[Bug middle-end/88897] Bogus maybe-uninitialized warning on class field (missed CSE)

[Bug middle-end/88897] Bogus maybe-uninitialized warning on class field (missed CSE)

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |msebor at gcc dot gnu.org
      Known to fail|                            |10.2.0, 11.0, 8.3.0, 9.3.0
   Last reconfirmed|2019-01-22 00:00:00         |2021-4-6

--- Comment #7 from Martin Sebor <msebor at gcc dot gnu.org> ---
The IL hasn't appreciably changed in GCC 11 so it still issues the warning,
albeit with more detail.

$ gcc -O2 -S -Wall pr88897.C
In constructor ‘temporary_buffer::temporary_buffer(temporary_buffer&&)’,
    inlined from ‘optional::optional(optional&&)’ at pr88897.C:22:79,
    inlined from ‘void foo()’ at pr88897.C:42:57:
pr88897.C:14:63: warning: ‘*(temporary_buffer*)((char*)&_local_state +
offsetof(future_state,
future_state::value.optional::<unnamed>)).temporary_buffer::_buffer’ may be
used uninitialized [-Wmaybe-uninitialized]
   14 |  temporary_buffer(temporary_buffer &&x) noexcept : _buffer(x._buffer) {
      |                                                            ~~^~~~~~~

pr88897.C: In function ‘void foo()’:
pr88897.C:40:16: note: ‘_local_state’ declared here
   40 |   future_state _local_state;
      |                ^~~~~~~~~~~~

[Bug middle-end/88897] [9/10/11 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|Bogus maybe-uninitialized   |[9/10/11 Regression] Bogus
                   |warning on class field      |maybe-uninitialized warning
                   |(missed CSE)                |on class field (missed CSE)

--- Comment #8 from Martin Sebor <msebor at gcc dot gnu.org> ---
The warning first appeared in r245840 (GCC 7.0.1 20170302), ironically
committed to fix PR tree-optimization/79345 (passing yet-uninitialized member
as argument to base class constructor should warn (-Wunitialized)).  The IL
emitted just before that revision doesn't look all that different:

void foo() ()
{
  struct future_stateD.2397 _local_stateD.2411;
  struct future_stateD.2397 * _4;
  charD.10 * _6;
  voidD.45 * _7;
  boolD.2220 _8;

;;   basic block 2, loop depth 0, count 0, freq 10000, maybe hot
;;    prev block 0, next block 7, flags: (NEW, REACHABLE, VISITED)
;;    pred:       ENTRY [100.0%]  (FALLTHRU,EXECUTABLE)
  # .MEM_3 = VDEF <.MEM_2(D)>
  # PT = nonlocal escaped null 
  # USE = nonlocal null { D.2411 } (escaped)
  # CLB = nonlocal null { D.2411 } (escaped)
  _4 = _Z3barvD.2407 ();
  # .MEM_5 = VDEF <.MEM_3>
  # USE = nonlocal null { D.2411 } (escaped)
  # CLB = nonlocal null { D.2411 } (escaped)
  _ZN12future_stateC1EvD.2402 (&_local_stateD.2411);
  # .MEM_10 = VDEF <.MEM_5>
  MEM[(struct  &)&_local_stateD.2411] ={v} {CLOBBER};
  # .MEM_11 = VDEF <.MEM_10>
  MEM[(struct optionalD.2323 *)&_local_stateD.2411]._M_engagedD.2340 = 0;
  # PT = nonlocal escaped null 
  _7 = &_4->valueD.2401;
  if (_7 != 0B)
    goto <bb 4>; [73.26%]
  else
    goto <bb 7>; [26.74%]
;;    succ:       4 [73.3%]  (TRUE_VALUE,EXECUTABLE)
;;                7 [26.7%]  (FALSE_VALUE,EXECUTABLE)

;;   basic block 7, loop depth 0, count 0, freq 2674, maybe hot
;;    prev block 2, next block 3, flags: (NEW)
;;    pred:       2 [26.7%]  (FALSE_VALUE,EXECUTABLE)
;;    succ:       3 [100.0%]  (FALLTHRU)

;;   basic block 3, loop depth 0, count 0, freq 6044, maybe hot
;;    prev block 7, next block 4, flags: (NEW, REACHABLE, VISITED)
;;    pred:       7 [100.0%]  (FALLTHRU)
;;                8 [100.0%]  (FALLTHRU)
  # .MEM_17 = PHI <.MEM_11(7), .MEM_13(8)>
  goto <bb 6>; [100.00%]
;;    succ:       6 [100.0%]  (FALLTHRU,EXECUTABLE)

