[llvm-testresults] chinook-clangslave1__O3-plain__clang_DEV__powerpc64 test results

llvm-testresults at cs.uiuc.edu llvm-testresults at cs.uiuc.edu
Thu Dec 26 11:10:56 PST 2013


http://llvm.org/perf/db_default/v4/nts/19069?compare_to=19060&baseline=17660
Nickname: chinook-clangslave1__O3-plain__clang_DEV__powerpc64:21

Comparing:
     Run: 19069, Order: 198052, Start Time: 2013-12-26 17:31:18, End Time: 2013-12-26 19:15:28
      To: 19060, Order: 198045, Start Time: 2013-12-26 15:21:10, End Time: 2013-12-26 17:07:41
Baseline: 17660, Order: 195884, Start Time: 2013-11-27 22:43:57, End Time: 2013-11-28 00:31:03

===============
Tests Summary
===============

Performance Regressions: 10 (17 on baseline)
Performance Improvements: 15 (16 on baseline)
Unchanged Tests: 1085 (1077 on baseline)
Total Tests: 1110

===========================
Run-Over-Run Changes Detail
===========================

Performance Regressions - Compile Time
--------------------------------------
  MultiSource/Benchmarks/MiBench/office-ispell/office-ispell: 1.59% (1.9034 => 1.9337, std.dev.: 0.0086)

Performance Regressions - Execution Time
----------------------------------------
  MultiSource/Benchmarks/TSVC/Searching-dbl/Searching-dbl: 7.75% (6.3447 => 6.8364, std.dev.: 0.0593)
  MultiSource/Applications/spiff/spiff: 3.75% (2.8170 => 2.9225, std.dev.: 0.0343)
  MultiSource/Benchmarks/Trimaran/enc-rc4/enc-rc4: 3.46% (1.8052 => 1.8677, std.dev.: 0.0153)
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/gemm/gemm: 3.11% (21.0710 => 21.7271, std.dev.: 0.0966)
  MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt: 2.60% (6.5258 => 6.6953, std.dev.: 0.0388)
  MultiSource/Benchmarks/Prolangs-C++/life/life: 2.42% (2.4869 => 2.5471, std.dev.: 0.0062)
  MultiSource/Benchmarks/tramp3d-v4/tramp3d-v4: 2.31% (0.5148 => 0.5267, std.dev.: 0.0007)
  MultiSource/Applications/sgefa/sgefa: 1.81% (0.8930 => 0.9092, std.dev.: 0.0047)
  SingleSource/UnitTests/Vectorizer/gcc-loops: 1.13% (18.7543 => 18.9657, std.dev.: 0.0243)

Performance Improvements - Compile Time
---------------------------------------
  SingleSource/Benchmarks/Stanford/Queens: -38.40% (0.0711 => 0.0438, std.dev.: 0.0052)
  MultiSource/Benchmarks/Prolangs-C/allroots/allroots: -33.95% (0.1405 => 0.0928, std.dev.: 0.0037)
  SingleSource/Benchmarks/Stanford/Oscar: -28.36% (0.0952 => 0.0682, std.dev.: 0.0046)
  MultiSource/Applications/ALAC/encode/alacconvert-encode: -16.83% (1.4163 => 1.1780, std.dev.: 0.0715)
  MultiSource/Benchmarks/Prolangs-C/archie-client/archie: -15.79% (0.8678 => 0.7308, std.dev.: 0.0426)
  MultiSource/Applications/lua/lua: -12.68% (4.8636 => 4.2469, std.dev.: 0.1806)
  SingleSource/Benchmarks/Shootout-C++/objinst: -6.74% (0.2241 => 0.2090, std.dev.: 0.0053)
  SingleSource/Benchmarks/Adobe-C++/loop_unroll: -6.24% (4.5558 => 4.2716, std.dev.: 0.1062)
  SingleSource/Benchmarks/Adobe-C++/functionobjects: -4.78% (0.5455 => 0.5194, std.dev.: 0.0088)
  SingleSource/Benchmarks/Shootout-C++/wordfreq: -4.41% (0.3582 => 0.3424, std.dev.: 0.0030)
  SingleSource/Benchmarks/Adobe-C++/stepanov_abstraction: -1.94% (0.6598 => 0.6470, std.dev.: 0.0038)

Performance Improvements - Execution Time
-----------------------------------------
  SingleSource/Benchmarks/Stanford/Queens: -35.57% (0.0388 => 0.0250, std.dev.: 0.0050)
  SingleSource/Benchmarks/Shootout/ary3: -19.12% (5.8297 => 4.7151, std.dev.: 0.3565)
  MultiSource/Applications/JM/lencod/lencod: -4.07% (8.1643 => 7.8323, std.dev.: 0.0836)
  MultiSource/Benchmarks/Olden/tsp/tsp: -3.59% (1.6007 => 1.5432, std.dev.: 0.0178)

================================
Run-Over-Baseline Changes Detail
================================

