* * [CK_TILE] Add sequence padding and variable length support in fmha (and v3)
- Group Mode Padding: Introduces the `-s_qpad` argument to support
physically padded layouts. Kernels now use padded start pointers
(`seqstart_padded_*_ptr`) for memory addressing.
- Batch Mode Variable Length: Adds `-q_eff_lens` and `-kv_eff_lens`
arguments for efficient processing of variable-length sequences by
passing cumulative effective lengths (`cu_seqlen_*_ptr`) to the kernel.
- FMHA examples: Support padding and variable length both in
group and batch mode. Dispatcher is updated as well (dispatch to
kPadSeqLenK enabled pipeline).
- New padding test cases: Add padding test cases to `smoke_test_fwd.sh` and
`test_fmha_fwd.inc`, and add benchmarks to `benchmark_fwd.sh` and
`benchmark_fwd_v3.sh` as well. These test cases and benchmarks that
specifically validate/benchmark the new padding and variable-length
functionalities in both group and batch modes.
* [CK_TILE] Fix build error in fmha unit tests
* [CK_TILE] add mqa, gqa to sequence padding unit tests
* [CI_TILE] Reduce the number of padding seqlen unit tests in FMHA to avoid timeouts in CI
* [CK_TILE] remove unnecessary MageKArgs overload in FmhaFwdV3Kernel and FmhaFwdKernel
* Remove C++20 code
C++20 features should not be used in CK. Remove all C++20 code.
* fix c++17 build
* format
* fix merge issue
---------
Co-authored-by: Thomas Ning <Thomas.Ning@amd.com>
Co-authored-by: Max Podkorytov <4273004+tenpercent@users.noreply.github.com>
* Update grouped_gemm example and pipeline
* find the root cause error in did not enable the transpose in gfx950 correctly
* Fix v3 pipeline, row and col major
* Disable f8 datatype tests, it fails on gfx950
* fix the abd test by clear the runtime argument unsupported
---------
Co-authored-by: AviralGoelAMD <aviral.goel@amd.com>
Co-authored-by: Mateusz Ozga <mateusz.ozga@amd.com>
* disable cast_tile_pk_fp16_fp32 on gfx950
* fix wrong encoding when hdim is not exponentiation of 2
---------
Co-authored-by: asleepzzz <hanwen.chang@amd.com>
* Have a workable version for SGPR
* have a workable version for atomic add
* Revert "have a workable version for atomic add"
This reverts commit 792377a590c26cfff9c8f545d9a9e8484a7422eb.
* substitute with the new sgpr read api
* update the CHANGELOG
* have a workable version for atomic add
* Revert "have a workable version for atomic add"
This reverts commit 792377a590c26cfff9c8f545d9a9e8484a7422eb.
* change to static for logic
* have a workable version for atomic add
* Revert "have a workable version for atomic add"
This reverts commit 792377a590c26cfff9c8f545d9a9e8484a7422eb.
* rename gemm_group_quant to gemm_quant
* Add TensorWise quant mode
* Cshuffle epilogue tests with tensor scaling
* Add tensor quant to example
* Don't use readfirstlane for reading scales - doesn't work for some reason
* Add to changelog
* revert include - from a merge problem?
* revert common.hpp include
* revert host.hpp include
* remove unused utility function
* rename quant pipeline problem
* refactor quant tests
* remove aquant utils
* use TEST_F
* fix all tests by changing gemm config
* Use typed tests
* fix copyright
* [CK_TILE] Add sequence padding and variable length support in fmha (and v3)
- Group Mode Padding: Introduces the `-s_qpad` argument to support
physically padded layouts. Kernels now use padded start pointers
(`seqstart_padded_*_ptr`) for memory addressing.
- Batch Mode Variable Length: Adds `-q_eff_lens` and `-kv_eff_lens`
arguments for efficient processing of variable-length sequences by
passing cumulative effective lengths (`cu_seqlen_*_ptr`) to the kernel.
- FMHA examples: Support padding and variable length both in
group and batch mode. Dispatcher is updated as well (dispatch to
kPadSeqLenK enabled pipeline).
- New padding test cases: Add padding test cases to `smoke_test_fwd.sh`,
and add benchmarks to `benchmark_fwd.sh` and `benchmark_fwd_v3.sh` as well.
These test cases and benchmarks that specifically validate/benchmark the
new padding and variable-length functionalities in both group and batch modes.
