AARCH64 codegen comparison update

gcc7's performance on Cortex A53 (32kB L1)

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1703238-RI-GCCLATEST31
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

C/C++ Compiler Tests 5 Tests
CPU Massive 9 Tests
Creator Workloads 4 Tests
HPC - High Performance Computing 2 Tests
Common Kernel Benchmarks 2 Tests
Multi-Core 5 Tests
Raytracing 2 Tests
Renderers 4 Tests
Scientific Computing 2 Tests
Server 2 Tests
Server CPU Tests 4 Tests
Single-Threaded 4 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
A53 vectorize, pre-patch
January 11 2017
 
thunderx/vectorize, pre-patch
January 10 2017
 
A53 vectorize/LTO, pre patch
January 11 2017
 
A53, post patch
January 14 2017
 
A53 mtune/vectorize, post-patch
January 14 2017
 
A53 vectorize, updated
February 15 2017
 
A53 vectorize, earlier build
February 25 2017
 
A57 vectorize/unrolled GCC 7.0.1
March 22 2017
 
A53 vectorize GCC 7.0.1
March 23 2017
 
A57 vectorize/unrolled GCC 6.3
March 22 2017
 
Invert Hiding All Results Option
 

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


AARCH64 codegen comparison update, "RAMspeed SMP 3.5.0 - Type: Copy - Benchmark: Floating Point", Higher Results Are Better "thunderx/vectorize, pre-patch", "A53 vectorize, pre-patch", "A53 vectorize/LTO, pre patch", "A53, post patch", "A53 mtune/vectorize, post-patch", "A53 vectorize, updated", "A53 vectorize, earlier build", "A57 vectorize/unrolled GCC 6.3", "A57 vectorize/unrolled GCC 7.0.1", "A53 vectorize GCC 7.0.1", "RAMspeed SMP 3.5.0 - Type: Copy - Benchmark: Integer", Higher Results Are Better "thunderx/vectorize, pre-patch", "A53 vectorize, pre-patch", "A53 vectorize/LTO, pre patch", "A53, post patch", "A53 mtune/vectorize, post-patch", "A53 vectorize, updated", "A53 vectorize, earlier build", "A57 vectorize/unrolled GCC 6.3", "A57 vectorize/unrolled GCC 7.0.1", "A53 vectorize GCC 7.0.1", "C-Ray 1.1 - Total Time", Lower Results Are Better "thunderx/vectorize, pre-patch",148.535,152.551,148.359 "A53 vectorize, pre-patch",188.03,189.128,186.751 "A53 vectorize/LTO, pre patch",185.096,184.511,184.834 "A53, post patch",186.958,186.536,186.568 "A53 mtune/vectorize, post-patch",186.717,186.724,186.374 "A53 vectorize, updated",161.538,162.345,161.526 "A53 vectorize, earlier build",161.189,164.232,161.278 "A57 vectorize/unrolled GCC 6.3",148.131,148.149,152.544 "A57 vectorize/unrolled GCC 7.0.1",156.361,154.138,153.933 "A53 vectorize GCC 7.0.1",151.031,151.05,150.992 "FFTW 3.3.4 - Build: Stock - Size: 2D FFT Size 2048", Higher Results Are Better "thunderx/vectorize, pre-patch",193.37,188.62,189.54,193.22,188.41 "A53 vectorize, pre-patch",196.58,194.64,197.09,195.71,200.5 "A53 vectorize/LTO, pre patch",179.52,179.17,181.17,181.16,181.62 "A53, post patch",185.91,186.38,186.24,186.28,186.26 "A53 mtune/vectorize, post-patch",184.77,184.88,185.55,184.57,184.26 "A53 vectorize, updated",184.94,185.14,185.15,185.24,185.26 "A53 vectorize, earlier build",191.75,191.68,191.94,191.18,191.17 "A57 vectorize/unrolled GCC 6.3",157.06,157.96,157.86,158.54,158.43 "A57 vectorize/unrolled GCC 7.0.1",172.99,172.71,173.03,173.15,173.29 "A53 vectorize GCC 7.0.1",156.13,157.07,156.98,156.69,156.72 "PostMark 1.51 - Disk Transaction Performance", Higher Results Are Better "thunderx/vectorize, pre-patch",1351,1351,1351 "A53 vectorize, pre-patch",1366,1358,1366 "A53 vectorize/LTO, pre patch",1381,1381,1373 "A53, post patch",1388,1381,1373 "A53 mtune/vectorize, post-patch",1373,1373,1388 "A53 vectorize, updated",1219,1213,1219 "A53 vectorize, earlier build",1201,1219,1213 "A57 vectorize/unrolled GCC 6.3", "A57 vectorize/unrolled GCC 7.0.1", "A53 vectorize GCC 7.0.1",1190,1190,1201 "Redis 3.0.1 - Test: GET", Higher Results Are Better "thunderx/vectorize, pre-patch",318268.59,324044.06,314465.41 "A53 vectorize, pre-patch",322268.75,289435.59,310269.94,307976.56,317057.69,315059.84 "A53 vectorize/LTO, pre patch",309789.31,309310.22,316255.53 "A53, post patch",308832.59,310945.28,307314.06 "A53 mtune/vectorize, post-patch",314267.75,309693.41,316355.56 "A53 vectorize, updated",273298.72,278629.12,279876.84 "A53 vectorize, earlier build",284495,283687.94,283045.56 "A57 vectorize/unrolled GCC 6.3",280820,276014.34,269541.75 "A57 vectorize/unrolled GCC 7.0.1",275178.88,275938.19,277392.5 "A53 vectorize GCC 7.0.1",272628.12,276625.16,279642.03 "Timed MAFFT Alignment 6.864 - Multiple Sequence Alignment", Lower Results Are Better "thunderx/vectorize, pre-patch",35.555989027023,35.703128814697,32.4645819664,35.295182943344,31.907799959183,35.822311878204 "A53 vectorize, pre-patch",35.442006111145,35.485805988312,32.142671108246,35.52744102478,35.552447080612,38.352542161942 "A53 vectorize/LTO, pre patch",35.345814228058,32.107788085938,35.365938186646,31.839611053467,32.141003131866,32.173220872879 "A53, post patch",35.429574966431,31.607346057892,32.077352046967,35.145592927933,32.008775949478,32.061383962631 "A53 mtune/vectorize, post-patch",32.193259000778,32.176614046097,32.146126031876 "A53 vectorize, updated",35.645505905151,32.382761001587,35.64851808548,35.668017864227,31.765883922577,32.269311904907 "A53 vectorize, earlier build",32.295727014542,36.230386972427,32.395914077759,35.560294866562,32.371257066727,38.294654846191 "A57 vectorize/unrolled GCC 6.3",35.42445397377,35.36857008934,35.621080875397 "A57 vectorize/unrolled GCC 7.0.1",36.047016859055,32.676728963852,35.968631029129,35.368231058121,32.630671024323,32.645817041397 "A53 vectorize GCC 7.0.1",38.920295000076,36.325515031815,32.798367977142,36.11018204689,36.248899936676,32.697088003159 "Primesieve 5.4.2 - 1e12 Prime Number Generation", Lower Results Are Better "thunderx/vectorize, pre-patch",571.866,565.077,561.699 "A53 vectorize, pre-patch",547.078,537.243,545.157 "A53 vectorize/LTO, pre patch",532.142,557.787,532.925 "A53, post patch",536.696,568.287,554.411 "A53 mtune/vectorize, post-patch",575.859,560.028,583.514 "A53 vectorize, updated",561.723,557.949,597.033,581.887 "A53 vectorize, earlier build",531.079,522.491,516.725 "A57 vectorize/unrolled GCC 6.3",534.079,533.402,528.367 "A57 vectorize/unrolled GCC 7.0.1",531.29,521.421,522.288 "A53 vectorize GCC 7.0.1",540.155,565.682,535.524 "Tachyon 0.98.9 - Total Time", Lower Results Are Better "thunderx/vectorize, pre-patch",71.3229,71.4013,71.5203 "A53 vectorize, pre-patch",69.2188,69.4231,69.1755 "A53 vectorize/LTO, pre patch",67.5241,67.552,67.8578 "A53, post patch",69.6143,69.2001,69.3829 "A53 mtune/vectorize, post-patch",69.2987,69.1898,69.5358 "A53 vectorize, updated",69.9105,69.3795,70.4009 "A53 vectorize, earlier build",69.3188,69.8077,69.041 "A57 vectorize/unrolled GCC 6.3",72.3035,72.3151,72.232 "A53 vectorize GCC 7.0.1",69.3646,69.5951,69.9595 "TTSIOD 3D Renderer 2.3a - Phong Rendering With Soft-Shadow Mapping", Higher Results Are Better "thunderx/vectorize, pre-patch",23.0071,23.0118,23.0134 "A53 vectorize, pre-patch",23.1633,23.1442,23.1632 "A53 vectorize/LTO, pre patch",23.7745,23.7761,23.7446 "A53, post patch",23.4938,23.4395,23.4785 "A53 mtune/vectorize, post-patch",23.5076,23.4944,23.4683 "A53 vectorize, updated",23.477,23.2029,23.1832 "A53 vectorize, earlier build",23.4994,23.486,23.4945 "A57 vectorize/unrolled GCC 6.3",22.2988,22.2767,22.2904 "A57 vectorize/unrolled GCC 7.0.1",23.5839,23.5324,23.5573 "A53 vectorize GCC 7.0.1",23.1535,23.1011,23.1406 "Fhourstones 3.1 - Complex Connect-4 Solving", Higher Results Are Better "thunderx/vectorize, pre-patch",3210.7,3211.2,3208.7 "A53 vectorize, pre-patch",3211.5,3212.7,3212.1 "A53 vectorize/LTO, pre patch",3214.2,3213.6,3213.5 "A53, post patch",3207.2,3209.5,3212.3 "A53 mtune/vectorize, post-patch",3203.2,3204,3209 "A53 vectorize, updated",3227.7,3217,3226 "A53 vectorize, earlier build",3235.5,3235,3230.8 "A57 vectorize/unrolled GCC 6.3",3321.8,3326.6,3328.4 "A57 vectorize/unrolled GCC 7.0.1",3397.8,3402.1,3395.5 "A53 vectorize GCC 7.0.1",3418.9,3415.7,3410.6 "Smallpt 1.0 - Global Illumination Renderer; 100 Samples", Lower Results Are Better "thunderx/vectorize, pre-patch",167,167,167 "A53 vectorize, pre-patch",167,167,167 "A53 vectorize/LTO, pre patch",168,168,168 "A53, post patch",168,168,168 "A53 mtune/vectorize, post-patch",167,167,167 "A53 vectorize, updated",166,166,167 "A53 vectorize, earlier build",166,167,166 "A57 vectorize/unrolled GCC 6.3",167,167,174 "A57 vectorize/unrolled GCC 7.0.1", "A53 vectorize GCC 7.0.1", "OpenSSL 1.0.1g - RSA 4096-bit Performance", Higher Results Are Better "thunderx/vectorize, pre-patch",21.5,21.5,21.5 "A53 vectorize, pre-patch",21.5,21.5,21.5 "A53 vectorize/LTO, pre patch",21.5,21.5,21.5 "A53, post patch",21.5,21.5,21.5 "A53 mtune/vectorize, post-patch",21.5,21.5,21.5 "A53 vectorize, updated",21.4,21.4,21.4 "A53 vectorize, earlier build",21.4,21.4,21.4 "A57 vectorize/unrolled GCC 6.3",21.2,21.2,21.2 "A57 vectorize/unrolled GCC 7.0.1",21.5,21.4,21.5 "A53 vectorize GCC 7.0.1",21.5,21.4,21.5 "Sudokut 0.4 - Total Time", Lower Results Are Better "thunderx/vectorize, pre-patch",102.00668287277,104.26202511787,101.9764380455 "A53 vectorize, pre-patch",102.35151290894,101.75110793114,101.73476886749 "A53 vectorize/LTO, pre patch",101.55679512024,101.51205706596,102.17406892776 "A53, post patch",101.90601301193,101.69927287102,102.04617595673 "A53 mtune/vectorize, post-patch",102.1142680645,102.05784988403,102.34896993637 "A53 vectorize, updated",102.92323112488,102.31637191772,102.9164750576 "A53 vectorize, earlier build",102.697660923,102.569231987,102.49186205864 "A57 vectorize/unrolled GCC 6.3",102.92916488647,103.01142311096,103.07156801224 "A57 vectorize/unrolled GCC 7.0.1",102.97246599197,103.09440302849,103.04292106628 "A53 vectorize GCC 7.0.1",103.08997702599,102.70796394348,103.12341380119 "GMPbench 0.2 - Total Time", Higher Results Are Better "thunderx/vectorize, pre-patch", "A53 vectorize, pre-patch", "A53 vectorize/LTO, pre patch", "A53, post patch", "A53 mtune/vectorize, post-patch", "A53 vectorize, updated", "A53 vectorize, earlier build", "A57 vectorize/unrolled GCC 6.3", "A57 vectorize/unrolled GCC 7.0.1", "A53 vectorize GCC 7.0.1",