[llvm-testresults] pollyperf-O3-polly-detect__clang_DEV__x86_64 test results

llvm-testresults at cs.uiuc.edu llvm-testresults at cs.uiuc.edu
Sat Feb 1 00:05:30 PST 2014


http://llvm.org/perf/db_default/v4/nts/21619?compare_to=21610&baseline=19899
Nickname: pollyperf-O3-polly-detect__clang_DEV__x86_64:35

Comparing:
     Run: 21619, Order: 200605, Start Time: 2014-02-01 03:24:55, End Time: 2014-02-01 08:10:37
      To: 21610, Order: 200580, Start Time: 2014-01-31 22:21:45, End Time: 2014-02-01 03:07:08
Baseline: 19899, Order: 198654, Start Time: 2014-01-07 01:22:04, End Time: 2014-01-07 06:18:02

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

Performance Regressions: 15 (58 on baseline)
Performance Improvements: 1 (31 on baseline)
Unchanged Tests: 1094 (1021 on baseline)
Total Tests: 1110

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

Performance Regressions - Compile Time
--------------------------------------
  SingleSource/Regression/C++/global_ctor: 100.00% (0.0120 => 0.0240, std.dev.: 0.0045)
  MultiSource/Benchmarks/Prolangs-C++/vcirc/vcirc: 60.00% (0.0200 => 0.0320, std.dev.: 0.0040)
  SingleSource/Benchmarks/Stanford/FloatMM: 50.00% (0.0320 => 0.0480, std.dev.: 0.0032)
  SingleSource/Benchmarks/Misc/flops-3: 50.00% (0.0240 => 0.0360, std.dev.: 0.0036)
  SingleSource/Regression/C/sumarray: 50.00% (0.0240 => 0.0360, std.dev.: 0.0045)
  MultiSource/Benchmarks/BitBench/uudecode/uudecode: 40.00% (0.0400 => 0.0560, std.dev.: 0.0061)
  MultiSource/Benchmarks/mediabench/adpcm/rawdaudio/rawdaudio: 25.00% (0.0480 => 0.0600, std.dev.: 0.0032)
  SingleSource/Benchmarks/Polybench/medley/floyd-warshall/floyd-warshall: 17.65% (0.0680 => 0.0800, std.dev.: 0.0044)
  MultiSource/Benchmarks/FreeBench/fourinarow/fourinarow: 4.29% (0.2800 => 0.2920, std.dev.: 0.0037)
  MultiSource/Benchmarks/Prolangs-C/bison/mybison: 2.34% (1.7120 => 1.7520, std.dev.: 0.0136)
  MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt: 2.26% (1.2401 => 1.2681, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl: 1.26% (1.2721 => 1.2881, std.dev.: 0.0059)

Performance Regressions - Execution Time
----------------------------------------
  MultiSource/Benchmarks/Olden/treeadd/treeadd: 10.71% (0.2240 => 0.2480, std.dev.: 0.0089)
  MultiSource/Benchmarks/MiBench/consumer-typeset/consumer-typeset: 9.09% (0.1760 => 0.1920, std.dev.: 0.0054)
  SingleSource/Benchmarks/BenchmarkGame/nsieve-bits: 2.53% (0.9481 => 0.9721, std.dev.: 0.0065)

Performance Improvements - Compile Time
---------------------------------------
  SingleSource/UnitTests/2002-08-02-CastTest: -60.00% (0.0200 => 0.0080, std.dev.: 0.0045)

