[Bug tree-optimization/108997] New: GCC prediction on bool comparisons seems wrong

[Bug tree-optimization/108997] New: GCC prediction on bool comparisons seems wrong

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108997

            Bug ID: 108997
           Summary: GCC prediction on bool comparisons seems wrong
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pinskia at gcc dot gnu.org
  Target Milestone: ---

I noticed this while looking into 108992.
I suspect we look into this but I am not 100% sure this is a bug.
Testcase:
void use(int *);
void use2(int *);

void foo(bool cond, int * p)
{
    if (cond) {
        use(p);
    }
    use2(p);
}

GCC predicts the branch to be taken 66% of the time, but I would have assume it
should be 50% as it is a bool and there is no other information. I suspect this
is due to prediction when comparing against 0.

[Bug tree-optimization/108997] GCC prediction on bool comparisons seems wrong

[nok.raven at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108997

Nikita Kniazev <nok.raven at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |nok.raven at gmail dot com

--- Comment #1 from Nikita Kniazev <nok.raven at gmail dot com> ---
Is it about bool?

void use(int *);
void use2(int *);

void foo(int * p, int cond)
{
    if (cond == 789) {
        use(p);
    }
    use2(p);
}

[Bug tree-optimization/108997] GCC prediction on bool comparisons seems wrong

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108997

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Nikita Kniazev from comment #1)
> Is it about bool?
YES.

[Bug tree-optimization/108997] GCC prediction on bool comparisons seems wrong

[nok.raven at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108997

--- Comment #3 from Nikita Kniazev <nok.raven at gmail dot com> ---
For cond == 789
  if (cond_2(D) == 789)
    goto <bb 3>; [20.24%]
  else
    goto <bb 4>; [79.76%]

For cond != 789
  if (cond_2(D) != 789)
    goto <bb 3>; [48.88%]
  else
    goto <bb 4>; [51.12%]

[Bug tree-optimization/108997] GCC prediction on bool comparisons seems wrong

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108997

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Nikita Kniazev from comment #3)
> For cond == 789
>   if (cond_2(D) == 789)
>     goto <bb 3>; [20.24%]
>   else
>     goto <bb 4>; [79.76%]
> 
> For cond != 789
>   if (cond_2(D) != 789)
>     goto <bb 3>; [48.88%]
>   else
>     goto <bb 4>; [51.12%]

an intergal type equaling a specific # is handled this way on purpose. Bool
should be handled differently because it only has two values, 0/1.
Anyways != is not handled specially except for the zero value case ...

[Bug tree-optimization/108997] GCC prediction on bool comparisons seems wrong

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108997

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hubicka at gcc dot gnu.org

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
Predictions for bb 2
  DS theory heuristics: 33.00%
  combined heuristics: 33.00%
  call heuristics of edge 2->3: 33.00%

so it seems it's about "conditional call" which is reasonable I think.

Doing

void use2(int *);
int x = 1;
void foo(bool cond, int * p)
{
    if (cond) {
        x = 1;
    }
    use2(p);
}

shows even probabilities, so does

  if (cond) use1 (p); else use2 (p);

[Bug tree-optimization/108997] GCC prediction on bool comparisons seems wrong

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108997

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
                 CC|                            |marxin at gcc dot gnu.org
         Resolution|---                         |INVALID

--- Comment #6 from Martin Liška <marxin at gcc dot gnu.org> ---
Yes, one can see the heuristics with -fdump-tree-profile_estimate as Richi
showed.