AMD EPYC Compiler Testing GCC 9.0 SVN With PGO AMD EPYC compiler benchmarks for a future article on Phoronix.com. GCC 8.2 RC1: -O2: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 8.1.1 20180720, File-System: ext4, Screen Resolution: 1920x1080 GCC 8.2 RC1: -O3: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 8.1.1 20180720, File-System: ext4, Screen Resolution: 1920x1080 GCC 8.2 RC1: -O3 -march=native: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 8.1.1 20180720, File-System: ext4, Screen Resolution: 1920x1080 GCC 8.2 RC1: -O3 -march=native - PGO: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 8.1.1 20180720, File-System: ext4, Screen Resolution: 1920x1080 GCC 9.0 SVN: -O2: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 9.0.0 20180720, File-System: ext4, Screen Resolution: 1920x1080 GCC 9.0 SVN: -O3: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 9.0.0 20180720, File-System: ext4, Screen Resolution: 1920x1080 GCC 9.0 SVN: -O3 -march=native: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 9.0.0 20180720, File-System: ext4, Screen Resolution: 1920x1080 GCC 9.0 SVN: -O3 -march=native - PGO: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 9.0.0 20180720, File-System: ext4, Screen Resolution: 1920x1080 SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better GCC 8.2 RC1: -O2 ..................... 1464.06 |======================= GCC 8.2 RC1: -O3 ..................... 1874.09 |============================== GCC 8.2 RC1: -O3 -march=native ....... 1943.30 |=============================== GCC 8.2 RC1: -O3 -march=native - PGO . 1877.54 |============================== GCC 9.0 SVN: -O2 ..................... 1499.96 |======================== GCC 9.0 SVN: -O3 ..................... 1877.03 |============================== GCC 9.0 SVN: -O3 -march=native ....... 1940.65 |=============================== GCC 9.0 SVN: -O3 -march=native - PGO . 1870.83 |============================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better GCC 8.2 RC1: -O2 ..................... 220.52 |================================ GCC 8.2 RC1: -O3 ..................... 221.30 |================================ GCC 8.2 RC1: -O3 -march=native ....... 219.38 |================================ GCC 8.2 RC1: -O3 -march=native - PGO . 219.09 |================================ GCC 9.0 SVN: -O2 ..................... 220.80 |================================ GCC 9.0 SVN: -O3 ..................... 221.28 |================================ GCC 9.0 SVN: -O3 -march=native ....... 219.09 |================================ GCC 9.0 SVN: -O3 -march=native - PGO . 219.24 |================================ SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better GCC 8.2 RC1: -O2 ..................... 919.02 |================= GCC 8.2 RC1: -O3 ..................... 1429.81 |========================== GCC 8.2 RC1: -O3 -march=native ....... 1687.64 |=============================== GCC 8.2 RC1: -O3 -march=native - PGO . 1676.15 |=============================== GCC 9.0 SVN: -O2 ..................... 918.96 |================= GCC 9.0 SVN: -O3 ..................... 1429.31 |========================== GCC 9.0 SVN: -O3 -march=native ....... 1688.09 |=============================== GCC 9.0 SVN: -O3 -march=native - PGO . 1674.99 |=============================== SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better GCC 8.2 RC1: -O2 ..................... 561.79 |================================ GCC 8.2 RC1: -O3 ..................... 561.85 |================================ GCC 8.2 RC1: -O3 -march=native ....... 556.37 |================================ GCC 8.2 RC1: -O3 -march=native - PGO . 213.98 |============ GCC 9.0 SVN: -O2 ..................... 561.29 |================================ GCC 9.0 SVN: -O3 ..................... 552.08 |=============================== GCC 9.0 SVN: -O3 -march=native ....... 556.37 |================================ GCC 9.0 SVN: -O3 -march=native - PGO . 213.95 |============ SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better GCC 8.2 RC1: -O2 ..................... 2156.61 |=========================== GCC 8.2 RC1: -O3 ..................... 2473.98 |=============================== GCC 8.2 RC1: -O3 -march=native ....... 2421.38 |============================== GCC 8.2 RC1: -O3 -march=native - PGO . 