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

Performance Regressions - Compile Time
--------------------------------------
  SingleSource/Benchmarks/McGill/chomp: 16.67% (0.1920 => 0.2240, std.dev.: 0.0054)
  SingleSource/Benchmarks/SmallPT/smallpt: 11.76% (0.1360 => 0.1520, std.dev.: 0.0040)
  MultiSource/Benchmarks/Olden/power/power: 10.53% (0.1520 => 0.1680, std.dev.: 0.0048)
  MultiSource/Benchmarks/TSVC/LoopRerolling-dbl/LoopRerolling-dbl: 7.27% (1.1561 => 1.2401, std.dev.: 0.0077)
  MultiSource/Benchmarks/TSVC/NodeSplitting-flt/NodeSplitting-flt: 7.14% (1.1761 => 1.2601, std.dev.: 0.0128)
  MultiSource/Benchmarks/TSVC/Symbolics-flt/Symbolics-flt: 7.14% (1.1761 => 1.2601, std.dev.: 0.0090)
  MultiSource/Benchmarks/TSVC/IndirectAddressing-dbl/IndirectAddressing-dbl: 6.48% (1.1721 => 1.2481, std.dev.: 0.0110)
  MultiSource/Benchmarks/TSVC/Packing-dbl/Packing-dbl: 6.36% (1.1321 => 1.2041, std.dev.: 0.0125)
  MultiSource/Benchmarks/TSVC/Searching-flt/Searching-flt: 6.34% (1.1361 => 1.2081, std.dev.: 0.0097)
  MultiSource/Benchmarks/TSVC/Recurrences-flt/Recurrences-flt: 6.25% (1.1521 => 1.2241, std.dev.: 0.0088)
  MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt: 6.14% (1.1721 => 1.2441, std.dev.: 0.0113)
  MultiSource/Benchmarks/TSVC/IndirectAddressing-flt/IndirectAddressing-flt: 6.06% (1.1881 => 1.2601, std.dev.: 0.0088)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-flt/CrossingThresholds-flt: 6.00% (1.2001 => 1.2721, std.dev.: 0.0103)
  MultiSource/Benchmarks/TSVC/StatementReordering-dbl/StatementReordering-dbl: 5.96% (1.1401 => 1.2081, std.dev.: 0.0120)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl: 5.92% (1.2161 => 1.2881, std.dev.: 0.0059)
  MultiSource/Benchmarks/TSVC/Symbolics-dbl/Symbolics-dbl: 5.78% (1.1761 => 1.2441, std.dev.: 0.0136)
  MultiSource/Benchmarks/TSVC/Expansion-dbl/Expansion-dbl: 5.77% (1.2481 => 1.3201, std.dev.: 0.0069)
  MultiSource/Benchmarks/TSVC/Expansion-flt/Expansion-flt: 5.77% (1.2481 => 1.3201, std.dev.: 0.0102)
  MultiSource/Benchmarks/TSVC/ControlLoops-flt/ControlLoops-flt: 5.75% (1.2521 => 1.3241, std.dev.: 0.0084)
  MultiSource/Benchmarks/TSVC/NodeSplitting-dbl/NodeSplitting-dbl: 5.74% (1.1841 => 1.2521, std.dev.: 0.0066)
  MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt: 5.67% (1.2001 => 1.2681, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/Searching-dbl/Searching-dbl: 5.63% (1.1361 => 1.2001, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/Packing-flt/Packing-flt: 5.59% (1.1441 => 1.2081, std.dev.: 0.0092)
  MultiSource/Benchmarks/TSVC/LoopRestructuring-dbl/LoopRestructuring-dbl: 5.52% (1.2321 => 1.3001, std.dev.: 0.0088)
  MultiSource/Benchmarks/TSVC/Recurrences-dbl/Recurrences-dbl: 5.24% (1.1441 => 1.2041, std.dev.: 0.0117)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-dbl/GlobalDataFlow-dbl: 5.19% (1.2321 => 1.2961, std.dev.: 0.0120)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-flt/GlobalDataFlow-flt: 5.19% (1.2321 => 1.2961, std.dev.: 0.0106)
  MultiSource/Benchmarks/TSVC/StatementReordering-flt/StatementReordering-flt: 5.17% (1.1601 => 1.2201, std.dev.: 0.0091)
  MultiSource/Benchmarks/TSVC/LoopRestructuring-flt/LoopRestructuring-flt: 5.16% (1.2401 => 1.3041, std.dev.: 0.0078)
  MultiSource/Benchmarks/TSVC/Equivalencing-dbl/Equivalencing-dbl: 5.07% (1.1841 => 1.2441, std.dev.: 0.0124)
  MultiSource/Benchmarks/TSVC/LinearDependence-flt/LinearDependence-flt: 4.95% (1.2921 => 1.3561, std.dev.: 0.0114)
  MultiSource/Benchmarks/TSVC/InductionVariable-flt/InductionVariable-flt: 4.87% (1.2321 => 1.2921, std.dev.: 0.0129)
  MultiSource/Benchmarks/TSVC/ControlLoops-dbl/ControlLoops-dbl: 4.79% (1.2521 => 1.3121, std.dev.: 0.0080)
  MultiSource/Benchmarks/TSVC/ControlFlow-flt/ControlFlow-flt: 4.68% (1.3681 => 1.4321, std.dev.: 0.0113)
  MultiSource/Benchmarks/TSVC/InductionVariable-dbl/InductionVariable-dbl: 4.56% (1.2281 => 1.2841, std.dev.: 0.0088)
  MultiSource/Benchmarks/TSVC/LinearDependence-dbl/LinearDependence-dbl: 4.31% (1.3001 => 1.3561, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/Reductions-dbl/Reductions-dbl: 4.05% (1.2841 => 1.3361, std.dev.: 0.0100)
  MultiSource/Benchmarks/TSVC/Reductions-flt/Reductions-flt: 4.04% (1.2881 => 1.3401, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/ControlFlow-dbl/ControlFlow-dbl: 3.47% (1.3841 => 1.4321, std.dev.: 0.0116)
  MultiSource/Applications/sqlite3/sqlite3: 1.50% (24.7575 => 25.1295, std.dev.: 0.0543)

