Skylake Xeon GCC vs. Clang Compiler Optimization Tests Intel Broadwell-DE 1587 GCC: -O0: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: GCC 5.3.1 20160222, File-System: ext4, Screen Resolution: 3840x2160 GCC: -O1: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: GCC 5.3.1 20160222, File-System: ext4, Screen Resolution: 3840x2160 GCC: -O2: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: GCC 5.3.1 20160222, File-System: ext4, Screen Resolution: 3840x2160 GCC: -O2 -march=native: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: GCC 5.3.1 20160222, File-System: ext4, Screen Resolution: 3840x2160 GCC: -O3: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: GCC 5.3.1 20160222, File-System: ext4, Screen Resolution: 3840x2160 GCC: -O3 -march=native: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: GCC 5.3.1 20160222, File-System: ext4, Screen Resolution: 3840x2160 GCC: -Ofast -march=native: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: GCC 5.3.1 20160222, File-System: ext4, Screen Resolution: 3840x2160 Clang: -O0: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: Clang 3.8.0-+rc2-1~exp1ubuntu2, File-System: ext4, Screen Resolution: 3840x2160 Clang: -O1: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: Clang 3.8.0-+rc2-1~exp1ubuntu2, File-System: ext4, Screen Resolution: 3840x2160 Clang: -O2: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: Clang 3.8.0-+rc2-1~exp1ubuntu2, File-System: ext4, Screen Resolution: 3840x2160 Clang: -O2 -march=native: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: Clang 3.8.0-+rc2-1~exp1ubuntu2, File-System: ext4, Screen Resolution: 3840x2160 Clang: -O3: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: Clang 3.8.0-+rc2-1~exp1ubuntu2, File-System: ext4, Screen Resolution: 3840x2160 Clang: -O3 -march=native: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: Clang 3.8.0-+rc2-1~exp1ubuntu2, File-System: ext4, Screen Resolution: 3840x2160 Clang: -Ofast -march=native: Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: 120GB Samsung SSD 850, Graphics: MSI AMD Radeon R7 370 / R9 270/370 OEM 4096MB, Audio: Realtek ALC1150, Monitor: DELL P2415Q, Network: Intel Connection OS: Ubuntu 16.04, Kernel: 4.4.0-7-generic (x86_64), Desktop: Unity 7.4.0, Display Server: X Server 1.17.3, Display Driver: radeon 7.6.1, OpenGL: 4.1 Mesa 11.1.2 Gallium 0.4, Compiler: Clang 3.8.0-+rc2-1~exp1ubuntu2, File-System: ext4, Screen Resolution: 3840x2160 GCC 4.8.5 20150623: Processor: 06/56 @ 2.30GHz (32 Cores), Motherboard: Supermicro X10SDV-7TP8F v1.01, Chipset: Intel Xeon E7 v4/Xeon, Memory: 2 x 8192 MB DDR4-2133MHz Micron, Disk: 240GB Toshiba MKNSSDAT240GB-DX, Graphics: ASPEED ASPEED Family, Network: Intel Connection X552 10 GbE SFP+ OS: Red Hat Enterprise Linux 7.2, Kernel: 3.10.0-327.el7.x86_64 (x86_64), Compiler: GCC 4.8.5 20150623, File-System: xfs, Screen Resolution: 1024x768 Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better GCC: -O0 .................... 26.89 |========================================== GCC: -O1 .................... 8.86 |============== GCC: -O2 .................... 8.35 |============= GCC: -O2 -march=native ...... 8.35 |============= GCC: -O3 .................... 8.17 |============= GCC: -O3 -march=native ...... 8.16 |============= GCC: -Ofast -march=native ... 8.00 |============ Clang: -O0 .................. 23.23 |==================================== Clang: -O1 .................. 8.97 |============== Clang: -O2 .................. 8.29 |============= Clang: -O2 -march=native .... 8.29 |============= Clang: -O3 .................. 8.40 |============= Clang: -O3 -march=native .... 8.29 |============= Clang: -Ofast -march=native . 8.12 |============= GCC 4.8.5 20150623 .......... 9.14 |============== GraphicsMagick 1.3.19 Operation: Blur Iterations Per Minute > Higher Is Better GCC: -O0 .................... 107 |========================== GCC: -O1 .................... 170 |========================================== GCC: -O2 .................... 169 |========================================== GCC: -O2 -march=native ...... 173 |=========================================== GCC: -O3 .................... 170 |========================================== GCC: -O3 -march=native ...... 178 |============================================ GCC: -Ofast -march=native ... 177 |============================================ Clang: -O0 .................. 75 |=================== Clang: -O1 .................. 