[llvm-testresults] pollyperf-O3-polly-scev-codegen-isl__clang_DEV__x86_64 test results

llvm-testresults at cs.uiuc.edu llvm-testresults at cs.uiuc.edu
Tue Feb 18 21:52:41 PST 2014


http://llvm.org/perf/db_default/v4/nts/22432?compare_to=22421&baseline=21513
Nickname: pollyperf-O3-polly-scev-codegen-isl__clang_DEV__x86_64:39

Comparing:
     Run: 22432, Order: 201620, Start Time: 2014-02-19 00:57:41, End Time: 2014-02-19 05:58:10
      To: 22421, Order: 201595, Start Time: 2014-02-18 19:43:03, End Time: 2014-02-19 00:44:52
Baseline: 21513, Order: 200427, Start Time: 2014-01-29 23:42:17, End Time: 2014-01-30 04:50:23

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

New Passes: 0 (2 on baseline)
Performance Regressions: 15 (45 on baseline)
Performance Improvements: 0 (17 on baseline)
Unchanged Tests: 1095 (1046 on baseline)
Total Tests: 1110

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

Performance Regressions - Compile Time
--------------------------------------
  SingleSource/UnitTests/AtomicOps: 300.00% (0.0040 => 0.0160, std.dev.: 0.0039)
  MultiSource/Benchmarks/Prolangs-C++/family/family: 37.50% (0.0320 => 0.0440, std.dev.: 0.0040)
  MultiSource/Benchmarks/Olden/perimeter/perimeter: 14.29% (0.0840 => 0.0960, std.dev.: 0.0037)
  MultiSource/Benchmarks/Trimaran/netbench-crc/netbench-crc: 9.09% (0.2200 => 0.2400, std.dev.: 0.0060)
  SingleSource/Benchmarks/Shootout-C++/heapsort: 6.90% (0.2320 => 0.2480, std.dev.: 0.0059)
  MultiSource/Applications/ALAC/decode/alacconvert-decode: 3.80% (1.2640 => 1.3120, std.dev.: 0.0181)
  MultiSource/Benchmarks/TSVC/Equivalencing-flt/Equivalencing-flt: 2.23% (1.4361 => 1.4681, std.dev.: 0.0112)
  MultiSource/Benchmarks/TSVC/Searching-flt/Searching-flt: 1.99% (1.4041 => 1.4321, std.dev.: 0.0053)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-flt/GlobalDataFlow-flt: 1.87% (1.5001 => 1.5281, std.dev.: 0.0089)
  MultiSource/Benchmarks/TSVC/Symbolics-dbl/Symbolics-dbl: 1.82% (1.5401 => 1.5681, std.dev.: 0.0077)
  MultiSource/Benchmarks/TSVC/NodeSplitting-dbl/NodeSplitting-dbl: 1.81% (1.5441 => 1.5721, std.dev.: 0.0080)
  MultiSource/Benchmarks/TSVC/NodeSplitting-flt/NodeSplitting-flt: 1.63% (1.4761 => 1.5001, std.dev.: 0.0084)
  MultiSource/Benchmarks/PAQ8p/paq8p: 1.30% (2.7722 => 2.8082, std.dev.: 0.0120)
  SingleSource/Benchmarks/Polybench/stencils/seidel-2d/seidel-2d: 1.02% (1.5761 => 1.5921, std.dev.: 0.0059)

Performance Regressions - Execution Time
----------------------------------------
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/cholesky/cholesky: 8.20% (1.0241 => 1.1081, std.dev.: 0.0313)

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

New Passes - Compile Time
-------------------------
  SingleSource/Benchmarks/Misc/oourafft
  SingleSource/Benchmarks/Misc/oourafft: 745.88% (11.3687 => 96.1660, std.dev.: 0.2210)

New Passes - Execution Time
---------------------------
  SingleSource/Benchmarks/Misc/oourafft
  SingleSource/Benchmarks/Misc/oourafft: - (0.0000 => 6.8884, std.dev.: 0.0138)

