detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the

libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the smaller than the current index. 3. opt_size Optimize for size with minimum performance impact. Clang/GCC's -Os. 4. min_size Optimize only for size. Clang's -Oz. 29.8 Debugging and profiling JULIA_DEBUG

detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the

detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the

detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the

detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the

detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the

detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the

libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the smaller than the current index. 3. opt_size Optimize for size with minimum performance impact. Clang/GCC's -Os.CHAPTER 30. ENVIRONMENT VARIABLES 398 4. min_size Optimize only for size. Clang's -Oz. 30

detail, this is achieved under the hood by using LLVM's half type, which behaves similarly to what the GCC -fexcess-precision=16 flag does for C/C++ code. julia> sizeof(Float16(4.)) 2 julia> 2*Float16(4 libraries ship compiled as shared libraries already, but if you are compiling the code yourself using GCC (or Clang), you will need to use the -shared and -fPIC options. The machine instructions generated void say_y(int y) { printf("Hello from C: got y = %d.\n", y); } and compile it with gcc -fPIC -shared -o mylib.so mylib.c. It can then be called by specifying the (absolute) path as the