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

Apache apache at cs.uiuc.edu
Tue Sep 30 05:37:27 PDT 2008


http://llvm.org/nightlytest/test.php?machine=324&night=7916
Name: ozzy
Nickname: ozzy-clang-x86
Buildstatus: OK

New Test Passes:
None

New Test Failures:
None

Added Tests:
test/Assembler/2008-09-29-RetAttr.ll
test/CodeGen/X86/2008-09-29-ReMatBug.ll
test/CodeGen/X86/2008-09-29-VolatileBug.ll for PR2835
test/Transforms/InstCombine/2008-09-29-FoldingOr.ll for PR2844


Removed Tests:
None

Significant changes in test results:
LLC-BETA:
 singlesource/Benchmarks/BenchmarkGame/n-body: -21.49% (6.98 => 8.48)
 singlesource/Benchmarks/BenchmarkGame/nsieve-bits: -19.19% (1.72 => 2.05)
 singlesource/Benchmarks/CoyoteBench/fftbench: 11.26% (2.93 => 2.60)
 singlesource/Benchmarks/CoyoteBench/huffbench: -27.39% (25.96 => 33.07)
 singlesource/Benchmarks/McGill/queens: -12.83% (3.82 => 4.31)
 singlesource/Benchmarks/Misc-C++/llloops: -13.20% (3.94 => 4.46)
 singlesource/Benchmarks/Misc-C++/stepanov_v1p2: -23.34% (10.54 => 13.00)
 singlesource/Benchmarks/Misc/ReedSolomon: -8.31% (11.19 => 12.12)
 singlesource/Benchmarks/Misc/ffbench: -17.07% (1.23 => 1.44)
 singlesource/Benchmarks/Misc/himenobmtxpa: -20.42% (1.42 => 1.71)
 singlesource/Benchmarks/Misc/oourafft: -24.90% (10.20 => 12.74)
 singlesource/Benchmarks/Shootout-C++/matrix: -35.89% (6.13 => 8.33)
 singlesource/Benchmarks/Shootout-C++/methcall: 8.17% (9.18 => 8.43)
 singlesource/Benchmarks/Shootout-C++/sieve: -8.02% (2.62 => 2.83)
 singlesource/Benchmarks/Shootout/ary3: -7.13% (4.35 => 4.66)
 singlesource/Benchmarks/Shootout/hash: 7.66% (5.48 => 5.06)
 singlesource/Benchmarks/Shootout/matrix: -10.30% (5.34 => 5.89)
 singlesource/Benchmarks/Shootout/sieve: -23.16% (11.14 => 13.72)
 multisource/Applications/aha/aha: -9.66% (5.59 => 6.13)
 multisource/Applications/hexxagon/hexxagon: -5.24% (17.94 => 18.88)
 multisource/Applications/obsequi/Obsequi: -15.51% (3.74 => 4.32)
 multisource/Applications/siod/siod: -32.16% (4.82 => 6.37)
 multisource/Applications/viterbi/viterbi: -29.63% (17.38 => 22.53)
 multisource/Benchmarks/MallocBench/cfrac/cfrac: -8.40% (2.62 => 2.84)
 multisource/Benchmarks/Olden/bh/bh: -6.87% (4.66 => 4.98)
 multisource/Benchmarks/Olden/em3d/em3d: -28.67% (6.94 => 8.93)
 multisource/Benchmarks/Olden/tsp/tsp: -5.00% (4.80 => 5.04)
 multisource/Benchmarks/Ptrdist/anagram/anagram: -15.28% (2.16 => 2.49)
 multisource/Benchmarks/Ptrdist/yacr2/yacr2: -16.84% (1.90 => 2.22)
 multisource/Benchmarks/SciMark2-C/scimark2: 20.45% (34.57 => 27.50)
 multisource/Benchmarks/Trimaran/enc-3des/enc-3des: -18.43% (2.55 => 3.02)
 multisource/Benchmarks/Trimaran/enc-md5/enc-md5: -70.83% (2.16 => 3.69)
 multisource/Benchmarks/Trimaran/enc-rc4/enc-rc4: -40.23% (1.74 => 2.44)
 multisource/Benchmarks/VersaBench/8b10b/8b10b: -6.93% (8.66 => 9.26)
 multisource/Benchmarks/VersaBench/beamformer/beamformer: -10.55% (2.75 => 3.04)
 multisource/Benchmarks/VersaBench/ecbdes/ecbdes: -40.59% (3.40 => 4.78)
 multisource/Benchmarks/llubenchmark/llu: 11.40% (8.07 => 7.15)
CBE:
 singlesource/Benchmarks/CoyoteBench/fftbench: 10.56% (3.03 => 2.71)
 multisource/Benchmarks/Olden/tsp/tsp: 5.24% (4.96 => 4.70)
JIT:
 singlesource/Benchmarks/CoyoteBench/fftbench: 11.50% (2.87 => 2.54)
 singlesource/Benchmarks/Dhrystone/dry: -15.54% (3.86 => 4.46)
 singlesource/Benchmarks/Misc/ffbench: -18.10% (1.16 => 1.37)
 singlesource/Benchmarks/Shootout-C++/lists: 7.84% (9.82 => 9.05)
LLC:
 singlesource/Benchmarks/CoyoteBench/fftbench: 11.43% (2.80 => 2.48)
 singlesource/Benchmarks/Misc/richards_benchmark: -14.37% (1.60 => 1.83)
 multisource/Benchmarks/Trimaran/enc-rc4/enc-rc4: -22.30% (1.48 => 1.81)
LLC-BETA compile:
 multisource/Applications/siod/siod: 66.99% (0.4120 => 0.1360)
 multisource/Benchmarks/ASCI_Purple/SMG2000/smg2000: 60.19% (0.4320 => 0.1720)
 multisource/Benchmarks/MallocBench/gs/gs: 51.89% (0.4240 => 0.2040)
 multisource/Benchmarks/MiBench/consumer-typeset/consumer-typeset: 61.79% (1.4760 => 0.5640)
 multisource/Benchmarks/tramp3d-v4/tramp3d-v4: 82.73% (1.1120 => 0.1920)




More information about the llvm-testresults mailing list