[Bug c++/37057] New: 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[Bug c++/37057] New: 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

An attempt to build OpenFOAM 1.5 (www.opencfd.co.uk/openfoam/version1.5.html)
on
powerpc-ibm-aix5.3.0.0 resulted in 7 compiler internal errors (ICEs) with gcc
4.3.1 and 6 ICEs with gcc 4.2.0.

The 4.2.0 compiler suite is as downloaded from the IBM AIX RPM repository
(http://www-03.ibm.com/systems/p/os/aix/linux/toolbox/), while the 4.3.1 suite
is a rebuild with forward porting of the 4.2.0 patches.

At this time, only preprocessed files of the chemkinReader.C ICE is
provided. The 4.3.1 .ii file compiles w/o errors or warnings with gcc 4.3.0 on
a
x86_64 system. Similarly, as OF 1.5 builds on x86_64 with gcc 4.3, the other
ICEs should compile cleanly on this platform as well. Notice the difference in
size of the .s files between the 4.2 and 4.3 ICEs.

Any suggestions, patches and or development versions likely to reduce / remove
the ICEs are welcome. I have not checked if there are any common patterns
triggering the ICEs.

GCC431 ICEs
-----------
lnInclude/wrapper.cpp:320: internal compiler error: Segmentation fault
interpolation/surfaceInterpolation/limitedSchemes/vanAlbada/vanAlbada.C:36:
internal compiler error: Illegal instruction
lnInclude/Reaction.H:143: internal compiler error: Segmentation fault
chemistryReaders/chemkinReader/chemkinReader.C:230: internal compiler error:
Segmentation fault
RASModel/RASModel.C:205: internal compiler error: Segmentation fault
LESModel/LESModel.C:137: internal compiler error: Segmentation fault
lnInclude/ReactingCloud.C:101: internal compiler error: Segmentation fault
[lnInclude/Reaction.H: include from reaction/reactions/makeChemkinReactions.C]

Using built-in specs.
Target: powerpc-ibm-aix5.3.0.0
Configured with: ../configure --with-as=/usr/bin/as --with-ld=/usr/bin/ld
--enable-languages=c,c++,java --prefix=/opt/freeware --enable-threads
--enable-version-specific-runtime-libs --host=powerpc-ibm-aix5.3.0.0
--target=powerpc-ibm-aix5.3.0.0 --build=powerpc-ibm-aix5.3.0.0
--disable-libjava-multilib
Thread model: aix
gcc version 4.3.1 (GCC)
COLLECT_GCC_OPTIONS='-c' '-v' '-save-temps' '-maix32' '-Daix' '-DDP' '-Wall'
'-Wno-strict-aliasing' '-Wextra' '-Wno-unused-parameter' '-Wold-style-cast'
'-DNoRepository' '-ftemplate-depth-40' '-IlnInclude' '-I.'
'-I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OpenFOAM/lnInclude'
'-I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OSspecific/Unix/lnInclude'
'-shared-libgcc'
 /opt/freeware/libexec/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/cc1plus -E -quiet -v
-IlnInclude -I. -I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OpenFOAM/lnInclude
-I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OSspecific/Unix/lnInclude -D_ALL_SOURCE
-Daix -DDP -DNoRepository reaction/reactions/makeChemkinReactions.C -maix32
-Wall -Wno-strict-aliasing -Wextra -Wno-unused-parameter -Wold-style-cast
-ftemplate-depth-40 -fpch-preprocess -o makeChemkinReactions.ii
ignoring nonexistent directory
"/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include-fixed"
ignoring nonexistent directory
"/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/../../../../powerpc-ibm-aix5.3.0.0/include"
#include "..." search starts here:
#include <...> search starts here:
 lnInclude
 .
 /usr/local/OpenFOAM/OpenFOAM-1.5/src/OpenFOAM/lnInclude
 /usr/local/OpenFOAM/OpenFOAM-1.5/src/OSspecific/Unix/lnInclude
 /opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include/c++

/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include/c++/powerpc-ibm-aix5.3.0.0
 /opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include/c++/backward
 /usr/local/include
 /opt/freeware/include
 /opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include
 /usr/include
End of search list.
COLLECT_GCC_OPTIONS='-c' '-v' '-save-temps' '-maix32' '-Daix' '-DDP' '-Wall'
'-Wno-strict-aliasing' '-Wextra' '-Wno-unused-parameter' '-Wold-style-cast'
'-DNoRepository' '-ftemplate-depth-40' '-IlnInclude' '-I.'
'-I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OpenFOAM/lnInclude'
'-I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OSspecific/Unix/lnInclude'
'-shared-libgcc'
 /opt/freeware/libexec/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/cc1plus -fpreprocessed
makeChemkinReactions.ii -quiet -dumpbase makeChemkinReactions.C -maix32
-auxbase makeChemkinReactions -Wall -Wno-strict-aliasing -Wextra
-Wno-unused-parameter -Wold-style-cast -version -ftemplate-depth-40 -o
makeChemkinReactions.s
GNU C++ (GCC) version 4.3.1 (powerpc-ibm-aix5.3.0.0)
        compiled by GNU C version 4.3.1, GMP version 4.2.2, MPFR version 2.3.1.
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
Compiler executable checksum: 75e8825e7e5e448ca4cc214b313b85d6
lnInclude/Reaction.H: In static member function 'static
Foam::autoPtr<Foam::Reaction<ReactionThermo> >
Foam::Reaction<ReactionThermo>::addIstreamConstructorToTable<ReactionType>::New(const
Foam::speciesTable&, const Foam::HashPtrTable<ReactionThermo, Foam::word,
Foam::string::hash>&, Foam::Istream&) [with ReactionType =
Foam::NonEquilibriumReversibleReaction<Foam::sutherlandTransport<Foam::specieThermo<Foam::janafThermo<Foam::perfectGas>
> >, Foam::ArrheniusReactionRate>, ReactionThermo =
Foam::sutherlandTransport<Foam::specieThermo<Foam::janafThermo<Foam::perfectGas>
> >]':
lnInclude/Reaction.H:143: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

