[Bug c/111882] New: GCC: 14: internal compiler error: in get_expr_operands, at tree-ssa-operands.cc:940

[Bug c/111882] New: GCC: 14: internal compiler error: in get_expr_operands, at tree-ssa-operands.cc:940

[141242068 at smail dot nju.edu.cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

            Bug ID: 111882
           Summary: GCC: 14: internal compiler error: in
                    get_expr_operands, at tree-ssa-operands.cc:940
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: 141242068 at smail dot nju.edu.cn
  Target Milestone: ---

Compiler Explorer: https://gcc.godbolt.org/z/h9MMnxbvK

When compile this program with option -O2, gcc crashes:
```
static void __attribute__((noipa)) f(int n) {
  int i, j;
  struct S { char d[n]; int a; int b : 17; int c : 12; };
  struct S A[100][1111];
  for (i = 0; i < 100; i++) {
    asm volatile("" : : "g"(&A[0][0]) : "memory");
    for (j = 0; j < 1111; j++) A[i][j].b = 2;
  }
}

void g(void) { f(1); }
```

The stack dump:
```
unhandled expression in get_expr_operands():
 <save_expr 0x7fe6f6035440
    type <integer_type 0x7fe6f5ea55e8 int sizes-gimplified asm_written public
SI
        size <integer_cst 0x7fe6f5ea7180 constant 32>
        unit-size <integer_cst 0x7fe6f5ea7198 constant 4>
        align:32 warn_if_not_align:0 symtab:-167332928 alias-set 3
canonical-type 0x7fe6f5ea55e8 precision:32 min <integer_cst 0x7fe6f5ea7138
-2147483648> max <integer_cst 0x7fe6f5ea7150 2147483647>
        pointer_to_this <pointer_type 0x7fe6f5eadb28>>
    side-effects public
    arg:0 <var_decl 0x7fe6f5e94f30 n.0 type <integer_type 0x7fe6f5ea55e8 int>
        used ignored SI <source>:3:10 size <integer_cst 0x7fe6f5ea7180 32>
unit-size <integer_cst 0x7fe6f5ea7198 4>
        align:32 warn_if_not_align:0 context <function_decl 0x7fe6f6033b00 f>>>

during GIMPLE pass: ifcvt
<source>: In function 'f':
<source>:1:36: internal compiler error: in get_expr_operands, at
tree-ssa-operands.cc:940
    1 | static void __attribute__((noipa)) f(int n) {
      |                                    ^
0x231788e internal_error(char const*, ...)
        ???:0
0xa0002a fancy_abort(char const*, int, char const*)
        ???:0
0x12e792d operands_scanner::get_expr_operands(tree_node**, int)
        ???:0
0x12e71b8 operands_scanner::get_expr_operands(tree_node**, int)
        ???:0
0x12e71b8 operands_scanner::get_expr_operands(tree_node**, int)
        ???:0
0x12e71b8 operands_scanner::get_expr_operands(tree_node**, int)
        ???:0
0x12e7e50 operands_scanner::parse_ssa_operands()
        ???:0
0x12e8cfa operands_scanner::build_ssa_operands()
        ???:0
0x12e8f44 update_stmt_operands(function*, gimple*)
        ???:0
0xd66fec gsi_insert_before(gimple_stmt_iterator*, gimple*, gsi_iterator_update)
        ???:0
0x1196551 tree_if_conversion(loop*, vec<gimple*, va_heap, vl_ptr>*)
        ???:0
Please submit a full bug report, with preprocessed source (by using
-freport-bug).
Please include the complete backtrace with any bug report.
See <https://gcc.gnu.org/bugs/> for instructions.
```

[Bug tree-optimization/111882] [13/14 Regression] : internal compiler error: in get_expr_operand in ifcvt

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |13.3

[Bug tree-optimization/111882] [13/14 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays inside a struct

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2023-10-19
     Ever confirmed|0                           |1
            Summary|[13/14 Regression] :        |[13/14 Regression] :
                   |internal compiler error: in |internal compiler error: in
                   |get_expr_operand in ifcvt   |get_expr_operand in ifcvt
                   |                            |with Variable length arrays
                   |                            |inside a struct

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
We start out with:
  (*A.1_23)[i_33]{lb: 0 sz: _10 * 4}[j_34]{lb: 0 sz: _11 * 4}.b{off: _13 * 4} =
2;

And then we add:
  _ifc__49 = (*A.1_23)[i_33]{lb: 0 sz: _10 * 4}[j_34]{lb: 0 sz: _11 *
4}.D.2748{off: ((sizetype) SAVE_EXPR <n.0> + 3 & 18446744073709551612) + 4};

[Bug tree-optimization/111882] [13/14 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Looks like this was broken when bitfield expansion was added to ifcvt (I think
r13-3219-g25413fdb2ac24933214123e24ba165026452a6f2 ).