Performance Regressions - Execution Time
----------------------------------------
  SingleSource/Benchmarks/Shootout/sieve: 157.80% (6.6644 => 17.1811, std.dev.: 0.0027)
  MultiSource/Benchmarks/MiBench/telecomm-CRC32/telecomm-CRC32: 6.78% (0.2360 => 0.2520, std.dev.: 0.0060)
  MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000: 6.03% (3.9842 => 4.2243, std.dev.: 0.0199)
  MultiSource/Benchmarks/Ptrdist/yacr2/yacr2: 5.80% (0.8281 => 0.8761, std.dev.: 0.0027)
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/doitgen/doitgen: 4.83% (3.3122 => 3.4722, std.dev.: 0.0558)
  MultiSource/Benchmarks/Ptrdist/bc/bc: 4.73% (0.5920 => 0.6200, std.dev.: 0.0102)
  MultiSource/Applications/lua/lua: 3.74% (21.7974 => 22.6134, std.dev.: 0.0649)
  SingleSource/Benchmarks/Dhrystone/fldry: 3.33% (0.6000 => 0.6200, std.dev.: 0.0032)
  MultiSource/Benchmarks/MallocBench/cfrac/cfrac: 3.25% (1.3521 => 1.3961, std.dev.: 0.0026)
  MultiSource/Applications/sqlite3/sqlite3: 2.66% (4.0563 => 4.1643, std.dev.: 0.0090)
  SingleSource/Benchmarks/BenchmarkGame/fannkuch: 2.60% (3.2362 => 3.3202, std.dev.: 0.0022)
  SingleSource/Benchmarks/BenchmarkGame/Large/fasta: 2.43% (0.9881 => 1.0121, std.dev.: 0.0070)
  MultiSource/Benchmarks/Olden/power/power: 2.06% (1.3601 => 1.3881, std.dev.: 0.0057)
  MultiSource/Applications/JM/lencod/lencod: 1.82% (8.1245 => 8.2725, std.dev.: 0.0188)
  SingleSource/Benchmarks/McGill/chomp: 1.73% (0.9241 => 0.9401, std.dev.: 0.0057)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-flt/GlobalDataFlow-flt: 1.09% (2.9242 => 2.9562, std.dev.: 0.0042)
  MultiSource/Applications/SPASS/SPASS: 1.08% (9.2446 => 9.3446, std.dev.: 0.0344)
  MultiSource/Benchmarks/TSVC/ControlFlow-dbl/ControlFlow-dbl: 1.05% (4.5563 => 4.6043, std.dev.: 0.0037)

