[llvm-testresults] bullet regression on Grawp-NO-LTO-PIC

[Bob Wilson]

Does anyone know what happened to bullet?

I saw pr6663 but that has supposedly been fixed on trunk.

On Mar 31, 2010, at 9:27 AM, daniel_dunbar at apple.com wrote:

> http://smooshlab.apple.com/perf/db_nt_internal/nightlytest/6887/
> Nickname: Grawp-NO-LTO-PIC:4
> Name:  grawp.apple.com
> 
> Run: 6887, Start Time: 2010-03-31 06:25:05, End Time: 2010-03-31 09:24:10
> Comparing To: 6854, Start Time: 2010-03-30 06:21:09, End Time: 2010-03-30 09:21:19
> 
> --- Changes Summary ---
> New Test Passes: 0
> New Test Failures: 4
> Added Tests: 0
> Removed Tests: 0
> Significant Changes: 9
> 
> --- Tests Summary ---
> Total Tests: 3897
> Total Test Failures: 938
> 
> Total Test Failures By Type:
>  CBE: 64
>  JIT: 433
>  JIT codegen: 433
>  LLC: 2
>  LLC compile: 2
>  LLC-BETA: 2
>  LLC_BETA compile: 2
> 
> --- Changes Detail ---
> New Test Passes:
> 
> New Test Failures:
> MultiSource/Benchmarks/Bullet/bullet [LLC, LLC-BETA, LLC compile, LLC_BETA compile]
> 
> Added Tests:
> 
> Removed Tests:
> 
> Significant Changes in Test Results:
> LLC:
> MultiSource/Applications/spiff/spiff: 7.24% (4.5600 => 4.8900)
> MultiSource/Benchmarks/Trimaran/netbench-crc/netbench-crc: 7.53% (0.9300 => 1.0000)
> LLC-BETA compile:
> Externals/SPEC/CINT2000/197_parser/197_parser: 5.16% (2.5194 => 2.6494)
> LLC-BETA:
> MultiSource/Applications/kimwitu++/kc: -6.25% (0.3200 => 0.3000)
> MultiSource/Applications/spiff/spiff: 6.35% (4.5700 => 4.8600)
> MultiSource/Benchmarks/Trimaran/enc-pc1/enc-pc1: 5.07% (1.3800 => 1.4500)
> LLC compile:
> MultiSource/Applications/lemon/lemon: 5.97% (0.8405 => 0.8907)
> CBE:
> MultiSource/Applications/spiff/spiff: 6.29% (4.7700 => 5.0700)
> MultiSource/Benchmarks/Ptrdist/ft/ft: -6.48% (1.0800 => 1.0100)
> _______________________________________________
> llvm-testresults mailing list
> llvm-testresults at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-testresults