Re: branch guessing, Microarchitectural support for counting

On Thu, 26 Dec 2024 19:11:01 -0000 (UTC)
John Levine <johnl@taugh.com> wrote:

According to Michael S  <already5chosen@yahoo.com>:

Yes, compilers often generate such code.
When coding in asm, I typically know at least something about
probability of branches, so I tend to code it differently: 

 
The first version of FORTRAN had a FREQUENCY statement which let you
tell it the relative likelihood of each of the results of a three-way
IF, and the expected number of iterations of a DO loop.  It turned
out to be useless, because programmers usually guessed wrong.
 
The final straw was a compiler where they realized FREQUENCY was
implemented backward and nobody noticed.
 
Unless you've profiled the code and you data to support your branch
guesses, just write it in the clearest way you can.
 

My asm coding is mostly various math library routines.
Not a production, just fun.
I find the style illustrated above to be the clearest way of handling
less common conditions. And even I am wrong about commonality, the
performance cost of such mistake on the modern hardware is minimal.