Lubo Antonov questions

Lubo Antonov answers

• Does having multiple work dimensions in a work group/item provides a speedup?
• OpenMP and GSL RNG - Performance Issue - 4 threads implementation 10x slower than pure sequential one (quadcore CPU)
• Atomic Minimum on x86 using OpenMP
• Is C++ platform dependent?
• OpenCL: Manually throw an exception in kernel
• Casting float2 to double2 and vice versa
• OpenCL global memory fetches
• Ambiguous inheritance of function when multiple inheritance of classes that themselves have diamond inheritance in their hierarchy