2x4_s1x4x2x8_nested_merged_3d = ck::transform_tensor_descriptor(
+        desc_2x2x2x4_s1x4x2x8_nested,
+        ck::make_tuple(ck::make_merge_transform(ck::make_tuple(2, 2)),
+                       ck::make_pass_through_transform(2),
+                       ck::make_pass_through_transform(4)),
+        ck::make_tuple(ck::Sequence<1, 0>{}, ck::Sequence<2>{}, ck::Sequence<3>{}),
+        ck::make_tuple(ck::Sequence<0>{}, ck::Sequence<1>{}, ck::Sequence<2>{}));
+    const auto desc_2x2x2x4_s1x4x2x8_nested_merged_1d = ck::transform_tensor_descriptor(
+        desc_2x2x2x4_s1x4x2x8_nested,
+        ck::make_tuple(ck::make_merge_transform(ck::make_tuple(4, 2, 2, 2))),
+        ck::make_tuple(ck::Sequence<3, 2, 1, 0>{}),
+        ck::make_tuple(ck::Sequence<0>{}));
+    const auto desc_2x2x2x4_s1x4x2x8_nested_merged_2d = ck::transform_tensor_descriptor(
+        desc_2x2x2x4_s1x4x2x8_nested_merged_3d,
+        ck::make_tuple(ck::make_merge_transform(ck::make_tuple(2, 4)),
+                       ck::make_pass_through_transform(4)),
+        ck::make_tuple(ck::Sequence<1, 0>{}, ck::Sequence<2>{}),
+        ck::make_tuple(ck::Sequence<0>{}, ck::Sequence<1>{}));
+
+    std::cout << "dims:((2,2),2),4 strides:((1,4),2),8" << std::endl;
+    Print1d(desc_2x2x2x4_s1x4x2x8_nested_merged_1d);
+    Print2d(desc_2x2x2x4_s1x4x2x8_nested_merged_2d);
+    Print3dCustom(desc_2x2x2x4_s1x4x2x8_nested_merged_3d);
+
+    return 0;
+}

