mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-02 04:31:25 +00:00
2e4b8a8fc4
* Remove old CK Tile Stream-K implementation The original CK Stream-K implementation was based on old CK's Stream-K block to C tile map. However, this implementation did not align with the original Stream-K paper. Thus, we implemented a new tile partitioner and associated Stream-K kernel, which was placed in the reboot namespace. Now that the new Stream-K implementation is ready, this change removes all artifacts of the old implementation. Specifically, the following changes were made: - Removes old Stream-K tile partitioner from CK Tile - Removes the reboot namespace such that the new implementation resides in the ck_tile namespace only. - Adds tests for bf8 and fp8 using the new implementation - Removes tests for the old implementation - Remove the v2 suffix from the new CK Tile Tile Partitioner derived classes. - Updates Stream-K Kernel ops file to use /** commenting style. * Remove v2 from tile partitioner validation function names
Stream-K GEMM
This folder contains examples of Stream-K GEMMs using the ck_tile tile-programming implementation.
build
# in the root of ck_tile
mkdir build && cd build
# you can replace <arch> with the appropriate architecture (for example gfx942) or leave it blank
../script/cmake-ck-dev.sh ../ <arch>
# Compile the Stream-K kernels
make tile_example_streamk_gemm_basic -j
This will result in an executable build/bin/tile_example_streamk_gemm_basic
example
args:
-m m dimension (default:512)
-n n dimension (default:512)
-k k dimension (default:512)
-a_layout tensor A data layout (default: R)
-b_layout tensor B data layout (default: C)
-c_layout tensor C data layout (default: R)
-reduction_strategy strategy for storing results in C tensor. atomic/reduction (default:atomic)
-persistent_dp persistent strategy for data-parallel section. Set to 0 for non-persistent or to 1 for persistent. (default:0)
-stride_a tensor A stride (default:0)
-stride_b tensor B stride (default:0)
-stride_c tensor C stride (default:0)
-v validation strategy. 0. No validation, 1. Validation on CPU, 2. Validation on GPU (default:1)
-prec data type. fp16/bf16/fp8/bf8 (default:fp16)
-warmup number of iterations before benchmarking the kernel (default:50)
-repeat number of iterations to benchmark the kernel (default:100)
-timer timing mode. gpu:gpu timer, cpu:cpu timer (default:gpu)
-init data initialization strategy. 0:random, 1:linear, 2:constant(1) (default:0)
-flush_cache flush the cache before running the kernel (default:true)