[Bug bootstrap/56714] New: [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline f unction might not be inlinable

[Bug bootstrap/56714] New: [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline f unction might not be inlinable

[gerald at pfeifer dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

             Bug #: 56714
           Summary: [4.9 regression] Bootstrap failure
                    libitm/local_atomic:1580:3: error: always_inline f
                    unction might not be inlinable
    Classification: Unclassified
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: major
          Priority: P3
         Component: bootstrap
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: gerald@pfeifer.com


This appeared in the last 24-48 hours from what I can tell, both on
i386-unknown-freebsd10.0 and amd64-unknown-freebsd8.3.

In file included from /scratch2/tmp/gerald/gcc-HEAD/libitm/libitm_i.h:40:0,
                 from /scratch2/tmp/gerald/gcc-HEAD/libitm/aatree.cc:28:
/scratch2/tmp/gerald/gcc-HEAD/libitm/local_atomic:1580:3: error: always_inline
f
unction might not be inlinable [-Werror=attributes]
   atomic_flag_clear(volatile atomic_flag* __a) noexcept
   ^
/scratch2/tmp/gerald/gcc-HEAD/libitm/local_atomic:1576:3: error: always_inline
f
unction might not be inlinable [-Werror=attributes]
   atomic_flag_clear(atomic_flag* __a) noexcept
   ^
/scratch2/tmp/gerald/gcc-HEAD/libitm/local_atomic:1572:3: error: always_inline
f
unction might not be inlinable [-Werror=attributes]
   atomic_flag_test_and_set(volatile atomic_flag* __a) noexcept
   ^
:
cc1plus: all warnings being treated as errors
gmake[4]: *** [aatree.lo] Error 1
gmake[4]: Leaving directory
`/scratch2/tmp/gerald/OBJ-0324-1745/i386-unknown-fre
ebsd10.0/libitm'

For bootstrap-lean, in stage 3 it seems (lost the stage information).

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andi-gcc at firstfloor dot
                   |                            |org
   Target Milestone|---                         |4.9.0

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

Jan Hubicka <hubicka at gcc dot gnu.org> changed:

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

--- Comment #1 from Jan Hubicka <hubicka at gcc dot gnu.org> 2013-03-30 23:12:56 UTC ---
Well, the error message means that the function is not inline (either
!DECLARED_INLINE or UNINLINABLE) and has always_inline flag.
Can you attach the preprocessed source file?

Honza

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[gerald at pfeifer dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

--- Comment #2 from Gerald Pfeifer <gerald at pfeifer dot com> 2013-03-30 23:35:26 UTC ---
Created attachment 29756
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29756
Pre-processed source aatree.ii

All those always_inline attributes seem to come from libitm/local_atomic.

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

--- Comment #3 from Jan Hubicka <hubicka at gcc dot gnu.org> 2013-03-31 20:08:22 UTC ---
I guess it is really complaining about the lack of inline modifier.
  __attribute__((__always_inline__)) bool
  atomic_flag_test_and_set_explicit(atomic_flag* __a,
        memory_order __m) noexcept
  { return __a->test_and_set(__m); }

I think adding "inline" should do the trick?

Honza

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[gerald at pfeifer dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

Gerald Pfeifer <gerald at pfeifer dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |*-unknown-freebsd*
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2013-03-31
         AssignedTo|unassigned at gcc dot       |gerald at pfeifer dot com
                   |gnu.org                     |
     Ever Confirmed|0                           |1

--- Comment #4 from Gerald Pfeifer <gerald at pfeifer dot com> 2013-03-31 21:13:16 UTC ---
Thanks Honza!  I missed the obvious and the fact that Andi failed to
update ChangeLog didn't help.  Working on a patch...

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[kargl at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

kargl at gcc dot gnu.org changed:

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

--- Comment #5 from kargl at gcc dot gnu.org 2013-03-31 21:43:12 UTC ---
Was there a problem the other bug report about this?

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56784

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[gerald at pfeifer dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

--- Comment #6 from Gerald Pfeifer <gerald at pfeifer dot com> 2013-03-31 21:50:42 UTC ---
*** Bug 56784 has been marked as a duplicate of this bug. ***

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[gerald at pfeifer dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

--- Comment #7 from Gerald Pfeifer <gerald at pfeifer dot com> 2013-03-31 21:51:44 UTC ---
> Was there a problem the other bug report about this?

Not that I can see.  That other report was just a duplicate coming a
few days later.

[Bug bootstrap/56714] [4.9 regression] Bootstrap failure libitm/local_atomic:1580:3: error: always_inline function might not be inlinable

[gerald at pfeifer dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56714

Gerald Pfeifer <gerald at pfeifer dot com> changed:

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

--- Comment #8 from Gerald Pfeifer <gerald at pfeifer dot com> ---
I committed the following, which fixes this:

        PR bootstrap/56714
        * local_atomic (__always_inline): Always define our version.
        (__calculate_memory_order): Mark inline.
        (atomic_thread_fence): Ditto.
        (atomic_signal_fence): Ditto.
        (atomic_bool::atomic_flag_test_and_set_explicit): Ditto.
        (atomic_bool::atomic_flag_clear_explicit): Ditto.
        (atomic_bool::atomic_flag_test_and_set): Ditto.
        (atomic_bool::atomic_flag_clear): Ditto.