Strumenti Utente

Strumenti Sito


roberto.alfieri:pub:vectorization

Differenze

Queste sono le differenze tra la revisione selezionata e la versione attuale della pagina.

Link a questa pagina di confronto

Entrambe le parti precedenti la revisione Revisione precedente
Prossima revisione
Revisione precedente
roberto.alfieri:pub:vectorization [13/06/2017 19:52]
roberto.alfieri
roberto.alfieri:pub:vectorization [14/06/2017 11:30]
roberto.alfieri
Linea 1: Linea 1:
 ====== Vectorization ====== ====== Vectorization ======
 +
 +[[ http://​www.training.prace-ri.eu/​uploads/​tx_pracetmo/​intel_mic_optimization.pdf | Vectorization and Code Optimization ]]
 +
  
 Processor peak performance includes the speed-up provided by the vector instructions, ​ Processor peak performance includes the speed-up provided by the vector instructions, ​
Linea 15: Linea 18:
  
  
-Auto-vectorization is the easiest and more portable way to get vectorization ​, but not all loops can be vectorized.+ 
 +Auto-vectorization is the easiest and more portable way to get vectorization
 + 
 +The compiler recognize several vectiorization options. 
 + 
 +Main vectorization options:  
 + 
 +^     ​^ ​ Intel compiler ​  ​^ ​  
 +^ KNL      | -xMIC-AVX512 ​ |  
 +^ BDW      | -xCORE-AVX2 ​  | 
 +^ Disable ​ | -no-vec ​      | 
 + 
 +Not all loops can be vectorized:
  
 Some examples: ​ Some examples: ​
  
-  * Loop with dependencies between iterations ​can not be vectorized.+  * Loop with dependencies between iterations
  
 <​code>​ <​code>​
Linea 27: Linea 42:
 } }
 </​code>​ </​code>​
 +
 +
 +  * Complex loops
 +
 +  * Function calls inside the loop:
 +
 +      for (int i = 0; i < N; i++)   a[i] = foo(b[i]);
 +
 +  * Loops on data that are not aligned in memory ​   ​
 +  ​
roberto.alfieri/pub/vectorization.txt · Ultima modifica: 14/06/2017 11:30 da roberto.alfieri