[Bug c/93432] variable is used uninitialized, but gcc shows no warning

[Bug c/93432] variable is used uninitialized, but gcc shows no warning

[manu at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93432

Manuel López-Ibáñez <manu at gcc dot gnu.org> changed:

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

--- Comment #4 from Manuel López-Ibáñez <manu at gcc dot gnu.org> ---
It warns in gcc 10.1. It may good idea to add this one as a testcase, since it
seems it got fixed without noticing.

[Bug c/93432] variable is used uninitialized, but gcc shows no warning

[manu at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93432

Manuel López-Ibáñez <manu at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |easyhack

--- Comment #5 from Manuel López-Ibáñez <manu at gcc dot gnu.org> ---
This just needs a testcase to prevent regressing...

[Bug c/93432] variable is used uninitialized, but gcc shows no warning

[vincent-gcc at vinc17 dot net]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93432

Vincent Lefèvre <vincent-gcc at vinc17 dot net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |vincent-gcc at vinc17 dot net

--- Comment #6 from Vincent Lefèvre <vincent-gcc at vinc17 dot net> ---
(In reply to Manuel López-Ibáñez from comment #4)
> It warns in gcc 10.1. It may good idea to add this one as a testcase, since
> it seems it got fixed without noticing.

Has this really been fixed, or does it work now just by chance?

This looks like PR18501, where the concerned variable is initialized only at
one place in the loop. Here, "z = 1" is followed by "z=z+1", so that this is
equivalent to "z = 2". But if in the code, I do this change, the warning
disappears (tested with gcc-12 (Debian 12-20220222-1) 12.0.1 20220222
(experimental) [master r12-7325-g2f59f067610] and -O1, -O2, -O3).

[Bug tree-optimization/93432] variable is used uninitialized, but gcc shows no warning

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93432

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c                           |tree-optimization
           Keywords|easyhack                    |

--- Comment #7 from Richard Biener <rguenth at gcc dot gnu.org> ---
It works by chance when optimizing but not optimizing too much.  -O3 unrolls
the loop and runs into CCP aka PR18501, we get

  <bb 2> [local count: 97603129]:
  if (y_5(D) <= 9)
    goto <bb 18>; [33.00%]
  else
    goto <bb 19>; [67.00%]

  <bb 18> [local count: 32209032]:
  _11 = z_6(D) + 1;

  <bb 19> [local count: 97603129]:
  # prephitmp_17 = PHI <2(2), _11(18)>

and CCP says _11 is UNDEFINED and merges it with 2 to 2.  For me even GCC 9.3.1
warns, so does GCC 7.5, both even with -O3.

We do not warn at -O0 but that's by design - we think doing the predicate based
uninit analysis early (on the unoptimized IL) would be too noisy.