Performance Regressions - Compile Time
--------------------------------------
  SingleSource/Regression/C++/EH/simple_throw: 300.00% (0.0040 => 0.0160, std.dev.: 0.0039)
  SingleSource/Regression/C++/2011-03-28-Bitfield: 150.00% (0.0080 => 0.0200, std.dev.: 0.0043)
  SingleSource/Regression/C/sumarraymalloc: 60.00% (0.0200 => 0.0320, std.dev.: 0.0040)
  SingleSource/Benchmarks/Shootout/methcall: 42.86% (0.0280 => 0.0400, std.dev.: 0.0037)
  MultiSource/Benchmarks/Olden/perimeter/perimeter: 41.18% (0.0680 => 0.0960, std.dev.: 0.0037)
  MultiSource/Benchmarks/McCat/15-trie/trie: 21.43% (0.0560 => 0.0680, std.dev.: 0.0044)
  MultiSource/Benchmarks/Prolangs-C++/trees/trees: 13.64% (0.0880 => 0.1000, std.dev.: 0.0044)
  SingleSource/Benchmarks/Shootout-C++/random: 7.14% (0.2240 => 0.2400, std.dev.: 0.0058)
  MultiSource/Benchmarks/Prolangs-C/unix-tbl/unix-tbl: 6.19% (1.1640 => 1.2360, std.dev.: 0.0242)
  SingleSource/Benchmarks/Misc-C++/Large/sphereflake: 4.76% (0.3360 => 0.3520, std.dev.: 0.0051)
  MultiSource/Benchmarks/TSVC/Packing-dbl/Packing-dbl: 3.89% (1.4401 => 1.4961, std.dev.: 0.0117)
  MultiSource/Benchmarks/TSVC/ControlLoops-dbl/ControlLoops-dbl: 3.59% (1.5601 => 1.6161, std.dev.: 0.0128)
  MultiSource/Benchmarks/Prolangs-C/cdecl/cdecl: 3.39% (0.4720 => 0.4880, std.dev.: 0.0051)
  MultiSource/Benchmarks/TSVC/Expansion-dbl/Expansion-dbl: 3.39% (1.6521 => 1.7081, std.dev.: 0.0112)
  MultiSource/Benchmarks/TSVC/Recurrences-dbl/Recurrences-dbl: 3.29% (1.4601 => 1.5081, std.dev.: 0.0100)
  MultiSource/Benchmarks/TSVC/NodeSplitting-dbl/NodeSplitting-dbl: 3.15% (1.5241 => 1.5721, std.dev.: 0.0080)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-dbl/GlobalDataFlow-dbl: 3.10% (1.5481 => 1.5961, std.dev.: 0.0088)
  MultiSource/Benchmarks/TSVC/Searching-dbl/Searching-dbl: 3.09% (1.4241 => 1.4681, std.dev.: 0.0139)
  MultiSource/Benchmarks/TSVC/IndirectAddressing-dbl/IndirectAddressing-dbl: 2.96% (1.4841 => 1.5281, std.dev.: 0.0139)
  MultiSource/Benchmarks/TSVC/LoopRerolling-dbl/LoopRerolling-dbl: 2.96% (1.4841 => 1.5281, std.dev.: 0.0071)
  SingleSource/Benchmarks/Misc-C++/stepanov_container: 2.89% (0.9681 => 0.9961, std.dev.: 0.0073)
  MultiSource/Benchmarks/TSVC/StatementReordering-dbl/StatementReordering-dbl: 2.75% (1.4561 => 1.4961, std.dev.: 0.0095)
  MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl: 2.73% (1.6121 => 1.6561, std.dev.: 0.0079)
  MultiSource/Benchmarks/TSVC/Symbolics-dbl/Symbolics-dbl: 2.62% (1.5281 => 1.5681, std.dev.: 0.0077)
  MultiSource/Benchmarks/TSVC/ControlFlow-dbl/ControlFlow-dbl: 2.30% (1.7361 => 1.7761, std.dev.: 0.0125)
  MultiSource/Benchmarks/TSVC/Reductions-dbl/Reductions-dbl: 2.24% (1.6081 => 1.6441, std.dev.: 0.0122)
  MultiSource/Benchmarks/TSVC/LoopRestructuring-dbl/LoopRestructuring-dbl: 1.98% (1.8161 => 1.8521, std.dev.: 0.0138)
  MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000: 1.90% (9.4842 => 9.6642, std.dev.: 0.0456)
  SingleSource/Benchmarks/Polybench/stencils/seidel-2d/seidel-2d: 1.79% (1.5641 => 1.5921, std.dev.: 0.0059)
  SingleSource/Benchmarks/Adobe-C++/simple_types_loop_invariant: 1.72% (2.3281 => 2.3681, std.dev.: 0.0093)
  SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding: 1.59% (3.0282 => 3.0762, std.dev.: 0.0128)
  MultiSource/Benchmarks/MiBench/automotive-susan/automotive-susan: 1.57% (1.0201 => 1.0361, std.dev.: 0.0058)
  MultiSource/Benchmarks/TSVC/ControlLoops-flt/ControlLoops-flt: 1.56% (1.5401 => 1.5641, std.dev.: 0.0073)
  MultiSource/Benchmarks/TSVC/ControlFlow-flt/ControlFlow-flt: 1.42% (1.6921 => 1.7161, std.dev.: 0.0086)
  MultiSource/Benchmarks/mediabench/jpeg/jpeg-6a/cjpeg: 1.41% (6.5440 => 6.6360, std.dev.: 0.0345)
  MultiSource/Benchmarks/tramp3d-v4/tramp3d-v4: 1.28% (29.3338 => 29.7099, std.dev.: 0.0617)
  MultiSource/Applications/SPASS/SPASS: 1.22% (18.4404 => 18.6645, std.dev.: 0.0234)
  SingleSource/Benchmarks/Adobe-C++/loop_unroll: 1.15% (4.1843 => 4.2323, std.dev.: 0.0173)
  SingleSource/Benchmarks/Misc-C++-EH/spirit: 1.11% (6.1044 => 6.1724, std.dev.: 0.0123)
  MultiSource/Benchmarks/MiBench/consumer-typeset/consumer-typeset: 1.10% (16.3483 => 16.5283, std.dev.: 0.0490)
  MultiSource/Applications/kimwitu++/kc: 1.04% (19.6891 => 19.8931, std.dev.: 0.0332)
  MultiSource/Applications/siod/siod: 1.03% (2.7162 => 2.7442, std.dev.: 0.0101)