150 |===================================== Clang: -O2 .................. 150 |===================================== Clang: -O2 -march=native .... 150 |===================================== Clang: -O3 .................. 150 |===================================== Clang: -O3 -march=native .... 153 |====================================== Clang: -Ofast -march=native . 153 |====================================== GCC 4.8.5 20150623 .......... 113 |============================ GraphicsMagick 1.3.19 Operation: HWB Color Space Iterations Per Minute > Higher Is Better GCC: -O0 .................... 136 |========================= GCC: -O1 .................... 226 |========================================== GCC: -O2 .................... 213 |======================================== GCC: -O2 -march=native ...... 228 |========================================== GCC: -O3 .................... 228 |========================================== GCC: -O3 -march=native ...... 234 |=========================================== GCC: -Ofast -march=native ... 237 |============================================ Clang: -O0 .................. 100 |=================== Clang: -O1 .................. 192 |==================================== Clang: -O2 .................. 193 |==================================== Clang: -O2 -march=native .... 196 |==================================== Clang: -O3 .................. 192 |==================================== Clang: -O3 -march=native .... 196 |==================================== Clang: -Ofast -march=native . 198 |===================================== GCC 4.8.5 20150623 .......... 145 |=========================== GraphicsMagick 1.3.19 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better GCC: -O0 .................... 22 |========= GCC: -O1 .................... 92 |======================================= GCC: -O2 .................... 100 |========================================== GCC: -O2 -march=native ...... 103 |============================================ GCC: -O3 .................... 103 |============================================ GCC: -O3 -march=native ...... 104 |============================================ GCC: -Ofast -march=native ... 104 |============================================ Clang: -O0 .................. 22 |========= Clang: -O1 .................. 89 |====================================== Clang: -O2 .................. 100 |========================================== Clang: -O2 -march=native .... 99 |========================================== Clang: -O3 .................. 100 |========================================== Clang: -O3 -march=native .... 102 |=========================================== Clang: -Ofast -march=native . 101 |=========================================== GCC 4.8.5 20150623 .......... 59 |========================= Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better GCC: -O0 .................... 496.06 |======= GCC: -O1 .................... 1302.83 |=================== GCC: -O2 .................... 2429.37 |=================================== GCC: -O2 -march=native ...... 2653.07 |====================================== GCC: -O3 .................... 2485.14 |==================================== GCC: -O3 -march=native ...... 2689.91 |======================================= GCC: -Ofast -march=native ... 2757.77 |======================================== Clang: -O0 .................. 325.27 |===== Clang: -O1 .................. 1931.06 |============================ Clang: -O2 .................. 2036.00 |============================== Clang: -O2 -march=native .... 1923.64 |============================ Clang: -O3 .................. 2001.77 |============================= Clang: -O3 -march=native .... 1962.71 |============================ Clang: -Ofast -march=native . 2114.80 |=============================== GCC 4.8.5 20150623 .......... 1343.84 |=================== Timed Apache Compilation 2.4.7 Time To Compile Seconds < Lower Is Better GCC: -O0 .................... 11.08 |================ GCC: -O1 .................... 17.00 |======================== GCC: -O2 .................... 23.12 |================================= GCC: -O2 -march=native ...... 22.84 |================================ GCC: -O3 .................... 24.88 |=================================== GCC: -O3 -march=native ...... 25.17 |=================================== GCC: -Ofast -march=native ... 25.16 |=================================== Clang: -O0 .................. 