2413.24 |============================== GCC 9.0 SVN: -O2 ..................... 2365.87 |============================== GCC 9.0 SVN: -O3 ..................... 2458.50 |=============================== GCC 9.0 SVN: -O3 -march=native ....... 2458.23 |=============================== GCC 9.0 SVN: -O3 -march=native - PGO . 2400.46 |============================== SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better GCC 8.2 RC1: -O2 ..................... 3462.39 |====================== GCC 8.2 RC1: -O3 ..................... 4683.55 |============================== GCC 8.2 RC1: -O3 -march=native ....... 4831.75 |=============================== GCC 8.2 RC1: -O3 -march=native - PGO . 4865.24 |=============================== GCC 9.0 SVN: -O2 ..................... 3432.88 |====================== GCC 9.0 SVN: -O3 ..................... 4723.98 |============================== GCC 9.0 SVN: -O3 -march=native ....... 4781.45 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 4845.49 |=============================== VP9 libvpx Encoding 1.7.0 vpxenc Frames Per Second > Higher Is Better GCC 8.2 RC1: -O2 ..................... 12.43 |================================= GCC 8.2 RC1: -O3 ..................... 12.36 |================================= GCC 8.2 RC1: -O3 -march=native ....... 12.38 |================================= GCC 8.2 RC1: -O3 -march=native - PGO . 12.29 |================================= GCC 9.0 SVN: -O2 ..................... 12.31 |================================= GCC 9.0 SVN: -O3 ..................... 12.21 |================================ GCC 9.0 SVN: -O3 -march=native ....... 12.37 |================================= GCC 9.0 SVN: -O3 -march=native - PGO . 12.07 |================================ FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 1024 Mflops > Higher Is Better GCC 8.2 RC1: -O2 ............... 7194.37 |=============================== GCC 8.2 RC1: -O3 ............... 7176.77 |=============================== GCC 8.2 RC1: -O3 -march=native . 8518.67 |===================================== GCC 9.0 SVN: -O2 ............... 6852.40 |============================== GCC 9.0 SVN: -O3 ............... 7168.63 |=============================== GCC 9.0 SVN: -O3 -march=native . 8506.83 |===================================== FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 1024 Mflops > Higher Is Better GCC 8.2 RC1: -O2 ............... 5796.33 |================================= GCC 8.2 RC1: -O3 ............... 5835.90 |================================= GCC 8.2 RC1: -O3 -march=native . 6557.83 |===================================== GCC 9.0 SVN: -O2 ............... 5775.23 |================================= GCC 9.0 SVN: -O3 ............... 5601.00 |================================ GCC 9.0 SVN: -O3 -march=native . 6531.37 |===================================== PolyBench-C 4.2 Test: 3 Matrix Multiplications Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 3.62 |================================= GCC 8.2 RC1: -O3 ..................... 3.59 |================================= GCC 8.2 RC1: -O3 -march=native ....... 3.58 |================================= GCC 8.2 RC1: -O3 -march=native - PGO . 3.35 |=============================== GCC 9.0 SVN: -O2 ..................... 3.68 |================================== GCC 9.0 SVN: -O3 ..................... 3.63 |================================== GCC 9.0 SVN: -O3 -march=native ....... 3.66 |================================== GCC 9.0 SVN: -O3 -march=native - PGO . 3.30 |============================== PolyBench-C 4.2 Test: Correlation Computation Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 5.55 |================================== GCC 8.2 RC1: -O3 ..................... 5.40 |================================= GCC 8.2 RC1: -O3 -march=native ....... 5.34 |================================= GCC 8.2 RC1: -O3 -march=native - PGO . 5.33 |================================= GCC 9.0 SVN: -O2 ..................... 5.50 |================================== GCC 9.0 SVN: -O3 ..................... 5.46 |================================= GCC 9.0 SVN: -O3 -march=native ....... 5.38 |================================= GCC 9.0 SVN: -O3 -march=native - PGO . 