[Bug bootstrap/100246] New: [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[Bug bootstrap/100246] New: [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

            Bug ID: 100246
           Summary: [11/12 Regression] GCC will not bootstrap with clang
                    3.4/3.5 [xcode 5/6, Darwin 12/13]
           Product: gcc
           Version: 10.2.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: bootstrap
          Assignee: unassigned at gcc dot gnu.org
          Reporter: iains at gcc dot gnu.org
  Target Milestone: ---

using CC=clang and CXX="clang++ -stdlib=libc++" which is needed to get C++11
support.

In file included from /src-local/gcc-git-11/gcc/config/i386/i386-options.c:94:
/src-local/gcc-git-11/gcc/config/i386/x86-tune-costs.h:363:13: error: call to
implicitly-deleted default constructor of 'const struct stringop_strategy'
  {libcall, {{-1, rep_prefix_4_byte, false}}},
            ^
/src-local/gcc-git-11/gcc/config/i386/i386.h:254:15: note: default constructor
of 'stringop_strategy' is implicitly deleted because field 'max' of
const-qualified type 'const int' would not be initialized
    const int max;

====

when building in c++89 mode, (GCC10.x and earlier) this worked fine.  It does
look like a clang bug (but the rules are quite complex around this topic).  It
would seem it's trying to find a constructor for the empty elements in the
lists (and that is implicitly deleted).

====

A workaround is to ensure that all the elements have a complete initialisers
with dummy entries for those that are not needed.  Patch pending.

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

Iain Sandoe <iains at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
               Host|                            |x86_64-darwin12,
                   |                            |x86_64-darwin13
           Keywords|                            |build
              Build|                            |x86_64-darwin12,
                   |                            |x86_64-darwin13
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2021-04-24

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #1 from Iain Sandoe <iains at gcc dot gnu.org> ---
Created attachment 50669
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=50669&action=edit
Patch that works around the clang issue/bug

this is against GCC-11.1rc2, but not likely to need much/any amendment for
GCC12.

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |11.2
            Version|10.2.1                      |11.0

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[gcc@Denis-Excoffier.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

Denis Excoffier <gcc@Denis-Excoffier.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |gcc@Denis-Excoffier.org

--- Comment #2 from Denis Excoffier <gcc@Denis-Excoffier.org> ---
My linux (2.6.32-39-pve) fails bootstrapping recent GCC 11.1.0 using
gcc (Debian 4.9.2-10+deb8u2) 4.9.2, with the following errors:

In file included from
/tmp/lcl/tmp/gcc/gcc-11.1.0/gcc/config/i386/i386-options.c:94:0:
/tmp/lcl/tmp/gcc/gcc-11.1.0/gcc/config/i386/x86-tune-costs.h:32:56: error:
uninitialized const member 'stringop_algs::stringop_strategy::max'
   {rep_prefix_1_byte, {{-1, rep_prefix_1_byte, false}}}};
                                                        ^
/tmp/lcl/tmp/gcc/gcc-11.1.0/gcc/config/i386/x86-tune-costs.h:32:56: warning:
missing initializer for member 'stringop_algs::stringop_strategy::max'
[-Wmissing-field-initializers]
/tmp/lcl/tmp/gcc/gcc-11.1.0/gcc/config/i386/x86-tune-costs.h:32:56: error:
uninitialized const member 'stringop_algs::stringop_strategy::alg'
(...hundreds of similar messages...)

Applying this patch clears the issue.

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #3 from Iain Sandoe <iains at gcc dot gnu.org> ---
Alan Modra posted the following alternate patch (which I will test with a
clang-3.5 bootstrap):

diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h
index 97d6f3863cb..cc3b1b6d666 100644
--- a/gcc/config/i386/i386.h
+++ b/gcc/config/i386/i386.h
@@ -73,8 +73,8 @@ struct stringop_algs
{
  const enum stringop_alg unknown_size;
  const struct stringop_strategy {
-    const int max;
-    const enum stringop_alg alg;
+    int max;
+    enum stringop_alg alg;
    int noalign;
  } size [MAX_STRINGOP_ALGS];
};

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #4 from Iain Sandoe <iains at gcc dot gnu.org> ---
for the record the patch in comment #3 allowed bootstrap to succeed on Darwin12
with clang 3.4-based Xcode.

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[gcc@Denis-Excoffier.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #5 from Denis Excoffier <gcc@Denis-Excoffier.org> ---
same for me, the patch in comment #3 allowed bootstrap to succeed on linux
(2.6.32-39-pve) with gcc (Debian 4.9.2-10+deb8u2).

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #6 from Iain Sandoe <iains at gcc dot gnu.org> ---
yet another permutation: 
This one keeps the const-ness of the contents, but provides a CTOR.
Note the we build some of the gcov stuff with a C compiler and therefore this
has to be conditional on C++.


diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h
index 97d6f38..a417c93 100644
--- a/gcc/config/i386/i386.h
+++ b/gcc/config/i386/i386.h
@@ -73,6 +73,11 @@ struct stringop_algs
 {
   const enum stringop_alg unknown_size;
   const struct stringop_strategy {
+#ifdef __cplusplus
+    stringop_strategy(int _max = -1, enum stringop_alg _alg = libcall,
+                     int _noalign = false)
+      : max (_max), alg (_alg), noalign (_noalign) {}
+#endif
     const int max;
     const enum stringop_alg alg;
     int noalign;

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bug-reports.delphin@laposte
                   |                            |.net

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
*** Bug 100585 has been marked as a duplicate of this bug. ***

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #8 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Iain D Sandoe <iains@gcc.gnu.org>:

https://gcc.gnu.org/g:f99f6eb58e1f894dae024f63cc2fe30fa7605e59

commit r12-2052-gf99f6eb58e1f894dae024f63cc2fe30fa7605e59
Author: Iain Sandoe <iain@sandoe.co.uk>
Date:   Sat Jul 3 19:47:48 2021 +0100

    X86: Provide a CTOR for stringop_algs [PR100246].