;;   basic block 4, loop depth 0, count 0, freq 7326, maybe hot
;;    prev block 3, next block 8, flags: (NEW, REACHABLE, VISITED)
;;    pred:       2 [73.3%]  (TRUE_VALUE,EXECUTABLE)
  # .MEM_13 = VDEF <.MEM_11>
  MEM[(struct optionalD.2323 *)_4]._M_engagedD.2340 = 0;
  # VUSE <.MEM_13>
  _8 = MEM[(struct optionalD.2323 &)&_local_stateD.2411]._M_engagedD.2340;
  if (_8 != 0)
    goto <bb 5>; [54.00%]
  else
    goto <bb 8>; [46.00%]
;;    succ:       5 [54.0%]  (TRUE_VALUE,EXECUTABLE)
;;                8 [46.0%]  (FALSE_VALUE,EXECUTABLE)

;;   basic block 8, loop depth 0, count 0, freq 3370, maybe hot
;;    prev block 4, next block 5, flags: (NEW)
;;    pred:       4 [46.0%]  (FALSE_VALUE,EXECUTABLE)
  goto <bb 3>; [100.00%]
;;    succ:       3 [100.0%]  (FALLTHRU)

;;   basic block 5, loop depth 0, count 0, freq 3956, maybe hot
;;    prev block 8, next block 6, flags: (NEW, REACHABLE, VISITED)
;;    pred:       4 [54.0%]  (TRUE_VALUE,EXECUTABLE)
  # VUSE <.MEM_13>
  # PT = nonlocal escaped null 
  _6 = MEM[(struct temporary_bufferD.2301 &)&_local_stateD.2411]._bufferD.2303;
  # .MEM_15 = VDEF <.MEM_13>
  MEM[(struct temporary_bufferD.2301 *)_4]._bufferD.2303 = _6;
  # .MEM_16 = VDEF <.MEM_15>
  MEM[(struct optionalD.2323 *)_4]._M_engagedD.2340 = 1;
;;    succ:       6 [100.0%]  (FALLTHRU,EXECUTABLE)

;;   basic block 6, loop depth 0, count 0, freq 10000, maybe hot
;;    prev block 5, next block 1, flags: (NEW, REACHABLE, VISITED)
;;    pred:       5 [100.0%]  (FALLTHRU,EXECUTABLE)
;;                3 [100.0%]  (FALLTHRU,EXECUTABLE)
  # .MEM_1 = PHI <.MEM_16(5), .MEM_17(3)>
  # .MEM_9 = VDEF <.MEM_1>
  _local_stateD.2411 ={v} {CLOBBER};
  # VUSE <.MEM_9>
  return;
;;    succ:       EXIT [100.0%] 

}

[Bug middle-end/88897] [9/10/11 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |9.4

[Bug middle-end/88897] [9/10/11/12 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2

[Bug middle-end/88897] [9/10/11/12 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|9.4                         |9.5

--- Comment #9 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 9.4 is being released, retargeting bugs to GCC 9.5.

[Bug tree-optimization/88897] [10/11/12/13 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|9.5                         |10.4

--- Comment #10 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 9 branch is being closed

[Bug tree-optimization/88897] [10/11/12/13 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|10.4                        |10.5

--- Comment #11 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 10.4 is being released, retargeting bugs to GCC 10.5.

[Bug tree-optimization/88897] [10/11/12 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[law at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

Jeffrey A. Law <law at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |law at gcc dot gnu.org
            Summary|[10/11/12/13 Regression]    |[10/11/12 Regression] Bogus
                   |Bogus maybe-uninitialized   |maybe-uninitialized warning
                   |warning on class field      |on class field (missed CSE)
                   |(missed CSE)                |

--- Comment #12 from Jeffrey A. Law <law at gcc dot gnu.org> ---
We're catching the previously missed CSE opportunity in fre1 now, and (of
course) we no longer get the bogus warning.  Removing the 13 regression marker.
 Finding and backporting the specific change seems like it's not worth it, but
leaving open with the other regression markers for now.

[Bug tree-optimization/88897] [11/12 Regression] Bogus maybe-uninitialized warning on class field (missed CSE)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88897

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|10.5                        |11.5

--- Comment #13 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 10 branch is being closed.