Performance Improvements - Compile Time
---------------------------------------
  SingleSource/UnitTests/2003-04-22-Switch: -60.00% (0.0200 => 0.0080, std.dev.: 0.0041)
  SingleSource/Benchmarks/BenchmarkGame/nsieve-bits: -44.44% (0.0360 => 0.0200, std.dev.: 0.0061)
  SingleSource/Benchmarks/Shootout/methcall: -27.27% (0.0440 => 0.0320, std.dev.: 0.0041)
  MultiSource/Benchmarks/Prolangs-C/loader/loader: -10.48% (0.4200 => 0.3760, std.dev.: 0.0137)
  MultiSource/Benchmarks/Ptrdist/ks/ks: -10.20% (0.1960 => 0.1760, std.dev.: 0.0066)
  MultiSource/Benchmarks/FreeBench/pcompress2/pcompress2: -7.32% (0.3280 => 0.3040, std.dev.: 0.0078)
  MultiSource/Benchmarks/MiBench/telecomm-gsm/telecomm-gsm: -5.76% (1.5280 => 1.4400, std.dev.: 0.0211)
  MultiSource/Benchmarks/nbench/nbench: -4.44% (1.3521 => 1.2920, std.dev.: 0.0147)
  MultiSource/Benchmarks/MallocBench/gs/gs: -2.45% (5.3840 => 5.2520, std.dev.: 0.0436)
  MultiSource/Applications/SPASS/SPASS: -2.36% (18.7966 => 18.3524, std.dev.: 0.0636)
  MultiSource/Benchmarks/MallocBench/espresso/espresso: -2.34% (6.8280 => 6.6680, std.dev.: 0.0420)
  MultiSource/Applications/ClamAV/clamscan: -2.18% (19.2243 => 18.8043, std.dev.: 0.0400)
  SingleSource/Benchmarks/Misc-C++/stepanov_container: -1.96% (1.0201 => 1.0001, std.dev.: 0.0060)
  MultiSource/Benchmarks/MiBench/consumer-lame/consumer-lame: -1.87% (5.3360 => 5.2360, std.dev.: 0.0382)
  MultiSource/Applications/lemon/lemon: -1.86% (1.5041 => 1.4761, std.dev.: 0.0090)
  MultiSource/Benchmarks/mafft/pairlocalalign: -1.58% (11.3843 => 11.2043, std.dev.: 0.0337)
  MultiSource/Benchmarks/MiBench/consumer-typeset/consumer-typeset: -1.51% (16.7083 => 16.4563, std.dev.: 0.0271)
  MultiSource/Applications/d/make_dparser: -1.46% (4.1081 => 4.0481, std.dev.: 0.0164)
  MultiSource/Benchmarks/7zip/7zip-benchmark: -1.38% (49.1529 => 48.4769, std.dev.: 0.0667)
  MultiSource/Benchmarks/Bullet/bullet: -1.30% (39.0726 => 38.5646, std.dev.: 0.0892)
  SingleSource/Benchmarks/Adobe-C++/loop_unroll: -1.21% (4.2803 => 4.2283, std.dev.: 0.0145)

Performance Improvements - Execution Time
-----------------------------------------
  SingleSource/UnitTests/Vectorizer/gcc-loops: -6.44% (5.2763 => 4.9363, std.dev.: 0.1014)
  MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt: -4.47% (2.0601 => 1.9681, std.dev.: 0.0045)
  SingleSource/Benchmarks/Shootout-C++/matrix: -4.03% (3.4722 => 3.3322, std.dev.: 0.0103)
  SingleSource/Benchmarks/Dhrystone/dry: -2.86% (0.7000 => 0.6800, std.dev.: 0.0058)
  MultiSource/Benchmarks/TSVC/LinearDependence-flt/LinearDependence-flt: -1.95% (4.5123 => 4.4243, std.dev.: 0.0036)
  SingleSource/Benchmarks/Linpack/linpack-pc: -1.93% (7.8605 => 7.7085, std.dev.: 0.0062)
  SingleSource/Benchmarks/Adobe-C++/functionobjects: -1.49% (3.4802 => 3.4282, std.dev.: 0.0118)
  MultiSource/Benchmarks/TSVC/ControlFlow-flt/ControlFlow-flt: -1.26% (4.1243 => 4.0723, std.dev.: 0.0035)
  SingleSource/Benchmarks/Misc-C++-EH/spirit: -1.17% (4.8003 => 4.7443, std.dev.: 0.0141)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl: -1.14% (3.8562 => 3.8122, std.dev.: 0.0154)

Report Time: 3.97s
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-testresults/attachments/20140201/28b168fa/attachment.html>


More information about the llvm-testresults mailing list