[Bug c++/22284] New: [4.1 Regression] Many C++ testsuite failures

[Bug c++/22284] New: [4.1 Regression] Many C++ testsuite failures

[jsm28 at gcc dot gnu dot org]

Many C++ testsuite execution tests started failing on mainline between 20050627
and 20050630, on platforms including hppa2.0w-hpux, hppa64-hpux and ia64-hpux. 
gcc-testresults shows failures also on other platforms such as ia64-linux.

Binary searching for "g++.dg/eh/cond1.C execution test" as a test failing on all
the above platforms shows that on hppa2.0w-hp-hpux11.23 it started failing
between 2005-06-27 07:39 UTC and 2005-06-27 07:42 UTC, i.e. this is caused by
the stack protection patch.

http://gcc.gnu.org/ml/gcc-cvs/2005-06/msg00980.html

<http://gcc.gnu.org/ml/gcc-patches/2005-07/msg00116.html> might be relevant
except that the problem includes targets not using sjlj exceptions.

-- 
           Summary: [4.1 Regression] Many C++ testsuite failures
           Product: gcc
           Version: 4.1.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: jsm28 at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org,jakub at gcc dot gnu dot
                    org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |EH, sjlj-eh, wrong-code
   Target Milestone|---                         |4.1.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|sjlj-eh                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[jakub at redhat dot com]

------- Additional Comments From jakub at redhat dot com  2005-07-03 12:08 -------
Strange, I have certainly bootstrapped/regtested the initial -fstack-protector
patch on ia64-linux and there were no regressions, neither on HEAD
(20050625 tree + those patches) nor in the 4.0.1 backport.
g++.dg/eh/cond1.C compiles/runs just fine for me.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rth at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[joseph at codesourcery dot com]

------- Additional Comments From joseph at codesourcery dot com  2005-07-03 12:17 -------
Subject: Re:  [4.1 Regression] Many C++ testsuite failures

On Sun, 3 Jul 2005, jakub at redhat dot com wrote:

> Strange, I have certainly bootstrapped/regtested the initial -fstack-protector
> patch on ia64-linux and there were no regressions, neither on HEAD
> (20050625 tree + those patches) nor in the 4.0.1 backport.
> g++.dg/eh/cond1.C compiles/runs just fine for me.

Example ia64-linux results showing the failure are at:

http://gcc.gnu.org/ml/gcc-testresults/2005-07/msg00117.html
http://gcc.gnu.org/ml/gcc-testresults/2005-06/msg01730.html

I don't have access to ia64-linux to run a regression search there, but I 
can run one on ia64-hpux if necessary to see if it is the same patch 
causing the failures there.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[jakub at redhat dot com]

------- Additional Comments From jakub at redhat dot com  2005-07-03 13:40 -------
Ok, I can now reproduce the g++.dg/eh/cond1.C failure with trunk GCC.
It works well when run against 4.0 libstdc++.so, or 20050625 HEAD one
(but in that case built with the -fstack-protector patch in).
Now, g++ 4.0 compiled/linked g++.dg/eh/cond1 also works when run against 4.0
libstdc++.so, or 20050625 HEAD one, but segfaults when using 20050703 HEAD
libstdc++.so (FYI, libgcc_s exact version doesn't matter).
cond1.s compiled by 20050703 HEAD and 4.0 is in fact identical except
.ident and COMDAT section names.
I have also built 20050703 HEAD libstdc++.so with g++ 4.0 (had to add
#include <ia64intrin.h> into ia64/atomicity.h), and cond1 also segfaults
against such built libstdc++.so.  That g++ 4.0 was without -fstack-protector
patches FYI.
So I'm pretty positive this is caused by some libstdc++ EH change,
with http://gcc.gnu.org/ml/gcc-cvs/2005-06/msg01019.html being the suspected
one.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[joseph at codesourcery dot com]

------- Additional Comments From joseph at codesourcery dot com  2005-07-03 13:54 -------
Subject: Re:  [4.1 Regression] Many C++ testsuite failures

On Sun, 3 Jul 2005, jakub at redhat dot com wrote:

> Ok, I can now reproduce the g++.dg/eh/cond1.C failure with trunk GCC.
> It works well when run against 4.0 libstdc++.so, or 20050625 HEAD one
> (but in that case built with the -fstack-protector patch in).
> Now, g++ 4.0 compiled/linked g++.dg/eh/cond1 also works when run against 4.0
> libstdc++.so, or 20050625 HEAD one, but segfaults when using 20050703 HEAD
> libstdc++.so (FYI, libgcc_s exact version doesn't matter).
> cond1.s compiled by 20050703 HEAD and 4.0 is in fact identical except
> .ident and COMDAT section names.
> I have also built 20050703 HEAD libstdc++.so with g++ 4.0 (had to add
> #include <ia64intrin.h> into ia64/atomicity.h), and cond1 also segfaults
> against such built libstdc++.so.  That g++ 4.0 was without -fstack-protector
> patches FYI.
> So I'm pretty positive this is caused by some libstdc++ EH change,
> with http://gcc.gnu.org/ml/gcc-cvs/2005-06/msg01019.html being the suspected
> one.

I'm now running binary searches on ia64-hp-hpux11.23 and 
hppa64-hp-hpux11.23 to confirm the responsible patches on those platforms.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[danglin at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |danglin at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] Many C++ testsuite failures

[jsm28 at gcc dot gnu dot org]

------- Additional Comments From jsm28 at gcc dot gnu dot org  2005-07-04 10:47 -------
On hppa64-hp-hpux11.23, appeared between 2005-06-27 07:39 UTC and 2005-06-27
07:42 UTC, so stack protection patch is responsible.