client_example/25_tensor_transforms/tensor_transform_using_wrapper.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2023, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <iostream>
+
+#include "ck/ck.hpp"
+
+#include "ck/utility/number.hpp"
+#include "ck/utility/tuple.hpp"
+#include "ck/utility/sequence.hpp"
+
+#include "ck/wrapper/layout.hpp"
+
+using DataType = int;
+
+template <typename Layout>
+void Print1d(const Layout& layout)
+{
+    std::cout << "Print1d" << std::endl;
+    for(ck::index_t w = 0; w < ck::wrapper::size(layout); w++)
+    {
+        std::cout << layout(ck::make_tuple(w)) << " ";
+    }
+    std::cout << std::endl;
+}
+
+template <typename Layout>
+void Print2d(const Layout& layout)
+{
+    std::cout << "Print2d" << std::endl;
+    for(ck::index_t h = 0; h < ck::wrapper::size<0>(layout); h++)
+    {
+        for(ck::index_t w = 0; w < ck::wrapper::size<1>(layout); w++)
+        {
+            std::cout << layout(ck::make_tuple(h, w)) << " ";
+        }
+        std::cout << std::endl;
+    }
+}
+
+// Print in (x,y),z pattern
+template <typename Layout>
+void Print3dCustom(const Layout& layout)
+{
+    std::cout << "Print3dCustom" << std::endl;
+    for(ck::index_t d = 0; d < ck::wrapper::size<0>(ck::wrapper::get<0>(layout)); d++)
+    {
+        for(ck::index_t h = 0; h < ck::wrapper::size<1>(ck::wrapper::get<0>(layout)); h++)
+        {
+            for(ck::index_t w = 0; w < ck::wrapper::size<1>(layout); w++)
+            {
+                std::cout << layout(ck::make_tuple(ck::make_tuple(d, h), w)) << " ";
+            }
+            std::cout << std::endl;
+        }
+        std::cout << std::endl;
+    }
+}
+
+int main()
+{
+    // Layout traverse in row-major
+    std::cout << "Note: Layout traverse in column-major" << std::endl;
+    // Basic descriptor 0, 1, 2, ... 30, 31 (compile-time descriptor)
+    // (dims:4,8 strides:1,4)
+    const auto shape_4x8       = ck::make_tuple(ck::Number<4>{}, ck::Number<8>{});
+    const auto layout_4x8_s1x4 = ck::wrapper::make_layout(shape_4x8);
+    std::cout << "dims:4,8 strides:1,4" << std::endl;
+    Print2d(layout_4x8_s1x4);
+    using Cord1x1Type                = ck::Tuple<ck::Number<1>, ck::Number<1>>;
+    constexpr ck::index_t offset_1x1 = layout_4x8_s1x4.template operator()<Cord1x1Type>();
+    std::cout << "Constexpr calculated [1, 1] offset:" << offset_1x1 << std::endl;
+
+    // Basic descriptor 0, 1, 8, 9, 16, 17, ... 30, 31 (runtime descriptor)
+    // dims:4,(2,4) strides:2,(1,8)
+    const auto shape_4x2x4         = ck::make_tuple(4, ck::make_tuple(2, 4));
+    const auto strides_s2x1x8      = ck::make_tuple(2, ck::make_tuple(1, 8));
+    const auto layout_4x2x4_s2x1x8 = ck::wrapper::make_layout(shape_4x2x4, strides_s2x1x8);
+
+    std::cout << "dims:4,(2,4) strides:2,(1,8)" << std::endl;
+    Print2d(layout_4x2x4_s2x1x8);
+
+    // Basic descriptor 0, 1, 8, 9, 16, 17, ... 30, 31 (compile-time descriptor)
+    // dims:(2,2),(2,4) strides:((1,4),(2,8)
+    const auto shape_2x2x2x4    = ck::make_tuple(ck::make_tuple(ck::Number<2>{}, ck::Number<2>{}),
+                                              ck::make_tuple(ck::Number<2>{}, ck::Number<4>{}));
+    const auto strides_s1x4x2x8 = ck::make_tuple(ck::make_tuple(ck::Number<1>{}, ck::Number<4>{}),
+                                                 ck::make_tuple(ck::Number<2>{}, ck::Number<8>{}));
+    static const auto layout_2x2x2x4_s1x4x2x8 =
+        ck::wrapper::make_layout(shape_2x2x2x4, strides_s1x4x2x8);
+
+    std::cout << "dims:(2,2),(2,4) strides:(1,4),(2,8)" << std::endl;
+    Print2d(layout_2x2x2x4_s1x4x2x8);
+    Print3dCustom(layout_2x2x2x4_s1x4x2x8);
+
+    // Basic descriptor 0, 1, 8, 9, 16, 17, ... 30, 31 (compile-time descriptor)
+    // dims:((2,2),2),4 strides:((1,4),2),8
+    // Transform to 2d
+    const auto shape_2x2x2x4_nested = ck::make_tuple(
+        ck::make_tuple(ck::make_tuple(ck::Number<2>{}, ck::Number<2>{}), ck::Number<2>{}),
+        ck::Number<4>{});
+    const auto strides_s1x4x2x8_nested = ck::make_tuple(
+        ck::make_tuple(ck::make_tuple(ck::Number<1>{}, ck::Number<4>{}), ck::Number<2>{}),
+        ck::Number<8>{});
+    static const auto layout_2x2x2x4_s1x4x2x8_nested =
+        ck::wrapper::make_layout(shape_2x2x2x4_nested, strides_s1x4x2x8_nested);
+
+    std::cout << "dims:((2,2),2),4 strides:((1,4),2),8" << std::endl;
+    Print1d(layout_2x2x2x4_s1x4x2x8_nested);
+    Print2d(layout_2x2x2x4_s1x4x2x8_nested);
+    Print3dCustom(layout_2x2x2x4_s1x4x2x8_nested);
+
+    return 0;
+}

client_example/CMakeLists.txt

@@ -48,7 +48,7 @@ else()
     endif()
 endif()
 
-find_package(composable_kernel COMPONENTS device_operations)
+find_package(composable_kernel COMPONENTS device_other_operations device_gemm_operations device_conv_operations device_contraction_operations device_reduction_operations)
 find_package(hip REQUIRED PATHS /opt/rocm)
 message(STATUS "Build with HIP ${hip_VERSION}")
 

dev-requirements.txt

-ROCmSoftwarePlatform/rocm-recipes
+ROCm/rocm-recipes
 RadeonOpenCompute/rocm-cmake@04f694df2a8dc9d7e35fa4dee4ba5fa407ec04f8 --build
-danmar/cppcheck@2.9
\ No newline at end of file
+danmar/cppcheck@2.9

docs/conf.py

@@ -4,23 +4,34 @@
 # list see the documentation:
 # https://www.sphinx-doc.org/en/master/usage/configuration.html
 
-import subprocess
+import re
 
 from rocm_docs import ROCmDocs
 
+html_theme_options = {"flavor": "list"}
 
-name = "Composable Kernel"
-get_version = r'sed -n -e "s/^rocm_setup_version(.* \([0-9\.]\{1,\}\).*/\1/p" ../CMakeLists.txt'
-version = subprocess.getoutput(get_version)
-if len(version) > 0:
-    name = f"{name} {version}"
+with open('../CMakeLists.txt', encoding='utf-8') as f:
+    match = re.search(r'.*set\(version ([0-9.]+)[^0-9.]+', f.read())
+    if not match:
+        raise ValueError("VERSION not found!")
+    version_number = match[1]
+left_nav_title = f"Composable Kernel {version_number} Documentation"
+
+# for PDF output on Read the Docs
+project = "Composable Kernel Documentation"
+author = "Advanced Micro Devices, Inc."
+copyright = "Copyright (c) 2023 Advanced Micro Devices, Inc. All rights reserved."
+version = version_number
+release = version_number
 
 external_toc_path = "./sphinx/_toc.yml"
 