GCC420 ICEs
-----------
interpolation/surfaceInterpolation/limitedSchemes/Minmod/Minmod.C:38: internal
compiler error: Segmentation fault
reaction/reactions/makeChemkinReactions.C:70: internal compiler error:
Segmentation fault
chemistryReaders/chemkinReader/chemkinReader.C:903: internal compiler error:
Segmentation fault
RASModel/RASModel.C:207: internal compiler error: Segmentation fault
LESModel/LESModel.C:139: internal compiler error: Segmentation fault
clouds/derived/basicReactingCloud/basicReactingCloud.C:78: internal compiler
error: Segmentation fault
[lnInclude/Reaction.H: include from reaction/reactions/makeChemkinReactions.C]

Using built-in specs.
Target: powerpc-ibm-aix5.3.0.0
Configured with: ../configure --with-as=/usr/bin/as --with-ld=/usr/bin/ld
--enable-languages=c,c++,java --prefix=/opt/freeware --enable-threads
--enable-version-specific-runtime-libs --host=powerpc-ibm-aix5.3.0.0
--target=powerpc-ibm-aix5.3.0.0 --build=powerpc-ibm-aix5.3.0.0
--disable-libjava-multilib
Thread model: aix
gcc version 4.2.0
 /opt/freeware/libexec/gcc/powerpc-ibm-aix5.3.0.0/4.2.0/cc1plus -E -quiet -v
-IlnInclude -I. -I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OpenFOAM/lnInclude
-I/usr/local/OpenFOAM/OpenFOAM-1.5/src/OSspecific/Unix/lnInclude -D_ALL_SOURCE
-Daix -DDP -DNoRepository reaction/reactions/makeChemkinReactions.C -maix32
-Wall -fpch-preprocess -o makeChemkinReactions.ii
ignoring nonexistent directory
"/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.2.0/../../../../powerpc-ibm-aix5.3.0.0/include"
#include "..." search starts here:
#include <...> search starts here:
 lnInclude
 .
 /usr/local/OpenFOAM/OpenFOAM-1.5/src/OpenFOAM/lnInclude
 /usr/local/OpenFOAM/OpenFOAM-1.5/src/OSspecific/Unix/lnInclude
 /opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.2.0/include/c++

/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.2.0/include/c++/powerpc-ibm-aix5.3.0.0
 /opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.2.0/include/c++/backward
 /usr/local/include
 /opt/freeware/include
 /opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.2.0/include
 /usr/include
End of search list.
 /opt/freeware/libexec/gcc/powerpc-ibm-aix5.3.0.0/4.2.0/cc1plus -fpreprocessed
makeChemkinReactions.ii -quiet -dumpbase makeChemkinReactions.C -maix32
-auxbase makeChemkinReactions -Wall -version -o makeChemkinReactions.s
GNU C++ version 4.2.0 (powerpc-ibm-aix5.3.0.0)
        compiled by GNU C version 4.2.0.
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
Compiler executable checksum: 1996167973c219369b345460d5c6f097
reaction/reactions/makeChemkinReactions.C:70: internal compiler error:
Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.


-- 
           Summary: 7 Internal Compiler Errors when compiling OpenFOAM-1.5
           Product: gcc
           Version: 4.3.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: jorn dot amundsen at ntnu dot no
 GCC build triplet: powerpc-ibm-aix5.3.0.0
  GCC host triplet: powerpc-ibm-aix5.3.0.0
GCC target triplet: powerpc-ibm-aix5.3.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #1 from jorn dot amundsen at ntnu dot no  2008-08-08 08:48 -------
Created an attachment (id=16043)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16043&action=view)
g++ preprocessed file

g++ preprocessed file, ICE with powerpc-ibm-aix5.3.0.0 compiles cleanly with
x86_64-redhat-linux


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #2 from jorn dot amundsen at ntnu dot no  2008-08-08 08:53 -------
Created an attachment (id=16044)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16044&action=view)
makeChemkinReactions.C assembler output 

GCC 4.3.1 powerpc-ibm-aix5.3.0.0 assembler output from compiling
makeChemkinReactions.ii with -save-temps


