[llvm-testresults] Grawp-PIC i386 nightly tester results

Apache apache at cs.uiuc.edu
Thu Apr 23 07:30:53 PDT 2009


http://llvm.org/nightlytest/test.php?machine=372&night=10536
Name: grawp.apple.com
Nickname: Grawp-PIC
Buildstatus: OK

New Test Passes:
Applications/ClamAV/clamscan [] 
Applications/JM/ldecod/ldecod [LLC compile, ] 


New Test Failures:
None

Added Tests:
test/Analysis/ScalarEvolution/2009-04-22-TruncCast.ll
test/Analysis/ScalarEvolution/pr3909.ll
test/Analysis/ScalarEvolution/sext-inreg.ll
test/CodeGen/X86/2009-04-21-NoReloadImpDef.ll
test/FrontendC++/2009-04-21-DtorNames-dbg.cpp for PR4025
test/FrontendC/2009-04-22-UnknownSize.c for PR2958
test/TableGen/DefmInherit.td
test/TableGen/MultiClassInherit.td
test/TableGen/nameconcat.td
test/Transforms/ArgumentPromotion/pr3085.ll
test/Transforms/IndVarSimplify/2009-04-22-IndvarCrash.ll
test/Transforms/IndVarSimplify/casted-argument.ll for PR4009
test/Transforms/LoopRotate/pr2639.ll
test/Transforms/LoopStrengthReduce/pr2537.ll
test/Transforms/LoopStrengthReduce/pr3086.ll
test/Transforms/ScalarRepl/2009-04-21-ZeroLengthMemSet.ll


Removed Tests:
None

Significant changes in test results:
Bytecode:
 singlesource/Benchmarks/Adobe-C++/stepanov_vector: 6.75% (48832 => 45536)
 singlesource/Benchmarks/Shootout-C++/reversefile: 14.12% (9744 => 8368)
 singlesource/Regression/C/bigstack: -36.84% (2432 => 3328)
 multisource/Benchmarks/FreeBench/distray/distray: 18.34% (8112 => 6624)
 multisource/Benchmarks/FreeBench/mason/mason: 39.73% (9584 => 5776)
 multisource/Benchmarks/MallocBench/gs/gs: 5.34% (437504 => 414160)
 multisource/Benchmarks/MiBench/consumer-jpeg/consumer-jpeg: 8.93% (210992 => 192144)
 multisource/Benchmarks/tramp3d-v4/tramp3d-v4: 12.46% (2176912 => 1905696)
 external/SPEC/CFP2006/447.dealII/447.dealII: 11.37% (1688560 => 1496576)
 external/SPEC/CINT2000/252.eon/252.eon: 5.03% (511664 => 485904)
GCCAS:
 multisource/Applications/ClamAV/clamscan: 10.33% (5.5484 => 4.9752)
CBE:
 multisource/Benchmarks/Olden/bh/bh: 8.28% (3.26 => 2.99)




More information about the llvm-testresults mailing list