search for: loop_unroll

...take timing results with a pinch of salt (a pinch is about +-10% in my experience with this tester). Ciao, Duncan. > Significant changes in test results: > GCCAS: > singlesource/Benchmarks/Adobe-C++/functionobjects: -102.68% (0.5960 => 1.2080) > singlesource/Benchmarks/Adobe-C++/loop_unroll: -44.04% (5.3323 => 7.6804) > singlesource/Benchmarks/Adobe-C++/simple_types_constant_folding: -10.34% (5.6883 => 6.2763) > singlesource/Benchmarks/Adobe-C++/simple_types_loop_invariant: -19.15% (3.4882 => 4.1562) > singlesource/Benchmarks/Adobe-C++/stepanov_abstraction: -97.63...

Hi, I need help debugging a problem a problem that only shows up on PowerPC with one of my commits in the generic CodeGen parts of llvm. I'd apreciate it if someone with access to a PowerPC system could take a look at this: https://reviews.llvm.org/D23097 <https://reviews.llvm.org/D23097> - Matthias -------------- next part -------------- An HTML attachment was scrubbed... URL:

...olly-NoCodeGen" case is mainly used to compare the compile-time performance for the polly-detect pass. As shown in the results, our patch file could significantly reduce the compile-time overhead for some benchmarks such as tramp3dv4 (24.2%), simple_types_constant_folding(12.6%), oggenc(9.1%), loop_unroll(7.8%) The "Polly-opt" case is used to compare the whole compile-time performance of Polly. Since our patch file mainly affects the Polly-Detect pass, it shows similar performance to "Polly-NoCodeGen". As shown in results, it reduces the compile-time overhead of some benchmarks...

...> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.355=2> (24.2%), simple_types_constant_folding > <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.366=2>(12.6%), > oggenc > <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.331=2>(9.1%), > loop_unroll > <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.235=2>(7.8%) Very nice! Though I am surprised to also see performance regressions. They are all in very shortly executing kernels, so they may very well be measuring noice. Is this really the case? Also, it may be interesting...

.../188.40.87.11:8000/db_default/v4/nts/16/graph?test.355=2> (24.2%), simple_types_constant_folding >> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.366=2>(12.6%), >> oggenc >> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.331=2>(9.1%), >> loop_unroll >> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.235=2>(7.8%) > >Very nice! > >Though I am surprised to also see performance regressions. They are all >in very shortly executing kernels, so they may very well be measuring >noice. Is this really the case?...

...00/db_default/v4/nts/16/graph?test.355=2> (24.2%), simple_types_constant_folding >>> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.366=2>(12.6%), >>> oggenc >>> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.331=2>(9.1%), >>> loop_unroll >>> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.235=2>(7.8%) >> >>Very nice! >> >>Though I am surprised to also see performance regressions. They are all >>in very shortly executing kernels, so they may very well be measuring >>noice....

...5.03% > > SPEC2k6 > > 453.povray 17.11% > > 482.sphinx3 3.44% > > 444.namd 2.89% > > Improvements: execution time decrease > > LNT > > MultiSource/Benchmarks/BitBench/uudecode/uudecode -50.90% > > SingleSource/Benchmarks/Adobe-C++/loop_unroll > -27.75% > > SingleSource/Benchmarks/Misc/perlin > -21.35% > > MultiSource/Benchmarks/Olden/em3d/em3d > -19.12% > > MultiSource/Benchmarks/tramp3d-v4/tramp3...

...> often under > heavy load due to other users, so take timing results with a pinch > of salt (a > pinch is about +-10% in my experience with this tester). Awesome Duncan, thank you for working on this! From the LLC tests, I see: > LLC: > singlesource/Benchmarks/Adobe-C++/loop_unroll: 14.13% (3.75 => 3.22) > singlesource/Benchmarks/Adobe-C++/simple_types_constant_folding: > 69.51% (6.92 => 2.11) > singlesource/Benchmarks/CoyoteBench/fftbench: -563.16% (0.57 => 3.78) > singlesource/Benchmarks/Misc-C++/ray: 6.45% (7.60 => 7.11) > singlesource/Benchmar...

Justed wanted to drop folks a note in case they started investigating issues... Eric let me know that he was seeing a significant compile time regression (3x!!!) for O2 builds of Boost spirit on the nightly testers. The really weird thing is that this was only happening for the ARM targeted build. =/ Very strange, and makes it more likely that there is a smoking gun of "oh, oops". I

...; 453.povray 17.11% > > 482.sphinx3 3.44% > > 444.namd 2.89% > > Improvements: execution time decrease > > LNT > > MultiSource/Benchmarks/BitBench/uudecode/uudecode > -50.90% > > SingleSource/Benchmarks/Adobe-C++/loop_unroll > -27.75% > > SingleSource/Benchmarks/Misc/perlin > -21.35% > > MultiSource/Benchmarks/Olden/em3d/em3d > -19.12% > > MultiSource/Benchma...

.../nts/16/graph?test.355=2> (24.2%), simple_types_constant_folding >>>> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.366=2>(12.6%), >>>> oggenc >>>> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.331=2>(9.1%), >>>> loop_unroll >>>> <http://188.40.87.11:8000/db_default/v4/nts/16/graph?test.235=2>(7.8%) >>> >>>Very nice! >>> >>>Though I am surprised to also see performance regressions. They are all >>>in very shortly executing kernels, so they may very well be...

....5852 0.0206 > MultiSource/Applications/JM/ldecod/ldecod.compile 5.13% 7.1410 7.5074 0.0136 > External/SPEC/CINT2006/401_bzip2/401_bzip2.compile 5.09% 2.1246 2.2328 0.0027 > External/SPEC/CINT2006/473_astar/473_astar.compile 5.07% 1.2970 1.3628 0.0028 > SingleSource/Benchmarks/Adobe-C++/loop_unroll.compile 5.07% 3.2894 3.4561 0.0073 > SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding.compile 4.95% > 2.5774 2.7051 0.0062 > MultiSource/Applications/siod/siod.compile 4.95% 2.2539 2.3655 0.0032 > SingleSource/Benchmarks/Misc-C++-EH/spirit.compile 4.95% 8.5869 9.0116 0.00...

...;> MultiSource/Applications/JM/ldecod/ldecod.compile 5.13% 7.1410 7.5074 0.0136 >> External/SPEC/CINT2006/401_bzip2/401_bzip2.compile 5.09% 2.1246 2.2328 0.0027 >> External/SPEC/CINT2006/473_astar/473_astar.compile 5.07% 1.2970 1.3628 0.0028 >> SingleSource/Benchmarks/Adobe-C++/loop_unroll.compile 5.07% 3.2894 3.4561 0.0073 >> SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding.compile 4.95% >> 2.5774 2.7051 0.0062 >> MultiSource/Applications/siod/siod.compile 4.95% 2.2539 2.3655 0.0032 >> SingleSource/Benchmarks/Misc-C++-EH/spirit.compile 4.95% 8....

Hi all, I've started looking at the GlobalMerge pass, enabled by default on ARM and AArch64. I think we should reconsider that, at least for AArch64. As is, the pass just merges all globals together, in groups of 4KB (AArch64, 128B on ARM). At the time it was enabled, the general thinking was "it's almost free, it doesn't affect performance much, we might as well use it".

...ce/Benchmarks/BenchmarkGame/puzzle | * | * | SingleSource/Benchmarks/BenchmarkGame/fannkuch | * | * | SingleSource/Benchmarks/Adobe-C++/functionobjects | * | * | SingleSource/Benchmarks/Adobe-C++/loop_unroll | * | * | SingleSource/Benchmarks/Adobe-C++/stepanov_abstraction | * | * | SingleSource/Benchmarks/Adobe-C++/stepanov_vector | * | * | SingleSource/Benchmarks/Adobe-C++/simple_types_loop_invariant...

Greetings everyone! The new pass manager is getting extremely close to the point where I'm not aware of any significant outstanding work needed, and I'd like to see what else would be needed to enable it by default. Here are the current functionality I'm aware of outstanding: 1) Does not do non-trivial loop unswitching. Majority of this is in https://reviews.llvm.org/D34200 but will

...NT2000/253.perlbmk/Output/253.perlbmk -1.0% External/SPEC/CINT95/132.ijpeg/Output/132.ijpeg -1.0% MultiSource/Benchmarks/Bullet/Output/bullet +1.2% MultiSource/Benchmarks/FreeBench/pifft/Output/pifft +1.6% MultiSource/Benchmarks/McCat/18-imp/Output/imp +1.7% SingleSource/Benchmarks/Adobe-C++/Output/loop_unroll +3.2% SingleSource/Benchmarks/Misc/Output/oourafft Targeting thumbv7 PIC -O2: -6.8% MultiSource/Benchmarks/Ptrdist/yacr2/Output/yacr2 -5.8% SingleSource/Benchmarks/Adobe-C++/Output/simple_types_constant_folding -5.8% MultiSource/Benchmarks/Ptrdist/bc/Output/bc -5.6% External/SPEC/CINT2000/256.bzip...

...0.179816396| 41.32| 1403783168| 54.97|0.179490547| 41.57| 1403737509| 54.97| |SingleSource/Benchmarks/Adobe-C++/functionobjects.test | 40|2.173834838| 7456450808|2.153285393| 0.95| 7456450841| 0|2.152186046| 1.01| 7456450841| 0| |SingleSource/Benchmarks/Adobe-C++/loop_unroll.test | 55|0.631261876| 4991342551|0.631185655| 0.01| 4991342299| 0|0.631218693| 0.01| 4991342299| 0| |SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding.test | 70|0.493520528| 1690319795|0.493526469| 0| 1690319517| 0|0.4933663...

...36 | n/a     1.19    n/a          n/a > SingleSource/Benchmarks/Adobe-C++/functionobjects                        | > 0.2100 37048    0.1799      *                0.1500      |    3.97    3.53 >  3.43 *           3.66 | 1.12    1.16    n/a          n/a > SingleSource/Benchmarks/Adobe-C++/loop_unroll                            | > 1.0500 328448   1.6400      *                1.7599      |    2.01    1.55 >  1.81 *           3.63 | 1.30    1.11    n/a          n/a > SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding          | > 0.8100 185728   2.1699      *            ...

Hi Tanya, > 1) Compile llvm from source and untar the llvm-test in the projects > directory (name it llvm-test or test-suite). Choose to use a > pre-compiled llvm-gcc or re-compile it yourself. I compiled llvm and llvm-gcc with separate objects directories. Platform is x86_64-linux-gnu. > 2) Run make check, report any failures (FAIL or unexpected pass). Note > that you need to