Performance Regressions - Execution Time
----------------------------------------
  SingleSource/Benchmarks/Polybench/linear-algebra/solvers/gramschmidt/gramschmidt: 16.08% (2.1401 => 2.4842, std.dev.: 0.0328)
  MultiSource/Applications/sqlite3/sqlite3: 2.01% (3.9842 => 4.0643, std.dev.: 0.0149)
  MultiSource/Benchmarks/TSVC/GlobalDataFlow-flt/GlobalDataFlow-flt: 1.02% (2.7442 => 2.7722, std.dev.: 0.0036)

Performance Improvements - Compile Time
---------------------------------------
  MultiSource/Benchmarks/BitBench/drop3/drop3: -25.00% (0.0640 => 0.0480, std.dev.: 0.0057)
  SingleSource/Benchmarks/Misc/flops: -5.00% (0.4800 => 0.4560, std.dev.: 0.0059)
  SingleSource/Benchmarks/Adobe-C++/stepanov_abstraction: -4.17% (0.6720 => 0.6440, std.dev.: 0.0093)
  SingleSource/Benchmarks/Stanford/Puzzle: -3.48% (0.4600 => 0.4440, std.dev.: 0.0047)
  MultiSource/Benchmarks/mafft/pairlocalalign: -2.82% (13.0563 => 12.6883, std.dev.: 0.0292)

Performance Improvements - Execution Time
-----------------------------------------
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/gemm/gemm: -34.17% (26.0616 => 17.1571, std.dev.: 0.1884)
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/2mm/2mm: -34.08% (50.3031 => 33.1581, std.dev.: 0.5231)
  SingleSource/Benchmarks/Polybench/linear-algebra/kernels/3mm/3mm: -23.83% (91.8497 => 69.9644, std.dev.: 1.7178)
  SingleSource/Benchmarks/CoyoteBench/fftbench: -16.32% (3.1122 => 2.6042, std.dev.: 0.1140)
  MultiSource/Benchmarks/FreeBench/analyzer/analyzer: -13.79% (0.1160 => 0.1000, std.dev.: 0.0012)
  SingleSource/Benchmarks/Linpack/linpack-pc: -12.59% (7.7845 => 6.8044, std.dev.: 0.0045)
  MultiSource/Benchmarks/Olden/perimeter/perimeter: -8.24% (0.3400 => 0.3120, std.dev.: 0.0091)
  MultiSource/Benchmarks/TSVC/LinearDependence-flt/LinearDependence-flt: -2.50% (4.8083 => 4.6883, std.dev.: 0.0028)
  MultiSource/Benchmarks/BitBench/five11/five11: -1.70% (2.3561 => 2.3161, std.dev.: 0.0106)
  SingleSource/UnitTests/Vectorizer/gcc-loops: -1.68% (5.7084 => 5.6124, std.dev.: 0.0044)
  MultiSource/Benchmarks/mafft/pairlocalalign: -1.08% (26.6977 => 26.4096, std.dev.: 0.0628)
  SingleSource/Benchmarks/Misc/salsa20: -1.02% (7.4205 => 7.3445, std.dev.: 0.0235)

Report Time: 0.65s
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-testresults/attachments/20140218/4ea93a54/attachment.html>


More information about the llvm-testresults mailing list