* [CK_TILE] Fix build error in fmha unit tests
---------
Co-authored-by: Po Yen Chen <PoYen.Chen@amd.com>
Co-authored-by: Yi DING <yi.ding@amd.com>
* Run ctest with --output-on-failure
* Fix synchronization issues in bwd pipelines
The bwd kernel reuses the same area of LDS for ds (SGrad), bias and
dbias (BiasGrad). This means that there must be block_sync_lds between
loading one tensor and storing another to the same area.
Heavy instructions like MFMA/WMMA and global loads are executed between
reuses of the same memory so in MOST cases loading is finished by all
warps before storing is started. However, sometimes warps progress at
different speeds.
Running the tests multiple times and, preferably, with multiple
processes on the same GPU helps to trigger this issue:
bin/test_ck_tile_fmha_bwd_bf16 --gtest_repeat=-1 --gtest_shuffle --gtest_throw_on_failure
* change host using fp16 to check
* fp8 to fp8 compare
* rewrite input parameters
* add not squant
* remove some output code
* for scale = 1
* format
* saturates only for fp8
* add fp8bf16 data type
* add fp8bf16 data type
* fix test fp8 code
* add run_fp8bf16_tests
* change fmha fwd example parameter(adding fp8bf16)
* Support fp8bf16 for Aiter
* Support aiter fp8bf16 in c++
* fix comment about fp8 in readme.md
* add fp8fp32
* add fp8fp32 test
* remove range_q etc.
* format
* fix test parameters about squant and fmha example input fp8bf16 fp8fp32 data type
* add fp8bf16 to data_type function
* change colmajor to rowmajor in test_ck_tile_fmha_fwd_fp8
* format
* reset atol for fp8
* fix bug for atol
---------
Co-authored-by: rocking <ChunYu.Lai@amd.com>
Co-authored-by: asleepzzz <hanwen.chang@amd.com>
* Factor out the three separate copies of load_interleaved_pk_type into a common utility class
* Add preprocessing with optional cache flushing and clearing of output for k_batch > 1 to the weight preshuffle GEMM example
* Remove a duplicate function
* Add support for B tensor type pk_int4_t for the weight preshuffle GEMM, with tests included
* I4 support introduced more failing test cases that mirror the existing ones for F8
* Simplify the check for which tests to skip (they all have F8 as A tensor type)
* Add a changelog entry
* add the test for v2 wp pipeline, polish the code, add the support of int4 for v2 wp pipeline
* have a workable version for atomic add
* Revert "have a workable version for atomic add"
This reverts commit 792377a590c26cfff9c8f545d9a9e8484a7422eb.
---------
Co-authored-by: ThomasNing <thomas.ning@amd.com>
* fix(grouped_gemm): numerical errors on gfx950 by correctly calculating the tail num
* WIP: add temp config to stress test numerical error correction
* refactor: remove comments
* Change splitk_batch_offset parameter to k_size in UniversalGemmKernel::MakeGemmTensorViews function
Prior to this change, the splitk_batch_offset parameter of
MakeGemmTensorViews had type SplitKBatchOffset. But, the only member
variable of the SplitKBatchOffset class used in the MakeGemmTensorViews
function was splitted_k (an int32_t). The splitted_k value was used as
part of defining the dimensions of the tensor view. That said, for
Stream K, we do not need to use the SplitKBatchOffset class since we are
not using Split K. Thus, this commit changes the splitk_batch_offset
parameter to a int32_t called k_size. This will avoid the constraint of
requiring a caller of MakeGemmTensorViews to use the SplitKBatchOffset
class while still providing the same functionality. Calls to
UniversalGemmKernel::MakeGemmTensorViews have been updated accordingly.
* StreamK Kernel RunGemm Implementation
Stream K cannot simply use UniversalGemmKernel's RunGemm for the
following reasons:
1. The UniversalGemmKernel::RunGemm function computes num_loop based on
a static function of the TilePartitioner. That said, for Stream K,
num_loop must be computed using a member function (namely
GetCurrentIterLength from PR #2708).
2. The UniversalGemmKernel::RunGemm function requires the use of a
SplitKBatchOffset object which is not used for Stream K since we are
not using Split K.
Thus, this change adds a RunGemm function in the StreamKKernel class.