[Bug tree-optimization/111882] [13/14 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2
                 CC|                            |andre.simoesdiasvieira@arm.
                   |                            |com

[Bug tree-optimization/111882] [13/14 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[avieira at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

avieira at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |avieira at gcc dot gnu.org
           Assignee|unassigned at gcc dot gnu.org      |avieira at gcc dot gnu.org

--- Comment #3 from avieira at gcc dot gnu.org ---
Taking this, first time I see a SAVE_EXPR. It looks like it indicates
side-effects, I'm gonna see if I can detect the presence of side-effects and
reject lowering if so. Does that sound OK?

[Bug tree-optimization/111882] [13/14 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

--- Comment #4 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Andre Simoes Dias Vieira
<avieira@gcc.gnu.org>:

https://gcc.gnu.org/g:24cf1f600b8ad34c68a51f48884e72d01f729893

commit r14-4798-g24cf1f600b8ad34c68a51f48884e72d01f729893
Author: Andre Vieira <andre.simoesdiasvieira@arm.com>
Date:   Fri Oct 20 17:02:32 2023 +0100

    ifcvt: Don't lower bitfields with non-constant offsets [PR 111882]

    This patch stops lowering of bitfields by ifcvt when they have non-constant
    offsets as we are not likely to be able to do anything useful with those
during
    vectorization.  That also fixes the issue reported in PR 111882, which was
    being caused by an offset with a side-effect being lowered, but constants
have
    no side-effects so we will no longer run into that problem.

    gcc/ChangeLog:

            PR tree-optimization/111882
            * tree-if-conv.cc (get_bitfield_rep): Return NULL_TREE for
bitfields
            with non-constant offsets.

    gcc/testsuite/ChangeLog:

            * gcc.dg/vect/pr111882.c: New test.

[Bug tree-optimization/111882] [13 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[avieira at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

avieira at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[13/14/15 Regression] :     |[13 Regression] : internal
                   |internal compiler error: in |compiler error: in
                   |get_expr_operand in ifcvt   |get_expr_operand in ifcvt
                   |with Variable length arrays |with Variable length arrays
                   |and bitfields inside a      |and bitfields inside a
                   |struct                      |struct
      Known to work|                            |14.0

--- Comment #5 from avieira at gcc dot gnu.org ---
Fixed on gcc-14 (when it was trunk, so removing 14 and 15 tag. Still needs
backport to gcc-13

[Bug tree-optimization/111882] [13 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[ricbal02 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

Richard Ball <ricbal02 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ricbal02 at gcc dot gnu.org

--- Comment #6 from Richard Ball <ricbal02 at gcc dot gnu.org> ---
Confirmed error exists on gcc-13. Backport applies cleanly and fixes the error.
Will request permission to backport on the mailing list.

[Bug tree-optimization/111882] [13 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

--- Comment #7 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-13 branch has been updated by Richard Ball
<ricbal02@gcc.gnu.org>:

https://gcc.gnu.org/g:4950f6bcd3cce9deb630b76af42cd6d6968ba03f

commit r13-8678-g4950f6bcd3cce9deb630b76af42cd6d6968ba03f
Author: Andre Vieira <andre.simoesdiasvieira@arm.com>
Date:   Fri Oct 20 17:02:32 2023 +0100

    ifcvt: Don't lower bitfields with non-constant offsets [PR 111882]

    This patch stops lowering of bitfields by ifcvt when they have non-constant
    offsets as we are not likely to be able to do anything useful with those
during
    vectorization.  That also fixes the issue reported in PR 111882, which was
    being caused by an offset with a side-effect being lowered, but constants
have
    no side-effects so we will no longer run into that problem.

    gcc/ChangeLog:

            PR tree-optimization/111882
            * tree-if-conv.cc (get_bitfield_rep): Return NULL_TREE for
bitfields
            with non-constant offsets.

    gcc/testsuite/ChangeLog:

            * gcc.dg/vect/pr111882.c: New test.

    (cherry picked from commit 24cf1f600b8ad34c68a51f48884e72d01f729893)

[Bug tree-optimization/111882] [13 Regression] : internal compiler error: in get_expr_operand in ifcvt with Variable length arrays and bitfields inside a struct

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111882

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

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

--- Comment #8 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed.