Collective Knowledge Aggregator proof-of-concept
Crowd results Raw CK browser Graphs Reports Datasets Models Home

This page is outdated! New version is available here.


Distinct solutions after online classification (auto/crowd-tune LLVM compiler flags (minimize execution time))

Scenario UID2aaed4c520956635 (experiment.tune.compiler.flags.llvm.e)
Data UID4ebcb9dcce4fabb9
Discuss (optimizations to improve compilers,
semantic/data set/hardware features
to improve predictions, etc):		GitHub wiki, Google group
Download:[ All solutions in JSON ], [ Solutions' classification in JSON ]
Reproduce all (with reactions):ck replay 2aaed4c520956635:4ebcb9dcce4fabb9
CompilerLLVM 3.6
CPUMT6572
Objectivemin
Improvement key IK1Main kernel execution time speedup [min]
Improvement key IK2Code size improvement

Improvements (<4% variation) Distinct workload for highest improvement
# Solution UID IK1 IK2 New distinct optimization choices Ref Best species Worst species Touched Iters Program CMD Dataset Dataset file CPU freq (MHz) Cores Platform OS Replay
S1 8b4f933dc4f3ab65 1.14 1.05 -O3 -ffast-math -fno-lax-vector-conversions -mstrict-align -mno-movt -O3 1 0 2 1 milepost-codelet-mibench-consumer-lame-src-quantize-codelet-7-1 default 1 ENSPERT OZZY Android 4.2.2
S2 7ed3464d8ba5c561 1.12 1.02 -O3 -fno-unroll-loops -ftree-vectorize -mstrict-align -fshort-wchar -ffixed-r9 -O3 1 0 2 1 cbench-automotive-susan smoothing image-pgm-0006 data.pgm 1300, 1300 1 HUAWEI Y330-U01 Android 4.2.2



[ Participated users, platforms, OS, CPU, GPU, GPGPU, NN, NPU ] [ How to participate ] [ Motivation (PPT) (PDF) ] [ Papers 1 , 2 , 3] [ Android app ] [ Collective training set ] [ Unified AI ]
View entry in raw format

Developed by Grigori Fursin           
Implemented as a CK workflow 		                         
   
                      Hosted at