* initial implementation for operator() for StreamKKernel: adding stream-k algorithm and calls to RunGemm
* Fix indexing and offset issues for StreamK
These changes do the following:
- Ensure offsets along the M and N dimensions are multiplied by
MPerblock or NPerBlock, respectively. This ensures tile window origins
are at the correct locations.
- Fix bug in the tile partitioner's GetTileIdxWithOffset. Now, we apply
divmod to the given references to ensure correct values are available
to the caller.
- Added documentation in the Stream-K operator()
* Initial gtests for Stream-K
These changes add an initial gtest suite for the CK Tile Stream-K
kernel. Currently, due to bugs in the StreamKTilePartitioner (which will
be handled in a future PR), there are validation issues for certain
cases which may differ on different architectures. Thus, we opted to run
cases that are only fully data-parallel (skipping others). A guard was
added to Stream-K's IsSupportedArgument method to ensure that callers
are aware of this constraint. Additionally, to ensure testing
reproducibility, options for setting the number of CUs and occupancy
were added to MakeKernelArgs.
* Use GemmPipeline operator() variant that takes hot loop and tail num
In Stream-K, the num_loop value varies per WG and per iteration of a
Stream-K loop. So instead, we use the version of the GemmPipeline's
operator() function that takes in has_hot_loop and tail_num. This is
similar to what is done in Grouped GEMM.
* changes from review: comments, move readfirstlane, remove ifndef
* Switch direction of C tensor traversal & add padding guard
Prior to this change, WGs travelled backwards through their assigned
macro tiles in the C tensor. For instance, if WG0 is responsible for C
tiles 0 and 1, it would first visit tile 1 then tile 0. This means that
the iter_end decrements in each iteration of the stream-K while loop.
Since we are working with unsigned integers, the subtraction operation
may not be safe. Thus, this change makes is such that WGs travel forward
so that their iter_start is incremented and their iter_end remains
fixed.
Additionally, we added a guard against WGs that are neither sk_blocks
nor dp_blocks to ensure such WGs do not participate in the GEMM.
Together, these changes make is such that the algorithm is correct when
sk_blocks is greater than zero.
* Disable StreamK_M256_N256_K256_SKBlocks12 test case
This instance involves >=3 WGs contributing to each macro tile in C. Due
to the use of atomics, this is resulting in precision errors. These
errors will not persist once the reduction strategy is implemented. We
will re-enable this test then.
---------
Co-authored-by: Astha Rai <astha.rai713@gmail.com>
* [CK_TILE][REGRESSION] Correct blockSize in Generic2dBlockShape (c254f3d7b4 )
WarpPerBlock_M * WarpPerBlock_N are not equal with ThreadPerBlock_M * ThreadPerBlock_N /warpSize. we should calculate BlockSize from WarpPerBlock_M * WarpPerBlock_N
To compatible with wave32, function GetBlockSize is added to calculate correct size in host side.
* fix blocksize for all kernel related with generic2dblockshap
* remove constexpr for blocks
## What's New
Add Split-N support for grouped convolution forward to handle tensors >2GB by splitting the batch dimension.
## Bug Fix
Fixed 32-bit integer overflow that caused crashes with 6+ splits:
- Use `long_index_t` for batch offset calculations
- Remove redundant GemmM initialization in constructors
## How It Works
- Automatically splits batch dimension when tensor exceeds 2GB
- Uses grid.z dimension for parallel processing of splits
- Each split processes a subset of batches independently
## Testing
Verified with tile_example_grouped_conv_fwd:
- n=3000 (6 splits) ✓
- n=3500 (7 splits) ✓
- n=10480 (40 splits) ✓
1. Refine Reduce2dShape to support both wave32 and wave64
2. Fix example reduce, permute and elementwise on gfx11 and gfx12
---------
Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>
* initial commit
* remove extra files
* fixing errors
* updated ReadMe file for mapping of diff quants with diff configs
* addressing review comments
* addressing review comments
* Resolved merge conflicts
* [CK TILE GEMM] Replace get_preshuffle_or with is_quantpreshuffle_enabled
The get_preshuffle_or was not working as expected, which led to incorrect behavior
in the quantization preshuffle process. This change replaces it with the more reliable
is_quantpreshuffle_enabled function to properly determine when preshuffle should be applied.
