AVX AVX 2.0 vs. 1.0 on Linux
intel Microprocessors Core fourth generation “Haswell” are just for their first month since it was released, and among its novelties have a slightly higher yield per cycle of its predecessor and new instruction sets :
- Transactional Synchronization eXtensions “TSX” (not available in the CPU K Series).
- Advanced Vector Extensions 2.0 “AVX2″.
- Fused Multiply-Add 3.0 “FMA3″.
Unfortunately no existing application leverages these new instruction sets (except the benchmark Linpack 11.0.3 ), so it has not been proven how beneficial they are.
From Phoronix we get some tests done to an Intel core i7-4770K “Haswell-DT”, which proved with some Linux applications compiled with gcc 4.8.1 compiler (supports AVX 1.0) and GCC 126.96.36.1993623 (supports AVX 2.0 ), in order to check the performance gained when using the AVX 2.0 instruction set over its predecessor.
We have developed a graph gathering your results:
The test Himeno Pressure Solver is showing the largest increase performance by AVX 2.0 (51.45% higher), although performance is remarkable 8.14% higher on the test BLAKE2.
Much remains to make its appearance a large number of applications written to take advantage of AVX 2.0 and 3.0 FMA, which for now we have no benchmark showing performance.
Link: GCC 4.9 Running With Intel’s Core i7 “Core-AVX2″ (Phoronix)
.Tags: AVX, AVX 1.0, AVX 2.0, AVX2, benchmark, core, CPU, gcc, Haswell, Haswell-DT, i7-4770K, intel, Linux, microprocesador, microprocessor, performance, procesador, processor, rendimiento, test