1. Add base class GridwiseGemm_xdl_cshuffle_base for all gridwise_gemm_xdl classes.
- to select correct LDS layout and epilogue behavior , three additional parameters is added.
- ForceNaiveLdsLayout: disable XOR based LDS layout when it is true
- DirectLoad: pipeline only use directload, we need force naive layout and ignore any padding on gfx9
- IsMxGemm: epilogue has two addtional dimensions
2. Move all LDS descriptor layout related fucntion to base class, including
- GetABlockDescriptor_AK0PerBlock_MPerBlock_AK1
- GetBBlockDescriptor_BK0PerBlock_NPerBlock_BK1
- GetCShuffleBlockDescriptor_MBlock_MPerBlock_NBlock_NPerBlock
3. Move several LDS related helper funtions to base class, including
- GetSharedMemoryNumberOfByte
- GetABlockDescriptor_AKB_AK0PerBlock_MPerBlock_AK1
- GetBBlockDescriptor_BKB_BK0PerBlock_NPerBlock_BK1
- GetCBlockDescriptor_MBlock_NXdlPerWave_MWaveMPerXdl_NBlock_NXdlPerWave_NWaveNPerXdl
4. Move all c epilogue related code to base class, and 4 kind of implementation are provided
- RunEpilogueNoShuffle
- RunEpilogue
- RunMultiDEpilogue
- RunMoeEpilogue
[ROCm/composable_kernel commit: 23cefda140]
* removed the api reference
* updating to the latest rocm-docs-core min version
* fixed a formatting issue with buffer views
* removed reference links from code snippets
* removed reference links from code snippets
---------
Co-authored-by: John Afaganis <john.afaganis@amd.com>
[ROCm/composable_kernel commit: 0cc83cb8e8]
This document describes techniques for reducing C++ template instantiation
overhead in the Composable Kernel codebase, including:
- Replacing recursive templates with pack expansion (O(N) → O(1) depth)
- Using named functors instead of lambdas to share instantiations
- Replacing template recursion with constexpr loops
- Using fold expressions for accumulation operations
These techniques can significantly reduce build times for template-heavy code.
[ROCm/composable_kernel commit: b66597ed96]
* ck-builder: restructure testing conv
In order to prepare for bwd of conv testing, this commit moves some
files and types around so that we can reuse ckt::Args for both forward
and backwards convolution.
* ck-builder: decouple fwd_ck.hpp and fwd_reference.hpp from fwd.hpp
This will allow us to more easily include fwd.hpp from backwards
definitions, which is required for initializing bwd values.
* ck-builder: fix layout of test_ckb_conv_bwd_weight_xdl_cshuffle_v3
Turns out that the supplied layout isn't actually supported...
* ck-builder: ck and reference conv integration for bwd weight
* ck-builder: ck bwd weight execution test
* ck-builder: ckt::run support for ck-tile bwd weight
* ck-builder: ck tile bwd weight execution test
* ck-builder: extra debug printing in MatchesReference
* ck-builder: make ckt::run return RunResult
This type is more convenient than std::tuple, as it will allow us to
use google test matchers with this in the future.
* ck-builder: RunResult matcher
Using EXPECT_THAT(..., SuccessfulRun()) will generate a check and a nice error
message about how and why running an algorithm failed.
* ck-builder: doc fixes
* ck-builder: add missing headers
[ROCm/composable_kernel commit: cc75948d1c]
This PR introduces a Python toolkit for analyzing Clang's `-ftime-trace` build performance data. This is the foundation for our systematic effort to reduce CK and CK-Tile build times (#3575).
The toolkit provides fast parsing of trace JSON files into pandas DataFrames using orjson, with specialized functions for analyzing template instantiation costs and compilation phase breakdowns. It includes a core library (`trace_analysis/`), example scripts for quick analysis, a comprehensive README with usage documentation, and an interactive Jupyter notebook demonstration.
Key features include memory-efficient DataFrame schemas with optimized dtypes, recursive hierarchical phase analysis, automatic metadata extraction (source file, compilation timing), and template instantiation filtering. The design supports both standalone scripts and interactive Jupyter notebook workflows.
This single-file analysis capability lays the groundwork for future multi-file analysis across thousands of compilation units, enabling data-driven optimization and build time regression detection.
