[Bug libstdc++/40038] New: [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and tanhl@GLIBCXX_3.4 not exported

[Bug libstdc++/40038] New: [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and tanhl@GLIBCXX_3.4 not exported

[debian-gcc at lists dot debian dot org]

seen in

https://buildd.debian.org/fetch.cgi?&pkg=gcc-4.4&ver=4.4.0-3&arch=mips&stamp=1241547561&file=log

PR26926 mentions these exports as unnecessary, but afaiu these are not supposed
to be removed in libstdc++.so.6.

  Matthias


-- 
           Summary: [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and
                    tanhl@GLIBCXX_3.4 not exported
           Product: gcc
           Version: 4.4.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: debian-gcc at lists dot debian dot org
GCC target triplet: mips-linux-gnu mipsel-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and tanhl@GLIBCXX_3.4 not exported

[paolo dot carlini at oracle dot com]

------- Comment #1 from paolo dot carlini at oracle dot com  2009-05-06 09:17 -------
Is this related to PR39491?


-- 

paolo dot carlini at oracle dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bkoz at redhat dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and tanhl@GLIBCXX_3.4 not exported

[debian-gcc at lists dot debian dot org]

------- Comment #2 from debian-gcc at lists dot debian dot org  2009-05-06 10:14 -------
same for arm-linux-gnu arm-linux-gnueabi

  Matthias


-- 

debian-gcc at lists dot debian dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 GCC target triplet|mips-linux-gnu mipsel-linux-|mips-linux-gnu mipsel-linux-
                   |gnu                         |gnu arm-linux-gnu arm-linux-
                   |                            |gnueabi


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and tanhl@GLIBCXX_3.4 not exported

[paolo dot carlini at oracle dot com]

------- Comment #3 from paolo dot carlini at oracle dot com  2009-05-06 11:02 -------
By the way, if you want to hear my advice, remember to regression test on those
much less widespread targets *before* major releases.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and tanhl@GLIBCXX_3.4 not exported

[paolo dot carlini at oracle dot com]

------- Comment #4 from paolo dot carlini at oracle dot com  2009-05-06 11:25 -------
Benjamin, can you have a look?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 and tanhl@GLIBCXX_3.4 not exported

[paolo dot carlini at oracle dot com]

-- 

paolo dot carlini at oracle dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.4.1


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 not exported

[jakub at gcc dot gnu dot org]

------- Comment #5 from jakub at gcc dot gnu dot org  2009-05-06 13:02 -------
Only ceill is actually missing in SVN.


-- 

jakub at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[4.4/4.5 regression] symbols|[4.4/4.5 regression] symbols
                   |ceill@GLIBCXX_3.4.3 and     |ceill@GLIBCXX_3.4.3 not
                   |tanhl@GLIBCXX_3.4 not       |exported
                   |exported                    |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 not exported

[paolo dot carlini at oracle dot com]

------- Comment #6 from paolo dot carlini at oracle dot com  2009-05-06 13:20 -------
Excellent. Then Matthias' suggested fix makes sense to me. Better if Benjamin
could also review it, however (lately, he touched this code, IIRC).


-- 

paolo dot carlini at oracle dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2009-05-06 13:20:29
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 not exported

[doko at gcc dot gnu dot org]

------- Comment #7 from doko at gcc dot gnu dot org  2009-05-07 06:55 -------
Subject: Bug 40038

Author: doko
Date: Thu May  7 06:55:15 2009
New Revision: 147217

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=147217
Log:
2009-05-07  Matthias Klose  <doko@ubuntu.com>

        PR libstdc++/40038
        * src/math_stubs_long_double.cc: Add ceill.

Modified:
    trunk/libstdc++-v3/ChangeLog
    trunk/libstdc++-v3/src/math_stubs_long_double.cc


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 not exported

[doko at gcc dot gnu dot org]

------- Comment #8 from doko at gcc dot gnu dot org  2009-05-07 07:03 -------
Subject: Bug 40038

Author: doko
Date: Thu May  7 07:03:17 2009
New Revision: 147218

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=147218
Log:
2009-05-07  Matthias Klose  <doko@ubuntu.com>

        PR libstdc++/40038
        * src/math_stubs_long_double.cc: Add ceill.

Modified:
    branches/gcc-4_4-branch/libstdc++-v3/ChangeLog
    branches/gcc-4_4-branch/libstdc++-v3/src/math_stubs_long_double.cc


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 not exported

[jakub at gcc dot gnu dot org]

------- Comment #9 from jakub at gcc dot gnu dot org  2009-05-07 07:19 -------
Fixed.


-- 

jakub at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 not exported

[bkoz at gcc dot gnu dot org]

------- Comment #10 from bkoz at gcc dot gnu dot org  2009-05-12 23:01 -------

Sorry for the delay Paolo, this fix looks fine.

I see tanhl missing too in that log, yet the gnu.ver exports have it and so
does src/math_compatibility_long_double.cc, although depending on
_GLIBCXX_HAVE_TANHL. What gives? Is this some kind of temporary build issue or
is this still indicative of configure/build issues?

Please note that there are no arm ABI baselines. Is this due to
multilib/configure issues and or large numbers of variants? Can Debian
volunteers fix this or elaborate? Hopefully the necessity of this is now
apparent. 

I appreciate the current efforts to get Debian builds back on track, but am a
bit alarmed at the releases w/ known ABI fails on secondary arches. What can be
done to improve this situation?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038

[Bug libstdc++/40038] [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3 not exported

[doko at debian dot org]

------- Comment #11 from doko at debian dot org  2009-05-13 16:28 -------
Subject: Re:  [4.4/4.5 regression] symbols ceill@GLIBCXX_3.4.3
 not exported

bkoz at gcc dot gnu dot org schrieb:
> ------- Comment #10 from bkoz at gcc dot gnu dot org  2009-05-12 23:01 -------
> 
> Sorry for the delay Paolo, this fix looks fine.
> 
> I see tanhl missing too in that log, yet the gnu.ver exports have it and so
> does src/math_compatibility_long_double.cc, although depending on
> _GLIBCXX_HAVE_TANHL. What gives? Is this some kind of temporary build issue or
> is this still indicative of configure/build issues?

no that was accidentally removed in your suggested workaround for PR39491.

> Please note that there are no arm ABI baselines. Is this due to
> multilib/configure issues and or large numbers of variants? Can Debian
> volunteers fix this or elaborate? Hopefully the necessity of this is now
> apparent.

I'll ask the Debian ARM port maintainers. would welcome feedback from GCC ARM
maintainers as well.

> I appreciate the current efforts to get Debian builds back on track, but am a
> bit alarmed at the releases w/ known ABI fails on secondary arches. What can be
> done to improve this situation?

baseline files would help. Debian itself does include symbol files as well, but
in this case I didn't update those when adding the new armeabi support. But I
only will notice such regressions when building the package for Debian or
Ubuntu, which I did for arm not before the 4.4 release.

  Matthias


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40038