CPU Branch Optimization

Just want to share collection of tricks to optimize branch in CPU.

Bounds Checking

Checking bounds [0,max)
// int i, max;
// if (i >= 0 && i < max) {}
if ((unsigned int) i < (unsigned int)max) {}

Checking bounds[min,max] 
// int i, min,max;
// if (i >= min && i <= max) {}
if ((unsigned int)(i - min) <= (unsigned int)(max - min)) {}

Comments

Post a Comment

Popular posts from this blog

World, View and Projection Matrix Internals

Image
This convention below is applicable to Direct3D and XNA matrices World Matrix Given a  position  and basis vectors  right ,  up  and  look  of an  object , a world matrix can be formed by the following arrangement: View Matrix Given a  position  and basis vectors  right ,  up  and  look  of a viewer, a view matrix can be formed by the following arrangement: Projection Matrix Given field of view  FOV ,  aspect ratio , near clip plane  Zn  and far clip plane  Zf , a perspective projection matrix can be formed by the following arrangment:
Read more

BASH: Reading Text File

BASH for C programmers: Read a text file word-by-word in BASH: #!/bin/bash for WORD in `cat filename` do echo $WORD done Read a text file line-by-line in BASH: #!/bin/bash while read LINE do echo $LINE done < filename Looping in BASH: i=0 while [ $i -lt 10 ] do echo $i # next i=$(($i+1)) done Update : another way to loop suggested in the comment.
Read more

GDC 2015 Links

Here are [growing] links to GDC 2015 slides/presentations: Update: GDC Vault for GDC 2015 is now up!  Programming Track Khronos Group ( Link ) Khronos OpenCL GDC Khronos Vulkan GDC Valve Vulkan Session GDC Live Stream of Vulkan Session (Youtube) The Future of High-Performance Graphics Virtual Reality VR Direct: How NVIDIA Technology is Improving the VR Experience ( pptx ,  pdf ) - Nathan Reed, Dean Beeler Advanced VR Rendering  - Alex Vlachos Far Cry 4 Fast Iteration for Far Cry 4 - Optimizing Key Parts of Dunia Pipeline  - Remi Quenin Rendering the World of Far Cry 4  - Stephen McAuley Adaptive Virtual Texture Rendering in Far Cry 4  - Ka Chen Mesh Cutting in Farming Simulator 15 - Gino van den Bergen Advanced Visual Effects in 2D Games - Viktor Lidholt Great Management of Technical Leads - Mike Acton Code Clinic: How to Write Code the Compiler Can Actually Optimize - Mike Acton Parallelizing the Naughty Dog Engine Using Fibers - Christian Gyrling
Read more