internal compiler error: Bus error

74 views
Skip to first unread message

Lev Karatun

unread,
Feb 1, 2017, 8:23:24 PM2/1/17
to deal.II User Group
Hi everyone,

I'm trying to install the latest version of dealII on a cluster now, got another error unfortunately. 
The output is below, and I attached the logs too. I would appreciate any help with this. 
In file included from /home/r/russ/lkaratun/aspect/dealii/include/deal.II/dofs/d                                                                             of_handler.h:27:0,
                 from /home/r/russ/lkaratun/aspect/dealii/include/deal.II/dofs/d                                                                             of_accessor.h:22,
                 from /home/r/russ/lkaratun/aspect/dealii/source/fe/fe_nedelec.c                                                                             c:23:
/home/r/russ/lkaratun/aspect/dealii/include/deal.II/dofs/block_info.h:233:34: in                                                                             ternal compiler error: Bus error
 types::global_dof_index BlockInfo::renumber (const unsigned int i) const
                                  ^
0x8d7a9f crash_signal
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/toplev.c:332
0x5d4dc0 pop_deferring_access_checks()
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/semantics.c:183
0x585848 cp_parser_parse_definitely
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:23820
0x59cc86 cp_parser_declarator
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:16227
0x5a28a0 cp_parser_init_declarator
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:15837
0x5a351e cp_parser_simple_declaration
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:10670
0x5a5020 cp_parser_block_declaration
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:10551
0x5acd6b cp_parser_declaration
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:10448
0x5abaad cp_parser_declaration_seq_opt
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:10334
0x5abcac cp_parser_namespace_body
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:15152
0x5abcac cp_parser_namespace_definition
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:15133
0x5accdd cp_parser_declaration
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:10436
0x5abaad cp_parser_declaration_seq_opt
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:10334
0x5ad2ba cp_parser_translation_unit
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:3813
0x5ad2ba c_parse_file()
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/cp/parser.c:28338
0x642a34 c_common_parse_file()
        /scinet/gpc/src/gcc/gcc-4.8.1/gcc/c-family/c-opts.c:1046
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See <http://gcc.gnu.org/bugs.html> for instructions.
make[2]: *** [source/fe/CMakeFiles/obj_fe_release.dir/fe_nedelec.cc.o] Error 1
make[1]: *** [source/fe/CMakeFiles/obj_fe_release.dir/all] Error 2
make: *** [all] Error 2

summary.log
detailed.log

Wolfgang Bangerth

unread,
Feb 1, 2017, 8:41:59 PM2/1/17
to dea...@googlegroups.com
On 02/01/2017 06:23 PM, Lev Karatun wrote:
>
> I'm trying to install the latest version of dealII on a cluster now, got
> another error unfortunately.
> The output is below, and I attached the logs too. I would appreciate any
> help with this.

Are you compiling with -j4 or similar? Does the same happen if you
compile with -j1? The kind of internal compiler error you have here
seems to happen most often if you run out of memory.

Best
W.

--
------------------------------------------------------------------------
Wolfgang Bangerth email: bang...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/

Lev Karatun

unread,
Feb 1, 2017, 9:06:26 PM2/1/17
to dea...@googlegroups.com
Hi Wolfgang,

I was first compiling with -j8 then tried again with -j1 but got the same error. 

Best regards,
Lev Karatun.

--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en
--- You received this message because you are subscribed to a topic in the Google Groups "deal.II User Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dealii/aZvol6kHWJI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dealii+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Matthias Maier

unread,
Feb 1, 2017, 11:03:45 PM2/1/17
to dea...@googlegroups.com
A bus error with gcc-4.8.1 on x86_64 is rare...

Is it reproducible? I.e. if you run the compilation with

$ make VERBOSE=1 -j1

you will see the full command invocation that lead to the internal
compiler error. Does it always stop at the same .cc file ?

Further, can you ask the administrators to update the compiler?
gcc-4.9.4 is a well tested and solid choice. (the newer gcc-5* and
gcc-6* variants might lead to an ABI nightmare if you do not recompile
every C++ library).

Best,
Matthias
>> dealii+un...@googlegroups.com.

Lev Karatun

unread,
Feb 2, 2017, 11:31:12 AM2/2/17
to dea...@googlegroups.com
Hi Matthias,

It doesn't seem to be reproducible. In fact, when I ran the compilation today again, it completed without errors. I'll check if there is an option to use gcc 4.9.4. But will I have to recompile Trilinos and BLAS in this case? I'd rather avoid doing it

Best regards,
Lev Karatun.


>> For more options, visit https://groups.google.com/d/optout.
>>

--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "deal.II User Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dealii/aZvol6kHWJI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dealii+unsubscribe@googlegroups.com.

Matthias Maier

unread,
Feb 2, 2017, 12:48:32 PM2/2/17
to dea...@googlegroups.com

On Thu, Feb 2, 2017, at 10:30 CST, Lev Karatun <lev.k...@gmail.com> wrote:

> It doesn't seem to be reproducible. In fact, when I ran the compilation
> today again, it completed without errors.

In that case a hardware issue might very likely be the problem.


> I'll check if there is an option to use gcc 4.9.4. But will I have to
> recompile Trilinos and BLAS in this case? I'd rather avoid doing it

I suggested this as a possible workaround if the bus error you saw is
reproducible. Given the fact that you could compile the library on
second try, there is not much you gain by upgrading to 4.9.4.

Best,
Matthias

Yi-Chung Chen

unread,
Feb 2, 2017, 1:27:28 PM2/2/17
to deal.II User Group
Hi Lev, 

I was wondering if you ran out of disk. In my experience, it looks like your previous compilation is only partially successful.

Best
YC Chen

Best regards,
Lev Karatun.


>> For more options, visit https://groups.google.com/d/optout.
>>

--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "deal.II User Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dealii/aZvol6kHWJI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dealii+un...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages