[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

--- Comment #3 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
This reduced test actually regressed with r241187:

struct A {
  A(A &);
};
struct B : A { };

A foo ();

int
main ()
{
  B{foo()};
}

$ ./cc1plus -quiet x.C -std=c++17
during RTL pass: expand
x.C: In function ‘int main()’:
x.C:11:3: internal compiler error: in assign_temp, at function.c:984
   11 |   B{foo()};
      |   ^~~~~~~~
0x105cdf1 assign_temp(tree_node*, int, int)
        /home/mpolacek/src/gcc/gcc/function.c:984
0xdf123d expand_call(tree_node*, rtx_def*, int)
        /home/mpolacek/src/gcc/gcc/calls.c:3825
0xfe98d5 expand_expr_real_1(tree_node*, rtx_def*, machine_mode,
expand_modifier, rtx_def**, bool)
        /home/mpolacek/src/gcc/gcc/expr.c:11135
0xfdc3d9 expand_expr_real(tree_node*, rtx_def*, machine_mode, expand_modifier,
rtx_def**, bool)
        /home/mpolacek/src/gcc/gcc/expr.c:8358
0xfbb233 expand_normal
        /home/mpolacek/src/gcc/gcc/expr.h:288
0xfd7588 store_field
        /home/mpolacek/src/gcc/gcc/expr.c:7102
0xfcf3c6 expand_assignment(tree_node*, tree_node*, bool)
        /home/mpolacek/src/gcc/gcc/expr.c:5374
0xe12534 expand_call_stmt
        /home/mpolacek/src/gcc/gcc/cfgexpand.c:2701
0xe15cab expand_gimple_stmt_1
        /home/mpolacek/src/gcc/gcc/cfgexpand.c:3682
0xe1634a expand_gimple_stmt
        /home/mpolacek/src/gcc/gcc/cfgexpand.c:3847
0xe1e4f9 expand_gimple_basic_block
        /home/mpolacek/src/gcc/gcc/cfgexpand.c:5887
0xe2013b execute
        /home/mpolacek/src/gcc/gcc/cfgexpand.c:6542
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See <https://gcc.gnu.org/bugs/> for instructions.

[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

--- Comment #4 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
But this one is accepted by icc/clang++ yet we ICE:

struct A {
  A();
  A(const A &);
};
struct B : A { };

A foo ();

int
main ()
{
  B b{foo()};
}

[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[xerofoify at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

Nicholas Krause <xerofoify at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |xerofoify at gmail dot com

--- Comment #5 from Nicholas Krause <xerofoify at gmail dot com> ---
Seems to be regressing  due to missing a check for another constructor type in
type_has_nontrivial_copy_init or somewhere up the stack.Both the other two in
the last reported case are fine. Therefore I'm assuming is not be found as a
trivial move or copy constructor somewhere and this looks to the compiler like
a different constructor type:
B b{foo()};

Seems to be a trivial constructor to me but the compiler seems to be ICEing
because its not or detected as such and therefore is not correctly building
special temp member functions. The original patch seems to causing some issue 
in this way.

[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-valid-code
           Assignee|unassigned at gcc dot gnu.org      |mpolacek at gcc dot gnu.org
             Status|NEW                         |ASSIGNED

[Bug c++/90254] [8/9/10/11 Regression] ice on aggregate initialization of unmovable base

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

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

https://gcc.gnu.org/g:320054784250e572cb75d6f69ab44b2330d61d8b

commit r11-2704-g320054784250e572cb75d6f69ab44b2330d61d8b
Author: Jason Merrill <jason@redhat.com>
Date:   Wed Aug 12 05:45:02 2020 -0400

    c++: Copy elision and [[no_unique_address]]. [PR93711]

    We don't elide a copy from a function returning a class by value into a
base
    because that can overwrite data laid out in the tail padding of the base
    class; we need to handle [[no_unique_address]] fields the same way, or we
    ICE when the middle-end wants to create a temporary object of a
    TYPE_NEEDS_CONSTRUCTING type.

    This means that we can't always express initialization of a field with
    INIT_EXPR from a TARGET_EXPR the way we usually do, so I needed
    to change several places that were assuming that was sufficient.