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #3 from jorn dot amundsen at ntnu dot no  2008-08-08 09:01 -------
Created an attachment (id=16045)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16045&action=view)
g++ preprocessed file

GCC 4.2.0 powerpc-ibm-aix5.3.0.0 preprocessed source, not tested on any other
GCC 4.2.x platform


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #4 from jorn dot amundsen at ntnu dot no  2008-08-08 09:02 -------
Created an attachment (id=16046)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16046&action=view)
makeChemkinReactions.C assembler output

GCC 4.2.0 powerpc-ibm-aix5.3.0.0 assembler output from compiling
makeChemkinReactions.ii with -save-temps


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[paul at epcc dot ed dot ac dot uk]

------- Comment #5 from paul at epcc dot ed dot ac dot uk  2008-08-18 15:09 -------
Hi,

Just to add, I too have seen this bug (using gcc 4.2.0, not tried other
versions yet) on a powerpc-ibm-aix5.3.0.0, so I'd be keen to hear of any
resolution.

Regards,
Paul Graham


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #6 from jorn dot amundsen at ntnu dot no  2008-08-22 06:18 -------
Compiling and testing against gcc 4.4 snapshot 20080808 still results in 7 ICEs
(1 and 3-7 as of 4.3.1):

lnInclude/wrapper.cpp:320: internal compiler error: Illegal instruction
interpolation/surfaceInterpolation/limitedSchemes/Gamma/Gamma.C:57: internal
compiler error: Segmentation fault
lnInclude/Reaction.H:143: internal compiler error: Segmentation fault
chemistryReaders/chemkinReader/chemkinReader.C:230: internal compiler error:
Segmentation fault
RASModel/RASModel.C:205: internal compiler error: Segmentation fault
LESModel/LESModel.C:137: internal compiler error: Segmentation fault
lnInclude/ReactingCloud.C:101: internal compiler error: Segmentation fault


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #7 from dje at gcc dot gnu dot org  2008-09-14 01:22 -------
This is a garbage collection bug.


-- 

dje at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dje at gcc dot gnu dot org
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2008-09-14 01:22:28
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #8 from dje at gcc dot gnu dot org  2008-09-14 01:45 -------
This starts when collecting a cgraph node:

#345 0x10049f1c in gt_ggc_mx_cgraph_node (x_p=0x31b4fa00) at gtype-desc.c:171
#346 0x10053290 in gt_ggc_m_P11cgraph_node4htab (x_p=0x300fba00)
    at gtype-desc.c:2065
#347 0x10005c00 in ggc_mark_roots () at /farm/dje/src/src/gcc/ggc-common.c:107
#348 0x1003063c in ggc_collect () at /farm/dje/src/src/gcc/ggc-page.c:1941
#349 0x10197848 in execute_todo (flags=2)
    at /farm/dje/src/src/gcc/passes.c:1047
#350 0x10197b88 in execute_one_pass (pass=0x2002dd80)
    at /farm/dje/src/src/gcc/passes.c:1301
#351 0x10197df8 in execute_pass_list (pass=0x2002dd80)
    at /farm/dje/src/src/gcc/passes.c:1327
#352 0x10342018 in tree_rest_of_compilation (fndecl=0x30d6bf80)
    at /farm/dje/src/src/gcc/tree-optimize.c:418
#353 0x10414c90 in cgraph_expand_function (node=0x30daf700)
    at /farm/dje/src/src/gcc/cgraphunit.c:1038

and ends up with pointers to garbage collected data:

#1  0x1002ecb0 in ggc_set_mark (p=0xa5a5a5a5)
    at /farm/dje/src/src/gcc/ggc-page.c:1319
#2  0x103214b4 in gt_ggc_mx_eh_status (x_p=0xa5a5a5a5) at ./gt-except.h:84
#3  0x1004c5f4 in gt_ggc_mx_function (x_p=0x301631b8) at gtype-desc.c:690


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #9 from dje at gcc dot gnu dot org  2008-09-14 01:51 -------
and I think related to static constructor/destructor functions.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #10 from dje at gcc dot gnu dot org  2008-09-14 02:01 -------
I'm not sure if I am tracking this back correctly, but:

#0  ggc_free (p=0x301631b8) at /farm/dje/src/src/gcc/ggc-page.c:1380
#1  0x1033fb54 in cgraph_release_function_body (node=0x31932300)
    at /farm/dje/src/src/gcc/cgraph.c:873
#2  0x103416a8 in cgraph_remove_node (node=0x31932300)
    at /farm/dje/src/src/gcc/cgraph.c:967
#3  0x10420f10 in cgraph_remove_unreachable_nodes (before_inlining_p=0 '\0', 
    file=0x0) at /farm/dje/src/src/gcc/ipa.c:191
#4  0x10417010 in cgraph_optimize () at /farm/dje/src/src/gcc/cgraphunit.c:1271
#5  0x105911d4 in cp_write_global_declarations ()
    at /farm/dje/src/src/gcc/cp/decl2.c:3608
#6  0x1000383c in toplev_main (argc=537008760, argv=0x109696c0)
    at /farm/dje/src/src/gcc/toplev.c:979
#7  0x100003a0 in main (argc=806760888, argv=0x0)
    at /farm/dje/src/src/gcc/main.c:35
#8  0x10000214 in __start ()

