[llvm-testresults] cfarm-32 i686 nightly tester results

Apache apache at cs.uiuc.edu
Tue Oct 7 16:54:43 PDT 2008


http://llvm.org/nightlytest/test.php?machine=316&night=8032
Name: gcc09
Nickname: cfarm-32
Buildstatus: OK

New Test Passes:
Benchmarks/Olden/bh/bh [LLC compile, LLC-BETA compile, JIT codegen, CBE, LLC, LLC-BETA, JIT] 
Benchmarks/Olden/bisort/bisort [LLC compile, LLC-BETA compile, JIT codegen, CBE, LLC, LLC-BETA, JIT] 
Benchmarks/Olden/health/health [, LLC-BETA compile, ] 
Benchmarks/Olden/mst/mst [LLC compile, LLC-BETA compile, JIT codegen, CBE, LLC, LLC-BETA, JIT] 
Benchmarks/Olden/perimeter/perimeter [LLC compile, LLC-BETA compile, JIT codegen, CBE, LLC, LLC-BETA, JIT] 
Benchmarks/Olden/tsp/tsp [LLC compile, LLC-BETA compile, JIT codegen, CBE, LLC, LLC-BETA, JIT] 
Benchmarks/Olden/voronoi/voronoi [LLC compile, LLC-BETA compile, JIT codegen, CBE, LLC, LLC-BETA, JIT] 


New Test Failures:
None

Added Tests:
test/CodeGen/Alpha/add128.ll
test/CodeGen/X86/2008-10-06-MMXISelBug.ll for PR2850
test/CodeGen/X86/2008-10-06-x87ld-nan-1.ll
test/CodeGen/X86/2008-10-06-x87ld-nan-2.ll
test/FrontendObjC/2008-10-3-EhValue.m
test/Other/2008-10-06-RemoveDeadPass.ll
test/Transforms/LoopIndexSplit/2008-10-06-Crash.ll
test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll


Removed Tests:
None

Significant changes in test results:
JIT:
 singlesource/Benchmarks/BenchmarkGame/n-body: -193.86% (11.23 => 33.00)
 singlesource/Benchmarks/Misc-C++/mandel-text: 22.93% (39.17 => 30.19)
 singlesource/Benchmarks/Shootout-C++/fibo: -30.53% (12.48 => 16.29)
 singlesource/Benchmarks/Shootout-C++/lists: -6.59% (42.46 => 45.26)
 singlesource/Benchmarks/Shootout/fib2: -30.63% (12.44 => 16.25)
 multisource/Applications/SIBsim4/SIBsim4: 11.29% (32.15 => 28.52)
 multisource/Benchmarks/Prolangs-C++/primes/primes: 13.30% (3.91 => 3.39)
 multisource/Benchmarks/VersaBench/8b10b/8b10b: -8.55% (16.26 => 17.65)
LLC:
 singlesource/Benchmarks/BenchmarkGame/n-body: 21.60% (14.21 => 11.14)
 singlesource/Benchmarks/Dhrystone/fldry: -40.93% (20.72 => 29.20)
 singlesource/Benchmarks/Misc-C++/mandel-text: 19.70% (37.66 => 30.24)
 singlesource/Benchmarks/Shootout-C++/fibo: -23.53% (12.45 => 15.38)
 singlesource/Benchmarks/Shootout/fib2: -23.67% (12.42 => 15.36)
 multisource/Applications/SIBsim4/SIBsim4: -8.81% (19.29 => 20.99)
 multisource/Applications/spiff/spiff: -9.77% (5.22 => 5.73)
 multisource/Benchmarks/FreeBench/analyzer/analyzer: -73.68% (0.38 => 0.66)
 multisource/Benchmarks/VersaBench/8b10b/8b10b: -15.69% (15.55 => 17.99)
 multisource/Benchmarks/tramp3d-v4/tramp3d-v4: 15.08% (7.36 => 6.25)
LLC-BETA:
 singlesource/Benchmarks/BenchmarkGame/n-body: 20.59% (14.13 => 11.22)
 singlesource/Benchmarks/BenchmarkGame/recursive: -5.51% (5.81 => 6.13)
 singlesource/Benchmarks/CoyoteBench/fftbench: 6.21% (14.01 => 13.14)
 singlesource/Benchmarks/Dhrystone/fldry: 29.36% (29.33 => 20.72)
 singlesource/Benchmarks/Misc-C++/oopack_v1p8: -58.23% (1.58 => 2.50)
 singlesource/Benchmarks/Misc/flops: -6.67% (34.49 => 36.79)
 singlesource/Benchmarks/Shootout-C++/fibo: -24.75% (13.05 => 16.28)
 singlesource/Benchmarks/Shootout/fib2: -24.96% (13.02 => 16.27)
 multisource/Applications/spiff/spiff: -8.47% (5.43 => 5.89)
 multisource/Benchmarks/FreeBench/analyzer/analyzer: -80.49% (0.41 => 0.74)
 multisource/Benchmarks/Prolangs-C++/primes/primes: 10.53% (3.42 => 3.06)
 multisource/Benchmarks/VersaBench/8b10b/8b10b: 7.83% (22.86 => 21.07)
CBE:
 singlesource/Benchmarks/Misc/flops: -6.48% (33.66 => 35.84)
 singlesource/Benchmarks/Shootout/fib2: -27.47% (13.58 => 17.31)
 multisource/Applications/sgefa/sgefa: 9.31% (6.98 => 6.33)
 multisource/Applications/spiff/spiff: -13.03% (5.45 => 6.16)
 multisource/Benchmarks/FreeBench/analyzer/analyzer: 19.81% (1.06 => 0.85)
GCCAS:
 multisource/Applications/siod/siod: 5.58% (18.6071 => 17.5693)
 multisource/Benchmarks/Prolangs-C/gnugo/gnugo: 8.43% (2.6785 => 2.4526)
LLC compile:
 multisource/Applications/spiff/spiff: -12.97% (1.5807 => 1.7857)
LLC-BETA compile:
 multisource/Benchmarks/MallocBench/espresso/espresso: -7.95% (2.8665 => 3.0945)




More information about the llvm-testresults mailing list