[Bug libstdc++/52128] New: [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[Bug libstdc++/52128] New: [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[mikpe at it dot uu.se]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

             Bug #: 52128
           Summary: [4.7 regression] r183788 caused massive libstdc++
                    testsuite regression on i686-linux
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: mikpe@it.uu.se


When bootstrapping gcc-4.7-20120203 on i686-linux I got massive regressions in
the C++ and libstd++ test suites, gcc-4.7-20120128 was OK:

                === g++ Summary ===

-# of expected passes           45969
-# of expected failures         284
-# of unsupported tests         350
+# of expected passes           37467
+# of unexpected failures       4114
+# of expected failures         277
+# of unresolved testcases      111
+# of unsupported tests         419

                === libgomp Summary ===

-# of expected passes           2948
+# of expected passes           289
+# of unexpected failures       477

                === libstdc++ Summary ===

-# of expected passes           8826
-# of unexpected failures       6
-# of expected failures         44
-# of unsupported tests         143
+# of expected passes           2269
+# of unexpected failures       2668
+# of expected failures         5
+# of unsupported tests         561

Bisection identified r183788 as the cause:

Author: redi
Date: Wed Feb  1 00:20:08 2012
New Revision: 183788

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=183788
Log:
    PR libstdc++/49204
        ...

Looking in the logs all new failures appear to be:

/mnt/work/objdir47/i686-pc-linux-gnu/./libstdc++-v3/src/.libs/libstdc++.so:
undefined reference to `std::_Bind_simple_helper<void
(std::thread::*)()<std::reference_wrapper<std::thread> > >::__type
std::__bind_simple<void (std::thread::*)(), std::reference_wrapper<std::thread>
>(void (std::thread::*&&)(), std::reference_wrapper<std::thread>&&)'
collect2: error: ld returned 1 exit status
compiler exited with status 1

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2012-02-06
         AssignedTo|unassigned at gcc dot       |redi at gcc dot gnu.org
                   |gnu.org                     |
     Ever Confirmed|0                           |1

--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> 2012-02-06 00:11:28 UTC ---
mine

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mario-baumann at web dot de

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> 2012-02-06 08:52:48 UTC ---
*** Bug 52133 has been marked as a duplicate of this bug. ***

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

--- Comment #3 from Jonathan Wakely <redi at gcc dot gnu.org> 2012-02-06 09:32:28 UTC ---
Created attachment 26579
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26579
does this patch fix it?

I can't reproduce this on 

i686-pc-linux-gnu
x86_64-unknown-linux-gnu
x86_64-unknown-netbsd5.1
powerpc64-unknown-linux-gnu

I don't see those failures in i686-pc-linux-gnu testresults:

http://gcc.gnu.org/ml/gcc-testresults/2012-02/msg00415.html 
http://gcc.gnu.org/ml/gcc-testresults/2012-02/msg00413.html
http://gcc.gnu.org/ml/gcc-testresults/2012-02/msg00410.html
http://gcc.gnu.org/ml/gcc-testresults/2012-02/msg00310.html


What configure command are you running?

Have you tried a clean build in an empty objdir?

what version of binutils are you using?

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.7.0

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[mario-baumann at web dot de]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

--- Comment #4 from Mario Baumann <mario-baumann at web dot de> 2012-02-06 12:41:11 UTC ---
Hi Jonathan,

the attached fix works for bug 52133. Thanks!

Mario.

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

--- Comment #5 from Jonathan Wakely <redi at gcc dot gnu.org> 2012-02-06 12:46:27 UTC ---
Great, thanks for testing it, I'll check it in this evening.

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[mikpe at it dot uu.se]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

--- Comment #6 from Mikael Pettersson <mikpe at it dot uu.se> 2012-02-06 16:03:09 UTC ---
(In reply to comment #3)
> Created attachment 26579 [details]
> does this patch fix it?

It does, thanks.

> What configure command are you running?

/tmp/gcc-4.7-20120204/configure --prefix=/tmp/install47
--with-gmp=/home/mikpe/pkgs/linux-x86/gmp-5.0.2
--with-mpfr=/home/mikpe/pkgs/linux-x86/mpfr-3.1.0
--with-mpc=/home/mikpe/pkgs/linux-x86/mpc-0.9
--disable-build-poststage1-with-cxx --disable-libquadmath --disable-plugin
--disable-lto --disable-nls --disable-shared --disable-libmudflap
--disable-libgomp --enable-threads=posix --enable-checking=release
--enable-languages=c,c++

> Have you tried a clean build in an empty objdir?

I always build gcc in an empty objdir.

> what version of binutils are you using?

binutils-2.20.51.0.7-8.fc14.i686

Fedora 14 user-space, but with a gcc-4.5.3 host compiler and a linux-3.2
kernel.

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

--- Comment #7 from Jonathan Wakely <redi at gcc dot gnu.org> 2012-02-06 20:51:57 UTC ---
Author: redi
Date: Mon Feb  6 20:51:52 2012
New Revision: 183942

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=183942
Log:
    PR libstdc++/52128
    * src/c++11/future.cc: Add explicit instantiation.

Modified:
    trunk/libstdc++-v3/ChangeLog
    trunk/libstdc++-v3/src/c++11/future.cc

[Bug libstdc++/52128] [4.7 regression] r183788 caused massive libstdc++ testsuite regression on i686-linux

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52128

Jonathan Wakely <redi at gcc dot gnu.org> changed:

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

--- Comment #8 from Jonathan Wakely <redi at gcc dot gnu.org> 2012-02-06 21:24:21 UTC ---
fixed