Is cgraph confused about static constructors and destructors on AIX?
AIX does not have init/fini sections.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #11 from dje at gcc dot gnu dot org  2008-09-14 02:04 -------
Honza,

I may not be analyzing this correctly, but it looks like
cgraph_remove_unreachable_nodes() may be removing something that is not dead. 
Is cgraph handling constructors and destructors on non-ELF systems correctly?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[hubicka at ucw dot cz]

------- Comment #12 from hubicka at ucw dot cz  2008-09-14 09:18 -------
Subject: Re:  7 Internal Compiler Errors when compiling OpenFOAM-1.5

> Honza,
> 
> I may not be analyzing this correctly, but it looks like
> cgraph_remove_unreachable_nodes() may be removing something that is not dead. 
> Is cgraph handling constructors and destructors on non-ELF systems correctly?

It ought to be.  I.e. as long as I remember, the constructors either
appear local but have DECL_STATIC_CONSTRUCTOR on ELF or they are
externally visible functions with specially mangled names.
Perhaps there is yet another way to handle it?  They should be
recognized by decide_is_function_needed predicate in cgraphunit.c

Honza


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[hubicka at ucw dot cz]

------- Comment #13 from hubicka at ucw dot cz  2008-09-14 09:24 -------
Subject: Re:  7 Internal Compiler Errors when compiling OpenFOAM-1.5

Looking at the log, it seems to be another leak where multiple
declarations points to single struct function.  This is of course quite
evil bug with various side effects (surprisingly often the sharing just
works, but it is always memory leak and tends to break various targets),
we had instance of it already in IPCP versioning and template
instantiation.  THis is why I added explicit ggc_free in cgraph code
now.

I am just leaving for US trip, so I am not sure how soon I will be able
to look, but debugging is quite easy.  You figure out the shared decls
(i.e. one is in the backtrace where garbagecollector crashes, other is
the one we call ggc_free on the struct function when removing it).  Then
breakpoint on the end of ggc_page with condition of result being either
of those addresses to see who builds them and the second one is the
wrong copy. 

Honza


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[hubicka at gcc dot gnu dot org]

------- Comment #14 from hubicka at gcc dot gnu dot org  2008-09-14 14:31 -------
On cross from today tree it seems fine:
Starting program: /home/jh/gcc-baseline/build-aix/gcc/cc1plus -O2 *.ii
-fpreprocessed  -quiet -dumpbase makeChemkinReactions.C -maix32 -auxbase
makeChemkinReactions -Wall -Wno-strict-aliasing -Wextra -Wno-unused-parameter
-Wold-style-cast -version -ftemplate-depth-40
GNU C++ (GCC) version 4.4.0 20080914 (experimental) (powerpc-ibm-aix5.3.0.0)
        compiled by GNU C version 4.1.2 20061115 (prerelease) (Debian
4.1.1-21), GMP version 4.2.1, MPFR version 2.2.1.
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 272438ec7d6ea2c49ce12d5bf798cf01
In file included from lnInclude/TroeFallOffFunction.H:109,
                 from reaction/reactions/makeChemkinReactions.C:37:
lnInclude/TroeFallOffFunctionI.H: In member function 'Foam::scalar
Foam::TroeFallOffFunction::operator()(Foam::scalar, Foam::scalar) const':
lnInclude/TroeFallOffFunctionI.H:75: warning: sorry: semantics of inline
function static data 'const Foam::scalar d' are wrong (you'll wind up with
multiple copies)
lnInclude/TroeFallOffFunctionI.H:75: warning:   you can work around this by
removing the initializer

Perhaps it was fixed by my yesterday commit on DECL_STRUCT_FUNCTION sharing?
Or am I doing something wrong? 
Honza


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 5168 bytes --]



------- Comment #15 from dje at gcc dot gnu dot org  2008-09-14 15:35 -------
Honza,

I think that your recent pt.c:tsubst_decl() fix coincidentally may have fixed
this as well.  Tracing the ggc_alloc_stat() calls, struct function only appears
to be allocated once, but likely is copied when it should not be.

Jørn,

Could you try rebuilding GCC 4.3.1 with Honza's patch and report if it fixes
your failures?

