[llvm-testresults] smoosh-01.apple.com nightly tester results

Eric Christopher echristo at apple.com
Sat Mar 6 01:05:49 PST 2010


On Mar 6, 2010, at 12:45 AM, daniel_dunbar at apple.com wrote:

> Significant Changes in Test Results:
> JIT codegen:
> MultiSource/Applications/hbd/hbd: 5.80% (0.4192 => 0.4435)
> MultiSource/Applications/lambda-0_1_3/lambda: 6.69% (0.3319 => 0.3541)
> CBE:
> MultiSource/Applications/ClamAV/clamscan: 8.33% (0.2400 => 0.2600)
> MultiSource/Applications/lemon/lemon: 7.82% (1.7900 => 1.9300)
> MultiSource/Benchmarks/OptimizerEval/optimizer-eval: -26.27% (39.9000 => 29.4200)
> LLC-BETA:
> MultiSource/Applications/lemon/lemon: 6.29% (1.7500 => 1.8600)
> LLC compile:
> MultiSource/Applications/hbd/hbd: 5.40% (0.5294 => 0.5580)
> MultiSource/Applications/lambda-0_1_3/lambda: 6.76% (0.4707 => 0.5025)
> GCCAS:
> MultiSource/Applications/hbd/hbd: 5.11% (0.2838 => 0.2983)
> JIT:
> MultiSource/Applications/hbd/hbd: 6.67% (0.4500 => 0.4800)
> LLC:
> MultiSource/Applications/ClamAV/clamscan: 8.33% (0.2400 => 0.2600)
> LLC-BETA compile:
> MultiSource/Applications/SIBsim4/SIBsim4: 5.90% (0.7964 => 0.8434)
> MultiSource/Applications/hbd/hbd: 7.74% (0.5337 => 0.5750)
> MultiSource/Applications/lambda-0_1_3/lambda: 6.40% (0.4735 => 0.5038)
> MultiSource/Applications/lemon/lemon: 7.93% (0.6899 => 0.7446)

This looks significant enough that someone should probably take a look...

-eric



More information about the llvm-testresults mailing list