-docs_core = ROCmDocs(f"{name} Documentation")
-docs_core.run_doxygen(doxygen_root="doxygen", doxygen_path="doxygen/docBin/xml")
+docs_core = ROCmDocs(left_nav_title)
+docs_core.run_doxygen(doxygen_root="doxygen", doxygen_path="doxygen/xml")
 docs_core.setup()
 
+external_projects_current_project = "composable_kernel"
+
 mathjax3_config = {
 'tex': {
     'macros': {

docs/doxygen/Doxyfile

@@ -58,7 +58,7 @@ PROJECT_LOGO           =
 # entered, it will be relative to the location where doxygen was started. If
 # left blank the current directory will be used.
 
-OUTPUT_DIRECTORY       = docBin
+OUTPUT_DIRECTORY       = .
 
 # If the CREATE_SUBDIRS tag is set to YES then doxygen will create 4096 sub-
 # directories (in 2 levels) under the output directory of each output format and
@@ -778,7 +778,9 @@ WARN_LOGFILE           =
 INPUT                  = ../../include/ck/tensor_operation/gpu/grid \
                          ../../include/ck/tensor_operation/gpu/block \
                          ../../include/ck/tensor_operation/gpu/thread \
-                         ../../library/include/ck/library/utility
+                         ../../library/include/ck/library/utility \
+                         ../../include/ck/wrapper
+
 
 # This tag can be used to specify the character encoding of the source files
 # that doxygen parses. Internally doxygen uses the UTF-8 encoding. Doxygen uses

docs/index.rst

@@ -34,6 +34,7 @@ Current CK library are structured into 4 layers:
 * "Templated Tile Operators" layer
 * "Templated Kernel and Invoker" layer
 * "Instantiated Kernel and Invoker" layer
+* "Wrapper for tensor transform operations"
 * "Client API" layer
 
 .. image:: data/ck_layer.png
@@ -50,6 +51,7 @@ The following is a list of CK documents in the suggested reading order:
 
    tutorial_hello_world
    dockerhub
+   wrapper
    Supported_Primitives_Guide
    API_Reference_Guide
    Contributors_Guide

docs/sphinx/_toc.yml.in

@@ -5,6 +5,6 @@ defaults:
   maxdepth: 6
 root: index
 subtrees:
-  - caption: About
-    entries:
-      - file: license
+- caption: About
+  entries:
+    - file: license

docs/sphinx/requirements.in

-rocm-docs-core>=0.20.0
+rocm-docs-core==0.30.1
 sphinxcontrib-bibtex==2.6.1

docs/sphinx/requirements.txt

@@ -16,7 +16,7 @@ beautifulsoup4==4.11.2
     # via pydata-sphinx-theme
 breathe==4.34.0
     # via rocm-docs-core
-certifi==2022.12.7
+certifi==2023.7.22
     # via requests
 cffi==1.15.1
     # via
@@ -26,7 +26,7 @@ charset-normalizer==3.1.0
     # via requests
 click==8.1.3
     # via sphinx-external-toc
-cryptography==40.0.2
+cryptography==41.0.6
     # via pyjwt
 deprecated==1.2.13
     # via pygithub
@@ -42,7 +42,7 @@ fastjsonschema==2.18.0
     # via rocm-docs-core
 gitdb==4.0.10
     # via gitpython
-gitpython==3.1.35
+gitpython==3.1.37
     # via rocm-docs-core
 idna==3.4
     # via requests
@@ -88,9 +88,9 @@ pydata-sphinx-theme==0.13.3
     # via
     #   rocm-docs-core
     #   sphinx-book-theme
-pygithub==1.58.2
+pygithub==1.58.1
     # via rocm-docs-core
-pygments==2.14.0
+pygments==2.15.0
     # via
     #   accessible-pygments
     #   pydata-sphinx-theme
@@ -109,11 +109,11 @@ pyyaml==6.0
     #   pybtex
     #   rocm-docs-core
     #   sphinx-external-toc
-requests==2.28.2
+requests==2.31.0
     # via
     #   pygithub
     #   sphinx
-rocm-docs-core==0.27.0
+rocm-docs-core==0.30.1
     # via -r requirements.in
 six==1.16.0
     # via
@@ -141,7 +141,7 @@ sphinx-book-theme==1.0.1
     # via rocm-docs-core
 sphinx-copybutton==0.5.1
     # via rocm-docs-core
-sphinx-design==0.3.0
+sphinx-design==0.4.1
     # via rocm-docs-core
 sphinx-external-toc==0.3.1
     # via rocm-docs-core
@@ -163,7 +163,7 @@ sphinxcontrib-serializinghtml==1.1.5
     # via sphinx
 typing-extensions==4.5.0
     # via pydata-sphinx-theme
-urllib3==1.26.15
+urllib3==1.26.18
     # via requests
 wrapt==1.15.0
     # via deprecated