http://gcc.gnu.org/ml/gcc-patches/2008-09/msg00999.html


#0  ggc_alloc_stat (size=84) at /farm/dje/src/src/gcc/ggc-page.c:1251
#1  0x100059ac in ggc_alloc_cleared_stat (size=84)
    at /farm/dje/src/src/gcc/ggc-common.c:132
#2  0x10072b10 in allocate_struct_function (fndecl=0x30119c00, 
    abstract_p=1 '\001') at /farm/dje/src/src/gcc/function.c:3944
#3  0x10562c28 in start_preparsed_function (decl1=0x30119c00, 
    attrs=0x20029648, flags=0) at /farm/dje/src/src/gcc/cp/decl.c:11544
#4  0x10581d78 in start_function (declspecs=0x2006e360, declarator=0x10, 
    attrs=0x0) at /farm/dje/src/src/gcc/cp/decl.c:11727
#5  0x107340e0 in cp_parser_init_declarator (parser=0x304ef230, 
    decl_specifiers=0x2ff21f20, checks=0x0, 
    function_definition_allowed_p=1 '\001', member_p=0 '\0', 
    declares_class_or_enum=806084960, function_definition_p=0x2ff21f18 "\001")
    at /farm/dje/src/src/gcc/cp/parser.c:17371
#6  0x10735bec in cp_parser_single_declaration (parser=0x304ef230, checks=0x0, 
    member_p=0 '\0', explicit_specialization_p=0 '\0', friend_p=0x2ff21fd8 "")
    at /farm/dje/src/src/gcc/cp/parser.c:17717
#7  0x10735ed4 in cp_parser_template_declaration_after_export (
    parser=0x304ef230, member_p=0 '\0')
    at /farm/dje/src/src/gcc/cp/parser.c:17570
#8  0x1073d484 in cp_parser_declaration (parser=0x304ef230)
    at /farm/dje/src/src/gcc/cp/parser.c:7885
#9  0x1073da88 in cp_parser_declaration_seq_opt (parser=0x304ef230)
    at /farm/dje/src/src/gcc/cp/parser.c:7816
#10 0x1073e7d0 in cp_parser_namespace_definition (parser=0x304ef230)
    at /farm/dje/src/src/gcc/cp/parser.c:12029
#11 0x1073d390 in cp_parser_declaration (parser=0x304ef230)
    at /farm/dje/src/src/gcc/cp/parser.c:7917
#12 0x1073da88 in cp_parser_declaration_seq_opt (parser=0x304ef230)
    at /farm/dje/src/src/gcc/cp/parser.c:7816
#13 0x1073e370 in c_parse_file () at /farm/dje/src/src/gcc/cp/parser.c:3031
#14 0x107abf98 in c_common_parse_file (set_yydebug=806760888)
    at /farm/dje/src/src/gcc/c-opts.c:1239
#15 0x10003800 in toplev_main (argc=537008760, argv=0x109696c0)
    at /farm/dje/src/src/gcc/toplev.c:968
#16 0x100003a0 in main (argc=806760888, argv=0x10)
    at /farm/dje/src/src/gcc/main.c:35
#17 0x10000214 in __start ()

(gdb) print fndecl
$6 = 0x30119c00
(gdb) pt
 <function_decl 30119c00 _M_destroy
    type <method_type 3017ab60
        type <void_type 300214e0 void type_6 VOID
            align 8 symtab 0 alias set -1 canonical type 300214e0
            pointer_to_this <pointer_type 30021548>>
        SI
        size <integer_cst 300073e8 constant 32>
        unit size <integer_cst 300070f0 constant 4>
        align 32 symtab 0 alias set -1 canonical type 301211a0 method basetype
<record_type 3011d9c0 _Rep>
        arg-types <tree_list 3017b1e0 value <pointer_type 30120680>
            chain <tree_list 3017b1a0 value <reference_type 3017a958>
                chain <tree_list 30014f60 value <void_type 300214e0 void>>>>
        throws <tree_list 300b3040>>
    nothrow public static external in_system_header no-static-chain SI file
/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include/c++/bits/basic_string.tcc
line 420 col 42 align 32 context <record_type 3011d9c0 _Rep> initial
<error_mark 30014aa0>
    arguments <parm_decl 30171d70 this
        type <pointer_type 301206e8 type <record_type 3011d9c0 _Rep>
            readonly unsigned type_0 type_6 SI size <integer_cst 300073e8 32>
unit size <integer_cst 300070f0 4>
            align 32 symtab 0 alias set -1 canonical type 301206e8>
        readonly unsigned SI file
/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include/c++/bits/basic_string.tcc
line 420 col 42 size <integer_cst 300073e8 32> unit size <integer_cst 300070f0
4>
        align 32
        chain <parm_decl 30171d20 __a type <reference_type 3017a958>
            unsigned SI file
/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include/c++/bits/basic_string.tcc
line 420 col 30 size <integer_cst 300073e8 32> unit size <integer_cst 300070f0
4>
            align 32>>
    result <result_decl 30171dc0 D.9306 type <void_type 300214e0 void>
        ignored VOID file
/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.3.1/include/c++/bits/basic_string.tcc
line 420 col 42
        align 8>
    template-info 3017b320 chain <function_decl 30119c80 _M_refcopy>>


-- 

dje at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2008-
                   |                            |09/msg00999.html


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1014 bytes --]



------- Comment #16 from jorn dot amundsen at ntnu dot no  2008-09-14 17:11 -------
(In reply to comment #15)

With 4.3.2, and by uncommenting a superfluous third party package, I am down at
4 ICE's. The makeChemkinReactions ICE is still present with 4.3.2, though. Now,
since cp/pt.c are identical in 4.3.1 and 4.3.2 I patched and started a
recompile of the official 4.3.2 release sources.

I hope this is ok. The build needs 4 hrs to complete. Of course, if there is
nothing to gain by patching 4.3.2, I will revert to 4.3.1.