10.93 |=============== Clang: -O1 .................. 18.85 |=========================== Clang: -O2 .................. 22.36 |=============================== Clang: -O2 -march=native .... 23.10 |================================ Clang: -O3 .................. 22.83 |================================ Clang: -O3 -march=native .... 23.07 |================================ Clang: -Ofast -march=native . 23.15 |================================= GCC 4.8.5 20150623 .......... 29.87 |========================================== Timed ImageMagick Compilation 6.9.0 Time To Compile Seconds < Lower Is Better GCC: -O0 .................... 12.23 |========== GCC: -O1 .................... 24.96 |===================== GCC: -O2 .................... 39.67 |================================= GCC: -O2 -march=native ...... 37.57 |================================ GCC: -O3 .................... 49.70 |========================================== GCC: -O3 -march=native ...... 48.97 |========================================= GCC: -Ofast -march=native ... 49.99 |========================================== Clang: -O0 .................. 9.78 |======== Clang: -O1 .................. 31.07 |========================== Clang: -O2 .................. 36.53 |=============================== Clang: -O2 -march=native .... 38.45 |================================ Clang: -O3 .................. 49.59 |========================================== Clang: -O3 -march=native .... 49.63 |========================================== Clang: -Ofast -march=native . 50.02 |========================================== GCC 4.8.5 20150623 .......... 32.26 |=========================== Timed PHP Compilation 5.2.9 Time To Compile Seconds < Lower Is Better GCC: -O0 .................... 7.27 |============== GCC: -O1 .................... 13.35 |========================== GCC: -O2 .................... 19.12 |===================================== GCC: -O2 -march=native ...... 18.68 |==================================== GCC: -O3 .................... 21.21 |========================================= GCC: -O3 -march=native ...... 21.68 |========================================== GCC: -Ofast -march=native ... 21.75 |========================================== Clang: -O0 .................. 6.44 |============ Clang: -O1 .................. 13.58 |========================== Clang: -O2 .................. 16.96 |================================= Clang: -O2 -march=native .... 17.92 |=================================== Clang: -O3 .................. 17.59 |================================== Clang: -O3 -march=native .... 17.90 |=================================== Clang: -Ofast -march=native . 17.95 |=================================== GCC 4.8.5 20150623 .......... 19.19 |===================================== C-Ray 1.1 Total Time Seconds < Lower Is Better GCC: -O0 .................... 61.53 |========================================== GCC: -O1 .................... 43.15 |============================= GCC: -O2 .................... 38.19 |========================== GCC: -O2 -march=native ...... 27.83 |=================== GCC: -O3 .................... 19.65 |============= GCC: -O3 -march=native ...... 14.60 |========== GCC: -Ofast -march=native ... 13.92 |========== Clang: -O0 .................. 57.09 |======================================= Clang: -O1 .................. 32.34 |====================== Clang: -O2 .................. 37.29 |========================= Clang: -O2 -march=native .... 36.22 |========================= Clang: -O3 .................. 25.61 |================= Clang: -O3 -march=native .... 25.20 |================= Clang: -Ofast -march=native . 20.64 |============== GCC 4.8.5 20150623 .......... 10.10 |======= FLAC Audio Encoding 1.3.1 WAV To FLAC Seconds < Lower Is Better GCC: -O0 .................... 37.56 |================================= GCC: -O1 .................... 5.97 |===== GCC: -O2 .................... 5.22 |===== GCC: -O2 -march=native ...... 4.87 |==== GCC: -O3 .................... 5.10 |==== GCC: -O3 -march=native ...... 4.88 |==== GCC: -Ofast -march=native ... 4.89 |==== Clang: -O0 .................. 47.95 |========================================== Clang: -O1 .................. 6.97 |====== Clang: -O2 .................. 6.66 |====== Clang: -O2 -march=native .... 5.26 |===== Clang: -O3 .................. 6.71 |====== Clang: -O3 -march=native .... 5.11 |==== Clang: -Ofast -march=native . 5.05 |==== GCC 4.8.5 20150623 .......... 