docs/wrapper.rst

+===============
+Wrapper
+===============
+
+-------------------------------------
+Description
+-------------------------------------
+
+.. note::
+
+    The wrapper is under development and its functionality is limited.
+
+
+CK provides a lightweight wrapper for more complex operations implemented in 
+the library. It allows indexing of nested layouts using a simple interface 
+(avoiding complex descriptor transformations). 
+
+Example:
+
+.. code-block:: c
+
+    const auto shape_4x2x4         = ck::make_tuple(4, ck::make_tuple(2, 4));
+    const auto strides_s2x1x8      = ck::make_tuple(2, ck::make_tuple(1, 8));
+    const auto layout = ck::wrapper::make_layout(shape_4x2x4, strides_s2x1x8);
+
+    std::cout << "dims:4,(2,4) strides:2,(1,8)" << std::endl;
+    for(ck::index_t h = 0; h < ck::wrapper::size<0>(layout); h++)
+    {
+        for(ck::index_t w = 0; w < ck::wrapper::size<1>(layout); w++)
+        {
+            std::cout << layout(ck::make_tuple(h, w)) << " ";
+        }
+        std::cout << std::endl;
+    }
+
+Output::
+
+    dims:4,(2,4) strides:2,(1,8)
+    0 1 8 9 16 17 24 25 
+    2 3 10 11 18 19 26 27 
+    4 5 12 13 20 21 28 29 
+    6 7 14 15 22 23 30 31 
+
+-------------------------------------
+Layout
+-------------------------------------
+
+.. doxygenstruct:: ck::wrapper::Layout
+
+-------------------------------------
+Layout helpers
+-------------------------------------
+
+.. doxygenfile:: layout_utils.hpp

example/01_gemm/gemm_xdl_fp16.cpp

@@ -30,7 +30,7 @@ using DeviceGemmInstance0 = ck::tensor_operation::device::DeviceGemmXdl
 // ######|      Type|      Type|      Type|        Type|        |        |        | Elementwise| Elementwise| Elementwise|Spacialization|  Size| Block| Block| Block|   |  XDL|  XDL|  Per|  Per|   ThreadCluster|  ThreadCluster| SrcAccessOrder|   SrcVectorDim|      SrcScalar|      DstScalar| AddExtraM|   ThreadCluster|  ThreadCluster| SrcAccessOrder|  SrcVectorDim|      SrcScalar|      DstScalar| AddExtraN| SrcDstVectorDim|       DstScalar|
 // ######|          |          |          |            |        |        |        |   Operation|   Operation|   Operation|              |      |      |      |      |   |     |     | Wave| Wave| Lengths_K0_M_K1|   ArrangeOrder|               |               |      PerVector|   PerVector_K1|          | Lengths_K0_N_K1|   ArrangeOrder|               |              |      PerVector|   PerVector_K1|          |                |       PerVector|
 // ######|          |          |          |            |        |        |        |            |            |            |              |      |      |      |      |   |     |     |     |     |                |               |               |               |               |               |          |                |               |               |              |               |               |          |                |                |
-         < ADataType, BDataType, CDataType, AccDataType, ALayout, BLayout, CLayout,  AElementOp,  BElementOp,  CElementOp,   GemmDefault,   256,   256,   128,     4,  8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,             2,              8,              8,      true,               7,               1>;
+         < ADataType, BDataType, CDataType, AccDataType, ALayout, BLayout, CLayout,  AElementOp,  BElementOp,  CElementOp,   GemmDefault,   256,   256,   128,     4,  8,   32,   32,    4,    2,     S<4, 64, 1>,     S<1, 0, 2>,     S<1, 0, 2>,              2,              8,              8,      true,     S<4, 32, 1>,     S<0, 2, 1>,     S<0, 2, 1>,             1,              4,              8,      true,               7,               1>;
 // // clang-format on
 
