[Bug tree-optimization/110475] New: [14 Regression] Wrong code at -O2/3/s on x86_64-pc-linux-gnu

[Bug tree-optimization/110475] New: [14 Regression] Wrong code at -O2/3/s on x86_64-pc-linux-gnu

[jwzeng at nuaa dot edu.cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110475

            Bug ID: 110475
           Summary: [14 Regression] Wrong code at -O2/3/s on
                    x86_64-pc-linux-gnu
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jwzeng at nuaa dot edu.cn
  Target Milestone: ---

Link to the Compiler Explorer: https://godbolt.org/z/qTn1MErqh

The following code snippet:

#include <stdio.h>
unsigned long long int res = 0;
unsigned int a = 4042837206U;
int b = 1558001202;
#define min(a, b) ((a) < (b) ? (a) : (b))

int main() {
    res = (unsigned long long int) min(1, b ? a : 1558001202);
    printf("%llu\n", res);
    return 0;
}

> $ /usr/gcc-trunk/bin/gcc -O0 bug.c; ./a.out
> 1
> $ /usr/gcc-trunk/bin/gcc -O1 bug.c; ./a.out
> 1
> $ /usr/gcc-trunk/bin/gcc -O2 bug.c; ./a.out
> 4042837207
> $ /usr/gcc-trunk/bin/gcc -O3 bug.c; ./a.out
> 4042837207
> $ /usr/gcc-trunk/bin/gcc -Os bug.c; ./a.out
> 4042837207

When compiled with -O2/3/s, it prints the wrong result 4042837207 instead of 1.
Earlier GCCs do not have this bug.

The following statement is in the code snippet above:
> res = (unsigned long long int) min(1, b ? a : 1558001202);
When the last number 1558001202 of this statement is changed to 0 or 1, it will
also output the wrong result 4042837207 under the optimization of -O1. Changing
to other numbers that are not 0 or 1 will not make this program output wrong
results at -O1.

> $ /usr/gcc-trunk/bin/gcc --version

Using built-in specs.
COLLECT_GCC=/home/compilers/gcc/gcc-trunk/bin/gcc
COLLECT_LTO_WRAPPER=/home/compilers/gcc/gcc-trunk/bin/../libexec/gcc/x86_64-pc-linux-gnu/14.0.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../gcc-trunk-source/gcc/configure
--enable-languages=c,c++,fortran --enable-checking=release
--enable-valgrind-annotations --disable-werror --disable-libstdcxx-pch
--enable-libgomp --enable-lto --enable-gold --with-plugin-ld=gold
--prefix=/usr/local/gcc-trunk
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 14.0.0 20230619 (experimental) [master r14-1917-gf8e0270272] (GCC)

[Bug tree-optimization/110475] [14 Regression] Wrong code at -O2/3/s on x86_64-pc-linux-gnu

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110475

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
           Assignee|unassigned at gcc dot gnu.org      |pinskia at gcc dot gnu.org
             Status|UNCONFIRMED                 |ASSIGNED
           Keywords|                            |wrong-code
   Last reconfirmed|                            |2023-06-29

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Mine. Dup of the flow sensitive phiopt issue.

[Bug tree-optimization/110475] [14 Regression] Wrong code at -O2/3/s on x86_64-pc-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110475

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |x86_64-*-*
   Target Milestone|---                         |14.0

[Bug tree-optimization/110475] [14 Regression] Wrong code at -O2/3/s on x86_64-pc-linux-gnu

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110475

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

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

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Dup of bug 110252.

  _10 = a.2_2 > 1;
  _11 = (long long unsigned int) _10;
  _12 = iftmp.3_5 | _11;

*** This bug has been marked as a duplicate of bug 110252 ***