* creation of install doc and refactor of doc in general
* updates based on review comments
* updated based on review comments
* updated readme and contributors markdown
* added extra note to not use -j on its own
* added note about smoke tests and regression tests
* made changes as per Illia's feedback
---------
Co-authored-by: Aviral Goel <aviral.goel@amd.com>
[ROCm/composable_kernel commit: a426f67301]
* Add conversion tests
* Fix ctor
* Fix nan logic
* Fix conversion logic
* Permute packed f4_t values
* Fix conversion to float, repack vector elements
* Fix device tests
* Permute elements in a vector
* Add a repro test
* Add a conversion for a repro test
* Update test vectors
* Update conversion
* Fix the test
* Update test vector generator
* Fix vector sr conversion
* Permute conversion args
* Update conversion
* Test
* Fix packing
* Simplify conversion function
* Pack conversion in a loop
* Pack conversion in a loop
* Pack another conversion in a loop
* Pack one more conversion in a loop
* Pack the last conversion in a loop
* Clean up
* Add printf to fix intrinsic
* Add a sw-based workaround
[ROCm/composable_kernel commit: 441343a23d]
* only build gemm_fp8_pk_i4 examples for gfx942/950
* fix cmake logic
* moved the architecture check to IsSupported function
* Revert "moved the architecture check to IsSupported function"
This reverts commit 056d2a08b3.
* disable all pk_i4 tests for targets other than gfx942/950
* fix cmake logic
[ROCm/composable_kernel commit: 23a949706c]
* only build gemm_fp8_pk_i4 examples for gfx942/950
* fix cmake logic
* moved the architecture check to IsSupported function
* Revert "moved the architecture check to IsSupported function"
This reverts commit 056d2a08b3.
[ROCm/composable_kernel commit: 99b2bbc1d6]
* enable ClangBuildAnalizer when doing ninja traces
* add branch and date to clang build log name
* fix jenkins syntax
* fix jenkins syntax once more
* fix jenkins syntax once more
* simplify the clang_build log name
* simplify the clang_build log name further
[ROCm/composable_kernel commit: 44c093ba0c]
* 50ms -> 28ms
* Fix bug in non fuse_add_store cases
* Fine tuned setting for 2 pass pipeline
* adjust workload
* remove unnecessary change
* add layernorm
* Adding output quant and unquant results at the same time.
* fix test
* fix format
* tune for cases 128x640 and 128x1024
* bug ifx
[ROCm/composable_kernel commit: d49abdaa87]
* Fix compile error on Windows (call to 'amd_wave_read_first_lane' is ambiguous)
* Fix compile error (no matching function for call to 'cast_to_f32_from_f8')
[ROCm/composable_kernel commit: c79bf11148]
* return value with macro and revert the return value
* [CK-TILE] no-macro launch api solution (#1992)
* no-macro solution
* address -Wcomma
---------
Co-authored-by: Max Podkorytov <4273004+tenpercent@users.noreply.github.com>
[ROCm/composable_kernel commit: e3c9886cdf]
* add one daily ci build on gfx908
* add redis invocation tag for gfx908
* make ci build for gfx908 conditional
* fix groovy logic
* add option to run perf tests for gfx908
* disable a few tests on mi100
[ROCm/composable_kernel commit: 1342ecf7fb]
* Added two kernel for M=32 problem
* Comment the first one
* Enable multiply_multiply for Scale_Block_M = 1 for deepseek
* Modify the a_thread offset since the A data load is different from B.
* edit fp8 ab scale for Scale_Block_M=1
* edit GemmSpec to MNKPadding
* enable blockwise pipelie v1 and v2. v1 is work for small K.
* add instance for gemm_ab_scale
* fix cmakelist of ckProfiler
* optimize blockscale gemm. todo: reduce vgpr usage
* fix a correctness bug
* sanity checked
* revert ckprofiler cmake changes
* clang format
* revert unnecessary changes.
* remove commented codes.
