[llvm-testresults] bwilson__llvm-gcc_PROD__i386 nightly tester results
Andrew Trick
atrick at apple.com
Tue Mar 8 15:39:39 PST 2011
* rdar://9105010 [pre-RA-sched][i386] 56% slowdown of skidmarks.MPEG with list-ilp
* PR9431 33% slowdown of Vector/multiplies with list-ilp on i386
* PR9432 22% slowdown of flops-7 with list-ilp on i386, -7% on x86_64
* PR9433 11% slowdown of enc-md5 with list-ilp on x86+64, 4.5% on i386
* rdar://9105031 [pre-RA-sched][x86_64] 10% slowdown of 470.lbm with list-ilp
* rdar://9105082 [pre-RA-sched][i386] Benchmark slowdowns with list-ilp
* rdar://9105070 [pre-RA-sched][x86_64] Assorted microbenchmark slowdowns with list-ilp
-Andy
On Mar 8, 2011, at 12:38 AM, llvm-testresults at cs.uiuc.edu wrote:
> bwilson__llvm-gcc_PROD__i386 nightly tester results
>
> URL http://llvm.org/perf/db_default/simple/nts/129/
> Nickname bwilson__llvm-gcc_PROD__i386:2
> Name curlew.apple.com
> Run ID Order Start Time End Time
> Current 129 127220 2011-03-08 07:58:26 2011-03-08 08:42:41
> Previous 128 127155 2011-03-07 08:11:16 2011-03-07 08:55:19
>
> Tests Summary
>
> Status Group #
> Performance Regressions 19
> Performance Improvements 17
> Unchanged Tests 913
> Total Tests 949
>
> Changes Detail
>
> Performance Regressions Δ Previous Current σ
> External/skidmarks10/skidmarks.MPEG.exec 127.75% 0.0436 0.0993 0.0001
> SingleSource/UnitTests/Vector/multiplies.exec 48.80% 2.0900 3.1100 0.0000
> SingleSource/Benchmarks/Misc/flops-7.exec 28.61% 3.3900 4.3600 0.0049
> SingleSource/Benchmarks/Shootout/matrix.exec 6.67% 3.3000 3.5200 0.0000
> MultiSource/Applications/hexxagon/hexxagon.exec 5.68% 13.7300 14.5100 0.0040
> SingleSource/Benchmarks/Misc/himenobmtxpa.exec 5.58% 5.5600 5.8700 0.0040
> External/SPEC/CFP2000/177_mesa/177_mesa.exec 5.43% 3.5000 3.6900 0.0040
> External/SPEC/CFP2006/470_lbm/470_lbm.exec 5.15% 10.6700 11.2200 0.0089
> External/SPEC/CINT2006/464_h264ref/464_h264ref.exec 5.12% 26.9400 28.3200 0.0392
> MultiSource/Benchmarks/Trimaran/enc-md5/enc-md5.exec 4.65% 1.2900 1.3500 0.0000
> MultiSource/Benchmarks/Olden/bh/bh.exec 4.32% 3.0100 3.1400 0.0040
> MultiSource/Benchmarks/Fhourstones-3_1/fhourstones3_1.exec 4.20% 2.3800 2.4800 0.0040
> MultiSource/Applications/JM/lencod/lencod.exec 3.91% 13.5600 14.0900 0.0569
> MultiSource/Benchmarks/sim/sim.exec 2.03% 7.9000 8.0600 0.0120
> SingleSource/Benchmarks/Misc-C++-EH/spirit.exec 1.50% 9.3100 9.4500 0.0080
> MultiSource/Benchmarks/ASCI_Purple/SMG2000/smg2000.compile 1.50% 7.3500 7.4600 0.0299
> MultiSource/Benchmarks/VersaBench/8b10b/8b10b.exec 1.23% 8.9100 9.0200 0.0160
> External/SPEC/CINT2000/175_vpr/175_vpr.exec 1.22% 4.9300 4.9900 0.0120
> External/SPEC/CFP2006/444_namd/444_namd.compile 1.16% 6.9000 6.9800 0.0228
>
> Performance Improvements Δ Previous Current σ
> SingleSource/Benchmarks/Dhrystone/fldry.exec -13.70% 1.4600 1.2600 0.0049
> External/SPEC/CFP2006/433_milc/433_milc.exec -9.78% 9.7100 8.7600 0.0063
> SingleSource/Benchmarks/Misc/fbench.exec -8.61% 2.0900 1.9100 0.0539
> SingleSource/UnitTests/Vector/build2.exec -5.49% 1.8200 1.7200 0.0000
> SingleSource/Benchmarks/Misc/salsa20.exec -4.88% 10.6500 10.1300 0.0360
> SingleSource/Benchmarks/Misc/flops-5.exec -3.65% 1.9200 1.8500 0.0040
> External/SPEC/CFP2000/188_ammp/188_ammp.exec -3.59% 17.8500 17.2100 0.0040
> External/Nurbs/nurbs.exec -2.94% 5.1100 4.9600 0.0080
> SingleSource/Benchmarks/CoyoteBench/almabench.exec -2.57% 12.0500 11.7400 0.0223
> External/SPEC/CFP2000/183_equake/183_equake.exec -2.48% 9.2700 9.0400 0.0167
> External/SPEC/CINT2006/458_sjeng/458_sjeng.exec -2.27% 6.6100 6.4600 0.0167
> MultiSource/Benchmarks/Bullet/bullet.exec -1.72% 6.4000 6.2900 0.0160
> MultiSource/Benchmarks/PAQ8p/paq8p.exec -1.64% 96.2500 94.6700 0.2567
> External/SPEC/CINT2006/473_astar/473_astar.exec -1.60% 11.2300 11.0500 0.0400
> MultiSource/Applications/sqlite3/sqlite3.exec -1.40% 5.0100 4.9400 0.0147
> External/SPEC/CINT2000/186_crafty/186_crafty.exec -1.18% 9.3400 9.2300 0.0224
> MultiSource/Applications/lua/lua.exec -1.09% 31.2500 30.9100 0.1076
> _______________________________________________
> llvm-testresults mailing list
> llvm-testresults at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-testresults
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-testresults/attachments/20110308/2d4d708c/attachment.html>
More information about the llvm-testresults
mailing list