--joern


> Honza,
> 
> I think that your recent pt.c:tsubst_decl() fix coincidentally may have fixed
> this as well.  Tracing the ggc_alloc_stat() calls, struct function only appears
> to be allocated once, but likely is copied when it should not be.
> 
> Jørn,
> 
> Could you try rebuilding GCC 4.3.1 with Honza's patch and report if it fixes
> your failures?
> 
> http://gcc.gnu.org/ml/gcc-patches/2008-09/msg00999.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #17 from dje at gcc dot gnu dot org  2008-09-14 19:52 -------
Because the problem is due to garbage collection, it is non-deterministic,
depending on what data GCC generates.  I was able to reproduce the problem in a
recent development version of GCC, prior to Honza's patch, but not in my
installed version of GCC 4.3.1, which is why I asked you to confirm that the
patch fixes the problem for you.  GCC 4.3.2 fixes numerous bugs over GCC 4.3.1,
although not necessarily ICEs that affect you.  I would recommend using a
patched version of GCC 4.3.2 over a patched version of GCC 4.3.1, but that is
your decision.


-- 

dje at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |WAITING
   Target Milestone|---                         |4.3.3


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #18 from jorn dot amundsen at ntnu dot no  2008-09-14 22:01 -------
(In reply to comment #17)
I tried with a patched 4.3.2, and it does unfortunately *not* fix the problem.

The compiler line (as read from -###) results in:

$ /opt/freeware/libexec/gcc/powerpc-ibm-aix5.3.0.0/4.3.2/cc1plus -fpreprocessed
chemkinReader.ii -quiet -dumpbase chemkinReader.ii -auxbase chemkinReader -o
chemkinReader.s
chemkinReader.ii: In member function 'void
Foam::chemkinReader::addReactionType(Foam::chemkinReader::reactionType,
Foam::DynamicList<Foam::Reaction<Foam::sutherlandTransport<Foam::specieThermo<Foam::janafThermo<Foam::perfectGas>
> > >::specieCoeffs, 0u, 2u, 1u>&,
Foam::DynamicList<Foam::Reaction<Foam::sutherlandTransport<Foam::specieThermo<Foam::janafThermo<Foam::perfectGas>
> > >::specieCoeffs, 0u, 2u, 1u>&, const ReactionRateType&) [with
ReactionRateType = Foam::FallOffReactionRate<Foam::ArrheniusReactionRate,
Foam::LindemannFallOffFunction>]':
chemkinReader.ii:91089: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

I could patch and recompile 4.3.1 if you think this might be useful.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #19 from dje at gcc dot gnu dot org  2008-09-14 22:42 -------
First, you reported seven ICEs.  You attached files for
makeChemkinReactions.ii.  That file failed before Honza's patch and does not
appear to fail after his patch.  I cannot reproduce the failure using my
version of GCC 4.3.1.

Saying that the patch does not fix "the" problem is not accurate.  You said
that the makeChemkinReactions.ii still was present in GCC 4.3.2, but you report
an ICE for chemkinReader.ii.  Please stop shifting the goalpost.  You have not
commented on makeChemkinReactions.ii.  No one said that the seven ICEs were do
to a single problem.  I suspect the makeChemkinReactions.ii ICE *is* fixed.

If you attach chemkinReader.ii, we can try to investigate it as well.  These
likely all are due to garbage collection issues, but not a single bug.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #20 from jorn dot amundsen at ntnu dot no  2008-09-15 04:42 -------
(In reply to comment #19)

I am sorry, You are right. I thought it fixed everything, though. I will check
4.3.2 + the cp/pt.c patch against every of the 7 original ICE's and give an
updated status on everyone in another update a little later today.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #21 from jorn dot amundsen at ntnu dot no  2008-09-15 20:06 -------
(In reply to comment #20)
David, I went through all cases, and it does indeed confirm your position. The
result of the walkthrough is summarized below. Also, I include the preprocessed
files of the remaining ICE's.


This overview describes current status with internal compiler errors in the
OpenFOAM v.1.5 CFD application (http://www.opencfd.co.uk/openfoam/) with
powerpc-ibm-aix5.3.0.0.

OVERVIEW OF PREVIOUSLY REPORTED ICE's
=====================================

--- #1 ---
lnInclude/wrapper.cpp:320: internal compiler error: Segmentation fault

 => part of hoard v.3.7.1. A linux x86 memory allocator. Not considered useful
    with AIX.

