[llvm-testresults] pollyperf-O3__clang_DEV__x86_64 test results

llvm-testresults at cs.uiuc.edu llvm-testresults at cs.uiuc.edu
Tue Feb 18 11:04:50 PST 2014


http://llvm.org/perf/db_default/v4/nts/22406?compare_to=22404&baseline=19898
Nickname: pollyperf-O3__clang_DEV__x86_64:34

Comparing:
     Run: 22406, Order: 201573, Start Time: 2014-02-18 14:27:13, End Time: 2014-02-18 19:10:05
      To: 22404, Order: 201570, Start Time: 2014-02-18 13:21:26, End Time: 2014-02-18 18:02:25
Baseline: 19898, Order: 198654, Start Time: 2014-01-07 01:21:49, End Time: 2014-01-07 06:07:29

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

Performance Regressions: 17 (62 on baseline)
Performance Improvements: 1 (23 on baseline)
Unchanged Tests: 1092 (1025 on baseline)
Total Tests: 1110

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

Performance Regressions - Compile Time
--------------------------------------
  SingleSource/UnitTests/2003-07-10-SignConversions: 400.00% (0.0040 => 0.0200, std.dev.: 0.0032)
  SingleSource/Regression/C/2003-05-22-VarSizeArray: 300.00% (0.0040 => 0.0160, std.dev.: 0.0030)
  MultiSource/Benchmarks/BitBench/uuencode/uuencode: 27.27% (0.0440 => 0.0560, std.dev.: 0.0030)
  MultiSource/Benchmarks/llubenchmark/llu: 27.27% (0.0440 => 0.0560, std.dev.: 0.0033)
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/mvt/mvt: 26.67% (0.0600 => 0.0760, std.dev.: 0.0041)
  MultiSource/Benchmarks/MiBench/automotive-basicmath/automotive-basicmath: 9.30% (0.1720 => 0.1880, std.dev.: 0.0057)
  MultiSource/Benchmarks/MiBench/security-blowfish/security-blowfish: 6.33% (0.3160 => 0.3360, std.dev.: 0.0059)
  SingleSource/Benchmarks/Shootout-C++/lists1: 4.76% (0.3360 => 0.3520, std.dev.: 0.0054)
  SingleSource/UnitTests/Vectorizer/gcc-loops: 3.05% (0.5240 => 0.5400, std.dev.: 0.0051)
  MultiSource/Applications/ALAC/encode/alacconvert-encode: 2.69% (1.1880 => 1.2200, std.dev.: 0.0121)
  MultiSource/Benchmarks/TSVC/InductionVariable-dbl/InductionVariable-dbl: 1.56% (1.2801 => 1.3001, std.dev.: 0.0075)
  MultiSource/Benchmarks/TSVC/ControlLoops-flt/ControlLoops-flt: 1.53% (1.3081 => 1.3281, std.dev.: 0.0078)

Performance Regressions - Execution Time
----------------------------------------
  MultiSource/Benchmarks/Trimaran/netbench-crc/netbench-crc: 4.97% (0.7240 => 0.7600, std.dev.: 0.0132)
  SingleSource/Benchmarks/Polybench/stencils/fdtd-2d/fdtd-2d: 4.55% (3.8722 => 4.0483, std.dev.: 0.0658)
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/syrk/syrk: 2.83% (3.9642 => 4.0763, std.dev.: 0.0362)
  MultiSource/Benchmarks/Olden/bisort/bisort: 1.63% (0.7360 => 0.7480, std.dev.: 0.0026)
  SingleSource/Benchmarks/Shootout-C++/sieve: 1.02% (2.3521 => 2.3761, std.dev.: 0.0079)

Performance Improvements - Execution Time
-----------------------------------------
  MultiSource/Benchmarks/BitBench/five11/five11: -1.03% (2.3281 => 2.3041, std.dev.: 0.0082)