    Several older compilers fail to build modern GCC because of missing
    or incomplete C++11 support.

    Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>

    PR bootstrap/100246 - [11/12 Regression] GCC will not bootstrap with clang
3.4/3.5 [xcode 5/6, Darwin 12/13]

            PR bootstrap/100246

    gcc/ChangeLog:

            * config/i386/i386.h (struct stringop_algs): Define a CTOR for
            this type.

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #9 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by Iain D Sandoe
<iains@gcc.gnu.org>:

https://gcc.gnu.org/g:5dd3fe90a5c0067e3e21d81aa904d13a4154eb4a

commit r11-8788-g5dd3fe90a5c0067e3e21d81aa904d13a4154eb4a
Author: Iain Sandoe <iain@sandoe.co.uk>
Date:   Sat Jul 3 19:47:48 2021 +0100

    X86: Provide a CTOR for stringop_algs [PR100246].

    Several older compilers fail to build modern GCC because of missing
    or incomplete C++11 support.

    Signed-off-by: Iain Sandoe <iain@sandoe.co.uk>

    PR bootstrap/100246 - [11/12 Regression] GCC will not bootstrap with clang
3.4/3.5 [xcode 5/6, Darwin 12/13]

            PR bootstrap/100246

    gcc/ChangeLog:

            * config/i386/i386.h (struct stringop_algs): Define a CTOR for
            this type.

    (cherry picked from commit f99f6eb58e1f894dae024f63cc2fe30fa7605e59)

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

Iain Sandoe <iains at gcc dot gnu.org> changed:

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

--- Comment #10 from Iain Sandoe <iains at gcc dot gnu.org> ---
so should be fixed for 11.2

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #11 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:858d7ee1a0cd97c01d844ea73ab81918da738344

commit r12-4944-g858d7ee1a0cd97c01d844ea73ab81918da738344
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Fri Nov 5 16:39:14 2021 +0100

    x86: Make stringop_algs::stringop_strategy ctor constexpr [PR100246]

    > Several older compilers fail to build modern GCC because of missing
    > or incomplete C++11 support.
    >
    >       * config/i386/i386.h (struct stringop_algs): Define a CTOR for
    >       this type.

    Unfortunately, as mentioned in my
    https://gcc.gnu.org/pipermail/gcc-patches/2021-November/583289.html
    mail, without the new dyninit pass this causes dynamic initialization of
    many variables, 6.5KB _GLOBAL__sub_I_* on x86_64 and 12.5KB on i686.

    The following patch makes the ctor constexpr so that already the FE
    is able to statically initialize all those.

    I have tested on godbolt a reduced testcase without a constructor,
    with constructor and with constexpr constructor.
    clang before 3.3 is unhappy about all the 3 cases, clang 3.3 and 3.4
    is ok with ctor and ctor with constexpr and optimizes it into static
    initialization, clang 3.5+ is ok with all 3 versions and optimizes,
    gcc 4.8 and 5+ is ok with all 3 versions and no ctor and ctor with
constexpr
    is optimized, gcc 4.9 is unhappy about the no ctor case and happy with the
    other two.

    2021-11-05  Jakub Jelinek  <jakub@redhat.com>

            PR bootstrap/100246
            * config/i386/i386.h
            (stringop_algs::stringop_strategy::stringop_strategy): Make the
ctor
            constexpr.

[Bug bootstrap/100246] [11/12 Regression] GCC will not bootstrap with clang 3.4/3.5 [xcode 5/6, Darwin 12/13]

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100246

--- Comment #12 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by Jakub Jelinek
<jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:04d786a7add8f824af3b0c0d29884770810f7001

commit r11-9329-g04d786a7add8f824af3b0c0d29884770810f7001
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Fri Nov 5 16:39:14 2021 +0100

    x86: Make stringop_algs::stringop_strategy ctor constexpr [PR100246]

    > Several older compilers fail to build modern GCC because of missing
    > or incomplete C++11 support.
    >
    >       * config/i386/i386.h (struct stringop_algs): Define a CTOR for
    >       this type.

    Unfortunately, as mentioned in my
    https://gcc.gnu.org/pipermail/gcc-patches/2021-November/583289.html
    mail, without the new dyninit pass this causes dynamic initialization of
    many variables, 6.5KB _GLOBAL__sub_I_* on x86_64 and 12.5KB on i686.

    The following patch makes the ctor constexpr so that already the FE
    is able to statically initialize all those.

    I have tested on godbolt a reduced testcase without a constructor,
    with constructor and with constexpr constructor.
    clang before 3.3 is unhappy about all the 3 cases, clang 3.3 and 3.4
    is ok with ctor and ctor with constexpr and optimizes it into static
    initialization, clang 3.5+ is ok with all 3 versions and optimizes,
    gcc 4.8 and 5+ is ok with all 3 versions and no ctor and ctor with
constexpr
    is optimized, gcc 4.9 is unhappy about the no ctor case and happy with the
    other two.

    2021-11-05  Jakub Jelinek  <jakub@redhat.com>

            PR bootstrap/100246
            * config/i386/i386.h
            (stringop_algs::stringop_strategy::stringop_strategy): Make the
ctor
            constexpr.

    (cherry picked from commit 858d7ee1a0cd97c01d844ea73ab81918da738344)