[Bug c/101512] New: [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[Bug c/101512] New: [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[gscfq@t-online.de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

            Bug ID: 101512
           Summary: [11/12 Regression] ICE in
                    maybe_trim_constructor_store, at tree-ssa-dse.c:379
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: gscfq@t-online.de
  Target Milestone: ---

Changed between 20200503 and 20200510 :


$ cat z1.c
int n();
typedef unsigned long V __attribute__ ((vector_size (64)));
V
foo (int i, V v)
{
  i = ((V)(V){n()})[n()];
  return v + i;
}


$ gcc-11-20200503 -c z1.c -O2 -mavx512f
$
$ gcc-12-20210718 -c z1.c -O2 -mavx512f
during GIMPLE pass: dse
z1.c: In function 'foo':
z1.c:8:1: internal compiler error: in maybe_trim_constructor_store, at
tree-ssa-dse.c:379
    8 | }
      | ^
0xcf24be maybe_trim_constructor_store
        ../../gcc/tree-ssa-dse.c:379
0xcf24be maybe_trim_partially_dead_store
        ../../gcc/tree-ssa-dse.c:605
0xcf24be dse_optimize_stmt
        ../../gcc/tree-ssa-dse.c:1146
0xcf24be execute
        ../../gcc/tree-ssa-dse.c:1237

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |12.0

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|12.0                        |11.2

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2021-07-20
             Status|UNCONFIRMED                 |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
# .MEM_16 = VDEF <.MEM_15>
D.2844 = {_2};

that's not valid GIMPLE, but gimplification already produces it.  We have

  D.2844 = {_2};
  _3 = n ();
  _4 = VIEW_CONVERT_EXPR<long unsigned int[8]>(D.2844)[_3];

there, so maybe it's FEs doing wrong as well (failing to set TREE_ADDRESSABLE
or
DECL_NOT_GIMPLE_REG).  I will have a closer look.

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
So we have

VIEW_CONVERT_EXPR<long unsigned int[8]>(<<< Unknown tree: compound_literal_expr
    V D.2844 = {(long unsigned int) n ()}; >>>)[n ()]

but fail to mark D.2844 TREE_ADDRESSABLE.  That's because
c_common_mark_addressable_vec only looks through an outermost
C_MAYBE_CONST_EXPR
but we see

VIEW_CONVERT_EXPR<vector(8) long unsigned int>(<<< Unknown tree:
c_maybe_const_expr

  <<< Unknown tree: compound_literal_expr
    V D.2844 = {(long unsigned int) n ()}; >>> >>>)

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
While the ICE in DSE happens only with GCC 11+ the bogus GIMPLE IL is latent on
older branches as well, including GCC 8.5.

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

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

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

--- Comment #4 from Martin Liška <marxin at gcc dot gnu.org> ---
Btw. started with r11-165-geb72dc663e9070b2.

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|11.2                        |11.3

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 11.2 is being released, retargeting bugs to GCC 11.3

[Bug tree-optimization/101512] [11/12 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

--- Comment #6 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Richard Biener <rguenth@gcc.gnu.org>:

https://gcc.gnu.org/g:e63d76234d18cac731c4f3610d513bd8b39b5520

commit r12-2590-ge63d76234d18cac731c4f3610d513bd8b39b5520
Author: Richard Biener <rguenther@suse.de>
Date:   Wed Jul 21 09:14:24 2021 +0200

    c/101512 - fix missing address-taking in c_common_mark_addressable_vec

    c_common_mark_addressable_vec fails to look through C_MAYBE_CONST_EXPR
    in the case it isn't at the toplevel.

    2021-07-21  Richard Biener  <rguenther@suse.de>

            PR c/101512
    gcc/c-family/
            * c-common.c (c_common_mark_addressable_vec): Look through
            C_MAYBE_CONST_EXPR even if not at the toplevel.

    gcc/testsuite/
            * gcc.dg/torture/pr101512.c: New testcase.

[Bug tree-optimization/101512] [11 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[11/12 Regression] ICE in   |[11 Regression] ICE in
                   |maybe_trim_constructor_stor |maybe_trim_constructor_stor
                   |e, at tree-ssa-dse.c:379    |e, at tree-ssa-dse.c:379
      Known to work|                            |12.0
           Priority|P3                          |P2

[Bug tree-optimization/101512] [11 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by Richard Biener
<rguenth@gcc.gnu.org>:

https://gcc.gnu.org/g:129bf4f2efc0ec22ee14affd8c8a3bfe07896356

commit r11-8830-g129bf4f2efc0ec22ee14affd8c8a3bfe07896356
Author: Richard Biener <rguenther@suse.de>
Date:   Wed Jul 21 09:14:24 2021 +0200

    c/101512 - fix missing address-taking in c_common_mark_addressable_vec

    c_common_mark_addressable_vec fails to look through C_MAYBE_CONST_EXPR
    in the case it isn't at the toplevel.

    2021-07-21  Richard Biener  <rguenther@suse.de>

            PR c/101512
    gcc/c-family/
            * c-common.c (c_common_mark_addressable_vec): Look through
            C_MAYBE_CONST_EXPR even if not at the toplevel.

    gcc/testsuite/
            * gcc.dg/torture/pr101512.c: New testcase.

    (cherry picked from commit e63d76234d18cac731c4f3610d513bd8b39b5520)

[Bug tree-optimization/101512] [11 Regression] ICE in maybe_trim_constructor_store, at tree-ssa-dse.c:379

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101512

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
      Known to fail|                            |11.2.0
      Known to work|                            |11.2.1
             Status|ASSIGNED                    |RESOLVED

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