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This page is outdated!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 GCC compiler flags (minimize execution time))

Scenario UID8289e0cf24346aa7 (experiment.tune.compiler.flags.gcc.e)
Data UID66892a40c3ddc57e
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 8289e0cf24346aa7:66892a40c3ddc57e
CompilerGCC 5.4.0
CPUjetson_tx1
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 e7aebe217d6bab62 2.12 0.34 -O3 -flto -O3 3 0 22 53 milepost-codelet-mibench-security-pgp-e-src-mpilib-codelet-1-1 default 1132.8, 1132.8, 1132.8, 1132.8 4 - Ubuntu 16.04.1 LTS
S2 f3ce5a9d2224d528 1.90 0.45 -O3 -flto -O3 1 0 13 46 milepost-codelet-mibench-automotive-qsort1-src-qsort-codelet-1-1 default 1224, 1224, 1224, 1224 4 - Ubuntu 16.04.1 LTS
S3 f88ddabdce48c51f 1.76 0.84 -O3 -fno-merge-constants -fno-sched-interblock -funroll-all-loops -O3 4 1 23 53 milepost-codelet-mibench-consumer-lame-src-quantize-pvt-codelet-6-1 default 1326, 1326, 1326, 1326 4 - Ubuntu 16.04.1 LTS
S4 8943d41ae143d4a0 1.76 0.80 -O3 -fno-ssa-phiopt -funroll-all-loops -O3 2 0 25 53 milepost-codelet-mibench-consumer-lame-src-quantize-pvt-codelet-6-1 default 1555.5, 1555.5, 1555.5, 1555.5 4 - Ubuntu 16.04.1 LTS
S5 dc85dbd7f277e7c0 1.50 1.00 -O3 -fno-tree-fre -O3 1 1 10 44 milepost-codelet-mibench-consumer-jpeg-c-src-jchuff-codelet-9-1 default 1428, 1428, 1428, 1428 4 - Ubuntu 16.04.1 LTS
S6 75bd3c407982dad3 1.27 1.01 -O3 -fmodulo-sched -fno-tree-loop-vectorize -O3 1 3 24 53 cbench-security-sha default image-jpeg-dnn-cat cat.jpg 1132.8, 1132.8, 1132.8, 1132.8 4 - Ubuntu 16.04.1 LTS
S7 d3ab52846b9bb1d4 1.12 1.07 -O3 -fno-merge-constants -fno-tree-pre -O3 1 0 2 36 milepost-codelet-mibench-telecomm-adpcm-c-src-adpcm-codelet-1-1 default 1224, 1224, 1224, 1224 4 - Ubuntu 16.04.1 LTS



[ 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