• waste time to chase a ghost...
• Question: number of workers can slow down Xalancbmk????
• train input, merge used
• #1: 334.756721
• #2: 342.477593
• #3: 376.607939
• #1: 348.298319
• #1: 333.039942
• #3: 357.604264
• #1: 331.753310
• #3: 359.777587
• Generate code duration? NO!
• performance different, trace, i386
• -3.37, perlbench
• -1.25, bzip2
• 4.82, gcc
• -1.64, mcf
• 0.00, gobmk
• -2.72, 
• -2.54
• -6.59
• -6.90
• -4.65
• 2.50
• 3.63
• performance difference, region, i386
• -6.61, sjeng
• -6.40, h264ref
• -5.31, omnetpp
• -2.83, libquantum
• -2.30, bzip2
• -2.11, mcf
• -1.65, perlbench
• -1.49, gobmk
• -1.11, hmmer
• 7.11, gcc
• 9.26, astar
• 2.00, xalancbmk
• performance differece, trace, ARM
• 483.xalancbmk   -12.53
• 471.omnetpp     -9.11
• 456.hmmer       -3.41
• 429.mcf         -2.43
• 473.astar       -2.02
• 403.gcc         -1.33
• 445.gobmk       -1.21
• 400.perlbench -1.04
• 464.h264ref     0.00
• 401.bzip2       2.22
• 458.sjeng       2.84
• 462.libquantum 7.46
• performance differece, region, ARM
• 471.omnetpp     -16.53
• 445.gobmk       -15.13
• 483.xalancbmk   -12.62
• 458.sjeng       -10.03
• 462.libquantum -2.49
• 473.astar       -1.83
• 456.hmmer       -1.70
• 401.bzip2       -0.47
• 403.gcc         -0.16
• 429.mcf         0.00
• 400.perlbench 0.58