[llvm-testresults] Geryon-X86-64 nightly tester results
daniel_dunbar at apple.com
daniel_dunbar at apple.com
Sat Oct 17 02:12:49 PDT 2009
http://smooshlab.apple.com/perf.cgi/db_nt_internal/nightlytest//2412
Name: clattner2.apple.com
Nickname: Geryon-X86-64:3
Buildstatus: OK
Run: 2412, Start Time: 2009-10-16 23:11:28
Comparing To: 2385, Start Time: 2009-10-15 23:11:25
--- Changes Summary ---
New Test Passes: 0
New Test Failures: 0
Added Tests: 0
Removed Tests: 0
Significant Changes: 46
--- Tests Summary ---
Total Tests: 3834
Total Test Failures: 130
Total Test Failures By Type:
CBE: 58
JIT: 26
JIT codegen: 26
LLC: 5
LLC compile: 5
LLC-BETA: 5
LLC_BETA compile: 5
--- Changes Detail ---
New Test Passes:
New Test Failures:
Added Tests:
Removed Tests:
Significant Changes in Test Results:
Bitcode:
Externals/SPEC/CFP2000/179_art/179_art: -6.72% (28816.0000 => 26880.0000)
Externals/SPEC/CINT2000/300_twolf/300_twolf: -10.21% (499856.0000 => 448800.0000)
MultiSource/Benchmarks/McCat/08-main/main: 10.88% (13232.0000 => 14672.0000)
MultiSource/Benchmarks/Olden/power/power: -27.79% (11056.0000 => 7984.0000)
SingleSource/Benchmarks/BenchmarkGame/puzzle: 7.41% (1296.0000 => 1392.0000)
SingleSource/Benchmarks/Shootout/lists: 9.25% (2768.0000 => 3024.0000)
SingleSource/Regression/C/sumarray: 6.67% (960.0000 => 1024.0000)
SingleSource/UnitTests/2002-08-19-CodegenBug: 5.41% (592.0000 => 624.0000)
SingleSource/UnitTests/2003-07-09-LoadShorts: 9.38% (1536.0000 => 1680.0000)
JIT codegen:
Externals/SPEC/CINT2000/300_twolf/300_twolf: -6.72% (1.9818 => 1.8486)
CBE:
Externals/SPEC/CFP2000/179_art/179_art: 7.55% (1.0600 => 1.1400)
SingleSource/Benchmarks/BenchmarkGame/puzzle: -22.35% (0.8500 => 0.6600)
SingleSource/Benchmarks/Shootout-C++/matrix: 24.08% (2.9900 => 3.7100)
SingleSource/Benchmarks/Shootout/matrix: 5.57% (4.3100 => 4.5500)
LLC-BETA:
Externals/SPEC/CFP2000/179_art/179_art: 16.25% (0.8000 => 0.9300)
MultiSource/Applications/hexxagon/hexxagon: -8.00% (9.3800 => 8.6300)
MultiSource/Benchmarks/Olden/bh/bh: -5.62% (0.8900 => 0.8400)
MultiSource/Benchmarks/Olden/em3d/em3d: 40.74% (2.9700 => 4.1800)
MultiSource/Benchmarks/Ptrdist/ks/ks: 5.18% (1.9300 => 2.0300)
MultiSource/Benchmarks/Ptrdist/yacr2/yacr2: -9.52% (1.0500 => 0.9500)
SingleSource/Benchmarks/CoyoteBench/huffbench: -6.50% (17.0700 => 15.9600)
SingleSource/Benchmarks/Misc/ffbench: -6.06% (0.9900 => 0.9300)
SingleSource/Benchmarks/Misc/richards_benchmark: -7.41% (1.0800 => 1.0000)
SingleSource/Benchmarks/Shootout-C++/hash2: 7.47% (2.8100 => 3.0200)
SingleSource/Benchmarks/Shootout/sieve: 5.15% (5.0500 => 5.3100)
LLC compile:
Externals/SPEC/CINT2000/300_twolf/300_twolf: -6.36% (2.4310 => 2.2763)
Externals/SPEC/CINT2006/471_omnetpp/471_omnetpp: -5.10% (12.7157 => 12.0678)
JIT:
Externals/SPEC/CFP2000/179_art/179_art: 14.14% (0.9900 => 1.1300)
Externals/SPEC/CINT2006/429_mcf/429_mcf: -5.45% (3.8500 => 3.6400)
MultiSource/Benchmarks/Ptrdist/anagram/anagram: -6.67% (1.2000 => 1.1200)
SingleSource/Benchmarks/McGill/chomp: -27.20% (1.2500 => 0.9100)
SingleSource/Benchmarks/Shootout/lists: 8.75% (5.6000 => 6.0900)
LLC:
Externals/SPEC/CFP2000/179_art/179_art: 19.23% (0.7800 => 0.9300)
MultiSource/Applications/aha/aha: 5.80% (2.2400 => 2.3700)
MultiSource/Applications/hexxagon/hexxagon: -8.00% (9.3800 => 8.6300)
MultiSource/Benchmarks/Olden/em3d/em3d: 40.74% (2.9700 => 4.1800)
MultiSource/Benchmarks/Olden/voronoi/voronoi: -5.26% (0.3800 => 0.3600)
MultiSource/Benchmarks/Ptrdist/ks/ks: 5.18% (1.9300 => 2.0300)
MultiSource/Benchmarks/Ptrdist/yacr2/yacr2: -9.52% (1.0500 => 0.9500)
SingleSource/Benchmarks/CoyoteBench/huffbench: -6.44% (17.0700 => 15.9700)
SingleSource/Benchmarks/Misc/ffbench: -6.06% (0.9900 => 0.9300)
SingleSource/Benchmarks/Misc/richards_benchmark: -8.26% (1.0900 => 1.0000)
SingleSource/Benchmarks/Shootout-C++/hash2: 7.47% (2.8100 => 3.0200)
SingleSource/Benchmarks/Shootout/sieve: 5.36% (5.0400 => 5.3100)
LLC-BETA compile:
Externals/SPEC/CINT2000/300_twolf/300_twolf: -6.23% (2.4438 => 2.2916)
Externals/SPEC/CINT2006/471_omnetpp/471_omnetpp: -5.20% (12.7814 => 12.1162)
More information about the llvm-testresults
mailing list