FreeBSD 9.1 Benchmarks Of GCC vs. LLVM/Clang Compilers With FreeBSD/PC-BSD 9.1 comparing the performance of the stock compilers: GCC 4.2.1 versus LLVM/Clang 3.1 on an Intel Core i7 3960X Extreme Edition system. FreeBSD/PC-BSD 9.1 x64/amd64 with stock packages/settings. Testing for a future article on Phoronix.com. OCZ VERTEX2 1.10 - Intel Core i7-3770K: Processor: Intel Core i7-3770K @ 3.50GHz (8 Cores), Motherboard: ECS Z77H2-A2X 1.0, Memory: 8192MB, Disk: OCZ VERTEX2 1.10, Graphics: NVIDIA GeForce GTX 460 768MB (675/1800MHz), Audio: NVIDIA GT104 HDA OS: FreeBSD, Kernel: 9.1-RELEASE (x86_64), Desktop: KDE 4.9.3, Display Server: X Server 1.10.6, Display Driver: NVIDIA 304.64, OpenGL: 4.2.0 NVIDIA 304.64, Compiler: GCC 4.2.1 20070831 + Clang 3.1 (SVN 156863), File-System: zfs, Screen Resolution: 1920x1080 LLVM Clang 3.1: Processor: Intel Core i7-3770K @ 3.50GHz (8 Cores), Motherboard: ECS Z77H2-A2X 1.0, Memory: 8192MB, Disk: OCZ VERTEX2 1.10, Graphics: NVIDIA GeForce GTX 460 768MB (675/1800MHz), Audio: NVIDIA GT104 HDA OS: FreeBSD, Kernel: 9.1-RELEASE (x86_64), Desktop: KDE 4.9.3, Display Server: X Server 1.10.6, Display Driver: NVIDIA 304.64, OpenGL: 4.2.0 NVIDIA 304.64, Compiler: Clang 3.1 (SVN 156863), File-System: zfs, Screen Resolution: 1920x1080 BlogBench 1.0 Test: Read Final Score > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 655790 |============================= LLVM Clang 3.1 ......................... 683983 |============================== BlogBench 1.0 Test: Write Final Score > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 2065 |================================ LLVM Clang 3.1 ......................... 2027 |=============================== Compile Bench 0.6 Test: Compile MB/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 141.21 |======================= LLVM Clang 3.1 ......................... 182.96 |============================== Compile Bench 0.6 Test: Initial Create MB/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 112.31 |======================= LLVM Clang 3.1 ......................... 146.38 |============================== PostMark 1.51 Disk Transaction Performance TPS > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 390 |================================= LLVM Clang 3.1 ......................... 379 |================================ Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 13.05 |=============================== LLVM Clang 3.1 ......................... 12.70 |============================== Botan 1.10.3 Test: Tiger Mbytes/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 314.25 |========================== LLVM Clang 3.1 ......................... 368.14 |============================== Botan 1.10.3 Test: KASUMI Mbytes/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 68.48 |=============================== LLVM Clang 3.1 ......................... 61.49 |============================ Botan 1.10.3 Test: AES-256 Mbytes/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 175.12 |============================== LLVM Clang 3.1 ......................... 174.50 |============================== Botan 1.10.3 Test: Twofish Mbytes/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 189.44 |============================== LLVM Clang 3.1 ......................... 183.68 |============================= Botan 1.10.3 Test: CAST-256 Mbytes/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 89.07 |========================= LLVM Clang 3.1 ......................... 105.88 |============================== Botan 1.10.3 Test: X9.19-MAC Mbytes/s > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 83.50 |=============================== LLVM Clang 3.1 ......................... 77.99 |============================= SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 552.24 |============================ LLVM Clang 3.1 ......................... 585.47 |============================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 329.66 |============================== LLVM Clang 3.1 ......................... 329.92 |============================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 2175.48 |=========================== LLVM Clang 3.1 ......................... 2355.30 |============================= SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 2232.78 |====================== LLVM Clang 3.1 ......................... 2920.54 |============================= SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 1114.86 |==================== LLVM Clang 3.1 ......................... 1592.82 |============================= GraphicsMagick 1.3.16 Operation: Blur Iterations Per Minute > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 116 |================================= LLVM Clang 3.1 ......................... 74 |===================== GraphicsMagick 1.3.16 Operation: Sharpen Iterations Per Minute > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 90 |================================== LLVM Clang 3.1 ......................... 50 |=================== GraphicsMagick 1.3.16 Operation: Resizing Iterations Per Minute > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 149 |================================= LLVM Clang 3.1 ......................... 97 |===================== GraphicsMagick 1.3.16 Operation: HWB Color Space Iterations Per Minute > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 177 |================================= LLVM Clang 3.1 ......................... 127 |======================== GraphicsMagick 1.3.16 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 112 |================================= LLVM Clang 3.1 ......................... 42 |============ Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 792.30 |=========================== LLVM Clang 3.1 ......................... 891.34 |============================== 7-Zip Compression 9.20.1 Compress Speed Test MIPS > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 19059 |=============================== LLVM Clang 3.1 ......................... 18576 |============================== Timed PHP Compilation 5.2.9 Time To Compile Seconds < Lower Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 25.39 |=============================== LLVM Clang 3.1 ......................... 25.38 |=============================== C-Ray 1.1 Total Time Seconds < Lower Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 44.47 |=============================== LLVM Clang 3.1 ......................... 40.52 |============================ Primesieve 4.0 1e12 Prime Number Generation Seconds < Lower Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 92.04 |======= LLVM Clang 3.1 ......................... 372.22 |============================== dcraw RAW To PPM Image Conversion Seconds < Lower Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 54.17 |=============================== LLVM Clang 3.1 ......................... 54.31 |=============================== LAME MP3 Encoding 3.99.3 WAV To MP3 Seconds < Lower Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 14.90 |=============================== LLVM Clang 3.1 ......................... 14.94 |=============================== N-Queens 1.0 Elapsed Time Seconds < Lower Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 38.64 |===== LLVM Clang 3.1 ......................... 218.21 |============================== OpenSSL 1.0.1c RSA 4096-bit Performance Signs Per Second > Higher Is Better OCZ VERTEX2 1.10 - Intel Core i7-3770K . 122.80 |============================== LLVM Clang 3.1 ......................... 123.07 |==============================