[llvm-testresults] lauro-clang-x86 i686 nightly tester results

Apache apache at cs.uiuc.edu
Fri May 23 23:06:07 PDT 2008


http://llvm.org/nightlytest/test.php?machine=272&night=6172
Name: laurov-desktop
Nickname: lauro-clang-x86
Buildstatus: OK

New Test Passes:
None

New Test Failures:
None

Added Tests:
None

Removed Tests:
None

Significant changes in test results:
JIT:
 singlesource/Benchmarks/BenchmarkGame/fannkuch: 6.17% (4.86 => 4.56)
 singlesource/Benchmarks/McGill/chomp: 5.92% (3.55 => 3.34)
 singlesource/Benchmarks/Misc/ffbench: 16.83% (3.98 => 3.31)
 singlesource/Benchmarks/Misc/richards_benchmark: 10.57% (2.27 => 2.03)
 singlesource/Benchmarks/Shootout/methcall: -6.87% (7.42 => 7.93)
 singlesource/Benchmarks/Shootout/sieve: -5.30% (11.52 => 12.13)
 singlesource/UnitTests/Vector/multiplies: -12.47% (7.46 => 8.39)
 multisource/Applications/JM/lencod/lencod: -6.27% (22.66 => 24.08)
LLC:
 singlesource/Benchmarks/BenchmarkGame/fannkuch: 6.12% (4.90 => 4.60)
 singlesource/Benchmarks/Misc/ffbench: 18.48% (3.68 => 3.00)
 singlesource/Benchmarks/Shootout/methcall: -7.88% (6.47 => 6.98)
 singlesource/UnitTests/Vector/multiplies: -14.46% (7.40 => 8.47)
 multisource/Benchmarks/Olden/treeadd/treeadd: -5.71% (5.60 => 5.92)
LLC-BETA:
 singlesource/Benchmarks/BenchmarkGame/fannkuch: 6.15% (4.88 => 4.58)
 singlesource/Benchmarks/McGill/chomp: 8.81% (3.29 => 3.00)
 singlesource/Benchmarks/Misc/ffbench: 5.41% (3.70 => 3.50)
 multisource/Benchmarks/SciMark2-C/scimark2: 13.23% (31.59 => 27.41)
CBE:
 singlesource/Benchmarks/Dhrystone/fldry: 14.83% (14.09 => 12.00)
 singlesource/UnitTests/Vector/build2: 10.01% (9.79 => 8.81)
 singlesource/UnitTests/Vector/multiplies: 12.79% (5.24 => 4.57)
 multisource/Applications/viterbi/viterbi: -50.53% (14.05 => 21.15)
Bytecode:
 multisource/Benchmarks/FreeBench/fourinarow/fourinarow: -13.36% (30092 => 34112)
 multisource/Benchmarks/llubenchmark/llu: 5.84% (4384 => 4128)
GCCAS:
 multisource/Benchmarks/MiBench/security-rijndael/security-rijndael: -47.81% (0.9120 => 1.3480)




More information about the llvm-testresults mailing list