5.40 |================================= PolyBench-C 4.2 Test: Covariance Computation Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 5.53 |================================== GCC 8.2 RC1: -O3 ..................... 5.39 |================================= GCC 8.2 RC1: -O3 -march=native ....... 5.34 |================================= GCC 8.2 RC1: -O3 -march=native - PGO . 5.42 |================================= GCC 9.0 SVN: -O2 ..................... 5.50 |================================== GCC 9.0 SVN: -O3 ..................... 5.49 |================================== GCC 9.0 SVN: -O3 -march=native ....... 5.41 |================================= GCC 9.0 SVN: -O3 -march=native - PGO . 5.37 |================================= SQLite 3.22 Timed SQLite Insertions Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 2.87 |======================================== GCC 8.2 RC1: -O3 ............... 2.88 |======================================== GCC 8.2 RC1: -O3 -march=native . 2.89 |======================================== GCC 9.0 SVN: -O2 ............... 2.86 |======================================== GCC 9.0 SVN: -O3 ............... 2.85 |======================================= GCC 9.0 SVN: -O3 -march=native . 2.89 |======================================== OpenSSL 1.1.0f RSA 4096-bit Performance Signs Per Second > Higher Is Better GCC 8.2 RC1: -O2 ..................... 4609.30 |=============================== GCC 8.2 RC1: -O3 ..................... 4609.73 |=============================== GCC 8.2 RC1: -O3 -march=native ....... 4610.63 |=============================== GCC 8.2 RC1: -O3 -march=native - PGO . 4617.23 |=============================== GCC 9.0 SVN: -O2 ..................... 4612.20 |=============================== GCC 9.0 SVN: -O3 ..................... 4602.30 |=============================== GCC 9.0 SVN: -O3 -march=native ....... 4616.47 |=============================== GCC 9.0 SVN: -O3 -march=native - PGO . 4623.60 |=============================== AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 55.52 |================================= GCC 8.2 RC1: -O3 ..................... 52.60 |=============================== GCC 8.2 RC1: -O3 -march=native ....... 50.80 |============================== GCC 8.2 RC1: -O3 -march=native - PGO . 49.05 |============================= GCC 9.0 SVN: -O2 ..................... 55.52 |================================= GCC 9.0 SVN: -O3 ..................... 53.03 |================================ GCC 9.0 SVN: -O3 -march=native ....... 50.72 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 48.51 |============================= 7-Zip Compression 16.02 Compress Speed Test MIPS > Higher Is Better GCC 8.2 RC1: -O2 ..................... 98480 |=============================== GCC 8.2 RC1: -O3 ..................... 97917 |=============================== GCC 8.2 RC1: -O3 -march=native ....... 100684 |================================ GCC 8.2 RC1: -O3 -march=native - PGO . 99762 |================================ GCC 9.0 SVN: -O2 ..................... 98423 |=============================== GCC 9.0 SVN: -O3 ..................... 97663 |=============================== GCC 9.0 SVN: -O3 -march=native ....... 97385 |=============================== GCC 9.0 SVN: -O3 -march=native - PGO . 96774 |=============================== Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 166.12 |================================ GCC 8.2 RC1: -O3 ..................... 166.70 |================================ GCC 8.2 RC1: -O3 -march=native ....... 155.64 |============================== GCC 9.0 SVN: -O2 ..................... 168.14 |================================ GCC 9.0 SVN: -O3 ..................... 168.20 |================================ GCC 9.0 SVN: -O3 -march=native ....... 156.58 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 147.78 |============================ m-queens 1.1 Time To Solve Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 32.22 |================================= GCC 8.2 RC1: -O3 ..................... 32.09 |================================= GCC 8.2 RC1: -O3 -march=native ....... 32.02 |================================= GCC 8.2 RC1: -O3 -march=native - PGO . 32.44 |================================= GCC 9.0 SVN: -O2 ..................... 31.59 |================================ GCC 9.0 SVN: -O3 ..................... 