--- #2 ---
interpolation/surfaceInterpolation/limitedSchemes/vanAlbada/vanAlbada.C:36:
internal compiler error: Illegal instruction

 => compiles successfully with GCC 4.3.2
 => compiles successfully with GCC 4.3.2 + patch

--- #3 ---
lnInclude/Reaction.H:143: internal compiler error: Segmentation fault
[included from makeChemkinReactions.C]

 => ICE still present with GCC 4.3.2 + patch

--- #4 ---
chemistryReaders/chemkinReader/chemkinReader.C:230: internal compiler error:
Segmentation fault

 => ICE with GCC 4.3.2
 => compiles successfully with GCC 4.3.2 + patch

--- #5 ---
RASModel/RASModel.C:205: internal compiler error: Segmentation fault

 => compiles successfully with GCC 4.3.2
 => compiles successfully with GCC 4.3.2 + patch

--- #6 ---
LESModel/LESModel.C:137: internal compiler error: Segmentation fault

 => compiles successfully with GCC 4.3.2
 => compiles successfully with GCC 4.3.2 + patch

--- #7 ---
lnInclude/ReactingCloud.C:101: internal compiler error: Segmentation fault
[included from basicReactingCloud.C]

 => ICE still present with GCC 4.3.2 + patch

NEW ICE's WITH GCC 4.3.2
=====================================

--- #8 ---
lnInclude/multivariateSurfaceInterpolationScheme.H:107: internal compiler
error: Segmentation fault
[included from multivariateGamma.C]

 => compiles with GCC 4.3.1
 => ICE with GCC 4.3.2
 => ICE with GCC 4.3.2 + patch


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #22 from jorn dot amundsen at ntnu dot no  2008-09-15 20:22 -------
Created an attachment (id=16329)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16329&action=view)
g++ preprocessed file


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #23 from jorn dot amundsen at ntnu dot no  2008-09-15 20:25 -------
Created an attachment (id=16330)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16330&action=view)
g++ preprocessed file


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #24 from jorn dot amundsen at ntnu dot no  2008-09-15 20:26 -------
Created an attachment (id=16331)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16331&action=view)
g++ preprocessed file


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #25 from dje at gcc dot gnu dot org  2008-09-15 23:28 -------
I cannot reproduce the ICEs that you say remain.  I tried using both the
current development version of GCC and GCC 4.3.2+pt.c patch.  I also tried
compiling the testcases with additional options that stress the garbage
collection system

--param ggc-min-expand=0 --param ggc-min-heapsize=0

None of these attempts reproduced a failure with your pre-processed sources.

I am surprised by your report

--- #3 ---
lnInclude/Reaction.H:143: internal compiler error: Segmentation fault
[included from makeChemkinReactions.C]

 => ICE still present with GCC 4.3.2 + patch

because that is the file you originally attached.  The new version is
different, but Honza and I could not reproduce the failure after his pt.c
patch.

I am not sure if there is some different AIX system fix installed.  Or do you
have process limits (ulimit -a) set low?  My limits generally are "unlimited".


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #26 from jorn dot amundsen at ntnu dot no  2008-09-16 07:34 -------
(In reply to comment #25)
Thank you for the update. If I understand this correctly, compiling GCC 4.3.2
as published on ftp://ftp.gnu.org/pub/gnu/gcc/gcc-4.3.2, adding the pt.c patch
on http://gcc.gnu.org/ml/gcc-patches/2008-09/msg00999.html only, compiles all
of my preprocessed .ii files as published on september 15 w/o ICE's.

My cc1plus libintl.a is from gettext v.0.10.40, libmpfr.a from mpfr v.2.3.1 and
libgmp.a from v.4.2.2. The build host AIX revision is 5.3.0.0 TL 07, hardware
is power5+. The ulimit -a reads

core file size          (blocks, -c) 1048575
data seg size           (kbytes, -d) unlimited
file size               (blocks, -f) unlimited
max memory size         (kbytes, -m) unlimited
open files                      (-n) unlimited
pipe size            (512 bytes, -p) 64
stack size              (kbytes, -s) 4194304
cpu time               (seconds, -t) unlimited
max user processes              (-u) 262144
virtual memory          (kbytes, -v) unlimited

I am also adding the full lslpp-lc to the case. Please let me know if there is
any discrepancys compared to your build environment I should look closer into.

Also, I am currently building from RPM spec files. Next item is to try a plain
patch+configure+make of the exact source as previously described. It would be
helpful if you could supply the exact configure statement together with the
usual influental environment variables you are using.


-- 

jorn dot amundsen at ntnu dot no changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jorn dot amundsen at ntnu
                   |                            |dot no


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #27 from jorn dot amundsen at ntnu dot no  2008-09-16 07:36 -------
Created an attachment (id=16335)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16335&action=view)
list of installp packages

A list of all installp packeges on the AIX build system as gathered with lslpp
-lc.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #28 from dje at gcc dot gnu dot org  2008-09-16 15:52 -------
The AIX systems to which I have access also are AIX 5.3.0.0 TL 07.

My GCC configure command only specifies the languages and where to find GMP and
MRPF.  I had been using gmp-4.1.4 static archive library, but recently upgraded
to gmp-4.2.3 dynamic library.  I am using mpfr-2.3.1 static archive library.  I
build GCC as a 32 bit application, which is the default.

I assume that you are compiling the OpenFOAM as 32 bits.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #29 from dje at gcc dot gnu dot org  2008-09-16 17:40 -------
I bootstrap GCC using GCC 4.1.  I assume you use GCC 4.2 distributed by IBM.

I do not understand your comment in the initial report about "forward porting
of the 4.2.0 patches."


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #30 from jorn dot amundsen at ntnu dot no  2008-09-16 18:23 -------
(In reply to comment #29)
> I bootstrap GCC using GCC 4.1.  I assume you use GCC 4.2 distributed by IBM.
I usually use the currently installed gcc-version, which might be more recent
than 4.2.

