int operators != and == when comparing to zero

by

This is a great question, but I think you’ve fallen victim to the compiler’s dependency analysis.

The compiler only has to clear the high bits of eax once, and they remain clear for the second version. The second version would have to pay the price to xor eax, eax except that the compiler analysis proved it’s been left cleared by the first version.

The second version is able to “cheat” by taking advantage of work the compiler did in the first version.

How are you measuring times? Is it “(version one, followed by version two) in a loop”, or “(version one in a loop) followed by (version two in a loop)”?

Don’t do both tests in the same program (instead recompile for each version), or if you do, test both “version A first” and “version B first” and see if whichever comes first is paying a penalty.

Illustration of the cheating:

timer1.start();
double x1 = 2 * sqrt(n + 37 * y + exp(z));
timer1.stop();
timer2.start();
double x2 = 31 * sqrt(n + 37 * y + exp(z));
timer2.stop();

If timer2 duration is less than timer1 duration, we don’t conclude that multiplying by 31 is faster than multiplying by 2. Instead, we realize that the compiler performed common subexpression analysis, and the code became:

timer1.start();
double common = sqrt(n + 37 * y + exp(z));
double x1 = 2 * common;
timer1.stop();
timer2.start();
double x2 = 31 * common;
timer2.stop();

And the only thing proved is that multiplying by 31 is faster than computing common. Which is hardly surprising at all — multiplication is far far faster than sqrt and exp.

Related Contents:

  1. Is < faster than
  2. Replacing a 32-bit loop counter with 64-bit introduces crazy performance deviations with _mm_popcnt_u64 on Intel CPUs
  3. Why does C++ code for testing the Collatz conjecture run faster than hand-written assembly?
  4. Is inline assembly language slower than native C++ code?
  5. If statement vs if-else statement, which is faster?
  6. Why is this C++ program so incredibly fast?
  7. Why is the construction of std::optional more expensive than a std::pair?
  8. C++: Mysteriously huge speedup from keeping one operand in a register
  9. Difference between rdtscp, rdtsc : memory and cpuid / rdtsc?
  10. What is IACA and how do I use it?
  11. Why does GCC generate 15-20% faster code if I optimize for size instead of speed?
  12. Fastest way to determine if an integer is between two integers (inclusive) with known sets of values
  13. How to write a large buffer into a binary file in C++, fast?
  14. Is ‘switch’ faster than ‘if’?
  15. Is it better to use std::memcpy() or std::copy() in terms to performance?
  16. A fast method to round a double to a 32-bit int explained
  17. How expensive is RTTI?
  18. Why is processing an unsorted array the same speed as processing a sorted array with modern x86-64 clang?
  19. How can I see the assembly code for a C++ program?
  20. Using scanf() in C++ programs is faster than using cin?
  21. Any optimization for random access on a very big array when the value in 95% of cases is either 0 or 1?
  22. How to overload std::swap()
  23. C++ performance challenge: integer to std::string conversion
  24. Can const-correctness improve performance?
  25. Which is faster : if (bool) or if(int)?
  26. Why is memmove faster than memcpy?
  27. Is accessing data in the heap faster than from the stack?
  28. Why are Python Programs often slower than the Equivalent Program Written in C or C++?
  29. Why does adding inline assembly comments cause such radical change in GCC’s generated code?
  30. Why does my Intel Skylake / Kaby Lake CPU incur a mysterious factor 3 slowdown in a simple hash table implementation?
  31. Is using double faster than float?
  32. How to generate assembly code with clang in Intel syntax?
  33. Why is std::fill(0) slower than std::fill(1)?
  34. Will my iPhone app take a performance hit if I use Objective-C for low level code?
  35. What’s faster, iterating an STL vector with vector::iterator or with at()?
  36. What is the most efficient way to append one std::vector to the end of another?
  37. Why do C++ optimizers have problems with these temporary variables or rather why `v[]` should be avoided in tight loops?
  38. Why is there a significant difference in this C++ for loop’s execution time? [duplicate]
  39. Algorithm for finding the smallest power of two that’s greater or equal to a given value
  40. Strange uses of movzx by Clang and GCC
  41. what is the performance impact of using int64_t instead of int32_t on 32-bit systems?
  42. Why is pow(int, int) so slow?
  43. linux perf: how to interpret and find hotspots
  44. Speed accessing a std::vector by iterator vs by operator[]/index?
  45. Learning to read GCC assembler output
  46. Why is calling vector.reserve(required + 1) faster than vector.reserve(required)?
  47. Why doesn’t a compiler optimize floating-point *2 into an exponent increment?
  48. Does try-catch block decrease performance [duplicate]
  49. Why is such complex code emitted for dividing a signed integer by a power of two?
  50. Using AVX intrinsics instead of SSE does not improve speed — why?

Leave a Comment