 // clang-format off

example/15_grouped_gemm/grouped_gemm_xdl_fixed_nk_fp16.cpp

@@ -299,8 +299,8 @@ int main(int argc, char* argv[])
     for(int i = 0; i < problem_size.group_count; i++)
     {
         problem_size.Ms.push_back(256 + 256 * i);
-        problem_size.Ns.push_back(128 + 128 * i);
-        problem_size.Ks.push_back(128 + 64 * i);
+        problem_size.Ns.push_back(256);
+        problem_size.Ks.push_back(128);
 
         problem_size.stride_As.push_back(problem_size.Ks[i]);
         problem_size.stride_Bs.push_back(problem_size.Ks[i]);

example/15_grouped_gemm/grouped_gemm_xdl_fixed_nk_fp8.cpp

@@ -300,8 +300,8 @@ int main(int argc, char* argv[])
     for(int i = 0; i < problem_size.group_count; i++)
     {
         problem_size.Ms.push_back(256 + 256 * i);
-        problem_size.Ns.push_back(128 + 128 * i);
-        problem_size.Ks.push_back(128 + 64 * i);
+        problem_size.Ns.push_back(256);
+        problem_size.Ks.push_back(128);
 
         problem_size.stride_As.push_back(problem_size.Ks[i]);
         problem_size.stride_Bs.push_back(problem_size.Ks[i]);

example/27_layernorm2d_fwd/run_layernorm_example.inc

@@ -44,9 +44,9 @@ int run_layernorm2d_fwd_example()
         {0, 1},
         std::vector<ck::index_t>{y.mDesc.GetStrides().begin(), y.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         {1},
         1e-4,
         x_dev.GetDeviceBuffer(),

example/42_groupnorm_fwd/run_groupnorm_fwd_example.inc

@@ -65,9 +65,9 @@ int run_groupnorm_fwd_example(int argc, char* argv[])
         {0, 0, 0, C, 1},
         std::vector<ck::index_t>{y.mDesc.GetStrides().begin(), y.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         std::vector<ck::index_t>{save_mean.mDesc.GetStrides().begin(),
-                                    save_mean.mDesc.GetStrides().end()},
+                                 save_mean.mDesc.GetStrides().end()},
         {1, 2, 4}, // reduction dimension: [H, W, C]
         1e-6,
         x_dev.GetDeviceBuffer(),

example/44_elementwise_permute/CMakeLists.txt

@@ -5,4 +5,6 @@ add_example_executable(example_elementwise_permute_4D_fp16_row elementwise_permu
 add_example_executable(example_elementwise_permute_4D_fp32_col elementwise_permute_4D_fp32_col.cpp)
 add_example_executable(example_elementwise_permute_4D_fp16_col elementwise_permute_4D_fp16_col.cpp)
 add_example_executable(example_elementwise_permute elementwise_permute.cpp)
-add_example_executable(example_elementwise_permute_3d elementwise_permute_3d.cpp)
+if((NOT GPU_TARGETS MATCHES "gfx940") AND (NOT GPU_TARGETS MATCHES "gfx941") AND (NOT GPU_TARGETS MATCHES "gfx942"))
+    add_example_executable(example_elementwise_permute_3d elementwise_permute_3d.cpp)
+endif()

example/62_conv_fwd_activ/CMakeLists.txt

@@ -42,6 +42,8 @@ foreach(gpu IN LISTS GPU_TARGETS)
       # ScaleAdd ScaleAdd Relu
       add_example_executable(example_convnd_fwd_xdl_scaleadd_scaleadd_relu_fp16 convnd_fwd_xdl_scaleadd_scaleadd_relu_fp16.cpp)
       add_example_dependencies(example_convnd_fwd_activ_xdl example_convnd_fwd_xdl_scaleadd_scaleadd_relu_fp16)
+      add_example_executable(example_convnd_fwd_xdl_scaleadd_scaleadd_relu_bcasted_bias_fp16 convnd_fwd_xdl_scaleadd_scaleadd_relu_bcasted_bias_fp16.cpp)
+      add_example_dependencies(example_convnd_fwd_activ_xdl example_convnd_fwd_xdl_scaleadd_scaleadd_relu_bcasted_bias_fp16)
    set(target 1)
  endif()
 endforeach()

example/62_conv_fwd_activ/convnd_fwd_xdl_scaleadd_scaleadd_relu_bcasted_bias_fp16.cpp

