Skip to content
GPU-side implementation of the OpenCL standard-library for VC4CL
C++ Objective-C C CMake
Branch: master
Clone or download
doe300 Improves some function performance
For complex ?:-operators, clang generates if-else blocks,
e.g. if clang cannot determine that neither side has side-effects.
By moving some calculations out of the ?:-operator, we allow clang
to use a select-instruction saving instructions and often also
execution cycles for either case.

Fixes implementation of sqrt and native_sqrt
Latest commit ed74948 Jul 14, 2019
Type Name Latest commit message Commit time
Failed to load latest commit information.
include Improves some function performance Jul 14, 2019
.gitignore Several improvements and fixes Dec 27, 2018
CMakeLists.txt Add: add CROSS_COMPILE option for x86_64 environment (#4) Mar 1, 2018
LICENSE Initial commit Oct 8, 2017 Readme Oct 9, 2017


This repository contains the VC4CL implementation of the OpenCL standard-library and is required to build the VC4C compiler.

For more information about the VC4CL implementation, see the VC4CL project.

You can’t perform that action at this time.