---------
Co-authored-by: Cong Ma <congma13@amd.com>
* Improve random number generation
* use different seed for each input (Q, K, V...);
* use deterministic generation of:
* seqstart_q/k (for group mode);
* block_table (for paged-kvcahe);
* cache_batch_idx (for kvcache);
* Extract arg_parser-related code from run functions to use them as tests
* Split examples into main programs and fmha runners, build instances separately
* Add dummy tests that use instances and runners
* Fix a missed corner case of f32->f8 conversion
When value if < min f8 denormal but > min f8 denormal / 2, it must be
rounded to min f8 denormal (i.e. 0b1), not to 0.
* Fix incorrect fp8 scales for P and O in validation code
DataTypeConfig was incorrectly compared with fp8_t.
* Add host generation of dropout random values and use it for validation
Previously host validation (reference_batched_dropout) used random
numbers generated by BlockDropout of the kernel, meaning that incorrect
generation on device (bad distribution, repeated numbers, too many zeros,
etc.) would not trigger any validation errors.
* Implement tests from smoke_test_bwd.sh
* Return result as enum to distinguish failure and missing instance
* Add tests for bwd features: bias, alibi, dropout
* Implement tests from smoke_test_fwd.sh
* Pass seqlen_q/k as vectors to fwd and bwd runners
* Add tests for fwd features: bias, alibi, dropout
* Add tests for pagedkv and splitkv
* Fix conditions when to use splitkv and pagedkv kernels
splitkv was executed only when use_kvcache which == (need_append_kvcache || use_cache_batch_idx || 0 < page_block_size).
In the SplitKV tests: the regular fwd kernel was executed if use_cache_batch_idx was not requested even when num_splitkv > 1.
In the AppendKV tests: the pagedkv kernel was executed but it often failed to find an instance.
* Add tests for appendkv
* Use is_v_rowmajor = true because there are no instances with column layout anymore
* Split public and private compile options for instances
Tests and examples need to know only about CK_TILE_FMHA_FWD_*_API.
* Improve parsing validation in bias and mask
* Pass bias as string for consistency with mask
* Catch parsing and other exceptions
* Add bwd test for deterministic flag
* Initialize fp8 tensors (-init=ufq) similarly to uf
* Fix splitkv/pagedkv invocation: use padded sk when seqlen_k_ptr is not null
seqlen_k cannot be used to determine padding when seqlen_k_ptr is
provided. The actual seqlen_k is taken from seqlen_k_ptr[b].
Even seqlen_k values (% bn0 == 0) use padded seqlen_k while seqlen_k_ptr
may contain arbitrary values.
In the example or tests this produces incorrect results with appendkv
(for example, -d=32 -s=1 -s_k=64 -s_knew=7 -vlayout=c -b=8).
* Fix use_pagedkv value when kvcache = true but page_block_size = 0
In this case block_table_ptr is nullptr which is accessed in the kernel.
* Clean up bwd tests
* Unify fwd tests for f16/bf16 and fp8
* Use better explicit instantiation declaration for fmha_bwd<2>
* Use the same seed for all tests, allow to override it with env variable
* Undo clang-format of one irrelevant file
For some reason my local clang-format-18 and the one in CI work differently.
* Do not build instances and tests on unsupported archs
* Build instance libraries as OBJECT library
* CI: Enable sccache for HIP
There are source files with LANGUAGE HIP, they need
-DCMAKE_HIP_COMPILER_LAUNCHER=sccache
* Add tests to REGRESSION_TESTS
* Fix OOB accesses in deterministic bwd due to incorrectly assumed kN0
The runner assumes kN0 = (hdim_q <= 128) ? 128 : 64 but there are
smaller tiles (for tr_load or fp32). This can create too small dq_acc_buf.
* Pass CK_TILE_FMHA_FWD_*_API as INTERFACE compile options
The instances don't actually depend on them, only examples and tests do.
Passing these definitions as INTERFACE allows to change FMHA_FWD_ENABLE_APIS
without recompiling instances that are already in ccache.
* Fix formatting and names
BlockWarps, WarpTile in Generic2dBlockShape are wave size dependent, it causes mangled name mismatch between host and device side.
Solution: Replace them with ThreadPerBlock and move BlockWarps, WarpTile calculation into Generic2dBlockShape
- Add support for tensor A/B in both fp16+pk_int4_t and fp8+pk_int4_t formats
- Implement A(bf8) B(i4) support in universal GEMM
- Use new implementation for i4 to fp8 conversion in Block Scale