11.90 |========== LAME MP3 Encoding 3.99.3 WAV To MP3 Seconds < Lower Is Better GCC: -O0 .................... 30.50 |======================================== GCC: -O1 .................... 13.25 |================= GCC: -O2 .................... 12.90 |================= GCC: -O2 -march=native ...... 11.99 |================ GCC: -O3 .................... 12.10 |================ GCC: -O3 -march=native ...... 10.69 |============== GCC: -Ofast -march=native ... 9.51 |============= Clang: -O0 .................. 31.95 |========================================== Clang: -O1 .................. 12.04 |================ Clang: -O2 .................. 12.17 |================ Clang: -O2 -march=native .... 12.03 |================ Clang: -O3 .................. 12.16 |================ Clang: -O3 -march=native .... 11.66 |=============== Clang: -Ofast -march=native . 11.10 |=============== GCC 4.8.5 20150623 .......... 21.71 |============================= Redis 3.0.1 Test: GET Requests Per Second > Higher Is Better GCC: -O0 .................... 1663526.96 |==================== GCC: -O1 .................... 2953039.46 |=================================== GCC: -O2 .................... 3007445.96 |==================================== GCC: -O2 -march=native ...... 2951037.21 |=================================== GCC: -O3 .................... 2910194.79 |=================================== GCC: -O3 -march=native ...... 3116395.33 |===================================== GCC: -Ofast -march=native ... 3023321.58 |==================================== Clang: -O0 .................. 1864548.29 |====================== Clang: -O1 .................. 3067218.96 |==================================== Clang: -O2 .................. 2962310.25 |=================================== Clang: -O2 -march=native .... 3102899.83 |===================================== Clang: -O3 .................. 2923813.88 |=================================== Clang: -O3 -march=native .... 2862656.75 |================================== Clang: -Ofast -march=native . 3100373.25 |===================================== GCC 4.8.5 20150623 .......... 1363657.21 |================ Redis 3.0.1 Test: SET Requests Per Second > Higher Is Better GCC: -O0 .................... 1189110.86 |==================== GCC: -O1 .................... 2175575.75 |==================================== GCC: -O2 .................... 2136957.54 |=================================== GCC: -O2 -march=native ...... 2076614.15 |================================== GCC: -O3 .................... 2183433.83 |==================================== GCC: -O3 -march=native ...... 2104521.04 |=================================== GCC: -Ofast -march=native ... 2243907.58 |===================================== Clang: -O0 .................. 1263741.85 |===================== Clang: -O1 .................. 1977929.48 |================================= Clang: -O2 .................. 2127555.19 |=================================== Clang: -O2 -march=native .... 2064604.79 |================================== Clang: -O3 .................. 1962280.52 |================================ Clang: -O3 -march=native .... 2037699.96 |================================== Clang: -Ofast -march=native . 2078148.71 |================================== GCC 4.8.5 20150623 .......... 1078008.79 |================== Hierarchical INTegration 1.0 Test: FLOAT QUIPs > Higher Is Better GCC: -O0 .................... 127505415.48 |=========== GCC: -O1 .................... 295705864.24 |========================= GCC: -O2 .................... 370257620.55 |================================ GCC: -O2 -march=native ...... 408904948.74 |=================================== GCC: -O3 .................... 381719210.59 |================================= GCC: -O3 -march=native ...... 407575756.08 |=================================== GCC: -Ofast -march=native ... 392373251.20 |================================== Clang: -O0 .................. 129742642.75 |=========== Clang: -O1 .................. 335679881.79 |============================= Clang: -O2 .................. 387996153.74 |================================= Clang: -O2 -march=native .... 288801639.39 |========================= Clang: -O3 .................. 386593550.68 |================================= Clang: -O3 -march=native .... 296846752.57 |========================= Clang: -Ofast -march=native . 295042490.70 |========================= GCC 4.8.5 20150623 .......... 241905166.12 |=====================