admin/matmul-analysis
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add BMM_Theory: p2.csv

Browse Source
This commit is contained in:
admin
2026-09-03 12:37:12 +00:00
parent 32febc1f62
commit 98223e0e12
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
BMM/BMM_Theory/p2.csv Normal file
View File

@@ -0,0 +1,2 @@
case_id,branch,npu,op,used_core_num,split_b,m_cnt,n_cnt,grid_k,core_map,b_core,merge_b0,single_core_m,single_core_n,single_core_k,k_l1,b_l1,l1_form,base_m,base_n,base_k,l2_policy_in,l2_policy_out,swizzle_w,workspace_bytes,tail_strategy,tail_m_cnt,tail_n_cnt,tail_k_cnt,tail_m_main,tail_n_main,tail_block_cnt,tail_wave_num,fixpipe_unitflag,out_dtype_bytes,note
x,StreamK,Ascend950PR,batch_mat_mul_v3,32,1,1,1,3,"B/M/N切出51块, 每块3核切K归约 (归约组内核c负责K段[c*K/3,(c+1)*K/3))",1,1,255,42,228,128,1,K段标准分块流水,128,42,64,allocate(部分和驻留L2),"resident(部分和4B驻留L2, 防精度丢失不随C的fp16/fp8转换)",0,6554520,grid_K=3路切K+归约,1,1,3,0,0,0,0,True,4,"P=8.33, grid_K=3, 部分和驻留L2按4B写出, AIV归约后按C dtype=4B写最终"
1 case_id branch npu op used_core_num split_b m_cnt n_cnt grid_k core_map b_core merge_b0 single_core_m single_core_n single_core_k k_l1 b_l1 l1_form base_m base_n base_k l2_policy_in l2_policy_out swizzle_w workspace_bytes tail_strategy tail_m_cnt tail_n_cnt tail_k_cnt tail_m_main tail_n_main tail_block_cnt tail_wave_num fixpipe_unitflag out_dtype_bytes note
2 x StreamK Ascend950PR batch_mat_mul_v3 32 1 1 1 3 B/M/N切出51块, 每块3核切K归约 (归约组内核c负责K段[c*K/3,(c+1)*K/3)) 1 1 255 42 228 128 1 K段标准分块流水 128 42 64 allocate(部分和驻留L2) resident(部分和4B驻留L2, 防精度丢失不随C的fp16/fp8转换) 0 6554520 grid_K=3路切K+归约 1 1 3 0 0 0 0 True 4 P=8.33, grid_K=3, 部分和驻留L2按4B写出, AIV归约后按C dtype=4B写最终