Performance Regressions - Compile Time
--------------------------------------
  MultiSource/Applications/sqlite3/sqlite3: 40.26% (15.7419 => 22.0801, std.dev.: 0.9289)
  MultiSource/Benchmarks/tramp3d-v4/tramp3d-v4: 6.00% (25.9661 => 27.5231, std.dev.: 0.1679)
  SingleSource/Benchmarks/Adobe-C++/stepanov_abstraction: 2.34% (0.6322 => 0.6470, std.dev.: 0.0038)

Performance Regressions - Execution Time
----------------------------------------
  SingleSource/Benchmarks/Misc-C++-EH/spirit: 16.68% (5.9583 => 6.9520, std.dev.: 0.1249)
  MultiSource/Benchmarks/MiBench/automotive-bitcount/automotive-bitcount: 9.22% (0.1128 => 0.1232, std.dev.: 0.0001)
  SingleSource/Benchmarks/Misc-C++/Large/sphereflake: 8.25% (3.3810 => 3.6599, std.dev.: 0.0495)
  MultiSource/Applications/lua/lua: 4.96% (29.1357 => 30.5802, std.dev.: 0.2852)
  MultiSource/Applications/siod/siod: 4.94% (3.4258 => 3.5952, std.dev.: 0.0108)
  MultiSource/Applications/sgefa/sgefa: 3.97% (0.8745 => 0.9092, std.dev.: 0.0047)
  MultiSource/Applications/spiff/spiff: 3.89% (2.8130 => 2.9225, std.dev.: 0.0343)
  MultiSource/Benchmarks/tramp3d-v4/tramp3d-v4: 3.03% (0.5112 => 0.5267, std.dev.: 0.0007)
  MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt: 2.55% (6.5291 => 6.6953, std.dev.: 0.0388)
  MultiSource/Benchmarks/TSVC/ControlFlow-dbl/ControlFlow-dbl: 2.01% (5.8933 => 6.0118, std.dev.: 0.0045)
  MultiSource/Benchmarks/TSVC/Symbolics-flt/Symbolics-flt: 2.00% (5.9126 => 6.0309, std.dev.: 0.0379)
  SingleSource/UnitTests/Vectorizer/gcc-loops: 1.86% (18.6185 => 18.9657, std.dev.: 0.0243)
  MultiSource/Benchmarks/Prolangs-C++/life/life: 1.78% (2.5025 => 2.5471, std.dev.: 0.0062)
  MultiSource/Benchmarks/TSVC/Reductions-flt/Reductions-flt: 1.61% (11.7894 => 11.9788, std.dev.: 0.0696)

Performance Improvements - Compile Time
---------------------------------------
  MultiSource/Benchmarks/MiBench/office-stringsearch/office-stringsearch: -13.00% (0.1938 => 0.1686, std.dev.: 0.0042)
  MultiSource/Benchmarks/Bullet/bullet: -9.66% (36.6274 => 33.0878, std.dev.: 0.6119)
  MultiSource/Benchmarks/MiBench/security-blowfish/security-blowfish: -6.88% (0.3402 => 0.3168, std.dev.: 0.0054)
  MultiSource/Benchmarks/MiBench/automotive-bitcount/automotive-bitcount: -6.51% (0.2810 => 0.2627, std.dev.: 0.0025)
  MultiSource/Benchmarks/TSVC/NodeSplitting-flt/NodeSplitting-flt: -5.68% (1.3676 => 1.2899, std.dev.: 0.0203)
  MultiSource/Benchmarks/MiBench/automotive-susan/automotive-susan: -4.63% (1.1409 => 1.0881, std.dev.: 0.0020)
  MultiSource/Benchmarks/TSVC/LoopRestructuring-dbl/LoopRestructuring-dbl: -4.42% (1.3819 => 1.3208, std.dev.: 0.0058)
  MultiSource/Benchmarks/MiBench/office-ispell/office-ispell: -4.31% (2.0207 => 1.9337, std.dev.: 0.0086)
  MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt: -4.14% (1.3505 => 1.2946, std.dev.: 0.0191)
  MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000: -2.92% (7.5124 => 7.2934, std.dev.: 0.0489)

Performance Improvements - Execution Time
-----------------------------------------
  MultiSource/Benchmarks/nbench/nbench: -10.05% (23.4696 => 21.1120, std.dev.: 0.6288)
  MultiSource/Benchmarks/Bullet/bullet: -6.10% (7.4226 => 6.9700, std.dev.: 0.0871)
  MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000: -2.91% (2.5559 => 2.4815, std.dev.: 0.0242)
  MultiSource/Applications/sqlite3/sqlite3: -2.05% (4.2978 => 4.2097, std.dev.: 0.0112)
  MultiSource/Applications/lambda-0_1_3/lambda: -1.89% (7.9597 => 7.8094, std.dev.: 0.0164)
  SingleSource/Benchmarks/Shootout/methcall: -1.03% (8.6669 => 8.5777, std.dev.: 0.0287)

Report Time: 0.76s
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-testresults/attachments/20131226/9b3efefa/attachment.html>


More information about the llvm-testresults mailing list