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

Performance Regressions - Compile Time
--------------------------------------
  SingleSource/Benchmarks/Shootout/fib2: 300.00% (0.0040 => 0.0160, std.dev.: 0.0031)
  MultiSource/Benchmarks/McCat/01-qbsort/qbsort: 16.67% (0.0960 => 0.1120, std.dev.: 0.0056)
  MultiSource/Benchmarks/TSVC/InductionVariable-dbl/InductionVariable-dbl: 9.43% (1.1881 => 1.3001, std.dev.: 0.0075)
  MultiSource/Benchmarks/TSVC/StatementReordering-dbl/StatementReordering-dbl: 8.51% (1.1281 => 1.2241, std.dev.: 0.0091)
  MultiSource/Benchmarks/TSVC/Packing-dbl/Packing-dbl: 8.18% (1.1241 => 1.2161, std.dev.: 0.0130)
  MultiSource/Benchmarks/TSVC/LoopRestructuring-dbl/LoopRestructuring-dbl: 7.62% (1.2081 => 1.3001, std.dev.: 0.0083)
  MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt: 7.59% (1.1601 => 1.2481, std.dev.: 0.0113)
  MultiSource/Benchmarks/TSVC/ControlLoops-flt/ControlLoops-flt: 7.44% (1.2361 => 1.3281, std.dev.: 0.0078)
  MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt: 7.43% (1.1841 => 1.2721, std.dev.: 0.0104)
  MultiSource/Benchmarks/TSVC/IndirectAddressing-flt/IndirectAddressing-flt: 7.14% (1.1761 => 1.2601, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/Symbolics-dbl/Symbolics-dbl: 7.09% (1.1841 => 1.2681, std.dev.: 0.0078)
  MultiSource/Benchmarks/TSVC/Recurrences-dbl/Recurrences-dbl: 7.04% (1.1361 => 1.2161, std.dev.: 0.0065)
  MultiSource/Benchmarks/TSVC/Searching-flt/Searching-flt: 6.99% (1.1441 => 1.2241, std.dev.: 0.0059)
  MultiSource/Benchmarks/TSVC/LinearDependence-dbl/LinearDependence-dbl: 6.92% (1.2721 => 1.3601, std.dev.: 0.0075)
  MultiSource/Benchmarks/TSVC/InductionVariable-flt/InductionVariable-flt: 6.91% (1.2161 => 1.3001, std.dev.: 0.0104)
  MultiSource/Benchmarks/TSVC/LoopRestructuring-flt/LoopRestructuring-flt: 6.88% (1.2201 => 1.3041, std.dev.: 0.0104)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl: 6.71% (1.1921 => 1.2721, std.dev.: 0.0102)
  MultiSource/Benchmarks/TSVC/Equivalencing-dbl/Equivalencing-dbl: 6.71% (1.1921 => 1.2721, std.dev.: 0.0061)
  MultiSource/Benchmarks/TSVC/Recurrences-flt/Recurrences-flt: 6.62% (1.1481 => 1.2241, std.dev.: 0.0076)
  MultiSource/Benchmarks/TSVC/ControlLoops-dbl/ControlLoops-dbl: 6.45% (1.2401 => 1.3201, std.dev.: 0.0074)
  MultiSource/Benchmarks/TSVC/Searching-dbl/Searching-dbl: 6.34% (1.1361 => 1.2081, std.dev.: 0.0111)
  MultiSource/Benchmarks/TSVC/StatementReordering-flt/StatementReordering-flt: 6.27% (1.1481 => 1.2201, std.dev.: 0.0109)
  MultiSource/Benchmarks/TSVC/Packing-flt/Packing-flt: 6.23% (1.1561 => 1.2281, std.dev.: 0.0082)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-flt/GlobalDataFlow-flt: 6.19% (1.2281 => 1.3041, std.dev.: 0.0103)
  MultiSource/Benchmarks/TSVC/LoopRerolling-dbl/LoopRerolling-dbl: 6.16% (1.1681 => 1.2401, std.dev.: 0.0112)
  MultiSource/Benchmarks/TSVC/Reductions-dbl/Reductions-dbl: 6.03% (1.2601 => 1.3361, std.dev.: 0.0098)
  MultiSource/Benchmarks/TSVC/Symbolics-flt/Symbolics-flt: 6.02% (1.1961 => 1.2681, std.dev.: 0.0093)
  MultiSource/Benchmarks/TSVC/IndirectAddressing-dbl/IndirectAddressing-dbl: 5.72% (1.1881 => 1.2561, std.dev.: 0.0084)
  MultiSource/Benchmarks/TSVC/NodeSplitting-flt/NodeSplitting-flt: 5.70% (1.1921 => 1.2601, std.dev.: 0.0090)
  MultiSource/Benchmarks/TSVC/Reductions-flt/Reductions-flt: 5.66% (1.2721 => 1.3441, std.dev.: 0.0113)
  MultiSource/Benchmarks/TSVC/ControlFlow-flt/ControlFlow-flt: 5.57% (1.3641 => 1.4401, std.dev.: 0.0080)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-dbl/GlobalDataFlow-dbl: 5.52% (1.2321 => 1.3001, std.dev.: 0.0082)
  MultiSource/Benchmarks/TSVC/Expansion-dbl/Expansion-dbl: 5.43% (1.2521 => 1.3201, std.dev.: 0.0070)
  MultiSource/Benchmarks/TSVC/NodeSplitting-dbl/NodeSplitting-dbl: 5.39% (1.1881 => 1.2521, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/LinearDependence-flt/LinearDependence-flt: 5.31% (1.2801 => 1.3481, std.dev.: 0.0092)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-flt/CrossingThresholds-flt: 5.28% (1.2121 => 1.2761, std.dev.: 0.0080)
  MultiSource/Benchmarks/TSVC/ControlFlow-dbl/ControlFlow-dbl: 5.26% (1.3681 => 1.4401, std.dev.: 0.0111)
  MultiSource/Benchmarks/TSVC/Expansion-flt/Expansion-flt: 5.13% (1.2481 => 1.3121, std.dev.: 0.0088)
  MultiSource/Benchmarks/MiBench/security-blowfish/security-blowfish: 5.00% (0.3200 => 0.3360, std.dev.: 0.0059)
  SingleSource/Benchmarks/Shootout-C++/reversefile: 3.06% (0.3920 => 0.4040, std.dev.: 0.0039)
  MultiSource/Benchmarks/MiBench/automotive-susan/automotive-susan: 2.62% (0.9161 => 0.9401, std.dev.: 0.0054)
  MultiSource/Applications/lua/lua: 1.77% (4.5200 => 4.6000, std.dev.: 0.0260)
  MultiSource/Applications/sqlite3/sqlite3: 1.58% (23.7574 => 24.1335, std.dev.: 0.0545)
  MultiSource/Benchmarks/mafft/pairlocalalign: 1.46% (10.4082 => 10.5602, std.dev.: 0.0204)
  MultiSource/Benchmarks/PAQ8p/paq8p: 1.08% (2.5882 => 2.6162, std.dev.: 0.0100)
  MultiSource/Applications/JM/lencod/lencod: 1.02% (17.6165 => 17.7965, std.dev.: 0.0346)