[ROCm/composable_kernel commit: a213ce676b]
* Add padding support with transpose
Also move check before writing storing is_src_valid during reading
* Add/modify instances to use wave transfer for gemm universal
Condition is changed so now the vectorsize of vmem reading and lds
writing must be equal to 8 in order to use the wave transfer
* Fix clang format
* Modify example
* Fix bwd data
* Add restriction for wave transfer with padding and transpose
Add test case which shows this limitation
* Fix validity checks 8 bit types
* Add validity check gemm_bias_add_reduce
* Add validity check grouped gemm tile loop
* Fix validity checks new flavours
* Minor fixes
* Fix clang format
[ROCm/composable_kernel commit: 2e49b6b2f7]
* WIP: host level interwave pipeline compiles
* WIP: interwave implementation computes correct GEMM result when no aquant
* WIP: quantization works for subset of problem shapes
* WIP: quantization works for subset of problem shapes
* WIP: interwave memory pipeline passes local test
* feat: Add interwave pipeline implementation for memory pipline in aquant
* test: add unit test for aquant memory pipeline
* WIP: host level interwave pipeline compiles
* WIP: interwave implementation computes correct GEMM result when no aquant
* WIP: quantization works for subset of problem shapes
* WIP: quantization works for subset of problem shapes
* WIP: interwave memory pipeline passes local test
* feat: Add interwave pipeline implementation for memory pipline in aquant
* fix: compilation error on gfx950
* chore: remove debug statements from the code
* test: resolve merge conflict
* test: remove non rcr unit tests from test suite
[ROCm/composable_kernel commit: b8751e505d]
* Re-enable f8 x bf8 tests on CompV3 as they now pass
* On CompV4, fp8 x bf8 tests now pass with K_BlockSize I32
* Add a changelog entry
---------
Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>
[ROCm/composable_kernel commit: 834642202c]
* Added common struct to enable code reduction in gemm gemm and gemm multi_d gemm multi_d wmma implementation
This file includes all shared components. The (shared between the two implementations) kernel, the pointer offset computation struct, the grid descriptor creator and definitions, the invoker struct and the argument struct.
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
* Used the common struct in the batched gemm gemm wmma cshuffle v3 implementation
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
* Used the shared structs in the gemm multiple D gemm multiple D wmma cshuffle v3 implementation
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
* Boy-scout: IWYU paradigm in the gemm gemm and gemm multiple D gemm multiple D wmma cshuffle v3 implementations
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
---------
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
[ROCm/composable_kernel commit: 917f35553a]
This change significantly improves compile-time performance by reducing template
instantiation depth for sequence generation and merging operations:
Optimizations:
- sequence_gen: Reduce instantiation depth from O(log N) to O(1) by using
__make_integer_seq to generate indices in a single step, then applying the
functor via pack expansion
- uniform_sequence_gen: Similarly optimized to O(1) depth using __make_integer_seq
with a helper that applies a constant value via pack expansion
- sequence_merge: Reduce depth from O(N) to O(log N) using binary tree reduction
strategy. Added direct concatenation specializations for 1-4 sequences to
avoid recursion in common cases, falling back to binary tree merging for 5+
sequences
Documentation:
- Added extensive inline comments explaining why sequence_merge cannot achieve
O(1) depth like sequence_gen (requires computing cumulative sequence lengths
from heterogeneous inputs, inherently requiring recursion)
- Documented the binary tree reduction approach and why it's superior to fold
expressions for this use case
Testing:
- Added comprehensive unit tests for uniform_sequence_gen with different values,
sizes, and edge cases
- Added tests for sequence_gen with custom functors (double, square, identity,
constant) to verify the new implementation works with arbitrary functors
- Added tests for sequence_merge with 4, 5, and many sequences to verify both
the direct concatenation path and binary tree reduction path
- Added tests for empty sequence edge cases
[ROCm/composable_kernel commit: de59c0716c]
* Fix alignment issue in Stream-K workspace buffer
In CK Tile Stream-K, the workspace buffer is used to hold flags and
partials, where the first i bytes holds the flags and the remaining
bytes hold partials. This change adds padding to the flags prefix of the
workspace buffer to ensure the number of bytes is 128B-aligned. Without
this alignment, since workgroups do not skip cache when reading from
partials, they may read stale partials data in cache, leading to
incorrect results. The added padding avoids the stale data reading.
This change also re-enables the test_ck_tile_streamk_reduction tests.
* Compute reference GEMM on GPU for test verification to decrease testing time
[ROCm/composable_kernel commit: f5c2f09036]
* Moved device struct for batched gemm wmma to a common file
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
* Use the common device struct in the scaled batched gemm wmma implementation
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
* Boy-scout: Remove unused includes and ambiguous comment
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
* Moved pointer offset calculation and gridwise argument to common struct
This change enables further code reduction by re-using the common structs for the batched gemm and batched gemm b scale wmma implementations.
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
* Moved type string to the common struct of DeviceBatchedGemm_Wmma_CShuffleV3_Common"
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
---------
Signed-off-by: Chris Tsiaousis <chris.tsiaousis@streamhpc.com>
[ROCm/composable_kernel commit: e1c46ff548]
* Added bias_bnorm_clamp for WMMA conv fwd large tensor.
Following operations are added for FP16/BF16 data type and NHWGCxGKYXC layout.