+// SPDX-License-Identifier: MIT
+// Copyright (c) 2023, Advanced Micro Devices, Inc. All rights reserved.
+
+#include <algorithm>
+#include <cstdlib>
+#include <iostream>
+#include <numeric>
+#include <type_traits>
+
+#include "ck/ck.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
+
+#include "ck/library/utility/algorithm.hpp"
+#include "ck/library/utility/check_err.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
+#include "ck/library/utility/convolution_parameter.hpp"
+#include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
+#include "ck/library/reference_tensor_operation/cpu/reference_conv_fwd.hpp"
+#include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
+
+constexpr ck::index_t NDimSpatial = 3;
+using InDataType                  = ck::half_t;
+using WeiDataType                 = ck::half_t;
+using AccDataType                 = float;
+using CShuffleDataType            = ck::half_t;
+using OutDataType                 = ck::half_t;
+
+template <ck::index_t... Is>
+using S = ck::Sequence<Is...>;
+
+using InLayout  = ck::tensor_layout::convolution::NDHWGC;
+using WeiLayout = ck::tensor_layout::convolution::GKZYXC;
+using OutLayout = ck::tensor_layout::convolution::NDHWGK;
+
+using BiasLayout = ck::tensor_layout::convolution::G_K;
+
+using InElementOp  = ck::tensor_operation::element_wise::PassThrough;
+using WeiElementOp = ck::tensor_operation::element_wise::PassThrough;
+
+using OutElementOp = ck::tensor_operation::element_wise::ScaleAddScaleAddRelu;
+
+static constexpr auto ConvSpec =
+    ck::tensor_operation::device::ConvolutionForwardSpecialization::Default;
+
+static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecialization::MNKPadding;
+
+template <typename OutElementOp>
+using DeviceGroupedConvNDFwdInstance =
+    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<
+        NDimSpatial,
+        InLayout,
+        WeiLayout,
+        ck::Tuple<OutLayout, BiasLayout>,
+        OutLayout,
+        InDataType,
+        WeiDataType,
+        AccDataType,
+        CShuffleDataType,
+        ck::Tuple<OutDataType, OutDataType>,
+        OutDataType,
+        InElementOp,
+        WeiElementOp,
+        OutElementOp,
+        ConvSpec,    // ConvForwardSpecialization
+        GemmSpec,    // GemmSpecialization
+        1,           //
+        256,         // BlockSize
+        128,         // MPerBlock
+        256,         // NPerBlock
+        32,          // KPerBlock
+        8,           // AK1
+        8,           // BK1
+        32,          // MPerXdl
+        32,          // NPerXdl
+        2,           // MXdlPerWave
+        4,           // NXdlPerWave
+        S<4, 64, 1>, // ABlockTransferThreadClusterLengths_AK0_M_AK1
+        S<1, 0, 2>,  // ABlockTransferThreadClusterArrangeOrder
+        S<1, 0, 2>,  // ABlockTransferSrcAccessOrder
+        2,           // ABlockTransferSrcVectorDim
+        8,           // ABlockTransferSrcScalarPerVector
+        8,           // ABlockTransferDstScalarPerVector_AK1
+        1,           // ABlockLdsExtraM
+        S<4, 64, 1>, // BBlockTransferThreadClusterLengths_BK0_N_BK1
+        S<1, 0, 2>,  // BBlockTransferThreadClusterArrangeOrder
+        S<1, 0, 2>,  // BBlockTransferSrcAccessOrder
+        2,           // BBlockTransferSrcVectorDim
+        8,           // BBlockTransferSrcScalarPerVector
+        8,           // BBlockTransferDstScalarPerVector_BK1
+        1,           // BBlockLdsExtraN
+        1,
+        1,
+        S<1, 32, 1, 8>,
+        8>;
+
+using DeviceGroupedConvNDFwdActivInstance = DeviceGroupedConvNDFwdInstance<OutElementOp>;
+
+namespace {
+// Use custom implementation to pass two more tensors for post op
+template <ck::index_t NDimSpatial,
+          typename InDataType,
+          typename WeiDataType,
+          typename OutDataType,
+          typename InElementOp,
+          typename WeiElementOp,
+          typename OutElementOp,
+          typename DeviceConvNDFwdInstance>
+bool run_grouped_conv_fwd(bool do_verification,
+                          int init_method,
+                          bool time_kernel,
+                          const ck::utils::conv::ConvParam& conv_param,
+                          const HostTensorDescriptor& in_g_n_c_wis_desc,
+                          const HostTensorDescriptor& wei_g_k_c_xs_desc,
+                          const HostTensorDescriptor& out_g_n_k_wos_desc,
+                          const InElementOp& in_element_op,
+                          const WeiElementOp& wei_element_op,
+                          const OutElementOp& out_element_op)
+{
+    constexpr ck::index_t NumDs = 2;
+    const ck::index_t G         = out_g_n_k_wos_desc.GetLengths()[0];
+    const ck::index_t K         = out_g_n_k_wos_desc.GetLengths()[2];
