[Bug target/94704] New: [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[Bug target/94704] New: [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

            Bug ID: 94704
           Summary: [8/9/10 Regression] class with empty base passed
                    incorrectly with -std=c++17 on s390x/s390
           Product: gcc
           Version: 10.0
            Status: UNCONFIRMED
          Keywords: ABI, wrong-code
          Severity: blocker
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jakub at gcc dot gnu.org
                CC: jakub at gcc dot gnu.org, ktkachov at gcc dot gnu.org,
                    matmal01 at gcc dot gnu.org, mpolacek at gcc dot gnu.org,
                    rearnsha at gcc dot gnu.org, redi at gcc dot gnu.org,
                    rguenth at gcc dot gnu.org, rsandifo at gcc dot gnu.org
        Depends on: 94383
  Target Milestone: ---
            Target: aarch64-*-linux

+++ This bug was initially created as a clone of Bug #94383 +++

In https://gcc.gnu.org/pipermail/gcc-testresults/2020-April/559424.html
I see
FAIL: tmpdir-g++.dg-struct-layout-1/t032 cp_compat_x_tst.o-cp_compat_y_tst.o
execute 
FAIL: tmpdir-g++.dg-struct-layout-1/t055 cp_compat_x_tst.o-cp_compat_y_tst.o
execute 
FAIL: tmpdir-g++.dg-struct-layout-1/t056 cp_compat_x_tst.o-cp_compat_y_tst.o
execute 
FAIL: tmpdir-g++.dg-struct-layout-1/t057 cp_compat_x_tst.o-cp_compat_y_tst.o
execute 
FAIL: tmpdir-g++.dg-struct-layout-1/t058 cp_compat_x_tst.o-cp_compat_y_tst.o
execute 
FAIL: tmpdir-g++.dg-struct-layout-1/t059 cp_compat_x_tst.o-cp_compat_y_tst.o
execute 
That suggests similar ABI bug as on aarch64 (PR94383), where the presence or
absence of C++17 empty base artificial FIELD_DECL affects the argument passing
(or return?).
One should be able to do
make check-g++ RUNTESTFLAGS='--target_board=unix/-DDBG struct-layout-1.exp'
to get more information into the log file which exact tests fail, from that
number one can look up in gcc/testsuite/g++/g++.dg-struct-layout-1 subdir of
the build directory t032_test.h etc. the T(...) line with the reported
number(s),
and one should be able to remove all but the chosen one T(...) line from the
file to see a minimized testcase (+ preprocess to better see how exactly the
structure looks like).


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94383
[Bug 94383] [8/9/10 Regression] class with empty base passed incorrectly with
-std=c++17 on aarch64

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |10.0
                 CC|                            |krebbel at gcc dot gnu.org

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Completely untested guess:
--- gcc/config/s390/s390.c.jj   2020-03-14 08:14:47.097741411 +0100
+++ gcc/config/s390/s390.c      2020-04-22 14:24:17.980091105 +0200
@@ -11917,7 +11917,8 @@ s390_function_arg_vector (machine_mode m

       for (field = TYPE_FIELDS (type); field; field = DECL_CHAIN (field))
        {
-         if (TREE_CODE (field) != FIELD_DECL)
+         if (TREE_CODE (field) != FIELD_DECL
+             || cxx17_empty_base_field_p (field))
            continue;

          if (single == NULL_TREE)
@@ -11967,7 +11968,8 @@ s390_function_arg_float (machine_mode mo

       for (field = TYPE_FIELDS (type); field; field = DECL_CHAIN (field))
        {
-         if (TREE_CODE (field) != FIELD_DECL)
+         if (TREE_CODE (field) != FIELD_DECL
+             || cxx17_empty_base_field_p (field))
            continue;

          if (single == NULL_TREE)
(on top of the today posted cxx17_empty_base_field_p patch).
Though, of course that doesn't handle the -Wpsabi part if s390{,x} wants to
emit such diagnostics; for that the functions or their caller should determine
if they would make a different ABI decisions if the field wouldn't be skipped.

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[krebbel at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

Andreas Krebbel <krebbel at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2020-04-22
             Status|UNCONFIRMED                 |NEW

--- Comment #2 from Andreas Krebbel <krebbel at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #1)

You patch fixes the testcases on IBM Z. I'll run a full bootstrap test over
night.

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704
Bug 94704 depends on bug 94383, which changed state.

Bug 94383 Summary: [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on aarch64
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94383

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

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

--- Comment #3 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:9612a4833d761e3beda083a3e4dc92feba3b01bc

commit r10-7985-g9612a4833d761e3beda083a3e4dc92feba3b01bc
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Mon Apr 27 09:11:57 2020 +0200

    s390: Fix C++14 vs. C++17 ABI incompatibility on s390{,x} [PR94704]

    The following patch fixes the C++14 vs. C++17 ABI passing incompatibility
    on s390x-linux.