31.64 |================================ GCC 9.0 SVN: -O3 -march=native ....... 31.56 |================================ GCC 9.0 SVN: -O3 -march=native - PGO . 32.31 |================================= C-Ray 1.1 Total Time Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 6.00 |================================== GCC 8.2 RC1: -O3 ..................... 2.84 |================ GCC 8.2 RC1: -O3 -march=native ....... 2.52 |============== GCC 8.2 RC1: -O3 -march=native - PGO . 2.51 |============== GCC 9.0 SVN: -O2 ..................... 6.02 |================================== GCC 9.0 SVN: -O3 ..................... 2.83 |================ GCC 9.0 SVN: -O3 -march=native ....... 2.52 |============== GCC 9.0 SVN: -O3 -march=native - PGO . 2.50 |============== Crafty 25.2 Elapsed Time Nodes Per Second > Higher Is Better GCC 8.2 RC1: -O2 ..................... 6004700 |=============================== GCC 8.2 RC1: -O3 ..................... 5989207 |=============================== GCC 8.2 RC1: -O3 -march=native ....... 6014866 |=============================== GCC 8.2 RC1: -O3 -march=native - PGO . 6026655 |=============================== GCC 9.0 SVN: -O2 ..................... 5780824 |============================== GCC 9.0 SVN: -O3 ..................... 5892825 |============================== GCC 9.0 SVN: -O3 -march=native ....... 5876297 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 5841423 |============================== Smallpt 1.0 Global Illumination Renderer; 100 Samples Seconds < Lower Is Better GCC 8.2 RC1: -O2 ..................... 4 |===================================== GCC 8.2 RC1: -O3 ..................... 4 |===================================== GCC 8.2 RC1: -O3 -march=native ....... 4 |===================================== GCC 8.2 RC1: -O3 -march=native - PGO . 3 |============================ GCC 9.0 SVN: -O2 ..................... 4 |===================================== GCC 9.0 SVN: -O3 ..................... 3 |============================ GCC 9.0 SVN: -O3 -march=native ....... 4 |===================================== GCC 9.0 SVN: -O3 -march=native - PGO . 4 |===================================== Stockfish 9 Total Time Nodes Per Second > Higher Is Better GCC 8.2 RC1: -O2 ..................... 59007314 |============================= GCC 8.2 RC1: -O3 ..................... 59837512 |============================= GCC 8.2 RC1: -O3 -march=native ....... 59730290 |============================= GCC 8.2 RC1: -O3 -march=native - PGO . 61080869 |============================== GCC 9.0 SVN: -O2 ..................... 58372089 |============================= GCC 9.0 SVN: -O3 ..................... 58412975 |============================= GCC 9.0 SVN: -O3 -march=native ....... 59201545 |============================= GCC 9.0 SVN: -O3 -march=native - PGO . 56149550 |============================ TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better GCC 8.2 RC1: -O2 ..................... 852597 |============================== GCC 8.2 RC1: -O3 ..................... 851283 |============================== GCC 8.2 RC1: -O3 -march=native ....... 851019 |============================== GCC 8.2 RC1: -O3 -march=native - PGO . 919193 |================================ GCC 9.0 SVN: -O2 ..................... 856576 |============================== GCC 9.0 SVN: -O3 ..................... 856577 |============================== GCC 9.0 SVN: -O3 -march=native ....... 856576 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 916136 |================================ Redis 4.0.8 Test: SET Requests Per Second > Higher Is Better GCC 8.2 RC1: -O2 ............... 1278916.54 |================================ GCC 8.2 RC1: -O3 ............... 1332541.96 |================================= GCC 8.2 RC1: -O3 -march=native . 1375161.96 |================================== GCC 9.0 SVN: -O2 ............... 1319138.27 |================================= GCC 9.0 SVN: -O3 ............... 1291788.21 |================================ GCC 9.0 SVN: -O3 -march=native . 1284845.13 |================================ Redis 4.0.8 Test: GET Requests Per Second > Higher Is Better GCC 8.2 RC1: -O2 ............... 1612387.56 |============================== GCC 8.2 RC1: -O3 ............... 1813896.31 |================================== GCC 8.2 RC1: -O3 -march=native . 