    This also fixes 90254, the same problem with C++17 aggregate initialization
    of a base.

    gcc/cp/ChangeLog:

            PR c++/90254
            PR c++/93711
            * cp-tree.h (unsafe_return_slot_p): Declare.
            * call.c (is_base_field_ref): Rename to unsafe_return_slot_p.
            (build_over_call): Check unsafe_return_slot_p.
            (build_special_member_call): Likewise.
            * init.c (expand_default_init): Likewise.
            * typeck2.c (split_nonconstant_init_1): Likewise.

    gcc/testsuite/ChangeLog:

            PR c++/90254
            PR c++/93711
            * g++.dg/cpp1z/aggr-base10.C: New test.
            * g++.dg/cpp2a/no_unique_address7.C: New test.
            * g++.dg/cpp2a/no_unique_address7a.C: New test.

[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[jason at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

Jason Merrill <jason at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[8/9/10/11 Regression] ice  |[8/9/10 Regression] ice on
                   |on aggregate initialization |aggregate initialization of
                   |of unmovable base           |unmovable base

--- Comment #7 from Jason Merrill <jason at gcc dot gnu.org> ---
The patch fixes the testcase from comment #4 for GCC 11.  I agree with the
clang response that the original testcase is ill-formed, and will continue to
be rejected.  The testcase in comment #3 is still broken due to PR 66139.

[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

--- Comment #8 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-10 branch has been updated by Jason Merrill
<jason@gcc.gnu.org>:

https://gcc.gnu.org/g:97014e4ada448aa8978b3cd14ed95e0e56f375d9

commit r10-9168-g97014e4ada448aa8978b3cd14ed95e0e56f375d9
Author: Jason Merrill <jason@redhat.com>
Date:   Wed Aug 12 05:45:02 2020 -0400

    c++: Copy elision and [[no_unique_address]]. [PR93711]

    We don't elide a copy from a function returning a class by value into a
base
    because that can overwrite data laid out in the tail padding of the base
    class; we need to handle [[no_unique_address]] fields the same way, or we
    ICE when the middle-end wants to create a temporary object of a
    TYPE_NEEDS_CONSTRUCTING type.

    This means that we can't always express initialization of a field with
    INIT_EXPR from a TARGET_EXPR the way we usually do, so I needed
    to change several places that were assuming that was sufficient.

    This also fixes 90254, the same problem with C++17 aggregate initialization
    of a base.

    gcc/cp/ChangeLog:

            PR c++/90254
            PR c++/93711
            * cp-tree.h (unsafe_return_slot_p): Declare.
            * call.c (is_base_field_ref): Rename to unsafe_return_slot_p.
            (build_over_call): Check unsafe_return_slot_p.
            (build_special_member_call): Likewise.
            * init.c (expand_default_init): Likewise.
            * typeck2.c (split_nonconstant_init_1): Likewise.

    gcc/testsuite/ChangeLog:

            PR c++/90254
            PR c++/93711
            * g++.dg/cpp1z/aggr-base10.C: New test.
            * g++.dg/cpp2a/no_unique_address7.C: New test.
            * g++.dg/cpp2a/no_unique_address7a.C: New test.

[Bug c++/90254] [8/9/10 Regression] ice on aggregate initialization of unmovable base

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254
Bug 90254 depends on bug 98642, which changed state.

Bug 98642 Summary: [10 Regression] wrong "use of deleted function" error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98642

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

[Bug c++/90254] [9/10 Regression] ice on aggregate initialization of unmovable base

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|8.5                         |9.4

--- Comment #9 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 8 branch is being closed.

[Bug c++/90254] [9/10 Regression] ice on aggregate initialization of unmovable base

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

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

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

--- Comment #10 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 9.4 is being released, retargeting bugs to GCC 9.5.

[Bug c++/90254] [9/10 Regression] ice on aggregate initialization of unmovable base

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Blocks|                            |90319
                 CC|                            |language.lawyer at gmail dot com

--- Comment #11 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
*** Bug 90319 has been marked as a duplicate of this bug. ***


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90319
[Bug 90319] [C++17] aggregate initialization of base: ICE in assign_temp

[Bug c++/90254] [9 Regression] ice on aggregate initialization of unmovable base

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
   Target Milestone|9.5                         |10.3
             Status|ASSIGNED                    |RESOLVED
      Known to fail|                            |9.5.0

--- Comment #12 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed in GCC 10.3.

[Bug c++/90254] [9 Regression] ice on aggregate initialization of unmovable base

[m.cencora at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90254

m.cencora at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |m.cencora at gmail dot com

--- Comment #13 from m.cencora at gmail dot com ---
Slightly modified example still produces ICE on gcc-10.3 (works in gcc-11 and
later)

g++ -std=c++17

struct A {
  A();
  A(const A &);
};
struct B : A { };

A foo ();

int
main ()
{
  B{foo()};
}