> I do not understand your comment in the initial report about "forward porting
> of the 4.2.0 patches."
Probably means you don't do this. Might be a key point. I apply the patches
listed in 
ftp://ftp.software.ibm.com/aix/freeSoftware/aixtoolbox/SPECS/gcc-4.2.0-3.spec
to  a more recent release. Most of the patches install with a small fuzz
factor. I correct failing hunks, or remove a hunk if it appears to be present.
Hence the wording forward porting.


-- 

jorn dot amundsen at ntnu dot no changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |SUSPENDED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #31 from dje at gcc dot gnu dot org  2008-09-16 18:34 -------
No, I do not use any patches from IBM.  The patches seem to be focussed on
packaging and layout, but I would recommend using the standard FSF GCC 4.3.2
release plus the pt.c patch that now is upstream in the FSF GCC sources.  We
cannot help you with a compiler that diverges from the FSF tree.

Again, I am not sure which versions of GMP and MPFR you are using.  I would
recommend that you configure and build those packages with --disable-shared and
ABI=32.

You also can try running cc1plus inside GDB and compiling the files to see
where the ICE occurs.  Right now we cannot reproduce the failures and are
flying blind.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #32 from jorn dot amundsen at ntnu dot no  2008-09-16 19:09 -------
(In reply to comment #31)
> No, I do not use any patches from IBM.  The patches seem to be focussed on
> packaging and layout, but I would recommend using the standard FSF GCC 4.3.2
> release plus the pt.c patch that now is upstream in the FSF GCC sources.  We
> cannot help you with a compiler that diverges from the FSF tree.
I will step back to GCC 4.3.2 + pt.c only, leaving out any IBM RPM patches. So
a recompile is on its way. Will take some time.

> Again, I am not sure which versions of GMP and MPFR you are using.  I would
> recommend that you configure and build those packages with --disable-shared and
> ABI=32.
I use gmp 4.2.2 (compiling 4.2.3 just now) and mpfr 2.3.1. Both are compiled as
32- and 64-bit shared libs packaged in one .a-file. I usually compile
everything except 64-bit libs with OBJECT_MODE=32. I doubt gcc even is possible
to build 64-bit.

> You also can try running cc1plus inside GDB and compiling the files to see
> where the ICE occurs.  Right now we cannot reproduce the failures and are
> flying blind.
Yes, I have installed the latest gdb. Also, other GCC applications I've
compiled works just fine in the Etnus Totalview debugger, which we have
installed. However, main focus right now is to step back towards your build
configuration and duplicate your build effort.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #33 from dje at gcc dot gnu dot org  2008-09-16 20:00 -------
I have not tried MPFR as a shared library.

One thing to be careful about with shared libraries on AIX is libpath.  libtool
configured for gmp and mpfr decides to hardcode the path and finds all of the
GCC build directory paths.  I manually edit the configured libtool to cleanup
libpath.  libgmp.a and libmpfr.a only need /usr/lib:/lib because they do not
depend on shared libgcc.a.  libgmpxx.a depends on the directories where
libstdc++.a and libgcc_s.a are installed.

cc1, cc1plus, etc. generally end up with sane libpaths, but libstdc++ (again
built by libtool) includes a lot of junk directories.  These latter
dependencies would not affect the ICEs when building OpenFOAM, but you might
want to be aware for the application itself.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[jorn dot amundsen at ntnu dot no]

------- Comment #34 from jorn dot amundsen at ntnu dot no  2008-09-17 08:02 -------
(In reply to comment #33)
Thank you for the tip on libtool libpath. Looking into it with dump -H, I did
not know it was that ugly. Indeed, this should be cleaned up if when building
and installing software.

In order to reproduce your successful compilation, I did

0. shared mpfr 2.3.1, shared gmp 4.2.3, shared gettext 0.10.40
1. untar 4.3.2 gcc-core and gcc-g++
2. apply pt.c patch
3. environ settings:
        export CONFIG_SHELL=/opt/freeware/bin/bash
        export CONFIGURE_ENV_ARGS=/opt/freeware/bin/bash
        export OBJECT_MODE=32
4. mkdir objdir
5. cd objdir
6. ../configure --enable-languages=c,c++
7. gmake bootstrap
8. <objdir>/gcc/g++ -B<objdir>/gcc -c <icefiles>.ii

This works! I am able to reproduce your successful compilation of every .ii
file  attached to this bug. The reason for the remaining failures is any of

 - a bug in any of the IBM GCC 4.2 RPM patches
 - one or more of the 4.2 hunks are not applicable with 4.3
 - I made an error in porting any of the 4.2 hunks to 4.3

Having a build tree which breaks and another with works I am now able to limit
and eventually find the root cause of this problem.

Hence this issue is resolved.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057

[Bug c++/37057] 7 Internal Compiler Errors when compiling OpenFOAM-1.5

[dje at gcc dot gnu dot org]

------- Comment #35 from dje at gcc dot gnu dot org  2008-09-17 12:22 -------
User reports successful resolution.


-- 

dje at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37057