1691908.00 |================================ GCC 9.0 SVN: -O2 ............... 1580690.66 |============================== GCC 9.0 SVN: -O3 ............... 1682310.23 |================================ GCC 9.0 SVN: -O3 -march=native . 1586367.58 |============================== Redis 4.0.8 Test: LPUSH Requests Per Second > Higher Is Better GCC 8.2 RC1: -O2 ............... 1135840.67 |================================= GCC 8.2 RC1: -O3 ............... 1187656.29 |================================== GCC 8.2 RC1: -O3 -march=native . 1108304.54 |================================ GCC 9.0 SVN: -O2 ............... 1130545.22 |================================ GCC 9.0 SVN: -O3 ............... 1135535.00 |================================= GCC 9.0 SVN: -O3 -march=native . 1133637.46 |================================ Redis 4.0.8 Test: LPOP Requests Per Second > Higher Is Better GCC 8.2 RC1: -O2 ............... 1117326.83 |================================= GCC 8.2 RC1: -O3 ............... 1143259.50 |================================== GCC 8.2 RC1: -O3 -march=native . 1108268.08 |================================= GCC 9.0 SVN: -O2 ............... 1078502.41 |================================ GCC 9.0 SVN: -O3 ............... 1103652.21 |================================= GCC 9.0 SVN: -O3 -march=native . 1152018.46 |================================== Redis 4.0.8 Test: SADD Requests Per Second > Higher Is Better GCC 8.2 RC1: -O2 ............... 1500041.04 |================================= GCC 8.2 RC1: -O3 ............... 1517826.77 |================================= GCC 8.2 RC1: -O3 -march=native . 1430638.45 |================================ GCC 9.0 SVN: -O2 ............... 1542587.25 |================================== GCC 9.0 SVN: -O3 ............... 1477048.94 |================================= GCC 9.0 SVN: -O3 -march=native . 1423169.79 |=============================== PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Write TPS > Higher Is Better GCC 8.2 RC1: -O2 ............... 1991.71 |=================================== GCC 8.2 RC1: -O3 ............... 2044.30 |==================================== GCC 8.2 RC1: -O3 -march=native . 2071.78 |===================================== GCC 9.0 SVN: -O2 ............... 1973.07 |=================================== GCC 9.0 SVN: -O3 ............... 2093.81 |===================================== GCC 9.0 SVN: -O3 -march=native . 2015.18 |==================================== PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Only TPS > Higher Is Better GCC 8.2 RC1: -O2 ............... 14037.15 |=================================== GCC 8.2 RC1: -O3 ............... 13985.70 |=================================== GCC 8.2 RC1: -O3 -march=native . 14374.13 |==================================== GCC 9.0 SVN: -O2 ............... 14101.79 |=================================== GCC 9.0 SVN: -O3 ............... 14004.08 |=================================== GCC 9.0 SVN: -O3 -march=native . 14307.54 |==================================== PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write TPS > Higher Is Better GCC 8.2 RC1: -O2 ............... 36523.78 |==================================== GCC 8.2 RC1: -O3 ............... 36248.65 |==================================== GCC 8.2 RC1: -O3 -march=native . 36226.69 |==================================== GCC 9.0 SVN: -O2 ............... 36232.44 |==================================== GCC 9.0 SVN: -O3 ............... 36081.57 |==================================== GCC 9.0 SVN: -O3 -march=native . 36115.94 |==================================== PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only TPS > Higher Is Better GCC 8.2 RC1: -O2 ............... 396925.91 |================================== GCC 8.2 RC1: -O3 ............... 408599.13 |=================================== GCC 8.2 RC1: -O3 -march=native . 408873.25 |=================================== GCC 9.0 SVN: -O2 ............... 397574.74 |================================== GCC 9.0 SVN: -O3 ............... 395454.61 |================================== GCC 9.0 SVN: -O3 -march=native . 401369.61 |================================== libjpeg-turbo tjbench 1.5.3 Test: Decompression Throughput Megapixels/sec > Higher Is Better GCC 8.2 RC1: -O2 ............... 