+
+    // Logical broadcast bias (we have to pass bias lengths in the same format as output - GNKDHW)
+    std::array<ck::index_t, NDimSpatial + 3> bias_g_k_lengths;
+    std::array<ck::index_t, NDimSpatial + 3> bias_g_k_strides;
+    // Fill other lenghts than G,K with 1 and strides with 0
+    bias_g_k_lengths.fill(1);
+    bias_g_k_strides.fill(0);
+    bias_g_k_lengths[0]              = G;
+    bias_g_k_lengths[2]              = K;
+    bias_g_k_strides[0]              = K; // stride to G
+    bias_g_k_strides[2]              = 1; // stride to K
+    const auto broadcasted_bias_desc = HostTensorDescriptor(bias_g_k_lengths, bias_g_k_strides);
+
+    //  y = relu ( alpha1 * conv(x) + alpha2 * z + bias )
+    Tensor<InDataType> in(in_g_n_c_wis_desc);
+    Tensor<WeiDataType> wei(wei_g_k_c_xs_desc);
+    Tensor<OutDataType> out_host(out_g_n_k_wos_desc);
+    Tensor<OutDataType> out_device(out_g_n_k_wos_desc);
+    std::array<Tensor<OutDataType>, NumDs> d_tensors = {Tensor<OutDataType>(out_g_n_k_wos_desc),
+                                                        Tensor<OutDataType>(broadcasted_bias_desc)};
+
+    std::cout << "in: " << in.mDesc << std::endl;
+    std::cout << "wei: " << wei.mDesc << std::endl;
+    std::cout << "out: " << out_host.mDesc << std::endl;
+    std::cout << "z_tensor: " << d_tensors[0].mDesc << std::endl;
+    std::cout << "bias_tensor: " << d_tensors[1].mDesc << std::endl;
+
+    // Make sure that we allocated only G * K values for bias
+    assert(static_cast<ck::index_t>(d_tensors[1].mData.size()) == G * K);
+
+    switch(init_method)
+    {
+    case 0: break;
+    case 1:
+        in.GenerateTensorValue(GeneratorTensor_2<InDataType>{-2, 2});
+        wei.GenerateTensorValue(GeneratorTensor_2<WeiDataType>{-2, 2});
+        d_tensors[0].GenerateTensorValue(GeneratorTensor_2<OutDataType>{-2, 2});
+        d_tensors[1].GenerateTensorValue(GeneratorTensor_2<OutDataType>{-2, 2});
+        break;
+    default:
+        in.GenerateTensorValue(GeneratorTensor_3<InDataType>{-1.0, 1.0});
+        wei.GenerateTensorValue(GeneratorTensor_3<WeiDataType>{-0.05, 0.05});
+        d_tensors[0].GenerateTensorValue(GeneratorTensor_3<OutDataType>{-0.05, 0.05});
+        d_tensors[1].GenerateTensorValue(GeneratorTensor_3<OutDataType>{-0.05, 0.05});
+    }
+
+    DeviceMem in_device_buf(sizeof(InDataType) * in.mDesc.GetElementSpaceSize());
+    DeviceMem wei_device_buf(sizeof(WeiDataType) * wei.mDesc.GetElementSpaceSize());
+    DeviceMem z_buf(sizeof(OutDataType) * d_tensors[0].mDesc.GetElementSpaceSize());
+    DeviceMem bias_buf(sizeof(OutDataType) * d_tensors[1].mDesc.GetElementSpaceSize());
+    DeviceMem out_device_buf(sizeof(OutDataType) * out_device.mDesc.GetElementSpaceSize());
+
+    in_device_buf.ToDevice(in.mData.data());
+    wei_device_buf.ToDevice(wei.mData.data());
+    z_buf.ToDevice(d_tensors[0].mData.data());
+    bias_buf.ToDevice(d_tensors[1].mData.data());
+
+    std::array<ck::index_t, NDimSpatial + 3> a_g_n_c_wis_lengths{};
+    std::array<ck::index_t, NDimSpatial + 3> a_g_n_c_wis_strides{};
+    std::array<ck::index_t, NDimSpatial + 3> b_g_k_c_xs_lengths{};
+    std::array<ck::index_t, NDimSpatial + 3> b_g_k_c_xs_strides{};
+    std::array<ck::index_t, NDimSpatial + 3> e_g_n_k_wos_lengths{};
+    std::array<ck::index_t, NDimSpatial + 3> e_g_n_k_wos_strides{};
+    std::array<ck::index_t, NDimSpatial> conv_filter_strides{};
+    std::array<ck::index_t, NDimSpatial> conv_filter_dilations{};
+    std::array<ck::index_t, NDimSpatial> input_left_pads{};
+    std::array<ck::index_t, NDimSpatial> input_right_pads{};
+
+    auto copy = [](const auto& x, auto& y) { ck::ranges::copy(x, y.begin()); };
+
+    copy(in_g_n_c_wis_desc.GetLengths(), a_g_n_c_wis_lengths);
+    copy(in_g_n_c_wis_desc.GetStrides(), a_g_n_c_wis_strides);
+    copy(wei_g_k_c_xs_desc.GetLengths(), b_g_k_c_xs_lengths);
+    copy(wei_g_k_c_xs_desc.GetStrides(), b_g_k_c_xs_strides);
+    copy(out_g_n_k_wos_desc.GetLengths(), e_g_n_k_wos_lengths);
+    copy(out_g_n_k_wos_desc.GetStrides(), e_g_n_k_wos_strides);
+    copy(conv_param.conv_filter_strides_, conv_filter_strides);
+    copy(conv_param.conv_filter_dilations_, conv_filter_dilations);
+    copy(conv_param.input_left_pads_, input_left_pads);
+    copy(conv_param.input_right_pads_, input_right_pads);
+
+    const std::array<const void*, NumDs> ds = {z_buf.GetDeviceBuffer(), bias_buf.GetDeviceBuffer()};
+
+    auto conv     = DeviceConvNDFwdInstance{};
+    auto invoker  = conv.MakeInvoker();
+    auto argument = conv.MakeArgument(in_device_buf.GetDeviceBuffer(),
+                                      wei_device_buf.GetDeviceBuffer(),
+                                      ds,
