News for hackers. Stuff that matters.
Re:Subtraction is not comparison by cygx on Thursday December 25, 2014 @10:37PM (#8797932)

    return (x > y) - (x < y);
Re:Subtraction is not comparison by vardump on Thursday December 25, 2014 @10:41PM (#8797947)
That can be very expensive operation. Potentially two branches, not counting return from subroutine.
Re:Subtraction is not comparison by cremno on Thursday December 25, 2014 @10:58PM (#8797992)
If performance really is of concern, qsort() or similar functions probably shouldn't be used. Instead a dedicated function, which allows to choose a specific algorithm (qsort() doesn't have to use quicksort) and also doesn't involve calling a comparison function pointed to by a function pointer, can be used.
Re:Subtraction is not comparison by bluecalm on Friday December 26, 2014 @09:43AM (#8798928)
One interesting thing I have learned recently is that GCC can do inlining through function pointers. That doesn't make your point about dedicated function being better idea for performance but it's one thing "std::sort is faster by design" people often miss.

From GCC documentation:

>>-findirect-inlining Inline also indirect calls that are discovered to be known at compile time thanks to previous inlining. This option has any effect only when inlining itself is turned on by the -finline-functions or -finline-small-functions options.

    Enabled at level -O2.
Re:Subtraction is not comparison by cremno on Friday December 26, 2014 @06:42PM (#8800977)
But GCC likely isn't able to do that for libc functions like qsort(). Maybe if LTO is enabled and libc is linked statically, it might.
Re:Subtraction is not comparison by bluecalm on Saturday December 27, 2014 @04:01AM (#8802120)
I have no idea when it can and when it can't do that to be honest. I tested qsort vs std::sort on my machine on my data and performance was the same (Windows, MinGW, GCC 4.8, -flto enabled) but other people reported different results.
Re:Subtraction is not comparison by azakai on Thursday December 25, 2014 @11:48PM (#8798085)
I would hope that on modern compilers on modern architectures, this would be done with zero branches. That is the case in a quick test on my machine (gcc 4.6.3).
Re:Subtraction is not comparison by vardump on Friday December 26, 2014 @07:50PM (#8801186)
Good if compilers are that smart nowadays. Just a few years ago I did see two branches in a very similar piece of code.
Re:Subtraction is not comparison by userbinator on Friday December 26, 2014 @11:47AM (#8799260)
The simplest way to implement that would be two SETxx instructions and a subtract on 386+, SLT/SGT on MIPS, two subtracts (or one compare and one subtract) and two predicated moves on ARM. No branching required at all.
Re:Subtraction is not comparison by gonzo on Friday December 26, 2014 @01:31AM (#8798268)
Not really. Maybe on an old atom.