ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-05 14:11:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

Revert "[CK_tile] Add rotating buffer feature for universal gemm (#2200)" (#2256)

Browse Source 
This reverts commit 99857e10e6.
This commit is contained in:
Illia Silin
2025-05-28 08:46:52 -07:00
committed by GitHub
parent 4286eae09a
commit bbdaf79a52
17 changed files with 74 additions and 409 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tile_engine/ops/gemm/README.md
View File

@@ -42,8 +42,6 @@ rm -rf tile_engine/ && ninja benchmark_gemm # rebuild
-repeat The number of iterations to benchmark the kernel. Default is 100.
-timer Whether if the timer is gpu timer or not. Possible values are true or false. Default is true.
-init The method of tensor initialization. Set to 0 for random, to 1 for linear, or 2 for constant(1). Default is 0, random.
-flush_cache To flush cache in between different runs.Possible values are true or false. Default is false.
-rotating_count count to flush cache. Default is 5.
-metric Metric with which to measure kernel performance. Set to 0 for latency, 1 for tflops, or 2 for bandwidth. Default is 0, latency.
-csv_filename The filename of benchmark result. Default is gemm_kernel.
-structured_sparsity whether use sparsity kernel or not. Possible values are true or false. Default is false.