+                                      out_device_buf.GetDeviceBuffer(),
+                                      a_g_n_c_wis_lengths,
+                                      a_g_n_c_wis_strides,
+                                      b_g_k_c_xs_lengths,
+                                      b_g_k_c_xs_strides,
+                                      std::array<std::array<ck::index_t, NDimSpatial + 3>, NumDs>{
+                                          e_g_n_k_wos_lengths, bias_g_k_lengths},
+                                      std::array<std::array<ck::index_t, NDimSpatial + 3>, NumDs>{
+                                          e_g_n_k_wos_strides, bias_g_k_strides},
+                                      e_g_n_k_wos_lengths,
+                                      e_g_n_k_wos_strides,
+                                      conv_filter_strides,
+                                      conv_filter_dilations,
+                                      input_left_pads,
+                                      input_right_pads,
+                                      in_element_op,
+                                      wei_element_op,
+                                      out_element_op);
+
+    if(!conv.IsSupportedArgument(argument))
+    {
+        throw std::runtime_error("The device op with the specified compilation parameters does "
+                                 "not support this convolution problem.");
+    }
+
+    float avg_time = invoker.Run(argument, StreamConfig{nullptr, time_kernel});
+
+    std::size_t flop = conv_param.GetFlops() + G * K +
+                       conv_param.GetOutputByte<OutDataType>() / sizeof(OutDataType);
+    std::size_t num_btype = conv_param.GetByte<InDataType, WeiDataType, OutDataType>() +
+                            G * K * sizeof(OutDataType) + conv_param.GetOutputByte<OutDataType>();
+
+    float tflops     = static_cast<float>(flop) / 1.E9 / avg_time;
+    float gb_per_sec = num_btype / 1.E6 / avg_time;
+    std::cout << "Perf: " << avg_time << " ms, " << tflops << " TFlops, " << gb_per_sec << " GB/s, "
+              << conv.GetTypeString() << std::endl;
+
+    if(do_verification)
+    {
+        auto ref_conv =
+            ck::tensor_operation::host::ReferenceConvFwd<NDimSpatial,
+                                                         InDataType,
+                                                         WeiDataType,
+                                                         OutDataType,
+                                                         InElementOp,
+                                                         WeiElementOp,
+                                                         OutElementOp,
+                                                         0, /*Num A Elementwise Tensors*/
+                                                         0, /*Num B Elementwise Tensors*/
+                                                         NumDs>();
+
+        auto ref_invoker  = ref_conv.MakeInvoker();
+        auto ref_argument = ref_conv.MakeArgument(in,
+                                                  wei,
+                                                  out_host,
+                                                  conv_param.conv_filter_strides_,
+                                                  conv_param.conv_filter_dilations_,
+                                                  conv_param.input_left_pads_,
+                                                  conv_param.input_right_pads_,
+                                                  in_element_op,
+                                                  wei_element_op,
+                                                  out_element_op,
+                                                  {},
+                                                  {},
+                                                  d_tensors);
+
+        ref_invoker.Run(ref_argument);
+
+        out_device_buf.FromDevice(out_device.mData.data());
+
+        return ck::utils::check_err(out_device, out_host, "Error: incorrect results!");
+    }
+
+    return true;
+}
+
+} // namespace
+
+#include "run_convnd_fwd_activ_example.inc"
+
+int main(int argc, char* argv[]) { return !run_convnd_fwd_example(argc, argv); }

example/62_conv_fwd_activ/run_convnd_fwd_activ_example.inc

@@ -24,7 +24,7 @@ bool run_convnd_fwd_example(int argc, char* argv[])
     // Following shapes are selected to avoid overflow. Expect inf in case of
     // size increase for some elementwise ops.
     ck::utils::conv::ConvParam conv_param{
-        3, 1, 16, 128, 8, {3, 3, 3}, {17, 17, 17}, {2, 2, 2}, {1, 1, 1}, {1, 1, 1}, {1, 1, 1}};
+        3, 2, 16, 128, 8, {3, 3, 3}, {17, 17, 17}, {2, 2, 2}, {1, 1, 1}, {1, 1, 1}, {1, 1, 1}};
 
     if(argc == 1)
     {