Performance Regressions - Execution Time
----------------------------------------
  MultiSource/Benchmarks/MiBench/security-rijndael/security-rijndael: 50.00% (0.0240 => 0.0360, std.dev.: 0.0027)
  MultiSource/Benchmarks/MiBench/network-patricia/network-patricia: 12.50% (0.0960 => 0.1080, std.dev.: 0.0039)
  SingleSource/Benchmarks/Shootout-C++/ackermann: 12.50% (1.0881 => 1.2241, std.dev.: 0.0035)
  MultiSource/Applications/siod/siod: 8.05% (2.8802 => 3.1122, std.dev.: 0.0088)
  MultiSource/Benchmarks/FreeBench/pcompress2/pcompress2: 7.50% (0.1600 => 0.1720, std.dev.: 0.0035)
  MultiSource/Benchmarks/Ptrdist/bc/bc: 3.33% (0.6000 => 0.6200, std.dev.: 0.0057)
  MultiSource/Benchmarks/Fhourstones/fhourstones: 2.93% (1.0921 => 1.1241, std.dev.: 0.0105)
  SingleSource/UnitTests/Vectorizer/gcc-loops: 2.89% (4.5603 => 4.6923, std.dev.: 0.0069)
  SingleSource/Benchmarks/Dhrystone/fldry: 2.63% (0.6080 => 0.6240, std.dev.: 0.0044)
  SingleSource/Benchmarks/Shootout/lists: 2.54% (5.8164 => 5.9644, std.dev.: 0.0149)
  SingleSource/Benchmarks/BenchmarkGame/fannkuch: 2.11% (3.2242 => 3.2922, std.dev.: 0.0024)
  MultiSource/Applications/lambda-0_1_3/lambda: 2.07% (5.0123 => 5.1163, std.dev.: 0.0296)
  MultiSource/Benchmarks/TSVC/ControlFlow-dbl/ControlFlow-dbl: 2.04% (4.5203 => 4.6123, std.dev.: 0.0065)
  SingleSource/Benchmarks/Polybench/linear-algebra/solvers/dynprog/dynprog: 1.90% (0.6320 => 0.6440, std.dev.: 0.0018)
  SingleSource/Benchmarks/SmallPT/smallpt: 1.14% (14.0009 => 14.1609, std.dev.: 0.0607)
  MultiSource/Applications/SPASS/SPASS: 1.05% (9.1686 => 9.2646, std.dev.: 0.0270)