141.43 |===================================== GCC 8.2 RC1: -O3 ............... 141.74 |===================================== GCC 8.2 RC1: -O3 -march=native . 142.18 |===================================== GCC 9.0 SVN: -O2 ............... 141.18 |===================================== GCC 9.0 SVN: -O3 ............... 137.82 |==================================== GCC 9.0 SVN: -O3 -march=native . 145.66 |====================================== Apache Benchmark 2.4.29 Static Web Page Serving Requests Per Second > Higher Is Better GCC 8.2 RC1: -O2 ............... 15972.59 |=================================== GCC 8.2 RC1: -O3 ............... 15891.72 |=================================== GCC 8.2 RC1: -O3 -march=native . 16205.32 |==================================== GCC 9.0 SVN: -O2 ............... 15943.49 |=================================== GCC 9.0 SVN: -O3 ............... 16065.26 |==================================== GCC 9.0 SVN: -O3 -march=native . 15856.03 |=================================== LAME MP3 Encoding 3.100 WAV To MP3 Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 14.05 |====================================== GCC 8.2 RC1: -O3 ............... 11.13 |============================== GCC 8.2 RC1: -O3 -march=native . 10.51 |============================= GCC 9.0 SVN: -O2 ............... 14.28 |======================================= GCC 9.0 SVN: -O3 ............... 11.00 |============================== GCC 9.0 SVN: -O3 -march=native . 10.50 |============================= FLAC Audio Encoding 1.3.2 WAV To FLAC Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 12.15 |====================================== GCC 8.2 RC1: -O3 ............... 12.27 |====================================== GCC 8.2 RC1: -O3 -march=native . 12.63 |======================================= GCC 9.0 SVN: -O2 ............... 12.22 |====================================== GCC 9.0 SVN: -O3 ............... 12.13 |===================================== GCC 9.0 SVN: -O3 -march=native . 12.44 |====================================== ebizzy 0.3 Records/s > Higher Is Better GCC 8.2 RC1: -O2 ..................... 1106391 |============================= GCC 8.2 RC1: -O3 ..................... 1113356 |============================== GCC 8.2 RC1: -O3 -march=native ....... 1147009 |============================== GCC 8.2 RC1: -O3 -march=native - PGO . 1112448 |============================== GCC 9.0 SVN: -O2 ..................... 1124222 |============================== GCC 9.0 SVN: -O3 ..................... 1109148 |============================= GCC 9.0 SVN: -O3 -march=native ....... 1111662 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 1166998 |=============================== GraphicsMagick 1.3.28 Operation: HWB Color Space Iterations Per Minute > Higher Is Better GCC 8.2 RC1: -O2 ............... 202 |======================================== GCC 8.2 RC1: -O3 ............... 199 |======================================= GCC 8.2 RC1: -O3 -march=native . 209 |========================================= GCC 9.0 SVN: -O2 ............... 197 |======================================= GCC 9.0 SVN: -O3 ............... 201 |======================================= GCC 9.0 SVN: -O3 -march=native . 201 |======================================= GraphicsMagick 1.3.28 Operation: Blur Iterations Per Minute > Higher Is Better GCC 8.2 RC1: -O2 ............... 148 |======================================= GCC 8.2 RC1: -O3 ............... 147 |======================================= GCC 8.2 RC1: -O3 -march=native . 154 |========================================= GCC 9.0 SVN: -O2 ............... 146 |======================================= GCC 9.0 SVN: -O3 ............... 148 |======================================= GCC 9.0 SVN: -O3 -march=native . 149 |======================================== GraphicsMagick 1.3.28 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better GCC 8.2 RC1: -O2 ............... 109 |========================================= GCC 8.2 RC1: -O3 ............... 108 |======================================== GCC 8.2 RC1: -O3 -march=native . 110 |========================================= GCC 9.0 SVN: -O2 ............... 102 |====================================== GCC 9.0 SVN: -O3 ............... 104 |======================================= GCC 9.0 SVN: -O3 -march=native . 