[Bug tree-optimization/103152] New: wrong code at -O1 on x86_64-linux-gnu

[Bug tree-optimization/103152] New: wrong code at -O1 on x86_64-linux-gnu

[zhendong.su at inf dot ethz.ch]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

            Bug ID: 103152
           Summary: wrong code at -O1 on x86_64-linux-gnu
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zhendong.su at inf dot ethz.ch
  Target Milestone: ---

This appears to be a recent regression.

[577] % gcctk -v
Using built-in specs.
COLLECT_GCC=gcctk
COLLECT_LTO_WRAPPER=/local/suz-local/software/local/gcc-trunk/libexec/gcc/x86_64-pc-linux-gnu/12.0.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../gcc-trunk/configure --disable-bootstrap
--prefix=/local/suz-local/software/local/gcc-trunk --enable-languages=c,c++
--disable-werror --enable-multilib --with-system-zlib
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 12.0.0 20211109 (experimental) [master r12-5011-g2abd924f91e] (GCC) 
[578] % 
[578] % gcctk -O0 small.c; ./a.out
[579] % 
[579] % gcctk -O1 small.c
[580] % ./a.out
Aborted
[581] % 
[581] % cat small.c
struct {
  int a:1;
} b;
int *c = (int *)&b, d;
int main() {
  d = c && (b.a = (d < 0) ^ 3);
  if (d != 1)
    __builtin_abort();
  return 0;
}

[Bug tree-optimization/103152] [11/12 Regression] wrong code at -O1 on x86_64-linux-gnu since r11-5890-g7d9767cb8eea0f21

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |11.3
                 CC|                            |jakub at gcc dot gnu.org,
                   |                            |marxin at gcc dot gnu.org
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2021-11-09
            Summary|wrong code at -O1 on        |[11/12 Regression] wrong
                   |x86_64-linux-gnu            |code at -O1 on
                   |                            |x86_64-linux-gnu since
                   |                            |r11-5890-g7d9767cb8eea0f21

--- Comment #1 from Martin Liška <marxin at gcc dot gnu.org> ---
Started with r11-5890-g7d9767cb8eea0f21.

[Bug tree-optimization/103152] [11/12 Regression] wrong code at -O1 on x86_64-linux-gnu since r11-5890-g7d9767cb8eea0f21

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |pinskia at gcc dot gnu.org
             Status|NEW                         |ASSIGNED
           Keywords|                            |wrong-code

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Mine.

[Bug tree-optimization/103152] [11/12 Regression] wrong code at -O1 on x86_64-linux-gnu since r11-5890-g7d9767cb8eea0f21

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
This is another signed one bit integer issue.

[Bug tree-optimization/103152] [12 Regression] wrong code at -O1 on x86_64-linux-gnu since r11-5890-g7d9767cb8eea0f21

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|                            |12.0
      Known to work|                            |11.1.0, 11.2.0
   Target Milestone|11.3                        |12.0
            Summary|[11/12 Regression] wrong    |[12 Regression] wrong code
                   |code at -O1 on              |at -O1 on x86_64-linux-gnu
                   |x86_64-linux-gnu since      |since
                   |r11-5890-g7d9767cb8eea0f21  |r11-5890-g7d9767cb8eea0f21

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Martin Liška from comment #1)
> Started with r11-5890-g7d9767cb8eea0f21.

Yes and no.
Tthis was actually fixed for GCC 11 by r11-7587-b610c30453d8 (you can confirm
that if you want).

I messed up r12-4276-882d806c1a8 and added an extra TYPE_PRECISION (type) != 1
which should not have been there. Removing the extra TYPE_PRECISION (type) != 1
check fixes this issue too.

[Bug tree-optimization/103152] [12 Regression] wrong code at -O1 on x86_64-linux-gnu since r12-1055-gb6bdd7a4cb41ee0

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #4)
> (In reply to Martin Liška from comment #1)
> > Started with r11-5890-g7d9767cb8eea0f21.
> 
> Yes and no.
> Tthis was actually fixed for GCC 11 by r11-7587-b610c30453d8 (you can
> confirm that if you want).

Note the latent issue (been since ?-1:0 was added to GCC 9) on GCC 10 and 9
were fixed by r10-10177-g4111b36079fe and r9-9765-g77867cd8185 (and didn't have
the issue which I accidently did for the trunk). So this is trunk only issue
still.

[Bug tree-optimization/103152] [12 Regression] wrong code at -O1 on x86_64-linux-gnu since r12-1055-gb6bdd7a4cb41ee0

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Created attachment 51754
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51754&action=edit
Patch which I will commit after bootstrap/test

Here is the patch which I am testing to remove the accidenly committed check. 
I did check to make sure it fixed the problem.

Note the problem showed up recently due to threading differences I think; it
has been definitely latent others.

[Bug tree-optimization/103152] [12 Regression] wrong code at -O1 on x86_64-linux-gnu since r12-1055-gb6bdd7a4cb41ee0

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103152

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Had a typo in the testcase name and the bug #, forgot the 1 :).
https://gcc.gnu.org/pipermail/gcc-cvs/2021-November/356460.html

Fixed.