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

[Evan Cheng]

I am going to assume Dale's 82778 has fixed the performance  
regression. Dale, can you verify?

Thanks,

Evan

On Sep 25, 2009, at 7:38 PM, Evan Cheng wrote:

> All of these tests use sqrt. Dale?
>
> Evan
>
> On Sep 25, 2009, at 4:28 PM, Daniel Dunbar wrote:
>
>> Does anyone know what this change was? The performance changes seem
>> significant, and non-trivial.
>>
>> It might be worth looking into.
>>
>> - Daniel
>>
>> On Wed, Sep 23, 2009 at 2:58 PM,  <daniel_dunbar at apple.com> wrote:
>>> http://smooshlab.apple.com/perf.cgi/db_nt_internal/nightlytest//1958
>>> Name:  smoosh-01
>>> Nickname: smoosh-01.apple.com:2
>>> Buildstatus: OK
>>>
>>> Run: 1958, Start Time: 2009-09-23 13:41:08
>>> Comparing To: 1957, Start Time: 2009-09-23 10:22:29
>>>
>>> --- Changes Summary ---
>>> New Test Passes: 0
>>> New Test Failures: 0
>>> Added Tests: 0
>>> Removed Tests: 0
>>> Significant Changes: 17
>>>
>>> --- Tests Summary ---
>>> Total Tests: 3537
>>> Total Test Failures: 91
>>>
>>> Total Test Failures By Type:
>>> CBE: 35
>>> JIT: 24
>>> JIT codegen: 24
>>> LLC: 1
>>> LLC compile: 1
>>> LLC-BETA: 3
>>> LLC_BETA compile: 3
>>>
>>> --- Changes Detail ---
>>> New Test Passes:
>>>
>>> New Test Failures:
>>>
>>> Added Tests:
>>>
>>> Removed Tests:
>>>
>>> Significant Changes in Test Results:
>>> LLC:
>>> MultiSource/Applications/sgefa/sgefa: 6.74% (0.8900 => 0.9500)
>>> MultiSource/Benchmarks/Olden/bh/bh: 8.44% (1.5400 => 1.6700)
>>> MultiSource/Benchmarks/VersaBench/beamformer/beamformer: 9.52%
>>> (1.2600 => 1.3800)
>>> SingleSource/Benchmarks/BenchmarkGame/n-body: 12.59% (1.3500 =>
>>> 1.5200)
>>> SingleSource/Benchmarks/Misc-C++/ray: 22.74% (4.0900 => 5.0200)
>>> SingleSource/Benchmarks/Misc-C++/sphereflake: 6.23% (2.8900 =>
>>> 3.0700)
>>> Bitcode:
>>> SingleSource/Benchmarks/BenchmarkGame/n-body: -14.54% (3632.0000
>>> => 3104.0000)
>>> JIT:
>>> MultiSource/Applications/sgefa/sgefa: 9.09% (0.9900 => 1.0800)
>>> MultiSource/Benchmarks/Olden/bh/bh: 8.02% (1.6200 => 1.7500)
>>> MultiSource/Benchmarks/VersaBench/beamformer/beamformer: 9.03%
>>> (1.4400 => 1.5700)
>>> SingleSource/Benchmarks/BenchmarkGame/n-body: 7.19% (1.3900 =>
>>> 1.4900)
>>> SingleSource/Benchmarks/Misc-C++/ray: 26.83% (4.1000 => 5.2000)
>>> SingleSource/Benchmarks/Misc-C++/sphereflake: 5.03% (2.9800 =>
>>> 3.1300)
>>> LLC-BETA:
>>> MultiSource/Benchmarks/Olden/bh/bh: 11.26% (1.5100 => 1.6800)
>>> MultiSource/Benchmarks/VersaBench/beamformer/beamformer: 8.66%
>>> (1.2700 => 1.3800)
>>> SingleSource/Benchmarks/BenchmarkGame/n-body: 12.59% (1.3500 =>
>>> 1.5200)
>>> SingleSource/Benchmarks/Misc-C++/sphereflake: 5.88% (2.8900 =>
>>> 3.0600)
>>> _______________________________________________
>>> llvm-testresults mailing list
>>> llvm-testresults at cs.uiuc.edu
>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-testresults
>>>
>>
>> _______________________________________________
>> llvm-testresults mailing list
>> llvm-testresults at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-testresults
>
> _______________________________________________
> llvm-testresults mailing list
> llvm-testresults at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-testresults