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.
* 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
* 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
* 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>
* 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>
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
* 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
* 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>
* 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
* Additional flavors for WMMA conv fwd large tensor
- added F16/BF16 clamp operation
- added F16/BF16 bias_clamp operation
- small modification to the device code to accomodate extra tensors
* changed strategy to handle GemmArgs array
* Adding generic instance
* Added generic instance to clamp and bias_clamp ops
* 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
* 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.
* 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>
* 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>
* 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()
* wip: test suite for batched gemm multiple d gemm multiple d, working on gridwise implenentation
* wip: many fixes in implementation of batched gemm gemm multiple d
* wip: batched gemm gemm multiple d gridwise op compiling, not working yet
* fix: incorrect d0 grid indexing in batched gemm gemm multipled
* feat: add instances for batched gemm add relu gemm add
* chore: configure instance with low vector transfer size for odd sizes
* chore: add some more validation to device batched gemm gemm multiple d, and removed template parameter that didn't really make sense
* fix: upate device_batched_gemm_gemm_wmma to work with new gridwise changes
* fix: disable odd size tests on XDL archs
* chore: removed temporary logging
* chore: update some references to C tensor to E tensor
* Tentative fix for example template params
* Tentative fix for non-multi-D batched gemm gemm device impl.
* Tentative fix for xdl example template params
* Tentative fix for profiler build on gfx90a
* chore: improve device batched gemm gemm multi D comment to include all ops and dimensions
* chore: explicitly call ck::make_tuple to prevent issues when std::make_tuple would apply
* fix: make the gemm1 data types match what happens in the device op
* feat: add d0s/d1s datatypes and layouts to the device op type string
* chore: change element-wise op so addition happens in fp32
* chore: add static asserts for gemm0/gemm1 calculated wave sizes
* chore: also updated other element-wise ops to use fp32 calculations
* chore: log number of supported instances
* chore: update instance comment
* chore: disable kernel timing in example by default
* fix: gemm1 wave size calculation
* fix: make sure batched gemm multiple d gemm multiple d profiler performs correct type conversions
* chore: remove increased tolerance in batched gemm gemm multiple d example
* chore: add comment explaining that verification fails for certain input values
* chore: clarify instance comment
---------
Co-authored-by: kiefer <kiefer.van.teutem@streamhpc.com>
Add signal-based synchronization for persistent GEMM kernels where
input data becomes available incrementally. Uses modulo wraparound
(like PyTorch's AsyncMM) for chunk index calculation:
chunk_idx = ((tile_idx + tile_idx_pivot) / tiles_per_chunk) % num_chunks
Key components:
- PersistentAsyncInputScheduler struct with tiles_per_chunk_m,
chunk_signals, tile_idx_pivot_m, and num_chunks fields
- wait_eq_wave method using __builtin_amdgcn_s_sleep for power efficiency
- IsSupportedArgument validation for scheduler parameters
- Example demonstrating async input scheduling with simulated producer
- GTest unit tests covering all layout combinations
* Add support to fp16 + compute fp16 and bf16 + compute bf16 contractions
Enables hipTensor to access the WMMA HW functionalities
for these combinations of datatype on gfx11 and gfx12.
* Fix change to contraction scale tests
* Fix clang-format