104 |======================================= GraphicsMagick 1.3.28 Operation: Resizing Iterations Per Minute > Higher Is Better GCC 8.2 RC1: -O2 ............... 181 |======================================== GCC 8.2 RC1: -O3 ............... 175 |======================================= GCC 8.2 RC1: -O3 -march=native . 185 |========================================= GCC 9.0 SVN: -O2 ............... 176 |======================================= GCC 9.0 SVN: -O3 ............... 179 |======================================== GCC 9.0 SVN: -O3 -march=native . 180 |======================================== GraphicsMagick 1.3.28 Operation: Sharpen Iterations Per Minute > Higher Is Better GCC 8.2 RC1: -O2 ............... 181 |======================================= GCC 8.2 RC1: -O3 ............... 177 |======================================= GCC 8.2 RC1: -O3 -march=native . 188 |========================================= GCC 9.0 SVN: -O2 ............... 176 |====================================== GCC 9.0 SVN: -O3 ............... 179 |======================================= GCC 9.0 SVN: -O3 -march=native . 182 |======================================== Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 7.56 |======================================= GCC 8.2 RC1: -O3 ............... 7.71 |======================================== GCC 8.2 RC1: -O3 -march=native . 7.65 |======================================== GCC 9.0 SVN: -O2 ............... 7.40 |====================================== GCC 9.0 SVN: -O3 ............... 7.74 |======================================== GCC 9.0 SVN: -O3 -march=native . 7.63 |======================================= Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 5.06 |======================================= GCC 8.2 RC1: -O3 ............... 5.06 |======================================= GCC 8.2 RC1: -O3 -march=native . 5.02 |======================================= GCC 9.0 SVN: -O2 ............... 5.06 |======================================= GCC 9.0 SVN: -O3 ............... 5.16 |======================================== GCC 9.0 SVN: -O3 -march=native . 5.04 |======================================= Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 6.00 |======================================== GCC 8.2 RC1: -O3 ............... 6.02 |======================================== GCC 8.2 RC1: -O3 -march=native . 5.79 |====================================== GCC 9.0 SVN: -O2 ............... 5.99 |======================================== GCC 9.0 SVN: -O3 ............... 6.01 |======================================== GCC 9.0 SVN: -O3 -march=native . 5.78 |====================================== Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 3.14 |======================================= GCC 8.2 RC1: -O3 ............... 3.14 |======================================= GCC 8.2 RC1: -O3 -march=native . 3.09 |====================================== GCC 9.0 SVN: -O2 ............... 3.13 |======================================= GCC 9.0 SVN: -O3 ............... 3.22 |======================================== GCC 9.0 SVN: -O3 -march=native . 3.08 |====================================== Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 5.51 |======================================== GCC 8.2 RC1: -O3 ............... 5.52 |======================================== GCC 8.2 RC1: -O3 -march=native . 5.32 |====================================== GCC 9.0 SVN: -O2 ............... 5.49 |======================================== GCC 9.0 SVN: -O3 ............... 5.53 |======================================== GCC 9.0 SVN: -O3 -march=native . 5.29 |====================================== Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 1.12 |======================================== GCC 8.2 RC1: -O3 ............... 1.12 |======================================== GCC 8.2 RC1: -O3 -march=native . 1.13 |======================================== GCC 9.0 SVN: -O2 ............... 1.12 |======================================== GCC 9.0 SVN: -O3 ............... 1.12 |======================================== GCC 9.0 SVN: -O3 -march=native . 1.13 |======================================== Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 1.36 |======================================== GCC 8.2 RC1: -O3 ............... 