On ia64-hp-hpux11.23, appeared between 2005-06-28 19:33 UTC and 2005-06-28 20:08
UTC, so the mentioned series of exception handling patches is likely responsible.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pbrook at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] sjlj exception handling broken

[rth at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |rth at gcc dot gnu dot org
                   |dot org                     |
             Status|UNCONFIRMED                 |ASSIGNED
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2005-07-06 21:29:02
               date|                            |
            Summary|[4.1 Regression] Many C++   |[4.1 Regression] sjlj
                   |testsuite failures          |exception handling broken


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug c++/22284] [4.1 Regression] sjlj exception handling broken

[rearnsha at gcc dot gnu dot org]

------- Additional Comments From rearnsha at gcc dot gnu dot org  2005-07-06 21:39 -------
Also fails on arm-elf and arm-netbsdelf. Reverting the call of
sjlj_emit_function_exit_after (function.c revision 1.628) to its previous
location fixes the failures on arm-elf.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rearnsha at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug middle-end/22284] [4.1 Regression] sjlj exception handling broken

[jsm28 at gcc dot gnu dot org]

------- Additional Comments From jsm28 at gcc dot gnu dot org  2005-07-07 22:48 -------
sjlj problems seem mostly fixed as of 20050707 mainline compilers, but IA64
failures remain.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug middle-end/22284] [4.1 Regression] ia64 exception handling broken

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-07-16 21:15 -------
Only IA64 exception handling is broken.  IA64 exceptions almost always uses libunwind, right (even if 
GCC creates the libunwind)?

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[4.1 Regression] sjlj       |[4.1 Regression] ia64
                   |exception handling broken   |exception handling broken


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug middle-end/22284] [4.1 Regression] ia64 exception handling broken

[hjl at lucon dot org]

------- Additional Comments From hjl at lucon dot org  2005-07-22 05:45 -------
I have identified that

http://gcc.gnu.org/ml/gcc-patches/2005-06/msg01149.html

is the cause.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |critical


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug middle-end/22284] [4.1 Regression] ia64 exception handling broken

[hjl at lucon dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1448 bytes --]


------- Additional Comments From hjl at lucon dot org  2005-07-22 05:51 -------
It looks like the unwind info in libstdc++.so is bad.

(gdb) r
Starting program:
/export/build/gnu/gcc-next/build-ia64-linux/gcc/testsuite/cond1.exe

Program received signal SIGSEGV, Segmentation fault.
0x20000000001b9b90 in read_encoded_value_with_base (encoding=180 '´', base=0,
    p=0x40000000000021b0 "\002", val=0x60000fffffffa5d0) at unwind-pe.h:267
267                 result = *(_Unwind_Internal_Ptr *) result;
Current language:  auto; currently c++
(gdb) p result
$1 = 232
(gdb) bt
#0  0x20000000001b9b90 in read_encoded_value_with_base (encoding=180 '´',
    base=0, p=0x40000000000021b0 "\002", val=0x60000fffffffa5d0)
    at unwind-pe.h:267
#1  0x20000000001ba2e0 in get_ttype_entry (info=0x60000fffffffa608, i=Variable
"i" is not available.
)
    at
/net/gnu-3/export/gnu/src/gcc-next/gcc/libstdc++-v3/libsupc++/eh_personality.cc:209
#2  0x20000000001babb0 in __gxx_personality_v0 (version=Variable "version" is
not available.
)
    at
/net/gnu-3/export/gnu/src/gcc-next/gcc/libstdc++-v3/libsupc++/eh_personality.cc:550
#3  0x20000000002ec590 in _Unwind_RaiseException () from /lib/libgcc_s.so.1
#4  0x20000000001bb420 in __cxa_throw (obj=Variable "obj" is not available.
)
    at /net/gnu-3/export/gnu/src/gcc-next/gcc/libstdc++-v3/libsupc++/eh_throw.cc:72
#5  0x4000000000000d00 in main ()


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug libstdc++/22284] [4.1 Regression] ia64 exception handling broken

[hjl at lucon dot org]

------- Additional Comments From hjl at lucon dot org  2005-07-22 17:24 -------
A patch is posted at

http://gcc.gnu.org/ml/gcc-patches/2005-07/msg01510.html

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|critical                    |normal
          Component|middle-end                  |libstdc++


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug libstdc++/22284] [4.1 Regression] ia64 exception handling broken

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-07-22 17:53 -------
*** Bug 22614 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |sje at cup dot hp dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug libstdc++/22284] [4.1 Regression] ia64 exception handling broken

[sje at cup dot hp dot com]

------- Additional Comments From sje at cup dot hp dot com  2005-07-25 15:04 -------
I tested the proposed patch on IA64 HP-UX and Linux and it did fix the EH
problems I saw without the patch.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug libstdc++/22284] [4.1 Regression] ia64 exception handling broken

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-07-29 13:36 -------
Subject: Bug 22284

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	hjl@gcc.gnu.org	2005-07-29 13:35:59

Modified files:
	libstdc++-v3   : ChangeLog 
	libstdc++-v3/libsupc++: eh_personality.cc 

Log message:
	2005-07-29  H.J. Lu  <hongjiu.lu@intel.com>
	
	PR libstdc++/22284
	* libsupc++/eh_personality.cc (PERSONALITY_FUNCTION): Revert
	the change to info.ttype_base.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/ChangeLog.diff?cvsroot=gcc&r1=1.3066&r2=1.3067
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libstdc++-v3/libsupc++/eh_personality.cc.diff?cvsroot=gcc&r1=1.17&r2=1.18



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284

[Bug libstdc++/22284] [4.1 Regression] ia64 exception handling broken

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-07-29 13:50 -------
Fixed.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22284