    Bootstrapped/regtested on s390x-linux without and with the patch, the
    difference being:
    -FAIL: tmpdir-g++.dg-struct-layout-1/t032
cp_compat_x_alt.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t032
cp_compat_x_tst.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t032
cp_compat_x_tst.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t055
cp_compat_x_alt.o-cp_compat_y_alt.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t055
cp_compat_x_alt.o-cp_compat_y_tst.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t055
cp_compat_x_tst.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t055
cp_compat_x_tst.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t056
cp_compat_x_alt.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t056
cp_compat_x_alt.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t056
cp_compat_x_tst.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t056
cp_compat_x_tst.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t057
cp_compat_x_alt.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t057
cp_compat_x_alt.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t057
cp_compat_x_tst.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t057
cp_compat_x_tst.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t058
cp_compat_x_alt.o-cp_compat_y_alt.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t058
cp_compat_x_alt.o-cp_compat_y_tst.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t058
cp_compat_x_tst.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t058
cp_compat_x_tst.o-cp_compat_y_tst.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t059
cp_compat_x_alt.o-cp_compat_y_alt.o execute
     FAIL: tmpdir-g++.dg-struct-layout-1/t059
cp_compat_x_alt.o-cp_compat_y_tst.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t059
cp_compat_x_tst.o-cp_compat_y_alt.o execute
    -FAIL: tmpdir-g++.dg-struct-layout-1/t059
cp_compat_x_tst.o-cp_compat_y_tst.o execute
    when performing ALT_CXX_UNDER_TEST=g++ testing with a system GCC 10
compiler
    from a week ago.  So, the alt vs. alt FAILs are all expected (we know
before
    this patch there is an ABI incompatibility) and some alt vs. tst (or tst
vs.
    alt) FAILs too - that depends on if the particular x or y test is compiled
    with -std=c++14 or -std=c++17 - if x_tst is compiled with -std=c++14 and
    y_alt is compiled with -std=c++17, then it should FAIL, similarly if x_alt
    is compiled with -std=c++17 and y_tst is compiled with -std=c++14.

    2020-04-27  Jakub Jelinek  <jakub@redhat.com>

            PR target/94704
            * config/s390/s390.c (s390_function_arg_vector,
            s390_function_arg_float): Ignore cxx17_empty_base_field_p fields.

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

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

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

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed on the trunk, should not be backported as it is an ABI change.

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

--- Comment #5 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:dde5ce541e3258276848aee85229a71c0e5f6965

commit r10-8007-gdde5ce541e3258276848aee85229a71c0e5f6965
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Tue Apr 28 10:26:24 2020 +0200

    s390: -Wpsabi diagnostics for C++14 vs. C++17 ABI incompatibility on
s390{,x} [PR94704]

    > We probably have to look into providing a -Wpsabi warning as well.

    So like this?

    2020-04-28  Jakub Jelinek  <jakub@redhat.com>

            PR target/94704
            * config/s390/s390.c (s390_function_arg_vector,
            s390_function_arg_float): Emit -Wpsabi diagnostics if the ABI
changed.

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

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

https://gcc.gnu.org/g:48e54fea7ba4a7cb7b3d1505951383120220e394

commit r10-8057-g48e54fea7ba4a7cb7b3d1505951383120220e394
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Wed Apr 29 22:38:01 2020 +0200

    s390: Fix up -Wpsabi diagnostics + [[no_unique_address]] empty member fix
[PR94704]

    So, based on the yesterday's discussions, similarly to powerpc64le-linux
    I've done some testing for s390x-linux too.

    First of all, I found a bug in my patch from yesterday, it was printing
    the wrong type like 'double' etc. rather than the class that contained such
    the element.  Fix below.

    For s390x-linux, I was using
    struct X { };
    struct Y { int : 0; };
    struct Z { int : 0; Y y; };
    struct U : public X { X q; };
    struct A { double a; };
    struct B : public X { double a; };
    struct C : public Y { double a; };
    struct D : public Z { double a; };
    struct E : public U { double a; };
    struct F { [[no_unique_address]] X x; double a; };
    struct G { [[no_unique_address]] Y y; double a; };
    struct H { [[no_unique_address]] Z z; double a; };
    struct I { [[no_unique_address]] U u; double a; };
    struct J { double a; [[no_unique_address]] X x; };
    struct K { double a; [[no_unique_address]] Y y; };
    struct L { double a; [[no_unique_address]] Z z; };
    struct M { double a; [[no_unique_address]] U u; };
     #define T(S, s) extern S s; extern void foo##s (S); int bar##s () { foo##s
(s); return 0; }
    T (A, a)
    T (B, b)
    T (C, c)
    T (D, d)
    T (E, e)
    T (F, f)
    T (G, g)
    T (H, h)
    T (I, i)
    T (J, j)
    T (K, k)
    T (L, l)
    T (M, m)
    as testcase and looking for "\tld\t%f0,".
    While g++ 9 with -std=c++17 used to pass in fpr just
    A, g++ 9 -std=c++14, as well as current trunk -std=c++14 & 17
    and clang++ from today -std=c++14 & 17 all pass A, B, C
    in fpr and nothing else.  The intent stated by Jason seems to be
    that A, B, C, F, G, J, K should all be passed in fpr.

