Add SchedulerV2 (pe_accel), DPPolicy overrides, and new benchmarks
- Add cycle-accurate PE accelerator scheduler (SchedulerV2) with tiled GEMM/Math pipelines (DMA_IN → GEMM → MATH → DMA_WB) - Add DPPolicy num_pes/num_cubes/num_sips overrides for single-PE testing - Support tuple target_pe for targeting specific PE subsets - Add gemm_single_pe and gpt3_qkv benchmarks - Switch default topology to pe_scheduler_v2 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
+1
-1
@@ -50,4 +50,4 @@ components:
|
||||
pe_tcm_v1: kernbench.components.builtin.pe_tcm:PeTcmComponent
|
||||
|
||||
# Custom — add your implementations here
|
||||
# pe_cpu_v2: kernbench.components.custom.my_pe_cpu:MyPeCpuComponent
|
||||
pe_scheduler_v2: kernbench.components.custom.pe_accel.scheduler:SchedulerV2Component
|
||||
|
||||
Reference in New Issue
Block a user