Ci-dessous, les différences entre deux révisions de la page.
| Prochaine révision | Révision précédente |
| linux:avx [2020/11/06 12:04] – créée phsw | linux:avx [2025/10/25 20:12] (Version actuelle) – [Programmation] add link phsw |
|---|
| ====== Instructions AVX ====== | ====== Instructions AVX ====== |
| | |
| | |
| | * [[https://www.modular.com/blog/understanding-simd-infinite-complexity-of-trivial-problems|Understanding SIMD: Infinite Complexity of Trivial Problems]] |
| | * [[https://sanixdk.xyz/blogs/how-i-accidentally-created-the-fastest-csv-parser-ever-made|How I accidentally created the fastest CSV parser ever made]] |
| |
| Plusieurs niveaux de vectorisation: | Plusieurs niveaux de vectorisation: |
| * ''AVX2'' | * ''AVX2'' |
| * ''AVX512'' | * ''AVX512'' |
| * | |
| | |
| | Comment compiler avec la MKL : https://www.intel.com/content/www/us/en/developer/tools/oneapi/onemkl-link-line-advisor.html |
| | |
| | |
| | |
| | ==== Programmation ==== |
| | |
| | * [[https://lemire.me/blog/2018/09/07/avx-512-when-and-how-to-use-these-new-instructions/]] |
| | * [[https://albertherd.com/2018/01/11/performance-differences-when-using-avx-instructions/]] |
| | * [[https://www.codeproject.com/Articles/874396/Crunching-Numbers-with-AVX-and-AVX]] |
| | * [[https://travisdowns.github.io/blog/2020/08/19/icl-avx512-freq.html#fnref:tiring]] |
| | * [[https://lemire.me/blog/2018/09/04/per-core-frequency-scaling-and-avx-512-an-experiment/]] |
| | * [[https://software.intel.com/sites/landingpage/IntrinsicsGuide/]] |
| | * [[https://healeycodes.com/counting-words-at-simd-speed|Counting Words at SIMD Speed]] |
| |