tzutalin
Compile DLib with AVX or SSE4 enabled
I'd like to compile dlib for my android application with SSE4 or AVX enabled, but it doesn't seem like there's an option in the makefile to enable it.
SSE4 and AVX instruction sets are available on Intel and AMD processors, not ARM processors. Android devices are generally built with ARM processors, that's why you can't compile for, let's say armeabi-v7a with SSE4 or AVX instructions enabled.
Some devices do have Intel Atom processors and that's why I'd like to take advantage of those accelerations.
Do you know how to use the openBlas library on android studio? Just add an item on cmakelists.txt, -ddlib_use_blas =1? Doesn't seem to work? There is no good solution. I don't know how to add it right now, let's say I have the dlib code and I have the openblas library that's cross-compiled, okay?
I wanted to know how to add openblas to the android version of dlib for speed optimization of the matrix. thank you