- grouped_conv2d_fwd_bias_bnorm_clamp
- grouped_conv3d_fwd_bias_bnorm_clamp
* changed strategy to handle GemmArgs array
* Adding generic instance
* fixed last nits from reviewers and copilot
[ROCm/composable_kernel commit: 2e08a7e5ab]
* Addition of Stream-K tests using Tile Engine
This change adds an implementation for generating Stream-K tests using Tile Engine.
This will generate various test executables for different combinations based on the
config files. This addition has simple tests running for bf16 and fp16, with both
atomic and reduction strategies and compv3 pipeline. The tests rely on the implementation
of Stream-K in Tile Engine.
* integrating addition of tree reduction and editing the README
* temporarily removing parallel and tree reduction from configs while bugs regarding them are being resolved
[ROCm/composable_kernel commit: b9bb1db5d9]
* WIP POC of dispatcher
* Dispatcher python workflow setup.
* Dispatcher cleanup and updates.
Further dispatcher cleanup and updates.
Build fixes
Improvements and python to CK example
Improvements to readme
* Fixes to python paths
* Cleaning up code
* Improving dispatcher support for different arch
Fixing typos
* Fix formatting errors
* Cleaning up examples
* Improving codegeneration
* Improving and fixing C++ examples
* Adding conv functionality (fwd,bwd,bwdw) and examples.
* Fixes based on feedback.
* Further fixes based on feedback.
* Adding stress test for autogeneration and autocorrection, and fixing preshuffle bug.
* Another round of improvements based on feedback.
* Trimming out unnecessary code.
* Fixing the multi-D implementation.
* Using gpu verification for gemms and fixing convolutions tflops calculation.
* Fix counter usage issue and arch filtering per ops.
* Adding changelog and other fixes.
* Improve examples and resolve critical bugs.
* Reduce build time for python examples.
* Fixing minor bug.
* Fix compilation error.
* Improve installation instructions for dispatcher.
* Add docker based installation instructions for dispatcher.
* Fixing arch-based filtering to match tile engine.
* Remove dead code and fix arch filtering.
* Minor bugfix.
* Updates after rebase.
* Trimming code.
* Fix copyright headers.
* Consolidate examples, cut down code.
* Minor fixes.
* Improving python examples.
* Update readmes.
* Remove conv functionality.
* Cleanup following conv removable.
[ROCm/composable_kernel commit: 9e049a32a1]
* Added bias_bnorm_clamp instances.
* fwd_bias_bnorm_clamp comp instances
* fwd_bias_bnorm_mem_inter and mem_intra instances
* fwd_bias_bnorm_merged_group_instances
* fwd_bias_bnorm_clamp_conv3d_bf16 and f16 instances
* Device level changes for fwd_bias_bnorm_clamp
* Added the test to the regression test list.
* Removed the part 2 and 2x instances
* Removed the irrelevant checks in wmma
* Refactored the instances to adapt to new device implementation
* Updated the reference and include files
* enabling tests
* Added missing profiler
* Added missing instance entry , deleted by mistake
* Reduce bias bnorm clamp instances to only a single generic one.
* Clean up cmakelists file
* clang-format
* Change bias bnorm clamp tests to use monotone initialization values to avoid tiny off-integer gemm results on RDNA3 from blowing up.
* Renaming some instance lists and add functions to be more standardized.
* Commented out non default instances.
---------
Co-authored-by: kiefer <kiefer.van.teutem@streamhpc.com>
[ROCm/composable_kernel commit: 8daf6ea302]
* add block scale parameters to kernel
* add block scale to kernel
* add smoke test
* format
* Revert "format"
This reverts commit 356c3c9706.
* only format my code
* format py
* fix auto not allowd in function prototype
* change instance tttt to ttff
* fix structured binding issue
* change s_acc elementwise op
* async pipeline add block scale
* add quantation P using shift exp2
* precompute (m - shift) once per row
* change blk scale seqstrt ptr name
* fix some name
* fix for deduction guide
* fix some comments
* add P scale to qr_ksvs_pipeline
* add comment to idx_identity
* change the method of calculating descale block index
* unify naming style: use block_scale_ as name prefix
* unify naming style
* update the CHANGELOG.md
* Add FP8 block scale quantization support for FMHA forward kernel
---------
Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>
Co-authored-by: Po Yen Chen <PoYen.Chen@amd.com>
[ROCm/composable_kernel commit: dd0b4294af]
* ck-builder: remove SPATIAL_DIM parameter from ConvTensorLayouts
This information is already in the SIGNATURE, so its pointless to pass it
separately. This streamlines the interface of those functions a bit. Also
touches up the style of those files in general.
* ck-builder: implement reference conv using old ck
The old ck implementation is more featureful and better tested.
* ck-builder: replace test_reference_execution reference with old ck
This strips out the ck-tile gpu reference implementation completely.
* ck-builder: clean up test_reference_execution
- Remove unneccesary messages
- Replace EXPECT_TRUE(true) with EXPECT_NO_THROW()
[ROCm/composable_kernel commit: 1040d9b1f5]