    Attached are two (updated) versions of the patch on top of the
    powerpc+middle-end patch just posted.

    The first one emits two separate -Wpsabi warnings like powerpc, one for
    the -std=c++14 vs. -std=c++17 ABI difference and one for GCC 9 vs. 10
    [[no_unique_address]] passing changes, the other one is silent about the
    second case.

    2020-04-29  Jakub Jelinek  <jakub@redhat.com>

            PR target/94704
            * config/s390/s390.c (s390_function_arg_vector,
            s390_function_arg_float): Use DECL_FIELD_ABI_IGNORED instead of
            cxx17_empty_base_field_p.  In -Wpsabi diagnostics use the type
            passed to the function rather than the type of the single element.
            Rename cxx17_empty_base_seen variable to empty_base_seen, change
            type to int, and adjust diagnostics depending on if the field
            has [[no_unique_attribute]] or not.

            * g++.target/s390/s390.exp: New file.
            * g++.target/s390/pr94704-1.C: New test.
            * g++.target/s390/pr94704-2.C: New test.
            * g++.target/s390/pr94704-3.C: New test.
            * g++.target/s390/pr94704-4.C: New test.

[Bug target/94704] [8/9/10 Regression] class with empty base passed incorrectly with -std=c++17 on s390x/s390

[timturnerc at yahoo dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94704

Tim Turner <timturnerc at yahoo dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |timturnerc at yahoo dot com

--- Comment #7 from Tim Turner <timturnerc at yahoo dot com> ---
With this patch (not sure yet whether it's relevant) in place:  
... http://www-look-4.com/category/computers/
diff --git a/gdb/testsuite/lib/gdbserver-support.exp
b/gdb/testsuite/lib/gdbserver-support.
Exp https://komiya-dental.com/health/healthy-foods/
index a2cc80f28d..7b9c0eef6e 100644
--- a/gdb/testsuite/lib/gdbserver-support.exp
http://www.iu-bloomington.com/services/travel-services/
+++ b/gdb/testsuite/lib/gdbserver-support.exp
@@ -451,8 +451,10 @@ proc gdbserver_exit { is_mi } {
https://waytowhatsnext.com/technology/korean-technology/
            # We use expect rather than gdb_expect because
            # we want to suppress printing exception messages, otherwise,
http://www.wearelondonmade.com/category/health/
            # remote_expect, invoked by gdb_expect, prints the exceptions.
+           set read_prompt 0
            expect { http://www.jopspeech.com/category/property/
                -i "$gdb_spawn_id" -re "$gdb_prompt $" {
+                   set read_prompt 1 http://joerg.li/category/technology/
                    exp_continue
                }
                -i "$server_spawn_id" eof {
http://connstr.net/property/mars-researches/
@@ -463,6 +465,7 @@ proc gdbserver_exit { is_mi } {
                    warning "Timed out waiting for EOF in server after
$monitor_exit" http://the-hunters.org/category/technology/
                }
}       
+           gdb_assert {$read_prompt}
        }
     } http://embermanchester.uk/health/social-privacy/
     close_gdbserver
...
and running in parallel with:
http://fishingnewsletters.co.uk/category/shopping/
...
$ stress -c 5 http://www.slipstone.co.uk/category/services/
...
I ran into:
...
(gdb) PASS: gdb.multi/multi-target.exp: continue: non-stop=on: inferior 2
http://www.logoarts.co.uk/category/travel/
Remote debugging from host ::1, port 34088^M
Process build/gdb/testsuite/outputs/gdb.multi/multi-target/multi-target
created; pid http://www.go-mk-websites.co.uk/category/shopping/  = 8649^M
monitor exit^M
(gdb) Killing process(es): 8649^M
http://www.acpirateradio.co.uk/technology/facetime/
#9 0x16a2c57 in pop_all_targets_at_and_above(strata)
/home/vries/gdb_versions/devel/src/gdb/target.c:678
http://www.mconstantine.co.uk/category/shopping/
    #10 0x1442749 in remote_unpush_target 
http://www.compilatori.com/tech/xiaomi/
/home/vries/gdb_versions/devel/src/gdb/remote.c:5522
    #11 0x1458c16 in remote_target::readchar(int)
/home/vries/gdb_versions/devel/src/gdb/remote.c:9137
https://www.webb-dev.co.uk/sports/how-to-choose-sportwear/ 
    #12 0x145b25b in remote_target::getpkt_or_notif_sane_1(std::vector<char,
gdb::default_init_allocator<char, std::allocator<char> > >*, int, int, int*)