9908a87c31
* Add placeholder test. * Initial conv bwd weight factory. * Conv builder test refactoring. * Add missing pieces to bwd weight factory. * Improve compile time erros message when no matching factory is found. * Use amcro to ensure automatic macthing between concepts are their string representations. * Improve compile time diagnostics. * Small improvements. * Improve missing member/wrong type compile-time errors. * Improve compile time diagnostics. * Concept bug fixes. * Remove debug assert. * Update algorithm signature diagnostics. * Factory bug fixes. * First functional version of bwd weight conv factory. * Refactor handing of GEMM-K batch template parameter in conv bwd weight factory. * Concept improvements. * Improve concept diagnostics. * Introduve a common size type for concepts. * Update compiletime diagnostics to use the size type. * Update conv specialization enum. * Fix fwd conv builder tests. * Fix smoke tests. * Separate bwd weigth and bwd data tests into separate targets. * Clean-up CK Tile builder tests. * Add bwd weight XDL CShuffle V3 factory. * Build conv bwd weigth v3 instances successfully. * Add instance traits for DeviceGroupedConvBwdWeight_Xdl_CShuffleV3. * Test fix. * Add instance traits for bwd weight algorithms. * Add unit tests for instance strings. * Build new instance traits unit tests but exclude WMMA for now. * Added factory for DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle. * Conv bwd weight DL factory. * Final implementation for bwd weight DL factory. * Add test for creating DeviceGroupedConvBwdWeightMultipleD_Xdl_CShuffle instance. * Add factory for DeviceGroupedConvBwdWeightMultipleD_Xdl_CShuffle * Treat ref algorithm the same way as real algorithms in the dispatcher. * Refactor large tensor support and WMMA configuration. * Add factory and tests for DeviceGroupedConvBwdWeight_Wmma_CShuffleV3. * Update Readme. * Fix WMMA bwd weight tests. * Added factory and tests for DeviceGroupedConvBwdWeightTwoStage_Wmma_CShuffleV3. * Factory and tests for DeviceGroupedConvBwdWeight_Wmma_CShuffle. * Dispatching for DeviceGroupedConvBwdWeightMultipleD_Wmma_CShuffle. * Add factory for DeviceGroupedConvBwdWeightMultipleD_Wmma_CShuffleV3 * Fix DeviceGroupedConvBwdWeightMultipleD_Wmma_CShuffleV3 factory and compute types for input and output tensor in bwd weigth convs. * Fix fwd factories after refactoring. * clang-format * Move compile-time diagnostics to a separate branch. * Fix ref algorithm dispatching. * Fix smoke tests. * clang-format * Fix factory for regular WMMA conv bwd weight. * Clarify builder Readme. * Remove obsolete test file. * Fix test after merge. * clang-format * Remove the C++26 extensions. * Unify conv elementwise ops and layout definitions for fwd and bwd directions. * Remove old layout and elementwise ops. * Unify handling of conv tensor types between fwd and bwd directions. * Unify block transfer for fwd and bwd directions. Rename ThreadSliceDim to ThreadClusterRank. * Make BlockTransferDescriptor concept parametrized. Introduce a common TileTransferParameters concept for conv algorithms. * clang-format --------- Co-authored-by: Ville Pietilä <>
Convolution Builder Factory Directory
This directory implements compile-time dispatch from high-level signature algorithm descriptors to our exisitng specialized convolution kernel implementations.
See the main builder documentation for an overview.
Design Overview
The factory system operates in two phases:
-
Algorithm Classification: The function
make_conv_instanceinconv_dispatcher.hppinspects the signature and algorithm descriptors to determine which kernel variant they satisfy (XDL V3, XDL, WMMA, DL, or Large Tensor) -
Factory Instantiation: Each factory (
conv_fwd_*_factory.hpp) transforms builder descriptors into CK device operation template parameters and instantiates the corresponding kernel device operation.
Key Files
-
conv_dispatcher.hpp: Entry point withmake_conv_instance()function. Contains dispatch logic and algorithm classification predicates. Start here to understand the overall flow. -
conv_fwd_*_factory.hpp: Individual factories for each kernel variant. Each extracts configuration from descriptors, validates parameters, and instantiates the underlying CK device operation. -
helpers/: Transformation utilities that map builder types to CK device operation parameters (layouts, data types, elementwise ops, block configurations, etc.)
Usage
#include "ck_tile/builder/factory/conv_dispatcher.hpp"
using Factory = decltype(make_conv_instance<signature, algorithm, "v1">());
The dispatcher automatically selects the appropriate factory following explicit logic.