* split weight preshuffle library targets
* bring back enable-post-misched=0
* fix build issues for gemm_multiply_multiply_fp8 instances
* fix clang format
* add verbose build flag when building for all targets
* reduce path names for new instances
* fix paths in cmake
* refactor gemm_multiply_multiply library target
* fix a bug in example
* fix example 65 cmake
* reduce the number of threads when building libs for all targets to 50
* use ninja to build for all targets
* reduce teh number of threads when building for all targets
* reduce the number of threads to 32 when building libs for all targets to 50
---------
Co-authored-by: mtgu0705 <mtgu@amd.com>
Co-authored-by: chenjun <junchen2@amd.com>
Co-authored-by: illsilin <Illia.Silin@amd.com>
Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>
[ROCm/composable_kernel commit: cbd74c2d12]
* tempsave, not selected
* finish the feature and merge with develop
---------
Co-authored-by: aska-0096 <haocwang@amd.com>
[ROCm/composable_kernel commit: c954bd0cfa]
* Re-implement qr_ks_vs_async pipeline by using kLoadOnce
* Remove last block_sync_lds() in the loop
* Tiny adjustment in qr_ks_vs_async pipeline for better performance
* Rename MakeQDramTileDistribution to MakeQRegTileDistribution for QLoadOnce pipeline
* Use LDS as intermediary stop when loading Q from global memory for qr_ks_vs_async pipeline
* Use un-rolled gemm for Gemm-0
* Use k0_loops small tile load/store to replace the big tile load/store for K
* Remove the commented lines in qx_ks_vs_custom_policy.hpp
* Tune the prefetching of V in qr_ks_vs_async pipeline
* Move the codes for storing the first v_lds tile some later
* Let BlockDropout reuse LDS with V
* Switch to separate code blocks according to iteration index
* Interleave code blocks for better performance
* Move clear_tile(s_acc) for better interleaving
* Move code interleaving
* Use MakeQDramTileDistribution for q_dram_window
* Roll-back to load Q directly from global memory instead of using LDS as intermediary stop
* Let V reuse the LDS of K
* Use array of tiles to represent Q in vgprs
* Use QLoadOnce == false for qr_ks_vs_async pipeline
* Special treatment for hdim-96 to save vgprs in qr_ks_vs_async pipeline
* Define statically indexed array k_lds_windows[] to reduce the using of get_slice_tile()
* Move the definition of v_tiles out from the loop
* Define statically indexed array v_lds_windows[] to reduce using of get_slice_tile()
* Remove using KLoadOnce in qx_ks_vs_custom_policy
* Remove un-used get_slice_tile() call
* Move the code line of clear_tile(s_acc)
* Tune the lines of codes to make them more tidy
* Re-arrange the codes before the main-loop
* Add comments
* Unify the alignment to be 8 for Q/K/V Lds decriptors
* Tuning to K pre-loading
* Tune K Lds and V Lds reuse for kPreloadWholeNextIterationK == false
* Adjust the pipeline codes
* Use NumPrefetchV to separate from NumVLdsBuffers
* Tune the location of a scheduler barrier code line
* Prefetch first v_tile at earlier time for both kPreloadNextWholeIterationK true/false paths
* Adjust the using of kPadSeqLenQ and kPadSeqLenK in the kernel
* Use __builtin_amdgcn_sched_barrier(0x7f) in the pipeline
* Move the location for store_tile() of first v_tile
* Rename the qr_ks_vs_async pipeline to qr_ks_vs_whole_k_prefetch pipeline
* Re-add NumPrefetchK as template for BlockFmhaPipelineQXKSVSCustomPolicy<>
* Try to fix old bugs in qx_ks_vs_custom_policy
* Remove K_LDS_LOAD_USE_OFFSET_TRANSFORM code-path to make qr_ks_vs_async and qx_ks_vs_custom_policy simpler
* Fix in MakeKDramTileDistribution() in qx_ks_vs_custom_policy
* Update to LdsBufferSequence and introduce NumKVLdsBuffers for max(NumPrefetchK, NumPrefetchV)
* Tiny Fix (#1888)
* Ck tile/paged attention workaround (#1894)
* Correction in GetRangeAlongX()
* Work-around to solve the failures in test_paged_attention_ck in xformers
* Tiny code adjustment in the qr_ks_vs_whole_k_prefetch pipeline
* Remove one call of move_tile_window for q_dram_window
* Refine the codes in GetNumPrefetchV()/GetNumKLdsBuffers()
* Tiny fix in qr_ks_vs_whole_k_prefetch pipeline
* Adjust the location of codes for storing the first V tile to LDS
* Tiny fix and add comments
* Change GetSmemKPackK size to improve performance
* Move the codes related to K-Lds to the pipeline default policy due to some override on the generic custom_policy
* Update MakeKDramTileDistribution() and MakeKLdsDescriptor() to completely remove bank conflicts for K-Lds access
* Adjustment in intermediate iteration codes for tiny performance improvement
* Reduce the number of VLds buffers to 2 for whole_k_prefetch situtation
* Use IsFirstKLdsBufferOverlapLastVLdsBuffer() to avoid potential Lds issue
* Adjust the code location for calling IsFirstKLdsBufferOverlapLastVLdsBuffer()
* Remove useless AsyncopyV
* Rename MakeQDramTileDistribution to MakeQRegTileDistribution when LDS is not used
* Keep qx_ks_vs_custom_policy work for other pipelines and move whole_k_prefetch specific codes to whole_k_prefetch default policy
* Recover the qr_ks_vs_async pipeline
* Recover qr_ks_vs_async in fmha.hpp and tiny fix in qr_ks_vs pipeline
* Revert "Try to fix old bugs in qx_ks_vs_custom_policy"
This reverts commit 39b82ca194.
* Tiny fix with regard to whole_k_prefetch pipeline compiling
* Update kPadSeqLenK setting in fmha_fwd_kernel
* Use q_element_func and k_element_func
* Use single q_tile rather than multiple sliced q_tiles
* Codes refine according to the comments
* Re-format one file
* Mark qr_ks_vs_whole_k_prefetch as QLoadOnec == true
[ROCm/composable_kernel commit: 4f54fa3058]