Performance Improvements - Compile Time
---------------------------------------
  SingleSource/UnitTests/SignlessTypes/cast-bug: -100.00% (0.0120 => 0.0000, std.dev.: 0.0045)
  SingleSource/UnitTests/blockstret: -100.00% (0.0120 => 0.0000, std.dev.: 0.0040)
  SingleSource/Benchmarks/Misc-C++/mandel-text: -42.86% (0.0280 => 0.0160, std.dev.: 0.0045)
  SingleSource/Benchmarks/Misc/perlin: -22.22% (0.0720 => 0.0560, std.dev.: 0.0060)
  SingleSource/Benchmarks/Shootout-C++/hello: -7.27% (0.2200 => 0.2040, std.dev.: 0.0059)
  MultiSource/Benchmarks/Ptrdist/yacr2/yacr2: -5.11% (0.7040 => 0.6680, std.dev.: 0.0131)
  SingleSource/Benchmarks/Misc-C++/stepanov_container: -2.93% (0.9561 => 0.9281, std.dev.: 0.0108)
  MultiSource/Applications/d/make_dparser: -1.57% (3.8240 => 3.7641, std.dev.: 0.0224)
  MultiSource/Applications/ClamAV/clamscan: -1.40% (17.4082 => 17.1642, std.dev.: 0.0406)
  MultiSource/Benchmarks/MiBench/consumer-typeset/consumer-typeset: -1.34% (15.2282 => 15.0242, std.dev.: 0.0367)

Performance Improvements - Execution Time
-----------------------------------------
  MultiSource/Benchmarks/FreeBench/neural/neural: -16.90% (0.2840 => 0.2360, std.dev.: 0.0033)
  SingleSource/Benchmarks/Linpack/linpack-pc: -12.70% (7.7485 => 6.7644, std.dev.: 0.0041)
  MultiSource/Benchmarks/FreeBench/analyzer/analyzer: -11.11% (0.1080 => 0.0960, std.dev.: 0.0027)
  MultiSource/Benchmarks/Trimaran/netbench-url/netbench-url: -6.81% (3.3482 => 3.1202, std.dev.: 0.0037)
  MultiSource/Benchmarks/TSVC/LinearDependence-flt/LinearDependence-flt: -5.01% (4.4683 => 4.2443, std.dev.: 0.0066)
  MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt: -4.67% (2.0561 => 1.9601, std.dev.: 0.0032)
  MultiSource/Benchmarks/TSVC/Symbolics-flt/Symbolics-flt: -3.68% (2.3881 => 2.3001, std.dev.: 0.0032)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl: -2.45% (3.9202 => 3.8242, std.dev.: 0.0349)
  MultiSource/Benchmarks/TSVC/InductionVariable-flt/InductionVariable-flt: -2.17% (4.2403 => 4.1483, std.dev.: 0.0062)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-flt/CrossingThresholds-flt: -2.02% (2.9722 => 2.9122, std.dev.: 0.0045)
  MultiSource/Benchmarks/TSVC/Symbolics-dbl/Symbolics-dbl: -1.79% (4.2563 => 4.1803, std.dev.: 0.0040)
  SingleSource/Benchmarks/Adobe-C++/functionobjects: -1.72% (3.4842 => 3.4242, std.dev.: 0.0033)
  MultiSource/Benchmarks/BitBench/five11/five11: -1.37% (2.3361 => 2.3041, std.dev.: 0.0082)

Report Time: 1.97s
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-testresults/attachments/20140218/0baf87e4/attachment.html>


More information about the llvm-testresults mailing list