1.36 |======================================== GCC 8.2 RC1: -O3 -march=native . 1.33 |======================================= GCC 9.0 SVN: -O2 ............... 1.36 |======================================== GCC 9.0 SVN: -O3 ............... 1.36 |======================================== GCC 9.0 SVN: -O3 -march=native . 1.33 |======================================= Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better GCC 8.2 RC1: -O2 ............... 3.17 |======================================== GCC 8.2 RC1: -O3 ............... 3.18 |======================================== GCC 8.2 RC1: -O3 -march=native . 3.11 |======================================= GCC 9.0 SVN: -O2 ............... 3.17 |======================================== GCC 9.0 SVN: -O3 ............... 3.17 |======================================== GCC 9.0 SVN: -O3 -march=native . 3.12 |======================================= TTSIOD 3D Renderer 2.3a Phong Rendering With Soft-Shadow Mapping FPS > Higher Is Better GCC 8.2 RC1: -O2 ............... 400.36 |================================== GCC 8.2 RC1: -O3 ............... 441.58 |====================================== GCC 8.2 RC1: -O3 -march=native . 439.54 |====================================== Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better GCC 8.2 RC1: -O2 ..................... 1014.66 |=============================== GCC 8.2 RC1: -O3 ..................... 998.89 |=============================== GCC 8.2 RC1: -O3 -march=native ....... 982.67 |============================== GCC 8.2 RC1: -O3 -march=native - PGO . 852.63 |========================== GCC 9.0 SVN: -O2 ..................... 1014.38 |=============================== GCC 9.0 SVN: -O3 ..................... 993.22 |============================== GCC 9.0 SVN: -O3 -march=native ....... 1007.74 |=============================== GCC 9.0 SVN: -O3 -march=native - PGO . 869.07 |=========================== Tinymembench 2018-05-28 Standard Memcpy MB/s > Higher Is Better GCC 8.2 RC1: -O2 ..................... 12359.30 |============================== GCC 8.2 RC1: -O3 ..................... 12555.70 |============================== GCC 8.2 RC1: -O3 -march=native ....... 12497.20 |============================== GCC 8.2 RC1: -O3 -march=native - PGO . 12453.20 |============================== GCC 9.0 SVN: -O2 ..................... 12422.90 |============================== GCC 9.0 SVN: -O3 ..................... 12490.00 |============================== GCC 9.0 SVN: -O3 -march=native ....... 12377.60 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 12300.60 |============================= Tinymembench 2018-05-28 Standard Memset MB/s > Higher Is Better GCC 8.2 RC1: -O2 ..................... 9978.60 |============================== GCC 8.2 RC1: -O3 ..................... 10037.70 |============================== GCC 8.2 RC1: -O3 -march=native ....... 10063.10 |============================== GCC 8.2 RC1: -O3 -march=native - PGO . 9975.20 |============================== GCC 9.0 SVN: -O2 ..................... 9989.40 |============================== GCC 9.0 SVN: -O3 ..................... 10039.70 |============================== GCC 9.0 SVN: -O3 -march=native ....... 9982.80 |============================== GCC 9.0 SVN: -O3 -march=native - PGO . 9982.90 |============================== Sockperf 3.4 Test: Latency Under Load usec < Lower Is Better GCC 8.2 RC1: -O2 ............... 19.59 |================================== GCC 8.2 RC1: -O3 ............... 19.67 |================================== GCC 8.2 RC1: -O3 -march=native . 19.75 |================================== GCC 9.0 SVN: -O2 ............... 19.75 |================================== GCC 9.0 SVN: -O3 ............... 22.43 |======================================= GCC 9.0 SVN: -O3 -march=native . 19.88 |=================================== Sockperf 3.4 Test: Latency Ping Pong usec < Lower Is Better GCC 8.2 RC1: -O2 ............... 6.66 |======================================= GCC 8.2 RC1: -O3 ............... 6.76 |======================================== GCC 8.2 RC1: -O3 -march=native . 6.42 |====================================== GCC 9.0 SVN: -O2 ............... 6.53 |======================================= GCC 9.0 SVN: -O3 ............... 6.57 |=======================================