Enabling strict floating point mode in GCC

by

Compiling with -msse2 on an Intel/AMD processor that supports it will get you almost there. Do not let any library put the FPU in FTZ/DNZ mode, and you will be mostly set (processor bugs notwithstanding).

For other architectures, the answer would be different. Those achitectures that do not offer any convenient way to get exact IEEE 754 semantics (for instance, pre-SSE2 IA32 CPUs) would require the use of a floating-point emulation library to get the result you want, at a very high performance penalty.

If your target architecture supports the fmadd (multiplication and addition without intermediate rounding) instruction, make sure your compiler does not use it when you have explicit multiplications and additions in the source code. GCC is not supposed to do this unless you use the -ffast-math option.

More Related Contents:

  1. Running C — helloWorld runs but nothing else — Ubuntu
  2. How to disable GCC warnings for a few lines of code
  3. printf and long double
  4. Trying to understand gcc option -fomit-frame-pointer
  5. How to prevent GCC from optimizing out a busy wait loop?
  6. How can I use “sizeof” in a preprocessor macro?
  7. Why does the C preprocessor interpret the word “linux” as the constant “1”?
  8. difference between %ms and %s scanf
  9. How do I print uint32_t and uint16_t variables’ value?
  10. gcc will not properly include math.h
  11. How to turn off gcc compiler optimization to enable buffer overflow
  12. Linking against older symbol version in a .so file
  13. GCC left shift overflow
  14. What does -D_XOPEN_SOURCE do/mean?
  15. How do you get the start and end addresses of a custom ELF section?
  16. strict aliasing in C
  17. Can I make gcc tell me when a calculation results in NaN or inf at runtime?
  18. Overriding C library functions, calling original
  19. why does GCC __builtin_prefetch not improve performance?
  20. Linking a C program directly with ld fails with undefined reference to `__libc_csu_fini`
  21. execute binary machine code from C
  22. Does pthread_mutex_lock contains memory fence instruction? [duplicate]
  23. what “inline __attribute__((always_inline))” means in the function?
  24. The difference between asm, asm volatile and clobbering memory
  25. Compare 16 byte strings with SSE
  26. Why is the compiler throwing this warning: “missing initializer”? Isn’t the structure initialized?
  27. prevent gcc from removing an unused variable
  28. Why does static initialization of flexible array member work?
  29. What is the effect of second argument in _builtin_prefetch()?
  30. Declaring an array of negative length

Leave a Comment